Viral Etiology of Congenital Malformations May 19-20, 1967 fiw Mod of foc habe ar NN ‘ 0 LIBRARY IFORNIA U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE National Institutes of Health U.S.S.L. EO Tey Bie dni mn he Setters ln isn a i i cot + tela cn 2 nn iis id viral Etiology of Congenital Malformations May 19-20, 1967 The National Heart Institute and the National Institute of Child Health and Human Development National Institutes of Health Bethesda, Maryland 20014 For sale by the Superintendent of Documents, U.S. Government Printing Office Washington, D.C. 20402 - Price 65 cents WM Wat AL ~~ 1968 Preface ray As research in any area moves from superficial observation and manipula- tion to penetrating investigation and exploration conventional, disciplinary boundaries have to be crossed. But scientists who have specialized in one specific direction find that they neither know, nor have ready access to, areas of knowledge bounded by other disciplines. Nowhere is this more true than in the elucidation of agents and mechanisms underlying the pro- duction of congenital malformations. Attacking only one group of etiologic agents, viruses, approaches techniques and knowledge from at least eight different disciplines, virology, immunology, fetal pathology, placentology, epidemiology, comparative morphology, genetics and pediatrics are needed. Clearly no one investigator can be an expert in all these areas; equally clearly his own research may be blocked by lack of knowledge in one of them. To promote interchange between these disciplines, to define the state of the art and to explore ways in which needed progress might be expedited, a small working conference was held in Bethesda on May 19th and 20th, 1967 under the auspices of the National Institute of Child Health and Human Develop- ment and the National Heart Institute. Since information exchange was one of the underlying purposes of the meeting, a transcript of the conference was obtained and this has been edited and annotated with cited and back- ground references as guides for the interested reader to these possibly foreign, but highly relevant, fields. The participants themselves found the facts and vistas presented during the conference both informative and stimulating. It is the hope of the editors that the reading audience will do the same. G. L. WOODSIDE S. C. MITCHELL prey Ee dae Contents Page Preface iil List of Participants vi I—Virus-Cell Interaction, The Implications for Fetal Disease 3 by Samuel L. Katz II—Virologic Aspects of the Perinatal Research Study 8 by Lon R. White III—Immunologic Responses of the Fetus 20 by Arthur M. Silverstein IV—Fetal Growth Retardation 35 by Peter Gruenwald V—Fetal Pathology Due to a Group of DNA Viruses 46 by George Margolis VI—Function of the Placenta, Barrier or Bridge? 67 by Joseph Dancis VII—Viral Infection of the Placenta 74 by Kurt Benirschke VIII—The Baboon as a Primate Animal Model 90 by Leonard R. Axelrod IX—Cytogenetic Aspects of Virus and Cell Relations 102 by Paul S. Moorhead X—Epidemiologic Approaches to the Origins of Congenital 113 Malformations by Robert W. Miller XI—Summary of the Conference Presentations 132 by Heinz F. Eichenwald XII—Summary of the Conference Recommendations 159 by Kurt Benirschke Cited Bibliography 161 Reference Bibliography 173 List of Participants Chairman—Benirschke, Kurt, Professor and Chairman of the Depart- ment of Pathology, Dartmouth Medical School, Hanover, New Hampshire Axelrod, Leonard R., Director, Division of Biological Growth and Develop- ment, Southwest Foundation for Research and Education, San Antonio, Texas Dancis, Joseph, Professor of Pediatrics, New York University Medical Center, New York University School of Medicine, New York City, N.Y. Eichenwald, Heinz F., Professor and Chairman, Department of Pediatrics, The University of Texas, Southwestern Medical School, Dallas, Texas Gruenwald, Peter, Associate Professor of Pathology, Johns Hopkins University, and Associate Pathologist, Sinai Hospital of Baltimore, Baltimore, Maryland Katz, Samuel L., Professor and Chairman of the Department of Pediatrics, Duke University, Durham, North Carolina Margolis, George, Professor of Pathology, Dartmouth Medical School, Hanover, New Hampshire Miller, Robert W., Chief, Epidemiology Branch, National Cancer Institute, Bethesda, Maryland Mitchell, Shiela C., National Heart Institute, Bethesda, Maryland Moorhead, Paul S., Associate Member, The Wistar Institute, Philadelphia, Pennsylvania Silverstein, Arthur M., Professor of Ophthalmic Immunology, The Wilmer Institute, Johns Hopkins University School of Medicine, Baltimore, Maryland White, Lon R., Resident in Pediatrics, King County Harborview Hospital, Seattle, Washington Woodside, Gilbert L., Associate Director for Extramural Programs, National Institute of Child Health and Human Development, Bethesda, Maryland Visitors Corfman, Philip A., M.D., Assistant to Director for Population Research, National Institute of Child Health and Human Development Frank, Michael, M.D., Developmental Biology Program, National Institute of Child Health and Human Development Friedman, William F., M.D., Cardiology Branch, National Heart Institute Mayer, Florence, M.D., Growth and Development Program, National Institute of Child Health and Human Development McCracken, George, M.D., Medical Officer, National Institute of Child Health and Human Development Overall, James, M.D., Research Associate, National Institute of Child Health and Human Development Singer, Jack D., M.D., Laboratory of Biomedical Sciences, National Institute of Child Health and Human Development vi PROCEEDINGS Dr. Benirschke: May I call the meeting to order? First of all, thank you for coming. I don’t propose to run a tight meeting here. Most important is that we are all very much relaxed and everyone has his say. If he wishes to expand further subsequently, he should feel free to do so. We should try to keep to the business before us. This is to discuss, in some depth, the current thoughts on the relationship between maternal viral infec- tion and fetal malformations. Not only fetal malformations, as I understand it, but also failure of normal fetal development and here not only congenital abnormalities, but also what some choose to call the runt syndrome, or the small-for-age-babies. I understand this conference was conceived by the Heart Institute and the Child Health Institute, both of which wished to be advised about the “state of the art” in virus interactions with the fetal organism; how viruses get across the placenta and what damage they do in the baby; what the magnitude of the problem is; what can be done to ascertain it or correct it if possible; and to advise the directors of these Institutes largely of what goes on in the field, where the field is heading and what the current obstacles are to a deeper understanding of the interaction between maternally acquired viruses and fetal disease. As I see it, there is an increasing interest in this country in work concern- ing the possible interactions of viruses with the fetuses in causing latent diease, such as the difficulties in rubella and perhaps other cerebral problems not as yet recognized as being due to intra-uterine virus disease. The Institutes then wish to have a transcribed document which will be abstracted by Dr. Mitchell to be used by the Institute directors, by the staff and be consulted by anyone who wished to consider our record in making decisions concerning grant support. The Institutes wish to have advice as to the possibility of further conferences in this area. They wish to be prepared 1 to go to the Congressmen in their request for money and be in the position to justify this. They wish to perhaps have publication out of this, or other similar conferences. But, I am to assure you that while this is being stenotyped, it will not be circulated or printed as it stands. If we do decide that we want to publish it we will all have the opportunity to edit the transcript. Consequently, every- body should feel quite free to say what he wants since he can amend his comments later. I think that this is important for us to bear in mind. We are to consider the state of the art of our disciplines in this respect. We wish to concern ourselves with the immediate and future obstacles to progress in virus interactions with the fetus and possible ways that we can see by which such needs might be met, whether the Institutes can help in any way, whether investigations can march in new directions or whether we can all profit, one from the other. The plan is today to have presentations by each one here—maybe twenty or thirty minutes of reasonably formal, succinct presentations of what they feel is currently the problem, what the achievements have been and, hope- fully, with the thought in mind of some recommendations. After each presentation there will be a short discussion. I would like to urge everyone, during the presentations, to jot down the ideas that he gets while others are talking so that during the broader discussions, maybe later on in the afternoon, but particularly tomorrow, we can really come to grips with some of the problems that have been raised and possibly give some answers to these problems. Now, I would like to call on Dr. Sam Katz to start and give us an over- view from the virologist’s point of view. I—Virus-Cell Interaction, the Implications for Fetal Disease by Samuel L. Katz Dr. Katz: If the eventual goal of this meeting is to provide stimulation and directions for reserach efforts aimed at unravelling the highly complex interrelationships of mother, placenta, fetus and virus, we must first seek a more complete understanding of far simpler systems in which host and microbe interact. In this regard, I should like to take some time initially to review the general patterns that have been observed to follow virus infection of intact animal cells in vitro. To interpret these cellular responses as test- tube models of the exceedingly complicated relationships which exist in a pregnancy infection would be unjustified and naive. But an appreciation of the processes may provide some guide to the next steps in the quest for full elucidation of those mechanisms responsible for fetal malformations. Other participants in this conference will demonstrate that some of these succeed- ing steps have indeed already been taken, with discussion of placental organ cultures, pregnant primate models, and studies of human infections by clin- ical, virological, immunological, pathological, genetic and epidemiological methods. Some of these observations subsequently to be reported may be more readily understood or explained more logically in the light of infor- mation acquired from investigations of host cell-virus interactions. The results of chance (or deliberate) encounter of a virus and an intact animal cell are dependent on a number of factors which include the properties of the virus itself, the environment and the state of the cell. One common form of response is that in which the cell serving as host for the virus syn- thesizes within a short time (2-6 hours) new viral nucleic acid and proteins which are assembled to form thousands of new intact virions. Associated with release of the new virus, cell death occurs and the “replicative lytic” cycle is complete. Quite a different pattern is observed when infected cells synthesize and release new virus but continue to survive and, under favorable condi- tions, to divide, with transmission of the infection to daughter cells. This might be called a “replicative non-lytic” infection. The multiplication of cer- tain viruses is accompanied by a stimulatory effect on their host cells which may proliferate in a reliable fashion leading to the formation of typical lesions prior to cell death. This “replicative proliferative” behavior is seen with the pox viruses. Finally, there are increasingly well-studied examples of “non-replicative proliferative” response in which, after an initial brief interaction of virus and host cells, accelerated cell growth commences and persists indefinitely in the absence of any detectable virus. The continued presence of virus-specified new antigens in the cells through many generations attests to the continued influence of some viral genetic information, but no infective virus is produced or can be induced. In Table I these four types of response are listed, with examples of each as observed in some embryonal cell types. As already mentioned, these 3 Table I Examples of the interactions of virus and cells in vitro. Virus Host cells Pattern of infection Herpes Simplex human embryonic replicative lytic Rubella human embryonic replicative non-lytic Vaccinia chick embryo membrane replicative proliferative Adenovirus hamster embryonic non-replicative proliferative patterns are dependent upon at least three factors—virus, cell and environ- ment—so that change in any one may provide a completely different response. When rubella virus is cultured in certain rabbit embryonic cells, a lytic infection occurs with gross destruction of the monolayer, but human embryonic cells will support viral replication indefinitely without any destructive effect. Lowering their environmental temperature from 37° C to 33° C will convert a lytic herpes simplex virus infection of some cells to a non-lytic one. Some strains of herpes simplex virus will produce lytic infec- tion of cells but others will recruit the same cells for fusion into large syncytia or polykarions. From investigations of the biochemical events which occur in the replica- tion of animal viruses, an increasing body of knowledge has been acquired relevant to the manner and sequence in which genetic information carried into the cell by a single molecule of viral nucleic acid achieves its goal of directing the successfully infected cell to synthesize viral nucleic acid, new enzymes and structural proteins from which to assemble thousands of new virions. As yet, morphologic cellular changes cannot be correlated completely with these biochemical steps, but ultrastructural observations have already provided some of the information needed to supplement the findings pre- viously noted by conventional microscopy and histochemistry. It has been possible, for example, to visualize, as well as to quantitate, the disaggregation of host cell polysomes which occurs with an RNA virus (polio) infection and to note the assembly of new virus-specific polyribosomes. Such observa- tions coincide graphically with data on the marked inhibition of normal cellular RNA and protein synthesis which is found early in the cycle of RNA virus infection. A timely review by Sommers (1) presents much of this material in a concise and lucid fashion. It is interesting to recall that the virus progeny, even at the most efficient production, do not constitute as much as 1 percent of cell mass. With RNA viruses of the myxovirus family (measles, rubella, mumps, and influenza among others) a consistent finding has been the unique mechanism by which the virus is completed and released from cells. A budding process of the cellular membrane provides an envelope for the nucleocapsid of the virus so that there is an incorporation of host cell membrane constituents 4 into the viral envelope coincident with release of virus to the extracellular environment. Such a process alters the cell surface chemically and may be further expressed in such new biological activity as a propensity to fuse with neighboring cells to form syncytia. Other structural and functional changes which may occur in the virus infected cell have been noted although their correlation with the events in the cycle of viral replication is not fully appreciated. Leakage of destructive hydrolases into the intracellular milieu may result either from direct damage by virus particles of the lysosomal membrane or may be mediated by new viral coded lytic enzymes. The inclusion bodies visualized in nuclei or cyto- plasm of some infected cells may represent accumulated virions and/or their unassembled subunits in some instances, while in others they are host cell materials present at the site of earlier virus synthesi. Chromosomal altera- tions have been reported with a number of viruses particularly those involving cell nucleus and Dr. Moorhead will undoubtedly discuss these and their significance. At what stage in gestation the offending virus is encountered will influence greatly the eventual result. Congenital malformations are but one expression of a spectrum which includes intrauterine death, spontaneous abortion, pre- mature labor and perinatal disease. Direct viral invasion of the fetus may not always be a requisite to the induction of some of these latter results, and the severity of maternal illness may determine pregnancy outcome by less obvious actions on placental and fetal physiology which, in our ignorance, we label “toxic” or “general systemic” effects. The high fever and severe respiratory tract involvement which accompany measles in pregnancy are blamed for the increased incidence of abortion, intrauterine death and pre- maturity which accompany or follow the acute illness. Teratogenesis has not been a feature of first trimester infection with measles virus. However, there are no appropriate studies on which to base conclusions regarding patho- genesis of the effects noted, and it may well be that measles virus invades the placenta and damages it without passing further to the fetus. Other agents may not reach the fetus by crossing the placenta but may gain access through the maternal reproductive tract with invasion of the amniotic sac. The importance of this route in early pregnancy has not been established but its role at the time of labor and delivery, or earlier when premature rupture of membranes has occurred, seems well established. The Coxsackie B viruses, excreted for long periods of time from the enteric tract of an infected mother or circulating in the maternal blood for short periods may, thereby, have two possible routes of entry. Similarly, herpes simplex virus may cause intrauterine infection as a result of primary infection with maternal viremia and placental seeding or, perhaps, via direct extension from a vaginal or cervical lesion of the mother. Undoubtedly, rubella virus will be cited more often than any other agent by the participants in this meeting. Since it appears to fulfill all the criteria which underlie our deliberations, this is understandable and justified. The 5 virus may infect a susceptible woman in early pregnancy, establish maternal viremia, seed the placenta, infect the embryo or fetus and produce a con- genital malformation. Because the techniques for successful laboratory ma- nipulation of rubella virus were developed just two years before the 1964-65 epidemic in the U.S.A. a great amount of new information has been acquired from that outbreak. Many of these data have necessitated partial or complete revision of some past theories and “facts” which had been based solely upon clinical and epidemiological observations without the substantiating virology and immunology. Even now a number of important problems remain un- resolved. We are uncertain of either the role or the efficacy in preventing teratogenesis of human immune globulin (“gamma globulin”) administered to the susceptible, pregnant rubella contact. The actual mechanisms re- sponsible for the observed anomalies have yet to be elucidated. To return to our initial perspective of virus and cell interactions, experi- mental observations of cultured cells infected by rubella virus have disclosed a number of unusual events. These may differ, as previously noted, with the particular cell type under study, but in a variety of human cells, chronic infections have been readily established which slow cellular division and sharply reduce the growth potential of the cultures. Nearly identical effects have been noted in cell strains derived directly from the tissues of con- genitally infected infants or fetuses. To correlate these findings with the report of subnormal numbers of parenchymal cells in the organs of rubella- infected infants is very tempting, and a link between the in vitro observations and the in vivo analyses may stem from the inhibition of mitosis noted in the infected cell cultures. There already has been one report of an ill-defined protein produced in rubella infected human fibroblasts which serves as a mitotic-inhibitor when added to uninfected cells. Careful clinical virologic investigations have helped to resolve some of the misinterpretations of earlier clinical and epidemiological rubella work but have re-emphasized the need for good laboratory techniques upon which to build the models and hypotheses of pathogenesis. A second, though far less broad, example of unwarranted reliance on insufficient evidence has arisen recently in the premature indictment of intrauterine mumps virus infection as the cause of endocardial fibroelastosis (EFE). This stems from the reports of an unexpectedly high incidence among patients with EFE of delayed cutaneous hypersensitivity to mumps antigens. The failure to detect in the same patients either mumps virus or antibodies to the agent did not deter the more vigorous proponents of this etiologic relationship. However, it is yet to be proven that any infectious agent induces immunologic tolerance in humans as a result of infection in utero. The murine lymphocytic chorio- meningitis virus model has not found a human counterpart. The mumps skin test is not sufficiently reliable to warrant the construction of any new hypotheses based solely on a “positive” reaction, as the sole index of prior infection with mumps virus. There may be other clues to be garnered from the observation, but they remain occult at present. 6 Although viral infection of the embryo or fetus may lead to faulty develop- ment or organ damage by mechanisms which remain unclear, at least two contrasting patterns emerge from present observations. A destructive process with a definite inflammatory response results from cytomegalovirus infection. In contrast, rubella virus appears to alter metabolic pathways and to inhibit mitosis without any distinct inflammation or necrosis. Yet both viruses may produce major malformations when their effects are exerted at the proper time and sites. The disparity of these pathways is even more pronounced when one considers the similarity of the final result in certain areas, such as the central nervous system (CNS). Here again timing seems a critical factor as so much of CNS development continues through all of pregnancy and for many more months post-natally. This offers a vulnerable target over a lengthy period, whereas basic cardiac maturation may be essentially complete after the early weeks of gestation and thereby better protected against significant maldevelopments. For a number of years the veterinary virologists and pathologists have encountered some striking examples of viral effects on pregnancy outcome. In their experience with blue-tongue and hog-cholera viruses much informa- tion resides which could be extremely helpful to those considering human problems. More dialogues are needed between them and us. The establish- ment of primate centers in a number of locales should permit and encourage fresh studies of the apparently enhanced susceptibility to viral infection of the pregnant female as well as many of the other unsolved problems which are not amenable to detailed and controlled investigation in the human host. From Dr. Margolis we should learn of feline and rodent studies with a group of viruses which provide fascinating models and might be adaptable to animals more closely related to humans. Finally, there is a great need for some prospective epidemiological approach which could be adapted to the assessment of the possible role of many other human viruses in the etiology of malformations. Dr. Lon White, a member of Dr. John Sever’s group which has conducted the infectious disease studies in the Perinatal Collaborative Program is scheduled next to present to us some of the data and the problems encountered in conducting one such project. II—YVirologic Aspects of the Perinatal Research Study by Lon R. White Dr. White: As Dr. Katz mentioned, I am here today pinchhitting for Dr. John Sever as the liaison representative of the Collaborative Perinatal Research Study which is NIH’s major, directed effort at the problems of human perinatal morbidity and mortality. The study is operated by the peri- natal research branch of the National Institute of Neurological Diseases and Blindness. Problems related to infection are generally the responsibility of the Section on Infectious Diseases, with which I am associated. The Perinatal Research Study was initiated in 1959 as a longitudinal prospective investigation. It involved the cooperation of fifteen medical cen- ters located at several points throughout the United States. The goal of ap- proximately 60,000 pregnancies was recently reached, and no more preg- nancies are being registered at the present time. Although the study was initiated as an investigation into the causes of cerebral palsy, mental retard- ation, and other neurological disorders, it evolved into an attack on preg- nancy wastage in general. The definition of “wastage” included both the damaged and the destroyed product. The women registering at these fifteen institutions did so at various times during pregnancy. Only 20 percent registered in the first trimester. At the time of registration a series of standardized observations were made, stand- ardized laboratory and other clinical tests were done, and a serum specimen was obtained which was frozen and stored at NIH. At intervals throughout the remainder of the pregnancy the mothers were again examined and serum was again obtained. Maternal serum was also collected at delivery and at the post partum visit. Cord blood serum was collected after 1962. Specimens of placenta were collected from the last 10,000 pregnancies of the study for possible virus isolation studies. This approach has allowed us to arrive at frequencies for the various viral illnesses during pregnancy and to get an idea of how reliable these clinical diagnoses are. The prevalence of antibody to a specific viral antigen often indicates the percentage of the population resistant or susceptible to that agent. By studying paired sera it is possible to determine the incidence of primary or secondary infection with the agent in different years and at differ- ent times of the year. Such information allows one to determine the approxi- mate risk of infection during pregnancy. I might mention a few basic frequencies. In 30,059 pregnancies repre- senting approximately the first half of the collaborative project, we found that about 5 percent of the pregnancies were complicated by one or more clinically diagnosed viral infections excluding the common cold. Despite the exclusion of the common cold the majority were flu-like illnesses. Among these 30,059 pregnancies there were 359 instances of labial herpes simplex 8 (cold sore), 45 illnesses diagnosed clinically as mumps, 23 as varicella zoster (chickenpox, shingles), 11 as rubeola, and 34 as rubella. It was particularly interesting that the reliability of the clinical diagnosis of mumps was very good. Approximately 40 out of the 45 illnesses appeared to be mumps on the basis of our serum antibody studies. Dr. Eichenwald: 45 cases out of 30,000 represents a rate of 1.5 per 1,000. Dr. White: Right. There were 23 cases of chickenpox-zoster out of the 30,059 and again serum antibody testing indicated that the clinical diagnosis was quite reliable. Unfortunately, if any of these infections had caused early abortions we would not have seen them because of the late registration of the majority of patients in the study. We have very few first trimester mumps cases and fewer still of chicken pox. This is a very major disadvantage of the collabora- tive project in relation to the problem of viral teratology. A second way of identifying pregnancies complicated by infections is to test maternal sera for antibodies against a number of antigens. This is being done, but is a costly process. This can be done with a single serum from each mother but is more meaningful when done on “paired” sera—the first specimen having been obtained at registration and the second at delivery or the post-partum visit. If the first serum was obtained during the second or third trimester, we gain very little information relating to first trimester events. A third approach to the problem is to identify the child with an abnormality and then to test the stored maternal serum for evidence of virus infection. A fourth alternative is to screen sera with a non-specific test to identify cases of probable maternal or congenital infection. Determination of cord serum IgM concentration may be useful in this regard, since elevated levels have been reported in newborns with cogenital syphilis, toxoplasmosis, cyto- megalic inclusion disease and congenital rubella. Dr. Katz: These data show how rubella has brought new emphasis to the development of the immunologic response and has initiated the exploita- tion of serologic or immunologic techniques in the absence of more convenient virologic methods. As familiar as we may be with rubella virus presently, there is still no single standard technique for virus isolation. That each system has some drawbacks is attested to by the lack of uniformity and the many different techniques used in different laboratories. This is, of course, a sharp contrast to the situation with other viral agents, such as measles, herpes simplex or polio, where some uniformity exists because of the ease and relia- bility of technical procedures that have become standard in most laboratories. With the exception of some of the active research groups, most workers feel more confident in their ability to establish a diagnosis or serologic tests, and less secure in their results from attempts at direct isolation of virus from fetal tissues, placenta, secretions and maternal sources. Dr. White: Dr. John Sever has studied the frequency of elevated IgM levels in normal and infected children with specimens from the Collaborative Perinatal Research Study. Cord sera from 1,000 children at 10 collaborating institutions were tested. Twenty-nine had elevated gamma M; 14 of these children had abnormalities including unexplained jaundice with hyper- bilirubinemia; mental and motor retardation; hepatosplenomegaly; skeletal malformations; cataracts and strabismus with nystagmus; failure to thrive; and other significant findings. One of these children had congenital toxo- plasmosis. Tests of children with congenital infection showed high cord blood gamma M for rubella (6 of 9), CID (2 of 3), toxoplasmosis (2), and generalized herpes (1). Maternal infections also were associated with high gamma M in the cord of children for rubella in the first trimester (9 of 37, 6 of these were abnormal), and serological evidence for maternal toxoplasmosis (4 of 5,1 child abnormal). There was no elevation of gamma M in children with congenital leukemia (3), and mongolism (19 of 20). One mongoloid child had high gamma M and bronchial pneumonia and peritonitis. Only 1 of 36 children with congenital heart disease had high gamma M, and this child had congenital rubella. Dr. Katz: Can you give us any indication of what is an elevated value? Dr. White: Of the 1,000, 3 percent had concentrations above 22 mg. %. Dr. Eichenwald: This is by immunodiffusion ? Dr. White: Yes. Dr. Eichenwald: Has he taken babies who have documented erythro- blastosis to determine the proportion with elevated IgM levels? Dr. White: Dr. Sever’s comments to me indicated that erthroblastosis does not usually seem to be associated with elevated IgM’s. Dr. Eichenwald: What does “usual” mean? Is there a relationship between immunologic, non-infectious event and elevated IgM levels? Dr. White: He found no increase in IgM levels in 6 infants with erythroblastosis. Dr. Eichenwald: It was demonstrated years ago that congenitally infected babies may have elevated IgM, but we have also encountered normal babies with similar high levels. Is there any systematic investigation going on to find out why these normal babies have an elevated IgM? Dr. White: Dr. Sever is investigating this now—the study is incomplete at this time. Dr. Gruenwald: Can non-pathogenic infections produce this? Dr. Silverstein: Yes. Dr. Gruenwald : Might this be the answer ? 10 Dr. Benirschke: Are you looking at these 29 children with elevated globulins from a variety of parameters, like placental pathology, fetal out- come, etc., and putting these against the types of antigens? Dr. White: Yes. But if the patients registered in the third trimester, what do we do? Our serologic studies are seldom very helpful. If we had frozen placenta on all of these, we might have a chance of isolating something. The placental tissue which was frozen has been stored at —20° C. If the agent were cytomegalovirus it would not be recoverable now. Rubella virus, on the other hand, might be recovered. Dr. Benirschke: What is important is that, for instance, the new pro- spective study now going on in Germany be advised of the possibilities that exist in this area, that follow-up would be very important in some depth if further data can be obtained. Dr. Gruenwald: I think they are freezing half of each placenta. Dr. Benirschke: But at proper temperatures? Dr. Eichenwald: We know infections may produce elevated IgM levels, but one cannot assume that high levels always imply infection. Could this not be due to a variety of other causes? Perhaps a child with erythroblastosis or one with an inflammatory response not related to infection could produce this also. Dr. Silverstein: There is no question in my mind that appropriate non- infectious agents can induce active response in the fetus and can result in high IgM level. Dr. Benirschke: I grant the possibility there might be infectious agents not yet recognized as being harmful. Dr. Silverstein: Absolutely. Dr. Benirschke: You would admit though that microcephaly, if you can’t pin it down any further, might be the result of an infection by some agent which is in need of being identified. Dr. Silverstein: It is also possible that infectious agents in the fetus may fail to give rise to elevated IgM levels. Dr. Gruenwald: What do you consider as noninfectious? Dr. Silverstein: I can conceive of blood proteins such as egg albumin possibly getting across to a human fetus and stimulating an IgM level. Dr. Eichenwald: This is what I meant by an immunological event of a non-infectious nature. Dr. Katz: Non-replicating antigen. 11 293-596 0—68——2 Dr. Eichenwald: I doubt if this problem can be approached in retro- spective fashion. Rather, the answer will have to come from a well-designed prospective study. If you decide to approach the issue by looking closely at infants with elevated IgM levels, you will probably find something abnormal in a high proportion, simply because you looked so hard at them, not because there is a true association between elevated IgM levels and abnormality. Dr. Katz: If you select Sever’s data on IgM for discussion, I would call attention to another debatable topic which Lon mentioned. Herpes simplex virus and “cold sores” are ubiquitous but remain the focus of mixed enlight- enment and ignorance in our understanding of their pathogenetic mecha- nisms. Simplex is a human virus that has been with us for years and appar- ently has a counterpart in nearly every animal species, a great family of viruses distributed throughout the evolutionary spectrum. Despite the ease with which it can be isolated in the laboratory (if you can’t cultivate herpes simplex, you might as well close up shop) and the multiplicity of animal hosts, we still do not fully comprehend human pathogenesis. Does dissemina- tion occur only with primary infection, while recurrences are purely local? Can encephalitis result only from primary invasions or may they result from recurrent episodes? Does maternal-fetal spread require viremia or may local invasion occur from a vaginal or uterine cervical lesion? The studies of Nahmias and colleagues (2) at Emory and those of Charles Alford (3) at Birmingham have already shed new light on the “markers” that may distin- guish benign strains from virulent populations. Unanswered questions persist in considering human fetal infections which may be fatal or severely damaging. The pathologist recognizes a fairly characteristic lesion, especially in the liver, but in all organs of some cases. The virologist can confirm the histologic hints with recovery of virus from affected tissues. Serologically, there are pitfalls and I had hoped Dr. White might be able to comment upon them. Does Dr. Sever’s data help us in this regard? How much cross-reactivity did he find in the sero-survey among the herpes family of viruses, simplex, cytomegalo-, and varicella-zoster, DNA agents with some antigenic crossing? Can antibody response be distinguished sharply and specifically ? Dr. White: I’m sure his comment would be that it is an obvious headache, unavoidable at this time. There is antigenic crossing. If you are interested in chickenpox rather than herpes simplex and if you have a mother diagnosed as having chickenpox and if she has a seroconversion to varicella antigens, you wouldn’t worry too much about crossing. Dr. Katz: In the usual age group of pregnancy, varicella should be relatively rare. Even without overt buccal lesions, herpes simplex is probably far more common than we realize. Once the proper examination is done to look for vaginal and cervical lesions, its true incidence becomes more 12 apparent. Its prevalence demands further efforts to devise a completely specific serologic test. Dr. White: I think you are saying that antigenic cross reactions limit the usefulness of the serological approach. Eventually we may be able to overcome these limitations with better serological techniques and antigens which don’t cross, or we may be forced to approach the diagnosis by virus isolation or some other means. With even a single agent there may be different types with varying viru- lence for the fetus. The usual type of labial “cold sore” herpes may or may not be a significant pathogen for the fetus. It may not be adequate to in- vestigate the teratogenicity of an agent by lumping cases together on the basis of either clinical or serological characteristics of the infection alone. Dr. Katz: I agree with you. Dr. Benirschke and I can cite the same experience with newborns at the Boston Lying-In Hospital during his time there. Among 5,500 infants delivered annually there will be one or two with frank illness whom we are asked to study, usually after the diagnosis has been recognized post-mortem from the characteristic focal hepatic lesions. From tissues of such infants, it is relatively easy to grow herpes simplex virus in abundance and we then feel somewhat self-satisfied because we have co- roborated the pathologist’s suspicions. Probably our concern would better be directed to the infant who does not manifest overt illness but who may have become infected in the perinatal period and end up with a brain that does not develop fully or function normally as a result of an occult infection. Until the pathogenesis of herpes simplex virus infection is fully elucidated, we cannot disregard all the mothers with “cold sores” or cervical lesions by accepting the dogma that such recurrent lesions will not result in fetal in- fection. There may even be some form of infection which produces a sequence similar to that observed with the Kilham rat viruses in which a very selected cell type in the central nervous system may be affected postnatally. Perhaps the virulent, genital strains induce the fatal neonatal infections with virus disseminated throughout all organs and tissues. Less virulent strains could, in contrast, evoke no acute, fulminating illness but slowly exert their damage. Another analogy may be found in the attenuated hog cholera and blue tongue viruses which do not produce severe enough infection to kill the developing swine or lamb but permit him to survive and manifest his abnormal central nervous system, a result of non-lethal intrauterine infection. Dr. Benirschke: Do individuals who have recurrent herpes sores have a viremic period? Is this well-established? Dr. Katz: No, this is not well-established. We assume they do not, but I am unaware of any studies of pregnant women in particular. Viremia has been shown to occur only with a primary herpetic infection and not with the recurrent forms, in normal hosts. The question herein would be—is the preg- nant woman a normal host in the manner with which she copes with recurrent 13 herpes simplex? Of course, one does not have to invoke viremia if spread is a more direct one from recurrent or primary lesions involving the female reproductive tract. Dr. Eichenwald: Do these women have “cold sores” of the cervix or uterus? They can have cold sores of almost any epithelial surface, so why not of these organs? Dr. Katz: Dr. Nahmias and others are investigating the incidence and significance of lesions of the cervix which are apparently far more commonly due to herpes simplex virus than formerly appreciated. Dr. Benirschke: This is recognized in Papanicolaou smears (NAIB) and gives confusion with carcinoma because of the malignant appearance of the cells. You may have the lesions in the uterus. The simpler thing is to inquire whether pregnant women with recurring herpes, who feel it coming on for several days, have a period of viremia. Dr. Katz: It has been assumed that viremia does not occur in such cases, because specific herpes simplex virus antibody is present in the serum. In individuals with very low titers, insufficient antibody might be available to prevent a brief viremia which could easily escape detection. Ordinarily we would expect that virus reaching the blood would be rapidly neutralized by specific antibody. Dr. Eichenwald : How many women of child-bearing age have antibodies to herpes simplex? Dr. Katz: This is an important question because it illustrates our naiveté in accepting data of the past as gospel for the present. Older studies showed at least 80% of young adults to possess herpes simplex antibodies indicating acquisition of infection in childhood and the teens. Similar studies today demonstrate a marked reduction in the number of immune adults with some groups as low as 35% immunes at the child-bearing age. Many factors may have contributed to this changing epidemiology, at least one of which may be an increase in the general level of hygiene with less transmission in a society living in conditions of modern sanitation. This experience is not unique to herpes virus and parallels the pattern with polioviruses in the U.S.A. (prior to alteration of the picture by vaccines) and some of the present trends with infectious hepatitis and salmonella. Perhaps Dr. White can tell us something of the sero-profile among the women in Dr. Sever’s study. Dr. White: Data from the Perinatal Research Study can provide the sort of answer that Dr. Eichenwald wants as to the prevalence of elevated IgM in the cord blood of normal and abnormal infants. We can determine the prevalence of specific antibodies to specific viruses and the incidence of sero-conversions during the second and third trimesters. 14 The shortcomings of the Collaborative Perinatal Research Study relate principally to the late registration of the majority of mothers, the lack of a sharp focus on certain types of problems, the heterogenity of the population and data collection, and its sheer bulk. The study will allow us to answer a number of specific questions without too much difficulty. However, by virtue of its very broad scope and large size it will suggest many new relationships between certain obstetric and pediatric events. It seems to me that there is little need for more studies of this same type unless they can avoid these difficulties. If we want more prospective studies on the consequences of maternal infection, then these should be designed to efficiently detect and define infections very early in pregnancy. Such a study would probably have to begin before conception. An alternative approach might be a combined retrospective-prospective approach. For example, a study in which newborn infants who might have a congenital infection could be identified in the first day of life through the use of cord blood testing, examination of the infant (chorioretinitis, unexplained petechiae or hepatosplenomegaly), or examination of the placenta. During this time the placenta might be held at +4° If all was normal, the placenta might be discarded; if congenital infection were suspected, the infant would be examined closely and followed for an indefinite time while a sample of placenta and cord serum would be frozen at —70° C. for subsequent virus isolation and serologic studies. This is the sort of study which might well answer questions raised by the Perinatal Study and yet be finanaced on a much more modest scale. Dr. Benirschke: Would you fill us in on the 45 mumps cases? Dr. White: Sure. Looking at these 45 mumps cases initially, I was struck by the frequency of hepatomegaly, moderately elevated bilirubin levels and a number of minor abnormalities. However, these were described nearly as often in the matched control children. The only thing we could really find was that they were small babies. There was a definite skew of the distribution toward small size, most marked in head size at one year of age. Dr. Eichenwald: Are these infants small irrespective of the time at which mumps occurred during the pregnancy ? Dr. White: We had such small numbers that we could not group them by trimester. As best we could tell, the baby was as likely to be small if the mother had mumps in the third or the first trimester. I plotted the mumps infants and their controls on intrauterine growth grids and found no evidence of intrauterine growth retardation. However, the applicability of the currently available grids to the Perinatal Study popu- lation is questionable. Dr. Gruenwald: At high gestational ages the weights are a little low. 15 Dr. White: The Perinatal Study population is geographically, racially, and socioeconomically heterogeneous. Dr. Benirschke: This is a recurring question, do they have cardiac fibroelastosis? Dr. White: No. There were no cardiovascular abnormalities noted in these children through one year of age. There were three children with mental retardation, two suspect and one definite. One was a child with microcephaly. The most dramatic individual case was that of a child who at 15 months of age was hospitalized with diabetic acidosis and died. In early infancy this patient had hepatomegaly and hyperbilirubinemia not explaned on the basis of any blood group incompatibility. At autopsy there were interstitial pneumonitis, hepatitis, and a residual enterocolitis. Obviously, it is a very provocative single case. Dr. Katz: Not just as a single case, but the whole syndrome of “runted” babies after maternal mumps would be extremely provocative if documented by careful attention to the pitfalls you mentioned. There are some parallelisms between the viruses of rubella and mumps; both are myxoviruses sharing many physical, chemical and biological characteristics. It would be attractive to search for a mumps model that might show the same inhibition of cell proliferation and mitosis that had been found with rubella. Dr. White: We need data on more than five first trimester mumps infections. Dr. Katz: In considering other possible approaches for the perinatal collaborative program, it seems wise to bear in mind the experience of many laboratories, including our own, with an alternate approach to the search for viruses in fetal, placental or maternal tissues. There are a number of examples where the inoculation of disrupted cells or suspensions prepared from grinding tissues has failed to transmit any detectable virus to tissue culture systems. With the same initial specimens, however, the preparation of primary cell culture from explants or trypsinized cell suspensions of the tissue specimen itself has been possible. Cultivation of these cells, sometimes for lengthy periods, has eventually paid off when they spontaneously degen- erated to reveal the presence of virus which until then was not detectable, presumably in some form of latency. To freeze all specimens that are later to be studied may be acceptable with serum, but if one wishes the best results with the tissues it may be necessary to prepare cultures from them as soon as possible once obtained, or to prepare cell suspensions which could be stored at extreme cold and later thawed for the initiation of cell cultures. It seems that with some virus-cell systems, it is only by the culture of the infected cell itself that resumption of the replicative cycle is encouraged and, thereby, the release of infectious virus is obtained. 16 Another detail to which we may have paid insufficient attention is the source of the cells used in many laboratories as culture systems into which to inoculate specimens for virus isolation. Many groups purchase their cultures from biological firms and the investigator may not even know the initial source of the primary embryonic or adult human tissue he is employing and although we have become wary of monkey tissues with their high inci- dence of “passenger” viruses, less emphasis has been placed on similar findings in some human cells. Often such cells have been obtained, randomly, from premature infants and embryos. In the study of Rawls and Melnick (4) at Houston they included (in one group of 77 infants) autopsy kidneys from babies under 6 months of age with a variety of congenital malformations, leukemia, infectious diseases and other disorders. Dr. Benirschke: Newborns? Dr. Katz: Prematures in one series; from 1 day of age to 6 months in Houston; Boué’s represent mainly spontaneous abortions at 12 to 20 weeks’ gestation. (5) Dr. Benirschke: In contrast to Boué’s (5) study who ground up 74 abortuses. Dr. Katz: This is the crucial point! These are not specimens which are “ground up” but rather they are reduced to cellular suspensions which are then grown as primary cell cultures. If the organs are obtained within a few hours of death, they can be minced and explanted in plasma clot cultures or trypsinized and grown as monolayers. From these there is a better chance of the emergence of a hidden virus than from the “grinding up” and inocula- tion procedure. Dr. Moorhead: Has no one frozen them down with preservative the way they do for tissue cultures? With glycerol, or DMSO, in small percentage? Dr. Katz: Yes, with either glycerol or DMSO. Dr. Moorhead: That should be feasible. Dr. Katz: Do you mean as a cellular suspension or an intact tissue? Dr. Moorhead: Intact tissue, I thought. Dr. Katz: Then I can’t comment because our work has been confined to cells only. Dr. Eichenwald: We have attempted to freeze fetal tissues in dimethyl sulfoxide, and then grow them later. It hasnt worked well. Most people, I think, have given this up and tried to grow them as soon as possible. Dr. Silverstein : What tissues have you tried? 17 Dr. Eichenwald: We worked with whole minced fetuses. We tried to suspend them in dimethyl-sulfoxide, freeze them and thaw the tissue at some convenient time but they did not grow well. Dr. Axelrod: Many of these tissues can be shipped by Air Mail in appro- priate media composition for three or more days and then can be grown out in tissue culture. We received specimens from Japan by Air Mail and these make excellent substrate for the outgrowths of viruses. Dr. Katz: In what state are they shipped ? Dr. Axelrod: One by one by three millimeter pieces suspended in media. Dr. Katz: Tissue can be kept for weeks or months that way, minced and in the refrigerator under a shallow depth of some medium such as Hanks- Sims, a balanced salt solution plus bovine serum ultrafiltrate. Dr. Axelrod: We determined that the oxygen from the surrounding media will get to the central core of a one-by-one-by-three millimeter piece very nicely for three days at least at average room temperatures. Dr. Benirschke: I though Dr. Benyesh-Melnick directed her question to congenitally abnormal babies in the hopes of finding viruses in say, children with congenital heart disease or other congenital abnormalities, and grew preferentially only those tissues of individuals who died at one or two years of age. Dr. Katz: Actually the Houston material consisted of a variety of condi- tions and age groups, newborn to 15 years of age. They came from a neighbor- ing pediatric institution regardless of the clinical diagnosis. Their isolates included cytomegaloviruses, adenoviruses, measles, varicella-zoster and Cox- sackie B 1. In contrast, the Baltimore study by Klein and Huank (6) examined 100 premature infants without regard to any congenital malformation and revealed 5 adenoviruses covering 3 different serotypes. Boué’s (5) studies" were done on 74 embryos which included spontaneous abortions and some surgical interruptions. He and his co-workers detected no viruses, using a modified organ culture technique. Dr. Benirschke: Is that a valid approach? Dr. Katz: As long as one does not interpret beyond the limitations of his data, it is an interesting and valid approach. For optimum productivity, it demands the correlation of pathology, obstetrics and pediatrics, virology, genetics, and perhaps others to extract the utmost information from each specimen. To isolate a virus from a tissue or organ of an infant is most provocative but this does not establish its etiologic role in whatever malforma- tion or disorder the baby may have had. Kidneys are not the sole organ to be approached in this fashion. Lung, heart, skin and other tissues can be handled in a similar way. 18 Dr. Eichenwald: We need a study which compares the incidence of virus isolations in fetuses which abort spontaneously to that found in fetuses aborted therapeutically. This would seem to be one way to determine the role of infections in spontaneous abortion. The second point I would like to raise deals with the size of infants born from mothers with infections during pregnancy. There are a number of studies in the literature, mostly of very poor quality, on what happens to infants of mothers who had hepatitis, mumps, measles, etc. during pregnancy. Dr. Benirschke: Art, would you describe the role of immunology in these problems? 19 III—Immunologic Responses of the Fetus by Arthur M. Silverstein Dr. Silverstein: The best way to introduce my topic is to repeat what we have all become aware of : that immunologic mechanisms are normal physiologic ones, and as such, have both benign and deleterious aspects. I don’t think that I have to dwell too much on the protective role of immunologic mechanisms and of antibody, which we are all aware of, especially in viral diseases. I might indicate that, as we will restrict our discussion to intrauterine, intrafetal disease, antibody either derived from the mother across the placenta (primarily gamma-G, 7S antibodies) or that actively produced by the fetus (gamma-M and gamma-G) should, as a first approximation, provide as much protection as they do in an adult. There are one or two exceptions, perhaps. For instance, the fetus of some species lack complement components. In certain instances we are aware that complement may be involved in the mediation of the protective action of antibody. In man, there is certainly a complement deficiency throughout the greater part of gestation. We are aware, now, that the fetus can respond immunologically to a variety of antigenic stimuli with an active response, one that duplicates in many respects that of the adult. The last 5 or 10 years have witnessed the change of our old concept that the fetus is immunologically incompetent, incapable of any immune response. Some interesting aspects of the fetal responses have come out of recent studies, that may bear on the questions of protection against viruses, and possibly contribute to an understanding of some of the deleterious conse- quences of immunological responses in utero. I think we are now able to make the generalization that the fetus of most mammalian species that have been studied (with the unfortunate exception of our favorite small laboratory rodents) is capable of immunological responses in utero when appropriately stimulated by antigen. At surprisingly young gestational ages a fetus is able to respond immuno- logically to certain antigens, but not others. As we go through gestation, we find that prior to a certain critical age, each antigen seems incapable of inducing an active immune response on the part of the fetus. After these critical stages of gestation, different for each antigen, the fetus can respond to the given antigen with an active response, including antibody formation, hypersensitivity and all that these imply. I do not think it is worth our while to discuss the possible explanations of the significance of these critical ages. Now, we know little about the majority of antigens in any given species. Probably the fetal lamb is the best known, and in this species, we know about the timing of competence for perhaps 8 or 10 antigens. Some, as I indicated, arise very early in gestation, some of them very late in gestation, and some 20 of them after birth. There are certain antigens to which even the newborn, for many months, is incapable of responding. The point to be made here is that the course of an infectious disease may be affected by a fetal ability to respond immunologically, or by its inability to respond. Thus, also, the course of a fetal viral infection may depend upon when the virus enters, and when the fetus develops immunologic competence to the antigens of that virus. This may be reflected in contributions either to the pathology of the disease process, or in his own protection against the virus with active production of antibody. Dr. Benirschke: How long does it take for him to respond after challenge, once he becomes capable of responding? Dr. Silverstein: We have been very interested in this question, since one has been accustomed to hearing about immunologic immaturity in the very young animal. We looked for signs of this immunologic immaturity and have not found it. Of course, prior to the age at which a fetus can respond it is immature in that sense. It has not got the wherewithal to do anything. When this critical age is passed, however, and the fetus becomes competent to respond, it seems to exercise all the maturity of an adult. We have yet to find a good objective sign of immunologic immaturity in the developing fetus, with respect to the cellular kinetics of the response, the molecular sequence of the response, or its quantitative aspects. For in- stance, when we studied homograft rejection, the rapidity of the fetal re- jection paralleled that of the adult. In general, we can expect the immunization of a competent fetus to result in response within the first three to six days, depending obviously on the nature of the antigen, on its dose, and on the manner of presentation of the antigen to the responding host. Now, a word may be in order here about some of the technical approaches. I think that they may prove useful to anyone interested in studying any sort of fetal disease processes or embryopathy. Again, we were originally lead to believe that one could not really do anything actively with the mammalian fetus for fear of causing interruption of pregnancy and abortion. It has become abundantly clear in the last few years, that this is untrue for many species. One can perform a variety of surgical interventions through the uterus allowing work with the fetus in a wide variety of approaches. Suffice it to say that depending obviously on the species, and on the type of uterus and placentation involved, surgical intervention can be accompilshed. For instance, the fetal lamb can be immunized as early as about 35 days of a 150- day gestation when, by the way, it has no functioning lymphoid tissue that we could find. Intervention in the fetal rhesus monkey can be accomplished at the end of the first trimester, but not without some difficulty. Obviously, the success rate will vary with the species. It is high with the fetal lambs, and somewhat less with the fetal monkey. In the latter instance, we get about 50 or 60 percent success rate with fairly radical procedures at 21 the end of the first trimester and perhaps 80 or 90 percent success rate at mid-gestation with certain other types of procedures. Some of these are quite extensive operative procedures. It has always appeared remarkable to me, to conceive of a fetus surviving some of the surgical approaches which we pursue, but in fact they do. We have not only been able to immunize the fetus by almost any route including intraocular, during the first trimester or first half of gestation, but during the latter half of gestation in species with which we have been working, we have performed skin grafts, thymectomies, splenectomies, etc., all of which are rather extensive procedures. One can insert permanent indwelling catheters in the fetal lamb and, I feel, in the fetal rhesus monkey in the near future, bring these through the uterus, through the abdominal incision so that one has continuous access to the fetal circulation. I would think that this latter approach would be eminently useful to the virologist interested in transplacental passage, in the study of fetal viremia for example. I think this probably suffices to indicate what one can do with the fetus. Comparable interference has been conducted with the human fetuses, by intrauterine transfusions across the abdominal wall in erythroblastosis. Also, there have been attempts, sometimes succesful, sometimes not, at exchange transfusions of the fetus, at 25 to 28 weeks gestation. There are some people in Scotland who are trying to perform bone marrow transfusions in the human fetus at about 14 or 16 weeks for the production of chimeras. Although I haven't yet heard whether they have had any success, these are among the possibilities that exist. I think that it is clear, then, that the fetus can make an active response, can help to protect itself against congenital infectious diseases in utero under certain select circumstances. Now, with respect to the possible deleterious contributions of immunologic mechanisms in the fetus, we refer to the same phenomena that we talk about in any discussion of general immunopathology: the same range of hyper- sensitivity diseases, the same types of immunologic contributions to the pathogenesis of a variety of infectious and non-infectious diseases that we are used to thinking of in the human. There are, however, certain special advantages offered by the fetus, and peculiarities of its response, which deserve discussion here. In congenital infections in general, and in congenital viral infections specifically, the possibility exists of immunologic contributions to the patho- genesis of the disease. A given virus, infecting the fetus prior to its critical age of immunologic maturation, may give a completely different clinical and histopathologic picture than if the virus enters the fetus after it has become immunologically competent to respond. Obviously, if we are talking about a cytopathogenic virus which provokes little or no immunopathologic contribution to the development of the disease, then the situation will be altered. 22 One can divide the pathological types of diseases that develop in response to viral infection into a spectrum, at one end of which immunologic contribu- tions will be important. I think we have reason to believe that is true. On one end of this spectrum then, we may think of viruses that exert a direct cytopathogenic effect on cells, leading to the death of cells, or to hypoplasia or hyperplasia. This has already been alluded to earlier. On the other end of the spectrum we may have certain viruses that exert their effect not so much by a direct toxicity to cells or organs in interference with organogenesis as in provoking an inflammatory response. This is what I would like to discuss now, the possibility that antigen in the immunologically competent fetus, be it virus or other type of pathogen, may stimulate an immune response in the fetal host. This immune response, which we normally think of as being benign and protective, may in fact contribute to the development of the disease and may in some instances be the chief cause of fetal embarrassment and the termination of pregnancy. I think the two best examples of this, one good example and one not so good, but which I think has possibilities, lie in the story of the lymphocytic choriomeningitis in the mouse and congenital syphilis in the human. You are probably all aware of these aspects, but it might be worthwhile to reca- pitulate briefly, because I think it tells such an interesting biologic story. In lymphocytic choriomeningitis we have a virus that is toxic to mice, such that the infection of a normal mouse (one that has an active immunologic competence) leads to the development of the typical pathologic disease, and in many cases to the death of the animal. By a curious inversion of our normally accepted values, the mouse that is immunologically incompetent to respond to this virus (a tolerant or im- muno-suppressed animal) develops a viremia, but does not develop the disease. This is a disease that histopathologically does not look especially like an immunologic process, but by all of the phenomenologic observations that have been made, one would be hard put not to call it an immunologic disease, at the least one triggered by the virus. So we have a situation in which the absence of the host’s immune response, the absence of ability to protect itself, results in its salvation, and the presence of an adequate immune response is its death warrant. This has been abun- dantly demonstrated. The other interesting story which I think again bears on our present concern is that of congenital syphilis. I mention this for two reasons. I think it is a biologically interesting story and further, I think it has implications for our interest in how viruses work in the fetus. Congenital syphilis generally is considered, and I do not mean to suggest that it should not be so considered, not to be an immunologic disease. Tre- ponema pallidum has an extremely good set of antigens which very com- petently induce immune responses in the infected host. The typical curve one is in the habit of drawing for the incidence of Treponema pallidum infection 23 in the human fetus starts at about the fifth or sixth month of gestation and increases as gestation continues. One does not generally see congenital syphilis much before the fifth or sixth month of gestation. It has been suggested that this is because the treponeme doesn’t cross the placenta prior to this time. But I think it is of some interest that one occasionally sees cases of fetuses which have been aborted for one reason or another and have treponeme in their tissues, but no typical lesions of syphilis, that is no inflammatory reactions. The interesting situation of a treponematosis without disease obviously suggests possibilities of viremia with otherwse pathogenic agents without disease. Obviously we would not see many of these cases, because in the absence of inflammatory response by the host and in the absence of very much direct effect of the treponeme or other agent on the host tissues, there would be little cause for fetal embarrassment and, therefore, little reason for the fetus to be aborted. Should the fetus be aborted, however, it is not likely that it would be examined for treponemes, because it would not present the lesions that give the pathologist the clue to do a silver stain. As soon as the typical lesions of congenital syphilis arise, one sees the plasma cells which the immunologist now accepts as being the morphologic indicators of active immune response; the cells which produce the anti- bodies. Thus, the suggestion should perhaps be entertained that we are deal- ing here with a situation in which, in the absence of immunologic competence and the capacity for active immunologic response, the otherwise pathogenic organism may not be pathogenic in the fetus until immunologic and perhaps other host responses develop. These may render the host able to mount what we normally think of as a protective mechanism, but which in this case may lead to the embarrassment of the fetus and fetal death. Thus, I think that lymphocytic choriomeningitis (involving specific in- competence due to immunologic tolerance) and congenital syphilis (involving general incompetence due to youth of the fetus) indicate that some pathogens depend for their pathogenicity on host factors. Now as far as current approaches to this problem, I think I have indicated the range of approaches which one can take to the fetus, and what one then is able to do in the direct experimental study of congenital infectious diseases. We in our laboratory are interested, as you might have gathered by now, in the question of immunologic contribution to the pathogenesis of disease, and we have been looking for model systems other than lymphocytic chorio- meningitis to try to examine the possibility that a pathogenic organism in- troduced early in gestation may give a completely different picture than one put in at a later time. We have been studying Treponema pallidum in the fetal rhesus monkey and blue tongue virus in fetal sheep. I do not think we have yet made as much progress as we would have liked. One can, however, produce experimental syphilis in the rhesus monkey, al- though it appears to be a milder disease process in this species than in the 24 human. We have not yet timed the development of competence in the rhesus monkey to the treponemal organism, but preliminary results look interesting. The point I want to make here is that by direct access to the fetus rather than by trying to inoculate the mother and hoping for transplacental passage, we can get close to 100 percent fetal infection. I would guess that attempts at in- fection of the mother would result in a sharply lower incidence of fetal in- fections. Thus, the yield is pretty poor, and rhesus monkeys are not getting cheaper, so one wants to have a better approach to the animal. Our blue-tongue virus studies have also proved to be very interesting. As has been indicated, we are working with an attenuated virus. It produces some interesting lesions in the adult, and somewhat different lesions in the fetus, the attenuated virus crossing the placenta and being more pathogenic in the fetus than in the adult. We have been interested in it as a possible experimental model because the description of the pathology in the adult involves good plasma cells, and thus obviously some immunologic component of the disease process. Again we are trying to establish a model in the fetal animal to correlate the time at which the fetus becomes immunologically competent to respond to the pathogen with the time that the disease assumes its typical form. In this instance also, if one depends upon maternal infection and transuter- ine passage of the virus, one gets perhaps a 30 or 40 percent yield of fetal in- fection. However, one can introduce the virus directly into the fetus at almost all stages of gestation by some of the techniques I have described, and assure oneself of 100 percent success, at least in getting the virus into the fetal host. We have seen in this study some very fascinating brain lesions. We have animals that have no cerebral cortex, an occasional animal with markedly smaller cerebellum and animals who, in the absence of most of their cortex, go on to normal term and seem to be neurologically normal. Dr. Eichenwald : Did this occur with the attenuated strains ? Dr. Silverstein: Yes. Dr. Benirschke: Am I right that it is possible to infect already immune mothers and have the organisms cross the placenta? Dr. Silverstein: I think this has been described: whether it is true or not I'do not know. Essentially what happens is that when one immunizes the adult sheep subsequently in order to render it immune to the virus, there is a lag time of five or six days and a viremia develops, with a spike of fever in the mother. It is apparently during this critical time, during the virema stage, that the virus gets into the fetus. I suppose in a putatively immune adult, if one introduces enough virus, some of the virus has a chance to get to the fetus. I believe that our preliminary study of the ocular changes occurring in blue tongue infections of the fetal lamb highlights the point I wish to make about the relationship of the stage of maturation of the host to the nature of the disease process induced by the pathogen. In addition to some interesting 25 corneal lesions seen throughout, the uveal and retinal changes are most sig- nificant. Early in gestation infection often leads to a non-inflammatory retinal dysplasia, resembling somewhat that occasionally seen in the human. Later, however, inflammatory reactions of the uveal tract lead to quite a different set of ocular lesions, with the normal structural relationships of the retina often unaffected. I believe that this makes the general point well, that host mecha- nisms, including immunologic ones, may contribute significantly to the course of disease processes such as we have been discussing. Dr. Benirschke: We like your lambs and monkeys, but what is the factual knowledge, or should there be gained factual knowledge, of the devel- opment of immune response in the human embryo and how can it be obtained ? Dr. Silverstein: We know very little about analogous immunologic situations in the human. We can make certain predictions from the rhesus, but I don’t know how far these will carry us. The rhesus tells us that a primate can develop immunologic competence to certain antigens very early in gestation. We have seen homograft rejection in the rhesus immediately after the end of the first third of gestation. I assume, from the general absence of graft-vs- host responses, that the human fetus at the time one is in the habit of trans- fusing it (22+ weeks) is immunologically capable of rejecting the donor leukocytes present in the transfused blood. Congenital syphilis represents one of the indirect ways that one can get at the timing of immunologic competence in the human. We do not see plasmo- cytosis or any signs of antibody response prior to about 21 weeks in the human fetus. In toxoplasmosis the youngest case of which I am aware (which is, I think, based on a morphologic demonstration), was 30-odd weeks. Dr. Frank: What about blood as an antigen? Dr. Silverstein: The trouble is that many of the blood group substances are not very good antigens. There is one report from California of a new- born who had antibodies against a maternal gamma globulin type. Dr. Frank: I was thinking of different red cell types. Since you do not know the fetal type there must be situations where the fetus has been given red cell type which is different from the mother. Dr. Silverstein: We know nothing about the timing of competence to this kind of antigen in the fetus. Dr. Benirschke: There was a case presented recently at the pediatric meeting in Atlantic City (1967) which looked as though it provides pretty good evidence for runt disease due to intrauterine transfusion. The first one, I think, that might be acceptable. In connection with this I want to show a slide. (Fig. 1) 26 b 5% ve. ; Figure 1 - Rw Se © Lung of immature human fetus, approximately 12 weeks gestational age showing aspirated exudate in primitive bronchi and (at arrows) cells with the histological qualities of plasma cells. (H & E x 640) . This is the youngest human fetus I can find with a suggestion of plasma cell formation in the interstitium of the lung. This is at about 12 weeks gestation. There is fairly good plasmacytoid transformation and there is a reaction interstitially. The bronchi were cultured and yielded chromobacter. It seems to me that this is the kind of information one ought to look for to see when man can make, for the first time, plasma cells, and we infer antibodies, to some specific antigens. Dr. Silverstein: I don’t find this surprising. I would be very willing to say that the human fetus, by analogy, with the rhesus, should be capable of forming antibody from the middle of the first trimester onward against some antigen or other. Dr. Benirschke: If one had the same observations in rubella and with various other specific antigens it would be helpful. Dr. Silverstein: If one can show for a specific entity the earliest age of competence and then make some correlation with the pathology, this is then very important. Dr. Gruenwald: Doesn’t the published material contain any useful in- formation in this respect? Téndury (7) has written a book about this. He has a tremendous amount of information. It may interest you why this can be obtained in Europe and not here. They dilate the cervix for therapeutic abor- tions much more and hence come out with intact embryo. Here they do not 27 293-5696 0—68——3 even have big enough dilators. They go in with a curette and come out with mashed material. This is the reason why so much material exists in Europe and not here. I wonder whether one might find this kind of information in Tondury’s material. He probably has not looked for it too much. His book on early rubella has a tremendous amount of material. He gets it from all over Germany and Switzerland. Dr. Benirschke: I have the book here. The lesions are more destructive than inflammatory. Dr. Gruenwald: It would be nice to review his material. This is exactly what you are looking for now, isn’t it ? Dr. Benirschke: I think so. Dr. Eichenwald: He does not mention plasma cells in his material. He is primarily interested in morphologic changes. Some of his data on influenza, polio and mumps are, I think, not really definitive. Dr. Gruenwald: But I think his rubella material is good and it might very well provide the answer you were seeking. Dr. Dancis: There are studies reported on human fetuses in which tissue obtained from early gestations was incubated with radioactive amino acids and gamma globulin formation was demonstrated. This was reported during the past few years in a monograph in the Scandinavian literature. (8) Dr. Silverstein: Since the question of gamma—M gloublin has come up, it should be mentioned that the fetuses of some species, apparently normal and unstimulated, may produce significant amounts of gamma—M globulin. This is true of the lamb, where one can find it in the circulation of the normal fetus at almost all gestational ages, produced primarily in the spleen. Studies have demonstrated the production of gamma-M by the normal human fetus, and this presumably is the base line above which you are measuring. What the relationship of this globulin production is to antigenic stimulation, or perhaps to some normal physiologic process of the fetus, is currently unknown. Dr. Dancis: You mentioned in the course of your presentation, once the stage of development is reached at which antibodies to an antigen can be formed, it seems to be complete and maximal at that time. Years ago we measured the response to diphtheria toxoid in infants of different ages. (9) A single injection of toxoid was administered and antibody levels were determined in the rabbit skin. With increasing age, antibodies were detected earlier, and the final titre reached higher levels. These babies had no demonstrable passive transfer of antibodies. Dr. Silverstein: There are many data like this in the literature. It has been hard, really, to interpret what they mean. It may signify different kinds of antibody. It may be that in our studies with animals, we were measuring other types of antibodies, the wrong kinds. 28 We have essentially done studies of three types to examine this question. We have skin-grafted animals before and immediately after the develop- ment of homograft competence. We have found that whereas the fetuses do not even “see,” immunologically, a skin homograft in utero prior to this age, after this age they will reject the grafts with the rapidity and competence and the inflammatory exuberance of the adult. Secondly, studying the response in the fetal lamb to a viral antigen, we were able to demonstrate that the fetus would deal with the antigen as rapidly as an adult, and that the subse- quent 19-S and 7-S response appeared to be comparable to what the adult would produce. Finally, in the rhesus monkey we studied cellular kinetics of the response to sheep erythrocytes, and the proportion of the total number of lymphocytes that the fetus can bring to bear in response to this antigen in comparison with the proportion the adult can employ. When we plot these kinetics, by actually counting single antibody producing cells, the adult and fetal curves were superimposable. In other words, with a surprisingly morphologically immature and hypo- cellular spleen, the fetus could still bring to bear the same proportion of its total lymphoid cell population to make antibodies that the adult could, and just as rapidly. How this can be explained in terms of Dancis’ data, I don’t know. Dr. Eichenwald: In support to what Dr. Silverstein just said, back in 1960 or 1961 we reported, and this was not published but it is simply avail- able in abstract, the observations on toxoplasma antibodies in the fetus and up to, I think, 28 weeks of gestation, which was the earliest specimen we had, by fractionating the IGM and IGG antibodies by column technique. It was interesting that the fetus’ IGM dye test antibodies, which are neutralizing antibodies as far as one could determine, were in the same order of magni- tude as the mother’s. This is the protective antibody. These were the same or, sometimes, even higher than the mother’s. This says nothing about time of onset but all we know is that as early as 28 weeks the baby was immunologi- cally protected, by his own mechanism, as well as by transfer from the mother. Dr. Silverstein: Morphologically this seems to hold also, in that the study of autopsy cases of congenital syphilis indicates that the fetus has no trouble in making an abundant plasma cell response. We have not obtained serology in these retrospective cases, but morphologically they look as compe- tent as the adult. Some of the best plasmacytosis I have seen has been in the human fetus. Dr. Dancis: This may be related to the potency of the antigen stimulus, which is determined by the quantity and quality of the antigen. With a superb antigen, such as bacteriophage 6y 174, the premature, newborn and adult human all show a similar antibody response. (10) To demonstrate differences, it may be necessary to select less effective antigens. 29 Dr. Katz: Would Dr. Silverstein accept as further evidence of embryonic inability to respond immunologically to foreign antigens the reports by Drs. Alford and Tom Weller (10) on the rubella embryos ? Their immunoglobulin determinations on the sera obtained from intact 11 to 16 week old embryos with congenital rubella virus infection showed no detectable IgM. They con- cluded that a primary antibody production in utero apparently could not be initiated until sometime after the 16th gestational week. Dr. Silverstein: It is all that we have to go on. I think the data on the absence of positive serology may be just as important as the absence of positive cytology. Dr. McCracken: In dealing with an antigen, say bacteriophage in your fetal lambs where you know the timing of complete immunological com- petence, what happens if you introduce this antigen earlier, get nothing and then follow the fetus to birth? Dr. Silverstein: We don’t know the earliest age of onset of immunologic competence to anything in any species. In the fetal lamb we have seen re- sponses at 40 days gestation, in as immature an animal as one could believe, and yet, it made antibody. We do not know how much younger an animal can make antibody, because technically it has been impossible to get antigens into the fetus earlier than this. We had the feeling we might have to send them in on the sperm to really get the answer to this question. (Laughter.) In answer to the second question, it is interesting that if a non-replicating antigen is injected intravenously into an animal before competence has ap- peared, the antigen will be catabolized. If it is completely eliminated prior to the critical age of competence, the animal will never know it has been there, and will not give a booster response to a second dose. Now, with the replicating antigen we have two possibilities, especially if itis a “good” antigen and if it is present in adequate amounts. One possibility is that the antigen, put in prior to the achievement of competence, will induce specific immunologic tolerance. This has been seen with both the replicating and non-replicating antigens. On the other hand, the possibility exists that the fetus will not become toler- ant, for reasons unknown, and will start responding to the antigen when competence develops. We do not know precisely what governs the decision by the fetus either to become tolerant or to respond to the antigen. Dr. Dancis: I should like to return to a comment made by Dr. Katz. Did you suggest that attenuated viruses might develop in the course of a natural epidemic? Dr. Katz: No, that was not my implication, but it raises a valid point anyway. The population of viruses which are found in the course of an epi- demic of any viral infection is a heterogeneous one. When the epidemic is characterized by overt illness, the population is mainly a virulent one (that is, virulent for the human host). However, as with all organisms, there is a 30 degree of phenotypic variation and a definite amount of mutation and selection constantly in operation. The selective pressures exerted upon this “wild” virus population when it is put into the relatively artificial environment of cell cultures in vitro may favor the emergence of one of the naturally occurring mutants as the predominant type though it was initially the rare one. If this environmental pressure has selected a mutant which is less virulent for the original human host, then the investigator has found his attenuated virus. The attenuated poliovirus strains utilized in oral polio vaccine are fine examples of this selective mechanism. From the viewpoint of this conference, however, a crucial question remains unanswered: is the attenuated virus as benign for the embryo as for the mature host? It is in this area that the veteri- nary virologist has acquired experience with the attenuated hog cholera and blue tongue viruses which are used for vaccines. These attenuated strains appear harmless for the adult swine and sheep, but when administered to the pregnant hog or ewe they may produce an infection which does not destroy the fetus but permits survival with malformations. This is not a field I know well at all so that I must turn to you, Dr. Silver- stein, for elaboration of the problem. But it is certainly a fascinating one. Apparently you can induce with attenuated infection the type of embryonic effects which permit viability with damage, whereas the virulent virus is more apt to cause death so that none survive with anomalies for study. Dr. Benirschke: You have brought out very nicely that we have been remiss not to take into consideration some of the animal experiments known from veterinary medicine. It is for this reason that I asked Don Medearis last summer to summarize this for us. He has done this in our Hanover conference (11) and has done a very nice job in summarizing what has been achieved in the veterinary field and how it might possibly apply to human pathology, a least what we can learn from it and how we should march. I recommend you look at this later. I would like to ask Dr. Silverstein something that seems to me quite important and fundamental. In the blue tongue virus situation experimentally produced hydranencephaly is seen which, after all, is not too uncommon a malformation of fetal disease in man. Everyone of us pathol- ogists has autopsied two or three cases and in the pediatric literature pictures of transillumination of infants’ heads are published and one chides a resident who did not pick this up antemortem. But we do not know the etiology in man. Histologically, this thin membrane of remaining brain tissue shows usually an inflammatory reaction. Often hemosiderin-laden macrophages, plasma cells, and all kinds of evidence suggest it is a destructive lesion, but we have no way of adjudging what this is due to. Blue tongue virus, like many of the other viruses are around us. Some of the human cases may actually be due to blue tongue virus infection but has this ever been considered? If this is possible and if such viruses should be looked for, what should be done to identify whether some of these animal models affect man in a like manner or in a very unlike manner? 31 If Lon and John Sever go about studying rubella and herpes, they might not ever encounter any of the animal viruses like the cat virus and the rat virus. Many mothers have pets at home who very well may transmit these agents to the pregnant woman. Specifically with hydranencephaly, are there cases in the NINDB study? If so, have they ever been investigated? Should we not look for just this morphologic equivalent as a hint to search for new virus diseases in man of which we are not dreaming at present? How do we meaningfully investigate babies with hydranencephaly then? Dr. Eichenwald: How many of these veterinary diseases, aside from rabies and brucellosis or hound dog disease and such are known to be transmittable to humans? Has anyone studied veterinarians or sheepherders to find out if they have antibodies to blue tongue disease? Dr. Margolis: I plan to talk about this point a bit later. Dr. Frank: I have a question for Dr. Silverstein. We are talking about antibody response and looking for it in the fetal animal, especially in the presence of viremia. I was wondering if in the presence of viremia one could expect to find circulating free antibody that could be used in other viruses? Might not high gM be a lucky observation? Dr. Benirschke: Shiela puts a typical female touch into this: how many sheepherders are pregnant? Dr. Silverstein: This is an important point. The blue tongue attenuated vaccine makes it clear that a virus that is almost nonpathogenic in the adult may be highly pathogenic in the fetus. We should be aware that in several important respects the fetus is different from the adult. Thus there may be other viruses that we don’t normally identify with human disease that may be disease-producers in the fetus. Dr. Gruenwald : Coming back to the attenuated virus, about three months ago there was a scare in Briton that made the headlines in the papers: Women who had been immunized against polio were supposedly having malformed babies. I don’t know whether this has gotten across the Atlantic. I have since been informed by a “usually reliable” observer that this was entirely baseless. This was a small observation which probably was chance, and had somehow made its way into the press. On the other hand, I would cer- tainly think that if everybody is going to be immunized by attenuated virus, one would try to immunize people before child-bearing age. Dr. Katz: In general, recommendations for all live virus vaccines have included the proviso that pregnant women be excluded unless an epidemic situation exists or they are about to travel to an area where exposure is highly likely. There are documented cases, rare as they may be, of the spread of vac- cinia virus from mother to fetus resulting in intrauterine generalized vac- cinia and death. It is assumed that even this might be avoided by primary vaccination prior to the child-bearing age, since fetal infection resulted in 32 mothers who were the recipients of their first (i.e., primary) exposure to vac- cinia virus. With the attenuated polioviruses, viremia has been demonstrated in up to 40 percent of susceptibles following ingestion of vaccine. Whether such viremia could infect the placenta and subsequently the fetus is not known. There have been some newspaper reports of an increased incidence of con- genital malformations in the offspring of European mothers who were fed oral poliovirus vaccine during pregnancy, but I am unaware of any documen- tation of this in the medical literature. Such data are extremely difficult to ac- cumulate but the least rumor or suggestion provokes such heat that it be- comes impossible to find the light. Perhaps Dr. Miller can tell us of any epidemiological studies in this country or elsewhere. Dr. Miller: I don’t know of any such study. Dr. Gruenwald : The trouble is that when a woman is in the first month of pregnancy or perhaps even in the second month before she has gone to a doctor, no precautions can be taken. So I think all these warnings ought to concern the entire child-bearing age because one does not know when a woman has become pregnant. With polio vaccine we face the problem of a freely communicated infec- tion due to the prolonged excretion of vaccine virus by the susceptible recip- ient. Therefore, a pregnant mother may acquire the virus from her other infant or pre-schooler who sheds it in his stool for 4-6 weeks after his feed- ing. This is in marked contrast to an immunizing agent such as attenuated measles virus which is not excreted by the recipient and therefore is a “dead- end” infection. Dr. Benirschke: Is there anything in the data of the NINDB study? Dr. White: There is. The difficulty is that it is hard to retrieve because it is not yet coded in any coordinated way on cards or on tape. In looking through the records of 1,700 mothers I found several who had received polio vaccine during pregnancy. A large percentage had gotten inactivated flu vac- cine. If I am correct there was perhaps four years ago a U.S. Public Health Service recommendation that pregnant women receive flu vaccine. Dr. Katz: That is an inactivated antigen which may have carried along with it avian leucosis virus or something like that. Dr. Eichenwald: There is a study by MacArthur (124) which seems to indicate that perhaps smallpox vaccine may be teratogenic. Many women were vaccinated in England during a smallpox outbreak, and some were preg- nant at the time. An unusually high incidence of abnormal infants was noted among the off-spring of mothers vaccinated during the first 8 weeks of preg- nancy. This has not been observed in other studies. Another live vaccine widely used in adult human beings is yellow fever. There are rather good data which indicate that this material is not teratogenic. 33 Dr. Benirschke: If I am not mistaken in blue tongue there are antigenic strains of various severity in fetal infection. Dr. Margolis: Since the comment was made about viruses in the field becoming attenuated and producing malformations I have wanted to com- ment about hog cholera vaccination. It is known that cerebellar hypoplasia, congenital tremors and hypomyelinogenesis occur in sows vaccinated with modified live hog cholera virus (12) (13). Further, it now appears that abortion in cattle may be caused by both field outbreaks and by the use of live vaccines for infectious bovine rhinotracheitis. The involved strain, of very low virulence, produce minimal disease in adults (14). These are natural experiments of great significance. Dr. Axelrod: We should all be cognizant of the fact that it isn’t just the attenuated viruses that we are injecting. In the case of polio, of course, as a fait accompli we now carry about 100 Simian viruses through the rhesus monkey cells, some of which have already produced tumors in other species when injected. As many drugs bearing trace quantities of impurities may be effective causes of side effects and/or toxicity, we should realize that contaminant viruses from many sources of virus production might account for many of the difficulties we see. Dr. Benirschke: Peter, would you now proceed with our fetal story? 34 IV—Fetal Growth Retardation by Peter Gruenwald Dr. Gruenwald: I have been asked to talk about growth retardation which I like to do. I will give you a brief rundown on the problems in fetal growth retardation. I want to make it clear that much of what I will say does not particularly pertain to virus infection and I think at the end we ought to examine where the growth retardation in virus infections stands with regard to all this. Much of my information is unfortunately limited to the third trimester of human pregnancy. The number of births occurring at earlier gestational ages is too small to furnish reliable data. Also, it should be clear that what I will show in the place of growth curves are really birth weight curves but this is as close as we can come to information on fetal growth. Illustration (fig. 2) shows a group of these curves, gathered from many sources in the literature as well as my own data. Many sets of data, but not all, fell within the range marked on the right as “average”. Differences in birth weight at term are caused, according to my hypothesis, largely by the time at which fetal growth departs from the straight-line course shown in figure by a heavy line. This line, the extrapolated growth curve, indicates the growth potential which the fetus would follow if adequately supplied. The fetal supply line (maternal plus placental) does not keep up with the increasing requirements, and at one point becomes the limiting factor; shortly thereafter growth de- parts from the straight-line course. After adjusting to extrauterine life, the infant returns to this growth rate as was shown long ago by McKeown and Record (15). Compared with growth in body mass, qualitative maturation is much less affected. Illustrations (figs. 3 and 4) show two kidney sections of infants of the same birth weight, about 1,000 grams. The one on the left is from a nor- mally grown, pre-term infant and shows immature nephrons below the cap- sule as is normal at this age. The other infant was born at term, severely growth-retarded, and its kidney is mature as one would expect at that age. I cannot be sure that it is really up to 40-week standards and not to 38, because we do not have sufficiently detailed standards. Organ weights show characteristic patterns in fetal growth retardation, as is shown in the figure. Graph (fig. 5) expresses organ weights and body length of severely growth-retarded neonates (with a weight below mean minus 2 standard deviations) for their week of gestation as percent of those of normally grown pre-term infants (birth weight within one standard devi- ation from the mean). Most striking is the large brain, still too small for gestational age but much less growth-retarded than other organs. The thymus, on the other hand, is very small owing to involution before birth. Illustrations (figs. 6 and 7) show on the left a normal thymus, and on the right that of a growth-retarded neonate with small, lymphocyte-depleted lobules. I am not 35 4000 | / EXTRAPOLATED _ _——|SWEDISH = 3500 AVERAGE SMOKERS LOW SOCIO- ECONOMIC SMOKERS 3000 LOWER S-E. 2500 2000 1500 — x © w z ooo 28 32 40 44 WEEKS FROM LMP Figure 2 Birth weight curves of human population groups, indicating that all have similar growth characteristics early in the third trimester of pregnancy and then depart from the straight (extrapolated) line at different times depending on the adequacy of growth support from mother via placenta. After Gruenwald (Am. J. Obst. Gynec. 95: 1112, 1966). The C. V. Mosby Company, St. Louis, Missouri. implying that this involution has anything to do with the immunologic capacity of the infant, but this has never been explored. The literature abounds with references to experimentally malformed fetuses that are also growth-retarded, for example, after x-irradiation (16), azaserine treatment (17), or folic acid antagonists (18). I found the same in selenium- containing chicken eggs where severe malformations develop. Illustrations (figs. 8 and 9) show reconstructions of the outline of the body of a control 36 Figure 4 Kidney sections of infants weighing about 1,800 grams, (a) from a pre-term infant with a birth weight adequate for its gestational age showing immature nephrons under the capsule, and (b) from full-term, growth retarded infant showing in the same location mature nephrons approximately commensurate with gestational age. After Gruenwald, Dawkins and Hepner (Sainai Hosp. J. 11: 51-80, 1963) . 37 +20 +10 p= Liver Thymus 0 Brain Heart Body length =10 pom =20fp=— Figure 5 Organ weights and body length of growth retarded infants of low birth weight expressed nd cent of corresponding values for normally grown pre-term infants of similar body of 4 days and 5 hours (left) and a 5-day affected embryo (right; areas of necrosis are shown stippled), at the same magnification. Finally, with the help of a graph (fig. 10) we might explore the question where growth retardation due to viral infection in utero stands in relation to all this. On the left is shown teratogenesis in a very wide sense, including some forms of prenatal disease. This affects the embryo during the first half of pregnancy and makes it enter the phase of predominant growth in an abnormal state which has affected the internal clock of its growth, its growth potential. Shown on the right are those conditions which affect a fetus that was hitherto normal, and limit its growth late in gestation. This is not to say that these latter factors could not also reduce the future growth potential as, in fact, they often do. Where does virus disease as a cause of growth retardation fit into this? It does probably fit in on the left side of figure, not because malformations are produced, but because it is a disease of the young fetus. This fetus, in the face of normal supplies, could not grow properly. Now, this makes some difference in further development. In those on the left-hand side, the malformed ones, we 38 Figure 6 Sections of thymus of a normal newborn infant (a) and a severely growth retarded one (b) shown at the same magnification to indicate involution in the latter. may find more of a retardation of maturation, of differentiation than we find on the right. The example of the kidneys I showed you, would fit on the right side. I found in selenium effects on chick embryos a rather marked retardation in differentiation of the mesonephros at 5 days. It is possible then, if virus diseased fetuses belong in the same category, that they would be not only growth retarded, but also retarded in maturation. On the other hand, being ill, they might share with the other group, on the right, such damages as involution of the thymus. Being told that babies are small generally after such and such a condition, doesn’t help us and it was even worse in the times when all small infants were 39 Figure 7 called premature. The only way we can ascertain growth retardation, is to see whether the growth of the fetus has been adequate for its gestational age. Thank you. Dr. Eichenwald: May I present an off-the-cuff theory? Most of you are familiar with the fact that it has recently been demonstrated that an infant well nourished for the first 6 months or so of life, but then deprived of adequate nutrition for a period of three or four months, will fail to grow normally, even if optimal nutrition is offered. These individuals never gain normal height. They stay below whatever normal guidelines for growth you care to use, not just initially, but for the rest of their existence. 40 Figures 8 & 9 Reconstructions from serial sections of a normal chick embryo of 4 days and 5 hours (a) and one with the effects of selenium compounds (areas of necrosis shown stippled) after 5 days of incubation, showing the severe growth retardation of the latter which is not entirely accounted for by the malformations. After Gruenwald (Am. J. Path. 34: 77-103, 1958). 41 Environmental Genetic General Specific maternal N / deprivation condition TERATOGENESIS NON-SPECIFIC CHRONIC primary FETAL DISTRESS rative spurious / early late MALFORMATION GROWTH RETARDATION SPECIFIC DISEASE Figure 10 Diagram of the relationship of embryonic growth retardation to malformation on the one hand, and nonspecific or specific late fetal disease on the other hand. Is it possible that an analogous situation exists in utero, that any noxious stimulus applied to a fetus over a relatively short period of time will have a permanent growth effect on that fetus? In other words, if the mother has mumps or chicken-pox, or measles, or pyelonephritis, or gets typhoid vaccine during a critical stage of the fetus’ development, can this produce a permanent growth impairment in the fetus much as is seen in early childhood? Dr. Gruenwald: It is hard to answer. We both attended the same con- ference dealing with this in Boston and it is evident from data brought to- gether on that occasion that insults of the same severity and duration have more pronounced effects, and a greater probability of irreversible disturbance if they act early during growth and development. The fetus may enter the second half of pregnancy with a normal growth potential, but not come out of it with a normal growth potential. What I tried to differentiate, were factors inherent early in the fetus and unrelated to its nutrition on the one hand, and other factors imposed on the fetus later on by its nutrition. But, in the end, even a nutritional deprivation early in life will reduce the growth potential. What I tried to bring out, was that these diseased fetuses such as those with rubella probably enter the growth phase of late fetal life in a condition where they are not capable of growing normally, even given normal supplies. By the end of pregnancy the differences between these two groups may to some extent disappear; that is quite true. Dr. Benirschke: This is a pendulum that swings back and forth; whether or not there is any truth in this concept of the so-called “teratogenetic termi- nation period” of Schwalbe (19). If that one renoblast becomes defective, you won’t have a kidney, if that one embryoblast doesn’t make it, you don’t have an embryo, just a placenta. What we are saying is that if you are meant to start out with 50 muscle cells in the arm, and you kill 30 of them, you may have a small arm, and apply it to the whole embryo. 42 This brings us actually to Dr. Margolis’ presentation very shortly. Before this, however, let me ask Dr. Moorhead what he considers the validity of the recent debate in the literature namely, the possibility that one reason for the small-for-dates X O Turner syndrome is the elimination of the group of cells during the “Lyonization” event. Dr. Moorhead: I don’t know of any work on that, on cell elimination. Dr. Gruenwald: In any number of conditions of trisomy, Schutt (20) has compiled a whole list of them and shown their growth deficit in comparison with normal curves. Dr. Benirschke: What are the S periods and what are the cell cycles? Dr. Moorhead: If you are talking about in vitro studies, and trying to get some clue, there is no question that those not involving sex chromosome anomalies have less growth potential in vitro, but that is the autosomal cases, those with sex chromosome anomalies don’t seem to be restrained in this way. Dr. Benirschke: Concerning the length of mitotic cycle in trisomics— Dr. Moorhead: There have been only one or two studies on this. It is quite premature to say anything on that, but in terms of convenience for subculture, potential total growth in vitro, this is clear with the autosomal cases but not for sex chromosome cases. Dr. Eichenwald: Most of us assume that if any factor inhibits fetal growth without producing damage the fetus’ growth will catch up once this noxious factor is removed. Is this assumption correct? Dr. Gruenwald: No. Dr. Eichenwald : Even though the noxious stimulus is removed, once the fetus is inhibited from growing for, say one week, it will never catch up? Dr. Gruenwald: I don’t know about one week, but those that are severely growth retarded, have been behind for much more than one week and will, as a rule, not catch up. This has been shown by a number of people, that those born weighing less than a third percentile, will stay there in length and head circumference and weight for years, and probably forever, and this has also been done in animals particularly rats. Rats that were malnourished in utero, by feeding the mother half as much as she would like to have, would grow up and if you would see one sitting on the table, you would say it is a normal rat, but it weighs 20 or 30 percent less, so this is definite—they don’t catch up. I think this has been well established. Dr. Eichenwald: What we can then assume, correctly or incorrectly, is that a factor in the mother which does not directly invade the fetus, as perhaps is the case in mumps, may result in a small baby which will never grow normally. 43 293-596 0—68—4 Is it your observation that the small infant born from a mother with mumps during her pregnancy is due to the illness per se in the mother, rather than due to direct viral invasion of the fetus ? Dr. Benirschke: Of monozygous twins with disparate growth at birth, the small one will never quite catch up to the big one. These are normal individuals, who just suffered a circulatory imbalance, as it were. Dr. Gruenwald: Genetically identical. Dr. Benirschke: That is as good an example, as you can find. Naeye’s (21) study in Pediatrics is quite good on this. Dr. Margolis: Harlow’s (22) study on the maternally deprived infant, shows how tremendously labile the brain of the infant is. An animal so deprived will often develop bizarre autistic behavior, as do institutionalized and socially deprived children. Dr. Axelrod: Dr. Eichenwald’s commentaries touch on an almost absurdly complex situation, and I think he is plagiarizing Aldous Huxley from Brave New World in which, as you all remember Dr. Eichenwald: Unconsciously, but you are right. Dr. Axelrod: The test tube babies’ future position in society is deter- mined by the amount of oxygen and nutrition fed the developing fetus; so Mr. Huxley seemed to have some insight into this nutritional problem. Now, the terminology used here is not terribly familiar to me. But, mathe- matically, of course, the production of enzymes through transcriptional and translational mechanisms become an exponential of the moment in which they exist. It is conceivable that thousands of enzymes, which exist prenatally, make their appearance never again to appear as part of the metabolic scheme. An exponential of reduction, mathematically, as some theoretical biologists have put forth would be reflected in the subsequent generations qualitatively and quantitatively intracellularly. Quantitatively, as to the number of cells at any one time making appearance in the tissues, and qualitatively as to the metabolic schemes which result in the mathematical exponential. The retardation of any process is a function of the numbers of molecules active in any one metabolic state due to the transcriptional mechanism and reflects later on, very appropriately, in the size of animal, even though metabolically the organism is considered normal. Dr. Gruenwald: This depends a lot at the time at which this happens. We know DNA replication stops in some parts of the body shortly before birth, and in some a little after birth. This is a critical period. This is one very basic thing, of course. Does this happen? If it happens at the time before DNA multiplication stopped, and now it just doesn’t get anywhere, it stops anyway— 44 Dr. Axelrod: DNA never stops. Dr. Gruenwald: By feeding the animal afterward, you won’t get any way. Dr. Axelrod: DNA never stops. Only the multiplication of the cells may stop, but there are replacement systems. Dr. Gruenwald: Beyond replacement levels. Dr. Axelrod: In order to get the gross manifestation, you have to go earlier and earlier into the system, so that the final reflection, which is the total organism, gestalt organism, is only a reflection of the earlier retardation offered in a production, so the earlier you retard DNA, the more gross will be the manifestation later on. Dr. Mitchell: I want to ask a totally different kind of question. Sam mentioned earlier in his presentation on viral mechanism that some viruses kill cells. I wonder, once fetal cells are killed, may they not be completely resorbed thus mimicking agenesis? Dr. Gruenwald: One of the reasons why the teratogenic termination period of Schwalbe is not applicable in many cases, is that many malforma- tions do not develop by primary deviation of a developmental process, but rather by secondary degenerative changes. In normal development, cell death plays an important role, and many of these teratogenic processes are exag- gerations of this normal degeneration. In many malformations primary deviation of development cannot be the cause, in others it may or may not be, and in some it must be the cause. In some hereditary malformations necrosis occurs suddenly, at a predetermined time, as if regulated by a timeclock. The products of degeneration are cleaned up, there is no scarring in the conven- tional sense, and one cannot see any traces later on that would permit one to tell how the malformation developed. Dr. Benirschke: But fetal deaths occur. There is no study other than Boue’s (5) of attempts at virus isolation from a larger number of embryos. We will have to come back to this when we discuss the H-1 virus from human abortions. George, will you continue the presentations? 45 V—Fetal Pathology Due to a Group of DNA Viruses by George Margolis Dr. Margolis: I have prepared a statement which I shall read essentially verbatim to achieve an economy of time. Last spring, with Lawrence Kilham and a medical student (23), I prepared a statement of similar nature for the Dartmouth Conference on Reproductive Failure. I thought the opening theme was so much on target that, with Kurt’s permission, I shall use it again here, with minor changes: If one were to undertake to construct a virus designed to induce reproduc- tive failure, no better model could be conceived than that of an agent which has the ability to enter and to destroy mitotic cells. Such an agent would, theoretically, be infectious for the fetus throughout gestation. An attack upon sites of cell proliferation at critical stages of organogenesis could result in malformations incompatible with life, or induce severe developmental de- fects. Considering that critical phases of growth and differentiation of various organs occur sequentially throughout fetal development infections even in late gestation or in the perinatal period would be capable of producing dis- abling or fatal sequelae. In a relatively stable adult cell population only mild infection would be expected. Hence, a pregnant host could harbor an inap- parent or latent infection with an agent, which would be highly virulent or lethal for her progeny. Significantly, this hypothetical agent is not merely the stuff of science fic- tion. A group of small DNA-containing viruses with a selective affinity for the mitotic cell has recently been defined (24) (25) (26). This group in- cludes the Kilham rat virus, the H-1 virus of Toolan, the virus of feline panleucopenia and possibly the minute of mouse. All, except the latter, are viruses long known to virologists. These agents are infectious over such a broad range of animal hosts and, in some instances, cross species barriers so readily that a sweeping conjecture may be put forth (if for no other reason than to stimulate work in this field), namely: each animal species harbors such a virus or else may be susceptible to an agent of this class existing in nature or being nurtured in laboratories. The demonstrated host range already includes rodents such as the mouse, rat, Mastomys and hamster, and felidae such as the cat, ferret, leopard and mink. And, there are provocative provisional studies on primates (27). The pathologic action of these agents, characterized by formation of intra- nuclear inclusion bodies containing myriads of virus particles, and by death and lysis of affected cells, commonly proceeds in the absence of an inflamma- tory response. This holds in both pre- and postnatal infections. This picture illustrates (fig. 11) the typical action of this virus focused upon the developing cerebellum. The external germinal layer, which is a major contributor to the later stages of cerebellar histogenesis, is the site of 46 Figure 11 numerous intranuclear inclusion bodies, most of which are surrounded by a prominent halo. At a slightly later phase of this infection (fig. 12) the parasitized cells will be observed undergoing pyknosis and fragmentation. Figure 12 Hy, fe i A vi 7 Eas] * : 24 4 Figure 14 48 The spectrum of infections induced by these agents ranges from generalized tissue involvement, affecting essentially every site of high proliferative ac- tivity, to a selective attack on certain target organs (28) (29). The fulmi- nant fatal disease may have either an intrauterine or perinatal onset. Non- lethal infections occurring late in gestation or in the perinatal period have induced an arrest in cerebellar and renal development, and a severe hepatitis (30) (31). At this conference, sponsored in part by the National Heart Institute, it is pertinent to record that, in generalized infections, prominent cytopathic effects occur upon the myocardium, endocardium, and vascular smooth muscle. Further, in viable nurselings, significant myocardial involve- ment and arrest of renal development have been encountered. I shall now illustrate the myocardial lesions (fig. 13). Intranuclear inclusions are seen in several of the cardiac muscle fibers. Some have a prominent halo about the inclusion; others are a dense amorphous mass completely filling the nucleus. Infection of neonates and sucklings has produced also a “mongoloid” state characterized by stunting of growth and tooth dysplasia (29). The severity of the hepatitis is perhaps best illustrated by its sequelae in surviving animals. As this slide shows (fig. 14), the hepatitis may result in striking nodular hyperplasia and stromal collapse simulating cirrhosis. Dr. Katz: Is this a mouse? Dr. Margolis: This is a rat infected with H-1 or rat virus. Actually, I am demonstrating prototype lesions, found in one of several animals, in- fected with one of several viruses for a special class. Adult differentiated tissue appears receptive only when responding to a strong stimulus for proliferation, as exemplified by the attack of rat virus upon bone at the site of healing fractures (32) and by the action of H-1 upon the regenerating liver following subtotal hepatectomy (33), and administra- tion of hepatotoxic doses of carbon tetrachloride (34). It appears that viru- lent strains pass readily through the placenta and infect the fetus without in- juring this organ. Although H-1 and rat virus have been isolated from human embryos, no causal relationship has been established as yet between these agents and abnormal outcome of pregnancy in man (35). The selective affinity of these agents for the cerebellum is indeed remark- able. Commonly, they may destroy the developing external germinal layer of this organ in the absence of overt action upon other tissues, leaving the surviving animal with a severe intractable ataxia. Indeed, the selective af- finity of these viruses for the developing cerebellum has been the one attribute of these agents which has allowed us to formulate a new concept of virus-cell- host interaction, and to identify the various members of this unique group. At first consideration it may seem strange that the attack of a group of viruses upon the cerebellum of the neonatal host could lead to the recognition of the affinity of these agents for the mitotic cell. The cerebellum, however, is the last portion of the central nervous system to complete its cytoarchitectonic 49 development, undergoing its major growth and differentiation postnatally. In contrast to the early development of the Purkinje cells and neurones of the deep cerebellar nuclei, which undergo their final divisions midway in gesta- tion, in animals, the external germinal layer initiates a period of remarkable proliferative activity late in gestation or at birth, and coninues to divide rap- idly and continuously postnatally, until it becomes the richest in DNA of all brain issues. In essence, this postnatal growth drive lasts longer than the whole period of intrauterine gestation. An H*thymidine autoradiograph of the developing cerebellum demon- strates these features. In this 4-day-old hamster (fig. 15), sacrificed 24 hours after intracerebral inoculation of H’-TdR, the heavy label of the external germinal layer underscores the high mitotic activity of this zone. Figure 15 Figure 16 illustrates the end phase of the viral attack upon the cerebellum. Earlier, we saw how the viral attack will destroy the external germinal layer. An attack in late gestation will result in an animal being born without a granular layer of the cerebellum. This photomicrograph is from a ferret in- fected postnatally. Because the posterior folia of the cerebellum happen to be spared, a striking contrast is offered between this zone and the profoundly hypoplastic cerebellum resulting from the viral disease. Dr. Moorhead : What virus? Dr. Margolis: This lesion was produced by a virus I shall discuss in a moment. Actually, this prototype lesion may be induced by any one of a number of viruses, acting upon a susceptible host. 50 Figure 16 Interestingly, the action of the agent illustrated in figure 16, resulting in congenital cerebellar hypoplasia, masqueraded for over 75 years as a genet ically determined developmental defect, despite the fact that the virus is one of the most common infectious agents among the cat population, and its manifestation, spontaneous feline ataxia, is perhaps the most common severe neurologic disease in cats, described as long ago as 1888 (36). Here is the cerebellum of a cat with ataxia of spontaneous origin (fig. 17). The outstand- ing feature is the hypoplasia, approximating an aplasia, of the granular layer. Again, a zone of relative sparing provides a contrast of the near-normal with pathological. In 1966, however, it was discovered that the cause of this disease entity is the feline panleucopenia virus, an agent which produces a totally different disease picture (37) (38) (39) in the older cat. Here it is manifested as panleucopenia and enteritis. Another remarkable feature of these infections is the long persistence of virus in either a latent state, or acting as a “slow virus” (24). Spontaneously ataxic kittens have yielded infectious virus as late as two months postnatally. It is possible to characterize the virus-host-cell interactions operating in these infections, and to explain their diverse manifestations on a unitary basis by considering these agents as mitolytic viruses, or as agents which attack preferentially cells with active DNA synthesis attending mitosis. No Sl Figure 17 other hypothesis fits the observations. Age dependent variations in cell pop- ulation kinetics, host-dependent variations in tissue susceptibilities, and variations in virulence, dosage, and time of introduction of the infectious agent represent major determinants of the resultant disease picture. Much remains to be explored about these relationships. Further search should con- tinue for other agents of this class and for extended range of host suscepti- bility. The implications for man are provocative, since there is a variant of human cerebellar disease which clinically and pathologically is analogous to the animal disease (24). One may ask, in fact, would not available data favor an infectious etiology for the human analogue? Dr. Eichenwald: What is the human analogue? Dr. Margolis: Cerebellar ataxia, characterized clinically by onset early in life or infancy, associated with some degree of mental retardation and pathologically by absence of, or hypoplasia, of the granular layer. Dr. Eichenwald: Does this bear any resemblance pathologically Dr. Margolis: Kuru has such a tremendous glial response but in some of our older infected animals there is a lesion which I have not mentioned 52 heretofore, but which resembles more the scrapie lesions in sheep, with its huge vacuolated cells. Dr. Moorhead: What about cells other than these? Dr. Margolis: These viruses are not neurotropes, but rather pantropes. You will recall the kidneys with retarded development shown by Dr. Gruen- wald. Here is the same phenomenon induced by the virus infection (fig. 18). First is a photograph of the nephrogenic zone of the neonatal kidney in the normal cat. Note the progressive maturation of the glomeruli in the older, deeper area of the developing renal cortex. In contrast, this is the cortex of the kidney of a kitten infected in utero with one of this group of viruses (fig. 19). Not the arrest of nephrogenesis, the fibrosis of the subcapsular zone and the maturity of the single glomerulus shown. Panleucopenia is probably seen in the adult more commonly than in the neonate simply because in the newborn animal the bone marrow is quiescent rather than responding to a stimulus for mitotic activity (31). As for the gastrointestinal lesions, the relationship to mitotic activity is possibly not quite so easily ascertained. In the adult the attack upon the highly proliferative mucosa correlates very nicely, but the rarity of this involvement in the neonate is puzzling. In severe, widespread perinatal infections, however, multiple tissue elements in the gut are affected, including mucosa, connective tissue of the lamina propria, and the smooth muscle wall. With infections earlier in utero the muscularis and connective tissue may be heavily involved and the mucosa unaffected. Parenthetically, the cerebellum may not be involved in these early fetal infections. Another point of interest: when I presented this material at the Yale Uni- versity a year ago, I learned that investigators there (40) had been using neonatal animals in a screening test for toxic action of the antimetabolite, cytosine arabinoside. Again, they chose hamsters, and their hamsters had lesions in the cerebel- lum identical to ours. These parallel findings bring to mind a provocatively titled paper by Hicks and D’Amato (41) a few years back, “How to Design and Build Abnormal Brains Using Radiation During Development”. Further you may remember that one of the outstanding clinical toxic features of cytosine arabinoside treatment is the loss of hair. Here is another parallel between the action of virus and of antimetabolite. This is a cross section of hair follicle of an animal infected in utero (fig. 20). Note the many inclusion bodies in the germinal epithelium. Dr. Benirschke: These viruses are everywhere. Panleucopenia is one of the major problems of zoos with tigers, leopards and lions, where the veterinarians vaccinate against panleucopenia. The virus is widespread in a variety of animals, easily transmitted to all animals tackled so far. 53 Figure 18 54 Figure 19 55 Figure 20 What is interesting to me is that the veterinarians have known panleucopenia and yet, the same virus causes cerebellar ataxia. This disease was ascribed to a genetic disorder, perhaps, because it is usually a litter of cats that becomes ataxic, and actually, it turns out to be disease manifestation of a virus infection. I think what is so important is that excellent veterinarians attempted up to a few years ago to breed cats with this disorder in hopes of unraveling the genetics of the disease. There are possibly many human analogues which we are overlooking because we haven’t come around to think about this kind of a concept. I think it is important to consider the possibility of using this next in a variety of other species, say the baboon. Dr. Gruenwald: The moment of birth, of course, is very important. As was pointed out in the British Medical Bulletin last year by Davison and Dobbing (42), the degree of maturity at birth varies considerably be- tween species. This becomes very important in work like this. There are certain things you can do in a rodent after birth that would correspond to a moment in life of the human or the baboon or some other species long before birth. 56 While we are at the moment of birth, I wonder whether somebody would like to say something about the possibility of intranatal virus infection. This has been assumed without any evidence to happen in Herpes simplex, to which babies succumb about two weeks after birth. It was thought that this happened at the time of passage through the birth canal. So, there are two problems having to do with the moment of birth: virus infection at that time and, on the other hand, the tremendous difference between species in maturity at birth. Dr. Benirschke: I think the moment of birth is probably important in Coxsackie B virus infections. In South Africa they had to close the new hospital when an epidemic occurred, then the disease stopped, because there was no longer nursery contamination. Rooming-in actually solved the prob- lem eventually. (43) (44) Dr. Miller: If I remember correctly, 15 years ago, the feline panleuco- penia agent was used to determine if it had any therapeutic effect in child- hood leukemia. (45) I wonder what happened to the nurses or families of the patients, if among them there were any pregnant women? Dr. Margolis: That is interesting, because in a much earlier paper, a tangential allusion was made to relationship between this virus and mitotic cells, but was never followed up (46). Where was this agent used against leukemia and why? Certainly, at that time, the affinity of this virus for tissues with high mitotic activity wasn’t recognized. I presume it was just because of its attack upon white cells? Dr. Miller: I thought so. Dr. Margolis: I don’t know. One wonders whether pregnant hosts, in whose intestinal tract and bone marrow, there is a high rate of cell turnover, wouldn’t be harboring the virus. Dr. Katz: I know nothing about human studies with feline panleucopenia. I was unaware of this. Dr. Mitchell: Is there any evidence, in cats, of vertical transmission of this virus? Dr. Margolis: In some instances we have observed that cats have given litters in which one fetus of four is affected, suggesting a recessive gene action, but sometimes all four will be diseased. An affected litter might in- clude a tiger, black, white, calico, or any mixture, but all are ataxic. Dr. Mitchell: Does a mother, ataxic due to virus, pass that virus on when she breeds? Dr. Margolis: This hasn’t been demonstrated. 57 There is one reference, from Belgium, which describes an acute febrile disease in a normal pregnant cat, who then gave birth to kittens who were ataxic (47). Jones (48) bred spontaneously ataxic males to spontaneously ataxic females, with negative results. Similarly, our ataxic hamsters have become pregnant, but their offspring have been normal. Dr. White: What about subsequent litters in the same mother? Dr. Margolis: Are you referring to the ataxic animal? Dr. White: Yes. Dr. Margolis: This hasn’t been followed very well in the field; but observations indicate that it is a one-time event. I know of no instance of a cat producing two affected litters. As to laboratory infections, we have wor- ried about this; occasionally we do antibody studies on some of the labo- ratory personnel. One of the men who handles the virus most has not shown rat virus or H-1 antibodies. It might be worthwhile to comment about whether these viruses are human viruses or really animal viruses. Dr. Kilham has gone to the field and extensively studied this matter in trash and garbage disposal areas. He has found most of the rat population will have either H-1 or rat virus antibodies; sometimes on one side of the Connecticut river, one rat population may have antibodies for one virus rather than the other, and vice-versa. These are really animal viruses, not human agents. Kilham (29) has probably isolated more strains of H-1 virus than anyone else; these are all from animal sources. Dr. Eichenwald: I am not quite clear about one thing. Do humans become infected with these agents? In other words, if you screen a human population without laboratory contact with these agents, are antibodies to any of these viruses demonstrable in a significant proportion of people? Dr. Margolis: Dr. Sever has made a few studies and has concluded that there is yet no evidence for the teratogenic action of these agents in man (35). Dr. Gruenwald: There must be plenty of slum dwellers who have plenty of contact with rats. Dr. White: Someone reported acquisition of H-1 or H-3 antibody in a series of mothers, have they not? Dr. Margolis: Toolan on more than one occasion (49) has isolated this virus from tumors—human tumors—and this is where the confusion arises. But has it crept out of the laboratory walls as a contaminant, or does it exist in tissue culture cells as a latent agent ? For example, Kilham has isolated one virus from the placenta of a case of mumps. I don’t recall what the fetus—— Dr. Benirschke: It was a spontaneous abortion. 58 Dr. Margolis: Investigating the relationship between mumps virus and the abortion, he isolated a virus he called HHP (Hodge’s Human Placenta) named after the patient. He eventually concluded that the virus was of labora- tory rather than human origin. This one example illustrates the difficulties encountered in working with this agent—just how many of the isolations are laboratory artifacts? Dr. Benirschke: In the adult ataxic cat who has acquired the disease at birth, can you find antibodies? Dr. Margolis: We have not been able to find infectious virus beyond about two months. We haven’t done a large number of antibody studies, but spontaneously ataxic cats have high titers. Pooled or individual sera from ataxic cats, used in neutralization tests, will effectively prevent infection in our experimental laboratory model. Dr. Silverstein: In answer to an earlier question about whether the passage of a virus from mother to offspring can confound the geneticist, I am not aware of any case in animals in which this is true, but there is a lovely example of this in plants. I recommend very much an article by René Dubos, called Tulipomania, which appeared years ago in several publications. It is a fascinating biologic story about a virus which infects tulip plants and results in a pleasing “pathology,” the beautifully multicolored tulip, due to alterations of pigment production. When they tried to breed these from bulbs, they reproduced this beautiful tulip; but when they tried it from seed, they obtained only plain tulips because the virus couldn’t enter the seed. It confounded the breeders for a long while and is an interesting model. Dr. Gruenwald: Transmission by mother’s milk is well known in mouse mammary tumors. Dr. Katz: In the murine leukemia and polyoma systems, as well as the avian leukosis and sarcoma infections, there are excellent examples of verti- cal transmission from mother to embryo. This becomes complicated epide- miologically, however, as both the mother and her progeny excrete virus so that they are the source of horizontal transmission also. The congenitally in- fected chicks are immunologically tolerant whereas infection in mature contacts produces an immune response. If we are searching for human coun- terparts, it becomes apparent how complex a system one may encounter if tolerance ever occurs in human embryonic infection. Dr. Benirschke: Are newborns with congenital infections runted? Dr. Margolis: In our studies? With congenital infections, yes, and also with postnatal infections they may occasionally be very severely runted. This is commonly associated with profound hypoplasia or destruction of the thy- mus. Figure 21 illustrates the striking viral attack upon the thymus which is sometimes observed. Not only is the gland atrophic, but the presence of 59 293-596 0—68—5 3 1 - » Ye. - « rs ® - Pa - - ya - EE) y EE) bd Figure 21 numerous intranuclear inclusions in the remaining thymic epithelial cells demonstrates that this “hypoplasia” is a direct result of the viral attack. The implications of this process in relationship to the runting syndrome are, I think, quite significant. Dr. Silverstein: We did not find marked runting in our blue tongue study. Dr. Benirschke: Good runting was observed by Ferm with intrauterine herpes infections. If he injects the hamster in utero the virus does not pass the placenta and the fetuses are significantly growth-retarded. Dr. Eichenwald: This is true in humans also. The statement is usually made that congenital human infections result in a high incidence of prematures. When we reexamined our own data on Coxsackie cytomegalic inclusion disease, toxoplasmosis and hepatitis, we found a high incidence of small babies, but they were small for their degree of maturity, not really premature. Dr. White: Dr. Fabiyi in our laboratory has made some interesting observations with rubella in the ferret which have subsequently been con- firmed by Dr. Schiff in Cincinnati. The ferret given rubella during the neo- natal period has a duration of infection which is much longer than that of the adult ferret. Virus can be recovered occasionally for at least two months. These animals grow more slowly than uninfected ferrets. In addition, a few seem to have scoliosis. 60 Dr. Benirschke: There are many causes of growth retardation. I am just thinking of the baby with congenital heart disease who perhaps, because of the poor peripheral perfusion or because of virus disease may be retarded in growth. Would someone make a comment concerning Brown and Evans’ (50) study of attempting to correlate antibody titers in children, with congenital heart disease? They describe significant elevations of Coxsackie virus anti- bodies. Can a causal relationship be inferred, since no virus, as I understand it, can be recovered from these infants? Dr. Margolis: Could T just get this in before we lose this other subject? Sometimes, in the infected animals, animals who will be totally disabled be- cause they have no cerebellum are not runted. They grow up to be fat, lazy animals. Dr. Benirschke: But, they are infected as newborns? Dr. Margolis: Yes, but other neonatally infected animals have the same cerebral disability and are runted. Dr. Benirschke: You have never really looked at the marrow or their blood counts and we don’t really know whether they are leucopenic. There may be a marrow component. Dr. Margolis: We have studied the marrow to a limited extent only. It is variably involved, often essentially spared. Dr. Katz: Before leaving this topic, Dr. Margolis, I wonder if you know the reasons for so striking an affinity for the virus and the mitotic cell? Has any work been done to approach this experimentally? Dr. Margolis: We have been asking ourselves the same question. To pursue the research further we must now concentrate heavily in such ap- proaches as fluorescent antibody procedures and autoradiography, and the use of tissue culture as well as in vivo studies. Having elaborated on a work- ing hypothesis, we are only now in a position to test it more rigorously. Al- though we are dealing with DNA viruses, which replicate in the nucleus, these agents have a fundamentally different relationship to the cell than such agents as herpes virus, which attack stable population of cells. Dr. Katz: Will these viruses in vitro infect cells which are not in active mitosis? Is there some difference in cell membrane, receptor sites for virus or some other factor in resting cells and mitosing ones? Does the virus fail to penetrate the resting cell or does it penetrate only to find a host cell in- capable of acting upon its genetic message? Dr. Moorhead: According to unpublished work of Girardi with the SV-40, it seems to be simply that cells in division in culture somehow don’t permit maturation of virus, a kind of competition. So you get a different 61 effect with SV—40 virus; but except for this observation, I can’t imagine what the key difference might be. Dr. Margolis: Johnson, in Bristol, England, has done the same studies as we have on tissue cultures, and interestingly enough, he came to the same conclusions we did, on the basis of finding a very transient cytopathic stage; old, non-proliferating cell cultures were almost completely resistant to virus (25) (38). We have been in correspondence with him just about this point. We intend now to study this virus-cell interaction in synchronized cell popu- lations. We hope, thereby, to obtain useful data in such a model. Dr. Benirschke: Have you any suggestions how to approach this? Dr. Katz: It is relatively easy to suggest, but perhaps not so easy to investigate. Your reference to the Coxsackie viruses brought to mind the studies with them and other enteroviruses which suggested that the immature cell has receptor sites which permit enterovirus attachment and infection while the more mature cell has lost these sites and is therefore not susceptible to the agents. In this respect, the actively mitosing cells may be regarded as immature and the resting cell may represent the mature one. Were this cor- rect, it shouldn’t be too difficult to determine whether with the former you got virus attachment and not with the latter. Dr. Margolis: Another related question is whether the virus susceptible regenerating hepatocyte is a mature cell or is it again an immature cell ? Dr. Gruenwald: Isn’t it true that most of the mitotic poisons we know of interfere with the orderly process of mitosis, which seems to be a vulnerable process, and produce chromosomal aberrations which seem to be lethal for some cells? On the other hand, it has been suggested that following rubella infection, chromosomal abnormalities do occur. Could it be an interference with the process of mitosis which, after all, is an extremely complicated process that must be carried out with considerable diligence in order to lead to viable daughter cells? Does this apply to virus infections? I don’t know. Dr. Moorhead: That is rather complex. I will touch on all of that in 20 minutes, I hope. Dr. Margolis: Another observation comes out of our studies, related to the presence of twin inclusion bodies in hepatocyte nuclei. In periodic studies of such cells it is possible to reconstruct a cycle of parallel progressive matura- tion of these inclusions, followed by cell death and lysis. These observations make me wonder if these cells have been infected before they divide, i.e., if parasitized cells will divide normally, then undergo cytopathic changes. We plan to do this now by taking time lapse movies on synchronized tissue cul- tures to study this relationship. The observations of Sandberg, Sofini, Takagi 62 and Moore (51) that the nuclei of binucleated cells replicated DNA at differ- ent periods, offers an inviting approach to this problem. Dr. Benirschke: I believe that the question I asked in relation to the studies by Brown and Evans is directly related to this problem. These cats have antibodies and you can not recover the virus but they have a specific lesion. If there is anything to this concept, which certainly holds for rubella, namely that congenital heart disease may be caused by a virus in utero, and if, say congenital lesions could be caused by other viruses in utero then you might not find viruses at birth. If I am not mistaken, that is what Brown and Evans addressed themselves to. They endeavored to find out whether some babies with congenital heart and other malformations, had a higher than expected rise in antibodies at birth or later. This seems to me a worthwhile approach. Is it reasonable? They suggested Coxsackie viruses. Dr. White: Dr. Mitchell has data relating to that. Dr. Mitchell: We do have some data on this but I have a morphogenic concern with the Evans and Brown approach. Many viruses are cell specific. Many different types of cells form the completed human heart. Yet in their study Evans and Brown consider all congenital heart defects as one entity. Dr. Benirschke: Then let me ask: can you have a variety of congenital heart disease in the well-established rubella syndrome? Dr. Mitchell: I am not sure there is a simple answer to that. Patent ductus arteriosus and peripheral pulmonic stenosis are the most common, the most characteristic lesions. Indeed, they may be the only two causally related to rubella but this has yet to be established. Dr. Benirschke: There are some with patent ductus. Some of whom are deaf, others blind, but not all of them have exactly the same other complica- tions, or do they? Dr. Mitchell: Well, not all, but a great number of the rubella babies you can diagnose in the crib. Dr. Benirschke: But do they all have the same manifestations? Dr. Mitchell: No, but they usually have more than one lesion. Dr. Benirschke: It seems to me that we don’t really know if there are five variants going across the placenta to do a lot of damage to many systems, or one variant goes across which happens to land in the ductus. I don’t think we can rule out this possibility of quantity which was brought up before, even in congenital rubella infection, which is the only one which we know any- thing about. 63 Dr. Mitchell: I don’t think we have a documented rubella baby with just one lesion. Dr. White: Deafness. Dr. Mitchell: Just deafness? Not small baby? Not microcephaly? Dr. White: I see it all the time. Dr. Margolis: I don’t know whether this is a too simplified interpretation which might help you in your thinking, but first of all, the lesions we are talk- ing about probably haven't been seen in man nor has this virus-mitotic cell relationship been described in known viruses of man. Rubella represents a totally different virus-cell interaction if we can believe the literature. Here there is an inhibition of mitosis, or cell turnover (21) (5) rather than the destruction of dividing cells. It would appear to me that such a system would be far more teratogenic than fetocidal. Dr. Axelrod: This afternoon when I speak I will bring up drugs and the effect on the fetuses, and although drugs are not viruses, they are agents which attack certain developmental systems at particular times. I don’t think we should be confused concerning whether the final realization of an infection or a drug gives one, two, three or more symptoms, or one exclusively. I think this is due to the timing of the attack of the drug or the virus, so that the drug or virus or other infection which attacks a system of organogenesis or histogenesis at any one time will manifest certain final symptoms. As development progresses the deficit will manifest other metabolic dys- crasias so the timing of the initial defect and the system which is developing at that time, will determine what one sees ultimately. Dr. Eichenwald: To extend this a little further, I think one also has to consider the individual variation among humans. I am sure there are enor- mous differences in susceptibility to infection even in the fetus, depending, in part, on the genetic background of this individual. Not every mother with rubella bears an infected fetus. Even though two mothers were infected at the same time, the disease in the two fetuses may be quite different. Since we know there are individual differences in susceptibility in later life, why should this not exist in the fetus? Dr. Silverstein: It interests me that we are getting around to the fact that disease is a complicated business. It is obviously important to know when the pathogen gets in, where it goes, what is the state of the fetus, what the fetus can do, and what the organism can do. It is obvious that not all syphilitic fetuses have Hutchinson’s teeth, and not all have interstitial fibrosis of the liver, perhaps depending upon where the treponemes go. But to further com- plicate the matter, it is not merely adequate to have a susceptible fetus and susceptible tissue at an appropriate age, with the pathogen there, to get a lesion. In our study of congenital syphilis in the monkey fetus, for instance, we may find treponemes in both the kidney and the liver, but the liver has 64 lesions and the kidney has not, or we find the organism in the lung and the liver, with lung but no liver reaction. Again, there are situations where we can sample the lung and show that organisms are there, teeming, in all of the lobes of the lung, but one lobe may present a syphilitic lesion, and another lobe not. Dr. Gruenwald: Maybe I could paraphrase what you have suggested. I think what you really meant to say is that people do not see the trees for the woods. I think this would probably be the better way of putting it. Dr. Silverstein: There are others who don’t see the woods for the trees. (Laughter. ) Dr. Katz: I have not read the report by Brown and Evans which you cited. At what age were these serum samples obtained that showed a higher incidence? Dr. Benirschke: The sera were obtained at the first prepartum visit in the 2d or 3d month of pregnancy and again at delivery. However, just because we can’t understand it, I think we ought to keep an open mind. This is what went on with cerebellar ataxia in cats; “It’s genetic, period.” And actually, it was not. We should keep an open mind, particularly since Coxsackie virus is ubiquitous. In Ireland, I believe, it was suggested to women not to have babies during a certain period of the year because anencephaly was very much more frequent during those periods of the year, when embryogenesis proceeds. There was no other good reason than just such practical approach, no specific knowledge of the villain. An- encephaly is a destructive disease. We don’t know the etiology. We can pro- duce it in the animal quite readily by many teratogens and it is possible that it is due to infection. There are many possibilities which need to be looked into, and it is a very pragmatic approach, to tell women to conceive say in November. If we have a suggestive relationship to Coxsackie antibodies, so far as congenital heart disease is concerned, then I think it ought to be investigated. Dr. Katz: Once again, I find myself in the position of pleading the cause of the epidemiologist. In the interpretation of any data based solely on the presence of serum antibody it becomes of foremost importance to consider the total preceding experience of the individuals whose sera are evaluated. Except in the most unusual instances, this is not possible. Accepting the futility of that first step in retrospective deduction, how then does one as- semble a proper set of controls? Allow me one anecdotal example. In 1955, the last year in which we had a huge epidemic of Poliomyelitis in Massachusetts, there were several infants delivered on our wards from mothers who had been hospitalized with para- lytic disease. One such newborn, known to be excreting type I poliovirus, was admitted to our nursery unit after the personnel had been carefully in- structed as to the precautions to be followed. Nevertheless, within less than two 65 weeks, every other infant in that nursery was shedding in his feces the same virus as the initial case. The same rapid spread has been reported with Cox- sackie viruses in nurseries. A sero-survey of these babies at any later date would in no way indicate whether their antibodies had been acquired as a result of intrauterine, perinatal or later infection. Retrospective antibody screening may provide clues but it cannot es- tablish etiology. This is an obvious reason, one of many, for the establishment of prospective studies such as the perinatal collaborative project. Many other examples can be gathered of the pitfalls of basing conclusions on antibodies alone, including the increasing experience with immunosuppressed transplant recipients and the cytomegaloviruses. In the mumps-endocardial fibroelastosis story this type of approach is carried even further when a skin test, less reliable than antibody studies, is used to build a case for intrauterine infection with a virus. Dr. Benirschke: Now we come to the placenta. 66 VI—Function of the Placenta, Barrier or Bridge? by Joseph Dancis Dr. Dancis: A word of comment about the bibliography that I have prepared. Many are general introductions to the subject of placental physiol- ogy with useful references; others describe some of the techniques that have been used in studying placental function. If one accepts as the definition of the placenta, the organ in which the maternal and fetal circulations are brought into close approximation, then the placenta is not really completed until 40-50 days gestation. The fetal circulation is well-developed by about 3 weeks gestation but the maternal circulation is not for another 20 days. Many of the serious effects on the fetus induced by infection occur early in the development of the placenta. In further evaluating the importance of the placenta to this conference, one must recognize that there are other routes to the fetus. Infections may ascend through the vagina, a not uncommon route during labor and delivery. Infectious agents may also pass through the amniotic membranes into the amniotic fluid, and thus infect the fetus. I don’t know how often this occurs. However, we shall dwell on the transplacental route of infection. As an introduction, we may first consider mechanisms of placental transfer. 1. Diffusion—this appears to be the mechanism of transfer of gases, water and possibly small molecules such as electrolytes. 2. Active transfer—using this term in a broad sense, rather than with the more restricted meaning of the membrane physiologist, there are many mechanisms to aid the transfer of water-soluble molecules such as carbo- hydrates, amino acids, vitamins, etc. 3. Lipid-soluble materials may not require active transport mechanisms, transfer being facilitated by solubility in lipid membranes. Definitive informa- tion is not available on this subject. 4. Transfer of proteins—the transfer of proteins from mother to fetus, particularly antibody protein, occurs before or after birth in different species and the route of transfer also varies with the species. In the human, indirect evidence indicates that the route of transfer is across the allantoic placenta. It has been suggested that transfer is mediated by pinocytosis. This gives no information concerning the specificity of the transfer mechanism, certain proteins being transferred more efficiently than others. 5. “Leaks”—it is evident from the clinical entity, erythroblastosis, that bodies as large as red cells can traverse the placenta. Recent investigations make it clear that the major flux occurs during birth, presumably when many vessels are torn and leakage occurs. However, there is also evidence that small numbers of cells traverse the placenta during pregnancy, and the direction is almost entirely from fetus to mother. The structure of the placenta suggests a one-way valve arrange- 67 ment so that increases in fetal intravascular pressure may lead to a leakage of fetal cells into the maternal circulation, but the reverse is generally prevented. 6. Placental infection—For transplacental infection to occur, a maternal bacteremia or viremia must first exist. In other words, maternal defenses must first be breached. If, now, the placenta were to become infected, transfer of the infectious agent into the fetal circulation could easily follow. This may represent a common means of dissemination. Does the placenta serve any protective role against infectious agents cir- culating in the maternal blood stream ? This was investigated in the guinea pig using bacteriophage. Bacteriophage was used because it is not infectious in the guinea pig and so placental transfer could be evaluated in the absence of placental infection. Phage could be demonstrated in the fetal circulation with regularity, following the injection of phage intravenously into a preg- nant guinea pig. However, the titre was considerably lower than in the maternal circulation. In other words, the placenta was an effective but not a complete barrier to a particle as small as the bacteriophage. If the placenta were immunologically competent, it could further protect the fetus. This possibility was investigated in the mouse. The evidence is now clear that cells within mouse placenta are capable, under experimental conditions, of transforming into lymphoid cells with immunological function. There is a large gap between observations made experimentally in the mouse, and physiological functions in human placenta. There have been occasional reports of plasma cells in human placenta. Dr. Benirschke was responsible for one of those reports. Dr. Benirschke: The frequency of our finding is significant, however, we are reluctant to publish any more cases because it is now an established phe- nomenon that the placenta is capable of making plasma cells. I can show you numerous placentas with plasma cells. Dr. Silverstein: That doesn’t prove that it is capable of making them. Dr. Benirschke: As to the need for an investigation where these plasma cells come from, this is still not settled. One finds plasma cells in the placenta frequently. Dr. Dancis: I interpreted the lack of continuing reports on this from Hanover as an indication this was a rare finding. I would like to have you put that down somewhere. Is it in this book? (52) Dr. Katz: You have to pay $17. (Laughter. ) Dr. Dancis: Which chairman do I have to order? (Laughter. ) Dr. Katz: We can tear a book up. There are only several articles I want there. (Laughter. ) Dr. Dancis: This really makes far more interesting, to me at any rate, the extensive, annoying work that went on in the mouse for years when I wasn’t sure it had any meaning at all. In the mouse, the plasma cells are already in the placenta at 11 days’ gestation when there is very little vascular penetration. Dr. Silverstein: I believe. I even believe Kurt’s work. I just question his right to draw that conclusion. Dr. Dancis: I just like to talk to somebody. (Laughter.) Dr. Dancis: How can involvement of the placenta contribute to terato- logical effects? There are two obvious ways. One would be to serve as a nidus of infection. I think this is often true. Certainly in rubella it seems to be pain- fully true. It could also, through interference with some of the special roles of the placenta result in injury to the fetus. This second possibility is really an “in- tuitive” judgment, which means there is no evidence that it is so but it should be so. Certainly with considerable involvement of the placenta, you would assume that placenta function may be interfered with, and interference with placenta function is bound to deprive the fetus. However, these things have not been measured, and the methods for measuring them are not at hand. The effect of maternal viral infections on the pregnancy varies with the type and the time of infection. Early infection may lead to abortion. The mechanism is not known—whether it affects uterine motility, placental func- tion or infects and damages the fetus, or whether more than one factor is operative. Teratogenic effects may occur during a broad range of pregnancy, as in cytomegalic inclusion disease or toxoplasmosis, or they may be limited to the first trimester as with rubella. Infections late in pregnancy may lead to disease in the infant, though possibly modified as in disseminated herpes of the newborn. I have been told that rubella infections of the mother don’t affect the fetus after about the 4th month. The fetus doesn’t even get infected. Not only not affected, but not infected. Dr. Katz: That last point may be questioned somewhat. I believe there are a few documented cases in which an infant whose mother had rubella after the first trimester was found to excrete rubella virus postnatally, even though he was a completely normal baby. Among 14 normal infants studied by Dorothy Horstmann and her group at Yale (53) were 4 rubella shedders, two of whom were delivered to mothers whose rubella occurred after the first trimester. 69 Dr. Dancis: Would you want me to extend it to the 5th month? Dr. Eichenwald: There were a fair number of children infected during the second and third trimester who excreted the virus but were perfectly normal. Dr. Benirschke: Conversely, is it not true that Dr. Weller had some babies in whom only from the placenta they could identify virus? Dr. Katz: Yes, after the first 8 weeks’ gestation, rubella virus was more frequently isolated from the placenta and not the fetus. Dr. Gruenwald: What does isolation from the placenta mean? It contains fetal blood and maternal blood. Dr. Benirschke: This is now in patients with documented rubella prior to their delivery in whom at birth the baby didn’t have blood or urine excretion of virus, but in which the placenta had the histologic features of infection and from which virus could be recovered. I think this just points to the complexity of the system. Dr. Dancis: Dr. Benirschke has raised an important point. The placenta becomes a more effective transport organ later in pregnancy. Here we have something obviously not explicable in that way. It may be that the placenta is more susceptible to infection early in pregnancy than it is later in pregnancy. This may be a keynote to the early rubella in- fections. The placenta is bathed in viremic blood and may serve as an excellent culture medium and the fetus is secondarily infected. The cells of the more mature placenta may not favor propagation of the virus. I should now like to describe, briefly, some of the techniques that are used to study placental function. A time-honored approach is to measure the blood levels in the maternal and fetal circulation, either of naturally oc- curring materials, or of substances injected into the mother. It is crude, but can be informative. This is essentially the technique that we used to measure the permeability of the placenta to phage. An approach that is being pursued clinically is the measurement of placen- tal metabolites that are excreted in maternal urine. In this category, are the determinations during pregnancy of diamine oxidase and estriol. In situ perfusion of placenta has been used in the sheep, guinea pig and human. The fetus is removed and the umbilical vessels cannulated and per- fused. This greatly simplifies the experimental situation and has provided interesting information from short-term experiments. Indwelling catheters into the umbilical circulation of the sheep permits a more prolonged study with relatively minor disturbance of physiological rela- tions. As Dr. Silverstein has indicated, the technique is also applicable to the monkey. In vitro perfusion of human placenta has been explored without making, so far, significant contributions. 70 Various in vitro culture techniques have been developed. Stewart (54) transplanted a piece of human placenta to the anterior chamber of the rabbit eye where it survived and synthesized steroids. Thiede (55) has succeeded in growing placental cells in monolayer. And Tao (56) has used organ cul- ture for early placentas, maintaining a good-looking placenta capable of synthesizing hormones for a period of weeks. Alvarez (57) has demonstrated the safety of percutaneous biopsy of placenta. Recently he reported on a series done as early as the third month of gestation. Success in obtaining a biopsy is largely dependent on whether the placenta is anteriorly placed. This is a method that deserves further explora- tion. It may provide an approach to diagnosing rubella infection of the fetus. Culture of the placenta, or possibly the demonstration of placentitis may in- dicate the diagnosis in a questionable case. Dr. Gruenwald: This is being done in this country by Dr. Cibils at Western Reserve, now at the University of Chicago. Dr. Dancis: He worked with Alvarez for some years and is supposed to be quite expert. Incubation of tissue slices is an old technique and need not be discussed further. It is possible with human placenta to tease out villi which, under the microscope, appear to be relatively intact micro-organs. This preparation will concentrate amino acids more effectively than the slice and may present other advantages for study. Dr. White: Dr. William Marine (58) and his coworkers at Emory, did a study in mothers who had rubella in the first trimester. They studied 29 therapeutic abortions. All deliveries were vaginal. Although they were seldom able to recover a well preserved fetus, the placenta was usually intact and quite well preserved. The “fetal” side was represented by chorionic villi which had been washed, isolated, and dissected free prior to culture. Tissue from the washed dicidua represented the maternal side of the placenta. There were no instances of decidual infection except in association with positive virus cultures from the fetal side. Further, isolation of rubella from the maternal side could be accomplished in only 40 percent of those cases with virus in the villi. Overall, 76 percent of the placentas were infected. This was taken to indicate a predilection of the fetal side of the placenta for persistent in- fection with rubella virus. I would like to hear your comments on both the results and the methods used for collecting the specimens for virus isolation. Dr. Katz: Most studies have found that if the fetus is virus-positive, the amniotic fluid and placenta are also. But negative fetal cultures, especially after the 8th week, may nonetheless be accompanied by a positive virus isola- tion from the placenta. The type of study which you say Dr. Marine has con- ducted is not one with which I am familiar. By swabbing the “fetal” side, do you mean the amnion? 71 Dr. Benirschke: Either one of the two, probably. Dr. Dancis : How were these abortuses delivered ? Dr. White: They were all vaginal deliveries. Dr. Benirschke: It’s fair to say, as Selzer (59) points out in his photo- graphs, that it is almost invariably the decidua of a placenta in rubella infec- tion which contains inclusion bodies and this presumably would be swabbed if one speaks of the maternal surface. Dr. Dancis: After the maternal viremia. You say there is persistent infection? Dr. Benirschke: Yes, in rubella placentitis. I think the placenta does represent a barrier and a variable one, but it is incumbent upon us to find out just what these variables are. The placenta has a very peculiar organization which we have to understand better. It is a difficult organ to understand. If we approach it from a comparative view, say with the mouse, and then talk about the rhesus gestation, etc., then we must recall that the placenta is the most variable organ in its histology and probably also in function. The liver of a cat or mouse or man are very similar, but the structural differences of the placenta of various species are so tremendous that one can infer there must also be some functional differences about which we know very little. I have tried to look up some of the literature pertinent to this topic in animals; it is precious little. What has been observed, in a variety of well- studied animal models, is the pathology of blue-tongue virus disease. And even this is limited to a very superficial description of chronic nodules in the sheep, preparatory to the invasion of the virus. One study is on equine abortion in the placenta in which the organ is largely necrotic (11). Then there are Ferm’s more quantitative and histologic studies of a vari- ety of viruses only in the hamster. But we don’t know anything about, say, the ability of the mouse or the hamster placenta to make antibodies. In fact, surprising findings are made thus in the hamster. The viruses are not con- centrated in the yolk sac where antibodies pass through, but are invariably concentrated in the chorioallantoic placental portion. You can’t even recover them from the yolk sac in experimental infection. It is a difficult field, com- parative placentology, and unfortunately I must say the human placental afflictions are also still very confusing. Dr. Katz: In the hands of the virologist, the points which Lon White was stressing are especially cogent. Most of the time, if fortunate enough to re- ceive any specimen at all, what he gets is a mixture of all portions. This has been obtained at curettage or passed spontaneously and preserved in a bowl. It is rarely possible to distinguish the components of the conceptus, let alone sample them without contamination by the other tissues and secre- tions. If for some reason a hysterotomy or hysterectomy has been performed, then it is possible to dissect and sample identifiable materials. This affords a 72 far more reliable basis for assessing which tissues have been infected, titra- tion of their content of virus and assay of antibody present. Apparently this is what Dr. Marine have been enabled to do and I would hope his work will be published in detail. Dr. White: So far only in abstract form in Clinical Research (58). Dr. Gruenwald: I don’t know what species of animal people will use for these studies, but when you talk about monkeys, I think it’s important to know that the prosimiae, the primitive monkeys, have an epithelio-chorial placenta, not a hemochorial placenta. This has been known for some time. These should be treated separately from the higher monkeys. Dr. Dancis: What do you mean by primitives? Lemurs? Dr. Silverstein: I wonder whether we know precisely where the maternal- fetal barrier is and whether there may not be more than one such barrier. Is it at the epithelium, or is there possibly another barrier at the vascular endothelium? The reason I ask is that I wonder whether it’s reasonable to think in terms of a placental infection that may not involve the fetus; one that might lodge between the endothelium and epithelium. If there is such a barrier (and by the way there is precedent for this in the eye) its exist- ence might shed light on whether the placenta participates fully in all of the immunologic defenses of the fetus. Dr. Katz: Only one question: what basis is there for the unilateral flow from fetus to mother when there is leakage? Dr. Benirschke: May I answer that, Joe? Dr. Dancis: Please do. 73 VII—Viral Infection of the Placenta by Kurt Benirschke Dr. Benirschke: The problem particularly of specific comparison is very important in considering placentation. May I remind you of a finding quite startling to veterinary pathologists. In attempts to produce germ-free dog pups, all were infected with Toxocara worms. They prevented the pups from getting micro-organisms at birth, but all had trans-placental infections with Larva migrans (60) (61). Now this is a big organism that passes through the placenta quite readily. I would like to remind you also of evidence, both for the gut and for the placenta in guinea pigs and dogs, that starch particles of various sizes pass readily across the placenta (62). The question of placental leaks in the maternal direction, is an ever- recurring one. I have summarized what evidence there is quite completely in the book (62). There is no question that fetal red cells almost regularly leak into the maternal circulation at one stage or another, usually in quanti- ties less than 0.5 ml. This is conventionally accepted to be the minimal im- munizing dose in Rh disease. Conversely, there is good evidence, although much more spurious and less frequently occurring, that maternal elements reach the fetus. This occurs to the extent even that a case of neonatal plethora has been described. If one looks in patients with leukemia or sickle cell disease, the number of cells found in the fetuses is so small that there is no point in worrying about con- tamination at the time of delivery. Usually no transfer occurs in this direction. Some substances then go across the placenta readily but irregularly. I think there is something to the placental barrier, and I personally believe we need more evidence for the notion that the placenta interacts in an immuno- logic fashion with the fetal system. So far as micro-organisms are concerned, you are all familiar with the ascending route of bacteria, through the endocervical canal, and the fre- quency of the amniotic sac infection or inflammation syndrome. At times there is no question that this is due to prior infection. We assume that cervical lesions are hardly ever responsible for an ascend- ing viral infection. We assume this to be so because in the viral diseases of the fetus there is a placentitis which is totally different from the placentitis which we see with the bacterial infection. In particular, there is a villitis which is quite different from bacterial infections. Let me remind you of the electromicrographic appearance of this barrier as we see it at the present. Figure 22 is from Ashley’s paper. The little stars, depicting ferritin molecules can be traced through the trophoblast. They have not reached the fetus but are only in his appendages. From there these par- ticles have to be transported into the villous circulation, and this is an im- portant step which concerns us. 74 coated pits and vesicles CE @D basal folds Se >< >< 2 or Du basement Tam IHS Sgn Ler Figure 22 Diagram of trophoblastic surface of placental villus as seen electronmicroscopically. Top- intervillous space, mv-microvilli of syncytium, d-desmosomes, m-mitochondria, ger-granular endoplasmic reticulum, cyt-nucleus of cytotrophoblast, syn-nucleus of syncytium. Stars represent ferritin as seen in transport after incubation. (From Ashley, Archives of Pathology, 80: 377, 1965). I would next like to draw attention to the differences in size of villi during pregnancy. Compare the 3-month villus at the left in figure 23 with the term villi at the right. Notice the difference in size of these villi and the fibrin mate- rial, considered by Kline to represent the “healing” of leaks in the mature placenta. There are perhaps 10 viruses that have been shown to have crossed the pla- centa in man to infect the fetus. In figure 24 I show an 18-week pregnancy of a woman vaccinated during pregnancy. The placenta participates in this virus disease at times and I could have shown rubella placentitis, which is not too dissimilar. There are inclusion bodies and villitis. In other cases with good evidence of previous rubella in- fection, there is partial fibrosis of villi which may be important to fetal growth in utero. 75 293-596 0—68——6 Figure 23 pared with mature villi (right) at same Immature (3 mos.) placental villi at left com magnification. (H & E x 160). 76 Figure 24 77 In any event, in contrast to bacterial infections, viral infections of the baby are almost invariably associated with morphologic changes in the structure in the villus. No changes occur, however, in poliomyelitis and in Coxsackie virus infections. Polio has not been studied well. The cessation of fetal movements with the birth of a paralyzed fetus, shown to possess polio virus, is well known but the placenta has not been well described. In the three cases of established prenatal Coxsackie disease that I have seen, the placenta showed no discern- ible disease (fig. 25). SLE — be Fes oe Fh - EER, Re Ey FRR ' Avs CF mgt oy had, CST aE SAT RY > a. 9% 3, - a ‘ A ! “7 Sf ve — - START wee Cf Wa hee nd RET - i Cor hho RSE : “- T= | vr See "ping >. Te ad «Ye sr YS is Fels - Jie be “wh Lew dl - » - - oe ry ne 5. FS 8 1 ht ¥ _ ". ~ - LN A ” a 3 4 BE . iH . . ER > # ~y i y « - ow . Xe \ i» ay - - * . ! . v , A fy » i 2 : Ly bea iY bs Figure 25 Degenerative and inflammatory lesion of villi in fetal and placental vaccinia at 18 weeks gestation. (Courtesy B. Ivemark, Stockholm.) (H & E x 160.) The next picture (fig. 26) depict placenta plasma cell production in the placenta at 18 weeks and at term. Having been alarmed by this finding, Dr. Driscoll looked for similar find- ings and now we find plasma cells frequently in mature placentas; moreover, we find them in patients in whom we have no reason to suspect virus or other disease. Therefore, we don’t know what this finding really signifies. I have suggested that these cells come from the Hofbauer cells, an aspect which needs to be studied. I am not familiar with a morphologic equivalent cell type in other animals but would like to hear what others think about this. In figure 27 is a 3-month villus with large cells in its stroma, named for Hofbauer. They often contain some pigment and most of us consider them to be phagocytic cells. They certainly are in a strategic position; if they are macrophages then they are interposed between the syncytium and the fetal vasculature and thus may perform a vital function. They could well inter- act, immunologically speaking, with an invading virus, an aspect which needs to be studied in greater detail. 78 v a / @ ® y Ak Tess e a A 8 of / 20 Figure 26 Plasma cells (two are at arrows) in the villi of two random term pregnancies. There was no knowledge of virus disease and the babies were apparently normal. No further studies were undertaken. Note also the focal villous necrosis. (Unna-Pappenheim x 640.) Figure 27 Villus of three months pregnancy, slightly edematous with several Hofbauer cells, two at arrows. Their vacuolar structure and resemblance to macrophages is apparent. (H & E x 400.) 79 We accept transplacental transfer also for tuberculosis. Evidence should be found in the placenta and is seen rarely as typical granuloma. We find granulomas also in rubella placentitis. Coxsackie virus does not affect the placenta, but small pox causes typical lesions. To illustrate the complexity of placental transfer, I refer to toxoplasmosis. The cysts may be found in the umbilical cord of affected infants (63). On the other hand, in figure 28, toxoplasmosis of the fetal membranes was found by Dr. Driscoll incidentally, in a perfectly normal term placenta. The baby was enrolled in the NINDB study and it was found subsequently that the mother had become infected at about 31% months. At this time the mother showed an antibody rise. At 314 years of age the baby is free of toxoplasmosis completely, despite this placental lesion with numerous organisms. Dr. Eichenwald: Does this baby have a positive toxoplasma serology ? Dr. Benirschke: I don’t know. Dr. Miller: Why was this placenta studied ? Dr. Benirschke: It was in the study group. Dr. Miller: This was a routinely studied placenta? Dr. Benirschke: Yes. This placenta was studied not only histologically but the organism was identified in mice. I think viruses may perhaps pass through minute lesions of villi, numer- ous in term placentas and very infrequent in young placentas, and which Kline (64) has suggested are the morphologic equivalent of a healed break. Fibrin deposits over areas of absent syncytium, or trophoblast in general. These are present in every placenta, in numerous spots, and presumably they are also the site of exchange of fetal red cells. The one model really carefully studied is the hamster placenta. In figure 29 is a graphic representation of what Dr. Ferm found in this species. The stars represent lesions and the vertical arrows virus multiplication. Mumps virus, injected intravenously at various stages of pregnancy of the golden hamster, does not pass the placenta but accumulates and multiplies in the placenta in which it does not produce lesions. Mumps virus, if injected into newborn, however, does cause disease. The hamster fetus then is not immune to the disease, but for some unknown reason the virus does not pass the barrier. It accumulates and proliferates without causing any morphologic equivalent in the placenta. Dr. Dancis: Do they know what part of the placenta? Dr. Benirschke: The Chorionic part. Dr. Dancis: Throughout? It’s not only on the periphery ? Dr. Benirschke: No. 80 Figure 28 Amnion (top) and chorion of free membranes of term placenta showing two toxoplasma pseudocysts at arrows. This was an incidental finding, the infant is well 3 years later. (Courtesy Dr. S. G. Driscoll, Boston.) (H & E x 640.) Dr. Dancis: It’s all the way through the placenta? Dr. Benirschke: It is found in ground-up chorioallantoic tissues, the exact location you can’t tell. When one takes these small buttons of hamster placenta and separates chorioallantoic from yolk sac tissue the virus is found only in the former tissue and is assayed by titration. 81 MATERNAL PLACENTA EMBRYO MUMPS RAT VIRUS HERPES _ 5 SIMPLEX zp H-1 VIRUS Figure 29 Diagrammatic effect of certain viruses upon permeability of the hamster placenta, the histopathologic effect (star-like symbol) of the maternal, fetal and placental tissues. Relative titers are indicated by height of arrows. (Courtesy V. H. Ferm, Hanover and by permission of Springer-Verlag.) Dr. Dancis: You can’t see anything histologically. You don’t know where it is. Dr. Benirschke: Not in this disease. One does see lesions in the herpes and H-1 virus infections. Both are in the chorioallantoic placenta and here principally in the trophoblast. Rat virus, on the other hand, does not cause lesions in the placenta. It multiplies in the placenta and then passes very readily to the fetus in which it causes death. 82 Herpes simplex causes extensive lesions in the mother like adrenal necrosis, and it causes extensive placental lesions, mainly necrosis, but it does not pass into the fetus. If injected into one, two or three of the embryos directly by hysterotomy, it will infect only that particular embryo and placenta, but it will not pass to the mother to cause disease. It will kill that one fetus, but the adjacent fetuses will be normal. H-1 virus passes readily, damages the embryos, it damages the placenta, but causes no lesion in the mother. To answer Dr. Dancis’ question, in figure 30 I show the Herpes virus—infection of the hamster placenta. This shows inclusion bodies in trophoblast. Figure 30 Histologic appearance of trophoblast of hamster placenta in maternal infection with Herpes virus. Large intranuclear inclusion bodies can be seen at arrows. (Courtesy Dr. V. H. Ferm, Hanover) H & E x 1600) . Although we have addressed ourselves to the question of how one can explain the various differences of transfer of substances across the placentas of various animals, actually we have no good explanations as yet for the experimental differences found. We don’t even know whether the human placenta really makes the plasma cells shown and, if it does, whether this constitutes an additional and important variable to the virus or toxoplasma disease of the embryo. Dr. Silverstein: | am willing to believe it does. Whether it constitutes a barrier or not, I don’t know. I think what you have shown supports the contention. I naively understood that there was a barrier between mother and fetus, and had assumed without really knowing that this was at the level of the trophoblast. 83 But I think your data suggest that there may be more than one barrier, and that the epithelial barrier may operate in certain situations and the vascular endothelial barrier may operate in others. Certain viruses may be able to penetrate one and not the other, so you might see a placentitis without fetal infection. Dr. Dancis: The mouse placental cells that have immunological capacity are not phagocytic towards colloidal carbon. They transform into plasma cells and lymphocytic cells rather than macrophages. I don’t know what cells they originate from. Dr. Katz: Have those cells, which you suggested might be plasma cells in the human, been examined electron-microscopically ? Dr. Benirschke: The Hofbauer cells have been studied. They don’t look like plasma cells. They do not have this enormous mass of endoplasmic reticulum characteristic for plasma cells. Dr. Silverstein: No early plasma cells precursor looks like a plasma cell anyway. Dr. Benirschke: It would be interesting to study this in the rhesus monkey by electron microscopy. Dr. Miller: In the Collaborative Perinatal Study, is anyone studying the children who have a specific defect, say, congenital heart disease, and looking at their placentas as compared with those of normal children? Dr. White: Yes. The placental histology studies are being done by Dr. Fujikura primarily. Dr. Dancis: I think it would be worth keeping in mind while we are talk- ing about the placental barrier that this term has always been a troublesome one. In general terms, the barrier doesn’t permit intimate mixing, and that is as much as it says. It must be interpreted quantitatively—how complete is the barrier between the maternal and fetal circulation? We must know, not only how effective the barrier is, but how serious is the threat posed by the penetration of “X” number of organisms. The be- havior of the placenta towards macromolecules has been studied and a poorly-understood selectivity of transfer has been demonstrated. It is certain that size alone does not explain this phenomenon. And now, Dr. Benirschke tells us of worms getting through! How big are those things? How do they make their passage through the placenta? Dr. Benirschke: They are quite large like microfilariae. They travel in the lymphatic and in the blood systems and the diseases they cause are due to allergic manifestations, at least in part. In South American monkeys micro- 84 filariae appear to be ubiquitous and harmless and apparently they don’t pass through the placenta. Dr. Silverstein: Some of these have been found in the retina in places where it would seem they couldnt have been passively distributed. Dr. Benirschke: But inanimate objects do pass this barrier at times without question. Dr. Dancis: This is so. To return to the story of maternal to fetal transfer, Zuelzer (65), the hematologist, published a curious study in which he withdrew blood from the umbilical vein in cases of incompatibility be- tween mother and fetus and he found a low but appreciable incidence of maternal cells. But when he sampled blood from the infant’s heel he found none. Zero. Dr. Benirschke: One must be very careful of findings on passive transfer across the placenta. In the few well-studied cases of maternal sickle cell anemia, the newborn baby may not have a single sickle cell despite massive sickling in the maternal blood. Similar studies have been made in maternal leukemia. Dr. Dancis: When Dr. Jon Uhr and myself did the study on the transfer of phage across the placenta in the guinea pig, we were very much con- cerned about the possibility that even a trace contamination from the mother in the fetal sample would be disastrous in the interpretation of the experiments. There were something like 107 particles circulating in the mother, and we were detecting about 10* in the fetus. Dr. Uhr suggested a clever approach to exclude surface contamination of our fetal samples. Two cardiac punctures were done on each fetus, withdrawing 0.1 ml and 0.5 ml. If the phage in our samples resulted from contamination of the fetus with maternal blood or fluids, we would not have expected a quantitative rela- tion between these two samples. But there was. One possible explanation as to why there is a more frequent transfer of cells from fetus to mother than in the reverse direction may lie in the struc- ture of the villus. If one considers the X-section of the villus surrounded by the maternal “lake”, it is evident that increases in intravascular pressure on the fetal side will facilitate leakage of cells out of the fetal circulation. In- creases in vascular pressure on the maternal side will collapse the fetal vasculature and prevent ihe transfer of cells. A similar explanation has been offered for the function of the choroid plexus which shares these anatomical characteristics. Dr. Silverstein: The distinction has to be made between the normal and the abnormal barrier, the normal and abnormal placenta. The problem is the same with any so-called barrier; the blood-brain barrier, or the blood- aqueous barrier, for example. When it’s normal, it doesn’t pass things. But the difference between normality and abnormality may really be remarkably slight. 85 By tapping on my right eye, I have just now disrupted my blood-aqueous barrier in that eye. Dr. Dancis: That is a rash thing. Dr. Miller: With regard to the viral etiology of congenital defects, what needs most to be done now in studying the placenta? Dr. Benirschke: We know virtually nothing of the structure of the placenta in almost any congenital deformity, except for the number of umbilical vessels. Histologically, we need quantitative data on the structure of the placenta in such cases. I don’t know if this qualitative data shown is of help. They are of interest to me, but I feel that we must know whether the placenta constitutes an independent immunologic system, whether the placenta can make anti- bodies. For instance, give a pregnant monkey rubella virus and see if the placenta now concentrates or makes substantial amounts of antibodies. Dr. Silverstein: I think it would be fascinating if somebody took those four viral systems you indicated, two of which appear to stop in the placenta and two to pass through, and study the virus by fluorescent antibody tech- niques in the placenta and fetus. That would answer Joe’s question about where it goes, and where it grows. Dr. White: Does it persist in the placenta for an extraordinarily long time? Would this be a reason for collecting placentas for possible cultures, as has been suggested ? These studies with herpes and mumps weren’t followed ? Dr. Benirschke: These are hamster. Hamster pregnancies are so short, that we cannot draw any inference from studies in these short-lived rodents. Dr. Miller: In studies of human placenta, is the best resource the Collab- orative Perinatal Study and Dr. Yerushalmy’s study ? Dr. Benirschke: In placentas with plasma cells we must find the cause for their presence. Are they associated with viral diseases? Similarly, in all those with irregular foci of scarring, or whatever you want to call it. Some of this certainly can be seen in rubella. Are the scarred villi found so often related to intrauterine viral experiences? Are there morphological equivalents in the placenta of babies with congenital abnormalities that resemble virus diseases? Dr. Miller: Are these studies proceeding fast enough, or is there some- thing more that should be done? Dr. Benirschke: I don’t know how fast Dr. Fujikura is working up the NINDB study material. It is a formidable task. Dr. Mitchell: He has an almost impossible job. Available to him are hematoxylin and eosin stained sections from all the placentas delivered in the 86 study since May 15, 1962. The sections were obtained from blocks cut midway between the placental margin and the cord insertion. In some instances it is still possible to cut further sections from the blocks obtained, but in no case is it possible to go back and re-examine the entire placenta. Dr. Benirschke: I believe we really need a good animal model, reason- ably similar to man in order to follow up the suggestions from the collabora- tive study; to see if they can be duplicated and manipulated. This is the princi- pal reason I asked that Dr. Axelrod come and talk about just such a potential model. As far as I know, there is no study along these lines other than the one by Parkman (66) who studied rubella in rhesus. Dr. Katz: One of the simple matters of which we know little, and which may appear too mechanical to you, is the question of how viruses circulate in the body. This may be of considerable significance in determining how they infect the placenta and by what means they cross to the fetus. It has been too easy, and naive, to imagine them free in the fluid phase of blood and other extracellular fluids. Increasing inquiry into this question has revealed that a number of viruses may be circulating and replicating within white blood cells while others are adhering to platelets. Whether these associations have a bear- ing on their ability to cross the placenta has not yet been demonstrated. A great variety of viruses are able to adsorb to erythrocytes in vitro and this too offers a possible transport mechanism in vivo. What happens when any of these virus-cell complexes finds its way to the vicinity of a placental villus is not known. Dr. Dancis: I recommend the use of the isolated villus to investigate some of the things you have suggested, Sam. These are fascinating ques- tions. In the field of protein transport there is a strong suspicion that this may be influenced by the configuration of the protein, as well as the size. It is not clear whether the selectivity resides in the avidity of uptake or the transfer through the placenta. Dr. Katz: Your presentation illustrating maturation of the placenta also makes it quite apparent that findings in term placentas cannot be accepted as typical in any way of what might be seen in the first trimester placenta. They are not at all the same. Dr. Dancis: Indeed they are not. Just a couple of days ago, stimulated by the title of this conference Dr. Cooper and I examined a placenta from a therapeutic abortion in which there were sound doubts as to whether this woman actually had rubella. However, the risk was too great to ignore. Un- fortunately prior to the examination, the placenta had already been dropped into some sort of disinfectant. We teased out a villus and put it under a microscope. I don’t think there was a placentitis at all. It may take as little as that to make the diagnosis. Dr. Silverstein: Is one villus enough? 87 Dr. Dancis: At this point we don’t know. Dr. Benirschke: Erythroblastosis is the principal aim of Alvarez’ (57) study. These investigators have done phase contrast microscopy of the villus to see whether it could be used as a means to judge when to terminate an endangered pregnancy. They produced nice pictures and once showed me a placental biopsy which they interpreted as showing congenital syphilis. They treated this patient and took another biopsy two weeks later. The villus biopsy at that time looked really quite different. Dr. Miller: It sounds as if there is much to be done with the placenta that individual investigators are not likely to do. Is there something the Child Health Institute should undertake with regard to the development of a program, the integration or organization of effort, or the development of re- sources and personnel ? Dr. Silverstein: I think there are people interested in looking at placentas who are not getting all of the placentas they would like. More and more it’s becoming obvious that investigators will be dealing with congenital infectious diseases by infecting the fetus directly, as we are doing with the blue tongue. It would be a shame if we didn’t look at the placentas of these experimental animals to find out whether virus put directly into the fetus will result in placental plasmacytosis. It may be that the virus can’t go through in the other direction past the endothelial barrier: that would be fascinating. Dr. Gruenwald: Someone would have to learn how to look successfully at a variety of species, which may not be too bad with the sheep but gets so darn complicated with the rodents. As far as the biopsy of the placenta is concerned, I am still suspicious whether you can, in view of the considerable variation that one encounters in nearby portions of the placenta, really arrive at this kind of conclusion from biopsies. I would certainly say that if it were possible to study amniotic fluid, it would certainly be preferable. By the way, I have had Dr. Cibils show me some placenta pieces with phase-contrast microscopy and with ordinary light microscopy. One can see almost as much with ordinary light microscopy. Dr. Benirschke: Yes, but there is a whole variety of studies that could be carried out with fresh placental tissue. Tissue culture for instance, in search of aneuploidy would be a possibility and much easier than amniotic fluid cell culture. A question which I hoped Dr. Dancis would discuss is of concern, namely that lesions in the placenta may cause specific transport dysfunctions. At the moment we haven't a clue whether, for instance, by biopsy one could make the diagnosis of phenylketonuria. There isn’t even a description of the placenta in this disease, or in the adrenogenital syndrome, or in most other syndromes. I am very much interested in the adrenogenital syndrome, because the placenta is a steroid organ. We don’t know if it functions normally in this disease but surely it could be studied in a family with this trait. If 88 anomalous function can be identified then a placental biopsy might be of great help. Naturally, there are some real dangers of leakage of fetal cells with maternal immunization which would have to be weighed. Dr. Dancis: This technique could be very useful in the diagnosis of in- born errors of metabolism. We have looked for the presence of a series of enzymes in normal placenta—those associated with maple syrup urine dis- ease, hypervalinemia, galactosemia, isovaleric acidemia and hyperuricemia. They are all present. What we need now is to study placentas from babies with each of these diseases to see if the enzyme is now missing. In this type of disease any part of the placenta would be satisfactory. It may not be so with infections. Dr. Silverstein: Do you think there is a chance that they wouldnt be missing ? Dr. Dancis: There is a chance. Isoenzymes may be under different genetic control. Dr. Gruenwald: Even the delivered placenta would help you to some extent. Dr. Dancis: Yes. Dr. Gruenwald: When do these characteristics appear? Early enough? Dr. Dancis: We have found the enzymes in placentas from pregnancies of 2-3 months gestation. Dr. Benirschke: Lennie, would you now tell us about some animal models? 89 VIII—The Baboon as a Primate Animal Model by Leonard R. Axelrod Dr. Axelrod: I appreciate this invitation. Not being an expert, I can talk freely and think freely unburdened by knowledge. I guess it is fully under- stood that term placentas, term babies and spontaneous abortions are not quite the epitome of materials to be used to understand the normal growth and development of prenatal animals, particularly the human, nor the optimal materials for the understanding of the mechanisms of disease proc- esses throughout the course of a disease, whether it is infectious or nonin- fectious in nature. Due to the finding of atherosclerotic plaques and a sense of humor, we decided some years ago to study various aspects of baboon biology as baseline studies. One of these was embryology. Another was histology. Since then, numerous other baseline and physiologic studies have been undertaken by the personnel at the Southwest Foundation. Most pertinent to our discussion today are the concepts of baboons (primates) as models for human disease. There are at least three requisites for a good animal model. First, and I guess foremost, is that this animal, in the highly complex and integrated systems organization and disease, parallel the human in as many ways as possible. Whether one animal ever fits the entire bill is questionable, but one animal can fit many of the bills required for these parallelisms to the human. Secondly it is important that the animal be easy to handle, no matter what its size or ferocity. Thirdly it is crucial that the cost be within the limitation of our grants- in-aid and in-house programs. Today I would like to discuss a few parameters of the primate called Papio species. Years ago I could be more definitive and say Papio or other subspecies, but since then a committee has studied the taxonomy of the baboon by all methods, including blood studies, and come to the conclusion after 2 years of work that it is impossible to delineate into subspecies. There- fore, we are now calling it Papio species. First, I would like to give you some idea, rather briefly, of how these ani- mals are kept at the Foundation. To begin with, the climate of San Antonio is rather ideal for these animals since it is almost exactly that found in Kenya, East Africa, where most of these animals originate. Our outdoor facility con- sists of large cages covered with wire mesh. The rock piles in the center have caves in them, the floors of which are heated by steam pipe. The animals may exercise freely, climbing all over the cage. The middle cage has no animals in it and the cleaning of each section takes place by having the animals go into the middle cage through suitable tunnel systems. No one is allowed near the cages except the animal handlers to help prevent cross infection. In one of these outdoor cages we have a colony which has been inbred for close to 40 20 years at the San Antonio Zoo. We have over 75 of these animals whose genetics are documented and although they are relatives of the animals we bring in from Kenya, East Africa, they even look grossly different. The family arrangement of the baboon is troop-type, consisting of a domi- nant male, 10 to 12 females with young, as well as peripheral young males and juveniles. The male takes a fairly active role in the protection and play of the offspring. The gestational age of these animals is 175-11 days. Our indoor holding facilities have cages which are separately hung from a bar. This allows easy access and good hygienic conditions under the cage. A fork lift goes into the slots found at the bottom of each cage, lifts it off the bar, and places it on a platform with ball-bearing rollers, facilitating rapid move- ment of the cages from place to place. Each cage is fitted with a movable back. The back of the cage can be drawn to the front by means of a knurling screw at the front, pinning the animal for intravenous injections or other manipula- tions. Fach cage is fitted with a metabolic floor and a separate collection system for urine and feces. Most of our newborn animals are obtained late in the third trimester by Caesarian section. Many of our animals are sent to Dr. Melnick at Baylor who started a program with us for obtaining germ free newborns. Some of our incubator systems are truly operating-room sterile. Others are not. The feed we use was developed at the Southwest Foundation. It is now put out in large quantity by the Purina Chow people. The feeding of these ani- mals is not as simple as it might seem. On the veldt they eat a variety of grains, roots and leaves. By determining the typical composition of forage during the various seasons in Kenya, Dr. Kane, who was with the armed forces for many years in the field of nutrition, worked out the diet which we use at the Foundation. It is in pellet form, readily eaten by the animal and even after surgery it is still appealing. Isoniazid is added in small quantities as a tubercu- lostatic. While the baboon is not susceptible to human tuberculosis it often carries an avian type of T.B. However, the baboon does not normally carry monkey B virus. About two and a half years ago we started the embryology program, firmly believing that until we know what the normal looks like for this animal, it is quite hopeless to attempt to understand the pathology. This program is in the hands of Dr. Andrew Hendricks, Dr. Allan Katzberg, and their associates. After many observations and correlations we came out with the following information. The menstrual cycle of the baboon is 36 days long. Perineum or sex skin undergoes characteristic changes during this 36-day period. Based on these changes, the cycle can be divided into four phases. Now, as an adjunct to this we have developed electronic gadgetry to help determine the time of ovulation. We have used thermal methods, echo methods and vaginal smears. This has now been correlated with the perineal sex skin changes. If one studies the perineal sex skin in anticipation of determining ovulation time it is more than hopeless, but if first you correlate this with vaginal smears, laparotomies, and electronic gadgetry it becomes realistic. The be- 91 293-596 0—68—7 ginning turgescence is nine days. The onset of turgescence coincides with menses, which is three to four days long. Maximum turgescence occurs at nine days. The female is then, and only then, receptive to the male. With very few exceptions, in the thousands of matings that we have brought to- gether, this phase marks the period of ovulation which occurs between 15 and 18 days in most females. Ovulation occurs exactly 3 days before the onset of deturgescence. Mating for two to eight hours during this period yields over 30 percent of conceptions as a function of mating vs. conception. This is an extremely favorable yield since the baboon is not a seasonal breeder and matings are for such a short time. We do not leave them in for 24 or 48 hours. Only four 2 to 8 hours, and most of the recent ones in the last 6 months have been for less than 2 hours. Of course, this method is important so that we know the ages of the early stages of embryonic development and can talk about plus or minus a certain time. We have developed methods for Caesarean section, so that many of our animals have had as many as eight and nine Caesarean sec- tions in a period of 11 to 14 months. These animals can therefore serve as their own controls in many prenatal studies. The amount of variation in humans and in primates is such that for many types of exacting studies one likes to use the animal as his own control. This is particularly true con- cerning central nervous system experiments. In the last nine years, there have been 514 births and 78 fetal deaths. Stillbirths account for 9.7 percent of this and abortions 3.5 percent. This is 0.7 percent better than the Russian Sukumi colony, where their animals have been under observation since 1927. Dr. Miller: Isn’t that better than for people? Dr. Axelrod: Quite possibly. I am not a people person. (Laughter.) Dr. Gruenwald: Where do you draw the line between abortions and stillbirths? Dr. Axelrod: Abortion occurs anytime before term, from our viewpoint. Stillbirth is a baby born dead having gone through the normal term development. Dr. Gruenwald: Anything not full term is an abortion? This is different from human terminology. Dr. Axelrod: That's right. Fetal deaths include all pregnancies which terminated with a dead fetus. This is the way we defined fetal death. The World Health Organization has recommended this classification, by the way, for humans and we are following it in the baboon. The incidence of fetal deaths is decreasing in the colony as we gain more experience in the management of breeding colonies. Dr. S. S. Kalter, Director of our division of Microbiology and Infectious Diseases, has a 92 large program going in the study of the native fauna and flora of these animals compared to what they get from the animal handlers. The animal handlers are swabbed and studied along with the animals continuously. Serologic studies are done repeatedly. Dr. Kalter and his division have a good idea of what this animal carries in Africa and more particularly in the various portions of Kenya. Deaths in the colony are not only autopsied but studied for their potential infection and disease process. In a total of 508 live births, no malformed animals were found. Single umbilical artery was not considered a malformation. Malformations are determined by gross observation and complete histological and pathology work-up. All tissues are sectioned and reviewed by the embryologists and the pathologists. Of 50 fetuses, one showed a malformation giving an inci- dence of 0.16 percent. The incidence in man—this time I have a parallel result—is variously determined as 1 to 8 percent. Incidence of abortions in man is approximately 10 percent. Incidence of stillbirths in man has been variously determined as 0.3 to 1 percent in the U.S. Dr. Katz: What was that one malformation? Dr. Axelrod: Diaphragmatic hernia. Since we anticipated using the baboon for teratogenesis the animal was studied in the following manner. The normal embryology stagings were determined, the morphology and circulation of the placenta was described and the incidence of stillbirths, abortions and spontaneous malformations ascertained. About 4 years ago we decided to try to see if we could produce in these animals the so-called thalidomide syndrome as exemplified by the outbreak of phocomelia in Germany. So what we did was obtain thalidomide and administer it to the mothers at various times on exactly the per kilogram basis that it was given therapeutically to the human female. As we have reported (67) phocomelia was produced. Indeed we can now predict exactly whether hindlimb, forelimb or both will be involved depending upon the time of administration of the drug with the teratogenic dose being approxi- mately 2.5 mg/kg. Dr. Mitchell: Did thalidomide produce any cardiac malformations in the baboon? Dr. Axelrod: No. The baboon serves as a good animal model because gestation takes approxi- mately 6 months. The first 39 days of gestation are exactly parallel to the human. You cannot tell them apart until the muzzle and tail start to grow. One does not want an animal whose gestation time is so short that all of the organogenic and histogenic stages go by in a fleeting moment. The sensitive or teratogenic period to thalidomide is between day 23 and 28 of the pregnancy. Treatment between 23 and 26 days gives forelimb malformation, amelia and phocomelia. Treatment between days 26 and 28 results in hindlimb malformations and amelia. Treatment between 23 and 28 days, continually, yields malformations of fore and hind limbs. 93 We have also attempted to study some of the interesting things involved with virus teratogenesis. The syndrome of cat scratch fever, not mentioned here today, is one in which Dr. Kalter is very interested. The human mani- festation of cat scratch fever does not yield any mortality figures but does yield a very interesting syndrome which could be in some ways mistaken for Hodgkin’s disease. Cat scratch fever is caused by an interesting virus which represents one of a larger group, as you know, and we have been successfully able to reproduce the exact syndrome of cat scratch fever in the baboon. Furthermore, taking immature animals of three months of age and sup- pressing their immuno-response system with either imuran or thymectomy and then administering adenovirus 12, a true “runt” syndrome has been implemented in these baboons. Finally six animals were given 5 cc. of rubella preparation between 49 and 98 days of gestation. Dr. Benirschke: Intravenously ? Dr. Axelrod: No, intramuscularly. In all but two cases, abortion occurred within 7 days post-inoculation. One was delivered by Caesarian section and died 7 days postnatally. Another one was of normal birth and died immedi- ately postnatally. These animals are now being studied in a multiplicity of ways. Dr. Dancis: Did the mother develop the disease? Dr. Axelrod: Yes. We also worked with Coxsackie and a number of other viruses. We have some special buildings which are set up for this type of work. Now I am going to read a short statement written by Dr. Kalter concerning simian viruses. “Many clinical studies have been performed within recent years which have been concerned primarily with teratogenesis in human patients. The majority of these reports have been related to the use of drugs in chemotherapeutic agents. The literature contains relatively few reports of in- fections of pregnant women, and of these only a portion relate to viral dis- ease. Most references are to rubella, infective hepatitis and cytomegalo virus, roughly in that order. Almost without exception these reports are limited to such laboratory animals as rabbits, guinea pigs, hamsters, and mice. Refer- ences to experimental use of primates are rare. Inasmuch as tumors and other anomalies did not occur in rodents when drugs such as thalidomide were tested, they were approved for human use. Following the discovery that birth defects in humans were produced as a result of these drugs, it was possible to demonstrate similar effects in primates. There is a need for the establishment of animal models for the study of con- genital anomalies related to virus infection in pregnant female primates for it is only by the employment of primates that this research can be associated with, or directly related to, conditions which exist in humans. Similarities between non-human primates and humans in in utero development, normal biological behavior, and response to infection substantiate the need for the use of primates in studies of this type. Secondly, this program would provide information relative to the anti- body response which can be produced in the mothers and in the babies by the inoculation of pregnant females during early gestation periods. By study- ing the peak levels attained and the duration of demonstrable antibody titres it is possible that a third goal may be achieved. That is the demonstration of a definite relationship of antibodies to congenital anomalies. It should be recognized that the necessity for research of this type stems from the fact that even though a relationship between infectious disease and birth defects has been known to exist for many years, an understanding of this relationship is very limited. Since Gregg (68) in 1941 showed that rubella infection in pregnant fe- males could be responsible for malformation of the fetus, many reports have been written which confirm this finding. The literature contains reports of in- fection with rubella, measles, mumps, chickenpox, hepatitis, Coxsackie, and cytomegalo virus infection in pregnant females. However, the use of special animals in an attempt to reproduce teratogenesis has been very limited. Fol- lowing the reports of Swann (69) in 1949, interest in the subject of maternal rubella as a cause of birth defects intensified. Stimulus for the study in the baboon was given by the finding of thalidomide defects. Congenital cataract, deafness, mental deficiency, and heart defects were recognized as being related to rubella in the pregnant female. Because of these observations, it was postulated that other viruses might also be re- sponsible for malformations of the fetus and that, whereas severe infection can cause death of the fetus and resulting abortion, less severe disease as rubella produces abnormal changes but allows the infant to survive. Cytomegalic inclusion disease has been reported in the literature with in- creasing frequency in the past 15 years, particularly since the production of laboratory techniques which can demonstrate the relationship between cytomegalo virus and clinical illness. Congenital infections have been shown to result in neonatal hepatitis, mental retardation and cerebral palsy. How- ever, this has been studied very little in primates. Additional studies in more recent years confirms these conclusions, but the manner in which these changes occur is still unknown. The opportunity to investigate these infections of primates is a responsi- bility which should not be neglected.” Thank you. Are there any questions? Dr. Gruenwald: Since so many people use the rhesus monkey for various studies, would you care to say anything about the comparison of the baboon and rhesus for special work ? Dr. Axelrod: Well, in the first place, as Dr. Benirschke would tell you, he is not comparing the two placentas. The placenta of the baboon is essen- tially like that of the human. I am afraid that from what these gentlemen tell me the placenta of the rhesus is not quite that way. Since that which gets 95 across the placenta is that which gets to the baby, for the most part, I should imagine that the baboon is superior in that respect for the study of placenta- transferred drugs and infectious agents. Dr. Gruenwald: I hate to bite the hand that feeds me—I have gotten very nice material from your laboratory but I would think the placenta of a baboon is more like that of a rhesus monkey than that of man. Dr. Axelrod: Is the placenta of the rhesus more like human or the placenta of the baboon more like human? Dr. Gruenwald: I would be hard put to answer that. Dr. Dancis: What are the differences? Dr. Gruenwald: The differences, as far as production of diseases and so on are concerned, may be nil. The point is that given a section of a baboon placenta, I would certainly not think for a moment it might be a human placenta. Dr. Axelrod: Is this because you are trained to see a certain type of morphological size and structuring, or is it because of real changes that you can see in the cells and membranes? Dr. Gruenwald: This applies to the mature placenta where rhesus and baboon are quite similar. I have not seen any early placentas, which is, of course, the more interesting thing. Concerning term placentas, I would certainly think that the two monkey species are more similar to each other, than either is to man. But this may have absolutely no bearing on the poten- tial for investigational use because, basically, they are all hemochorial placen- tas, and whether the lobule is constructed a little differently and whether the maternal arteries and veins come in and go out at somewhat different places, and what the villus stems look like, is probably quite irrelevant. So what I said should not detract from the usefulness of the material. I think you beautifully demonstrated that the baboon is useful, and I am not questioning this for a moment. My only question was, in view of the fact that so many people use rhesus, how you thought the two species compared. Dr. Axelrod: We feel that the baboon is a useful animal model, more useful than the rhesus monkey, for example, because it does seem to parallel the human in the teratogenesis of thalidomide. The metabolism of estrogens in the fetus and placenta, and the secretion of pregnanediol in the pregnant female baboon closely resembles that in the human. The reproductive physi- ology of the rhesus is quite different. Similarly polycystic ovarian syndrome with “constant estrus” and the testicular production of testosterone from precursors is exactly the same in the baboon and in the human. As far as I am aware the embryology of the baboon has been better documented than that of the rhesus but this, of course, could be corrected. 96 Dr. Eichenwald: It might be pertinent to ask why the rhesus was chosen as a model. Dr. Axelrod: Availability. Dr. Eichenwald: Because he was cheap? Or because his metabolism resembled the human? Dr. Benirschke: He was easy to catch, runnning all over the villages, and was a pest. He was much cheaper than an expedition to Nairobi, Kenya. Dr. Eichenwald: He was selected not on the basis of any thought, but on the basis of availability. Dr. Benirschke: I believe that is correct. Then it turned out the kidneys grew beautifully and such cultures were good for virus studies. Dr. Axelrod: They are not a pest in India now. The Indian Government doesn’t like to export them. There is a religious taboo on the export of too many of them. In Kenya or other parks the wildlife service goes around with poisoned meat to kill the baboons. Dr. Benirschke: I am interested in your terrible experience with rubella virus. It seems as though an excessive amount of virus was injected. Dr. Axelrod: That is what Dr. Kalter feels. It was an overload. He is titrating down now. Dr. Katz: It was given by the respiratory route, to mimic natural infection, in some, while others were infected parenterally. Dr. Benirschke: This must be a massive inoculation. Dr. Katz: The virus content of a 5 cc. inoculum obviously would depend on the infectivity titer of the pool from which it was taken. If this were typical of most rubella preparations available earlier, it might have contained be- tween 5,000 and 10,000 tissue culture infective doses. Dr. Benirschke: This is exploratory, in all fairness. They want to see if they get anywhere and I hope they will follow it up. Dr. Alexrod: Indeeed we are doing just that now. But in view of the many human abortions of unknown cause we feel that abortion, subsequent to even an admittedly massive dose of virus, is of interest. Dr. Benirschke: Did you recover the virus from your abortuses? Dr. Axelrod: I don’t know. They are working this up now. Some of these abortions were real messes. Dr. Benirschke: One of them may be long dead and whether you get the virus out of them is questionable. 97 Dr. Katz: Another important question—and I realize this work was not done in your own department so it’s unfair to expect you to produce all the answers—would deal with the use of appropriate control animals. Five cc. of control cell culture fluid, or inactivated virus preparation, would have to be administered by the same route to other pregnant animals. Was something of that sort done? Dr. Axelrod: Three controls were done with injections from day 32 to day 98 after conception. Obviously the cell culture had a very real effect for there was only one live birth from three pregnancies. However, there was only one abortion. The third pregnancy terminated in an overdue stillbirth. Dr. Katz: You said that Dr. Kalter had a lot of data on the microbial flora of the baboons and I wonder if you could comment on these. Are they host to a variety of indigenous agents such as the simian viruses found in monkeys? Dr. Axelrod: He has made the comment that we now have 100 rhesus monkey viruses from the kidney, whereas with the baboon, virologically speaking, is a very clean animal since there is even a question whether he was able to isolate two viruses from baboon kidney. He does use baboon kidney quite extensively for tissue culture. The animal, as I said, does not get monkey B, does not get human T.B., but does get an avian type, which is not found to be a killer in the colony. The colony is very healthy with only some extremely mild outbreaks of diarrhea. These animals are caught in Kenya, kept a minimum of 3 weeks there. They are TB tested, and vermicide is given them. Then they are shipped over to San Antonio where they are kept another 3 weeks in an isolation ward, acres away from our immediate plant, and only then allowed into the colony. Dr. Benirschke: What is the price tag of a pregnancy? Dr. Axelrod: Price tag of a pregnancy ? A pregnant animal or pregnancy ? Dr. Benirschke: A pregnant animal. Dr. Axelrod: $550 f.0.b. San Antonio for timed gestation. Dr. Benirschke: What do you think it will cost your institution to produce a pregnant animal from the time it’s captured, kept alive, etc. Dr. Axelrod: It is very difficult to say. I can tell you the price of the animals—$200 per female before established pregnancy. Dr. Benirschke: One can buy a pregnant rhesus for about $150. Dr. Silverstein: A dependable timed gestation rhesus costs about $550. Dr. Axelrod: A female of bearing age which has been in our colony 6 to 8 months is still amenorrheic as when first trapped. Apparently this is 98 a psychological reaction. They get upset. After they have been in the colonies 6 to 8 months, they can reproduce. An animal of that type would cost $180-$200. Dr. Benirschke: A non-pregnant adult female? Dr. Axelrod: Fully gone over, certified animal, so to speak. Dr. Benirschke: Would be $180-$200. If bred then—what does it cost to keep them alive? Dr. Axelrod: $1 a day covers the cost of animal handling, feed, and so forth. Dr. Benirschke: So you would reckon it is—$300 or $400 for a pregnant female? Dr. Mitchell: A comparison that might be interesting is that an NIH dog, a large dog such as a collie costs $150. I think you demonstrated that for some studies the baboon was better than the dog, and, in fact it is not much more expensive. Dr. Axelrod: You can do your work in Africa for $14, if you trap your own baboon. More practically though, many of our females in the embryol- ogy program have had eight or nine Caesarians. Dr. Gruenwald: That would make it much cheaper. Dr. Benirschke : How long do they live? Dr. Axelrod: We have no way of knowing how long they live in the wild, because from the ecological standpoint as soon as they fall behind the troop their major enemy, the leopard, gets them. We have animals in the zoo, from the zoo, rather, which we know are 34 years old. We estimate that the animal has lived 25 to 35 years in captivity with no problem. Dr. Gruenwald: Will they reproduce that long? Dr. Axelrod: No. We have now enough data on the eruption pattern of teeth and the bone growth from our dental research program to know that the ratio for this animal to the human is 2.87 to 1. This holds for their growth of bone structure, their dental eruption pattern, the onset of puberty and the stages of puberty. Dr. Silverstein: I would like to comment on this question of cost to the best of my ability. We have been interested in timed gestation pregnant ani- mals, which are very expensive no matter what species of primate you are deal- ing with. The timed gestation pregnant rhesus will run between $500 and $600, to the best of our knowledge, no matter who is growing it. We have calculated that the difference between a rhesus and a baboon timed gestation, involving well-conditioned animals free of overt disease, will be less than about 40 percent. My rough calculation, therefore (I am not a baboon expert), would 99 be somewhere around $700 to $800 for a baboon. Obviously, if the baboon is going to answer any given question better than the rhesus, then the price, while high, is not so very different. Dr. Axelrod: I think the philosophy we have to adopt is to try and avoid the pitfalls of a lot of artifactual results. Very often these cost much more in the end in time, effort, and money than a thoroughly studied animal. Dr. Moorhead: I want to second that. Dr. Katz: Dr. Jack Singer, sitting behind me here, wishes me to ask you if baboons devour their abortuses and also whether they deliver spontaneously. Dr. Axelrod: They do, and they eat the placenta if you don’t get to it quickly enough. We much prefer to take these babies by Caesarian section, and they survive beautifully. We then have the animal and know that it is clean. Dr. Katz: What do you feed the newborn Caesarian animals? Dr. Axelrod: They get a Similac preparation. We have now determined the composition of baboon milk. It’s more concentrated in protein and fat than human milk. This work is now being published. Dr. Benirschke: What is the age of puberty ? Dr. Axelrod: 2.87 to 1. It varies, but varies within this framework, just like humans vary. A 22- to 24-pound female is ready for reproduction. They go better by poundage. Dr. Miller: Do you go on expeditions to catch these animals? Dr. Axelrod: No. I have never been to Africa. We do have a small staff in Africa; veterinarian, trapper, white hunter, with backup laboratories. Dr. Benirschke: In part, this is the reason for their success, the weeding out in the bush of bad animals. They trap them and the bad ones are thrown back. Dr. Axelrod: If any of you are interested, we have a full write-up here on exactly what we go through for these animals. In Africa, for example, they pass two T.B. tests; Hemoglobin of 5 grams minimum, sedimentation rate of 20 millimeters per hour maximum, and so forth. These are our specifications. Our entire animal management division is handled by Dr. Robert Humber, who formerly was chief veterinarian for the Air Force, and he knows his business. It’s an expensive business, but he knows it. Dr. Gruenwald: It may not be all that expensive if you save yourselves the trouble of bringing animals in that aren’t all right. Dr. Axelrod: We don’t think it’s that expensive either. When we do experi- ments we have somewhat of a better confidence level as to what we are 100 doing, that it is not artifactual, and we don’t find some metabolic problem which, in fact, is due to some infection. Dr. Margolis: It’s hard to believe it’s that inexpensive unless it’s subsidized. Dr. Katz: Does Dr. Kalter maintain a sero-survey of the animals housed there? Dr. Axelrod: Constantly. Dr. Katz: Even though so few viruses have been found among the baboons, don’t they possess antibodies to a variety of agents? Dr. Axelrod: He has a serological profile. Immunoserological profile. Plus actual culturing. We have been asked for animals and we have shipped them out. They 2o with a brochure on them. Let me read this, if you are interested. This is a protocol for screening African animals. General body condition of newly trapped animals may be medium to lean, but not of a debilitated or emaciated condition. Animals’ teeth will give some indication of age and health. Should not be extremely worn or broken. As one examines the teeth, the color of the gums and mucous membranes are noted. Badly infected gums, very pale membranes are good reasons for culling. A low temperature, below 97, eliminates the animal. Dr. Moorhead: What portion are not kept? Dr. Axelrod: Less than 50 percent. Dr. Benirschke: Are there other questions? Dr. Dancis: They have a fetal adrenal, don’t they? Dr. Benirschke: Beautiful ones, nearly like man. Perhaps we should have Dr. Moorhead’s presentation now. 101 IX—Cytogenetic Aspects of Virus and Cell Relations by Paul S. Moorhead Dr. Moorhead : This should be called “cell genetic aspects” or cytogenetic aspects since the basic genetics of the cell and the virus of course underlie everything. I don’t know of anything pertinent regarding genetics in a strict sense, concerning humans and virus infection; someone else here may want to comment on that. However, it has only been about six years since studies have been initiated concerning that effect of viruses which causes chromo- some damage. This is the general emphasis here for discussion of the cyto- genetic aspects of virus and cell relations. This began with Hampar and Ellison’s finding (70) and with studies on transformation of cells in vitro with an orientation towards a model for malignant neoplasia (71, 72). The latter is not quite germane to our main interest here; but I will go into it in respect to our own work on SV—40 since it highlights many things which are helpful to keep in mind. The other aspect of virus-cell work that stimulated much of the interest in chromosome damage induced by viruses was that by Nichols et al. (73) in which they found that circulating leukocytes in persons with viremia, particularly measles, do have a significant amount of chromosome damage. This was followed by studies by Aula (74) and by many other people (75, 76) showing that chicken pox, yellow fever vaccine, infectious hepatitis, as well as a number of other viremic conditions in man, are associated with some chromosome breakage, significantly above the normal background rate. Also, in a number of these studies some particulars have not been con- firmed, which illustrates the “iffy” nature of this kind of study. First of all, certain people have said in publication that they are not in agreement regarding what the background level of chromosome damage and breakage is in circulating leukocytes. I should mention here that all this refers to a short-term culture of circulating leukocytes which are induced to undergo division in vitro by the addition of phytohemagglutinin or other mitogens, specific initiators of cell division. This division then reveals damage to the chromosomes in cells which had presumably been in a nondividing stage in the circulation of the organism, in this case, man. This is one major area of cytogenetic work, with the leukocyte. I should also mention the implication of hepatitis with mongolism. Bob Miller will talk about this tomorrow. This is a supposed epidemiologic asso- ciation between the occurrence of hepatitis epidemics and the incidence of Down’s syndrome births in Australia found by Stoller and Collman (77). Bob will, I think, go into this much more effectively. You must remember that the rate of occurrences of Down’s syndrome (mongoloids) in the population is about one in 600; so it is very difficult to prove anything with that sort of low frequency fluctuation. Their conclusion 102 was based on the number of births nine months after the periods of peaks of hepatitis epidemics. Down’s Syndrome is always associated with the pres- ence of an extra #21-22 chromosome in the zygote, from the beginning, so here we are not concerned with only the leukocytes. We are concerned with an effect on the egg or sperm, or possibly at the one- or two-cell stage of the zygote. Aside from the report by Stoller and Collman which makes a fairly good case, Robinson and Puck (72) have published a couple of papers which I think outline very well what to look for and the sort of orientation necessary for picking up clustering of abnormal individuals, people with abnormal numbers of chromosomes and/or translocations. They report that there was a clustering in their data, taken from a large clinical operation for studying newborns. A rate of zero abnormals in 1,500 births was observed, then in the next 4 or 5 months they obtained five in 1,000. However, this is statisti- cally questionable; also, in that period of time conditions may change in their laboratory regarding criteria, vigilance, etc., but at least these publica- tions provide a valuable beginning for this way of looking at the problem, in a setting equipped for studies of a large number of newborns. This involves non-disjunction, the five cases involved, I think, primarily sex chromosome anomalies, which are among the most common type of human chromosomal anomaly. The elimination or acquisition of an extra chromosome doesn’t involve breakage. Attention should be brought to the work by Carr (78) and by Szulman (79), and a few others (80, 81), on chromosomal studies of aborted mate- rial. There are about three or four such studies, Carr’s being among the best, which have shown that the incidence of zygotic chromosomal anomalies [and occasional mosaics which must have occurred at the two, four, or eight- cell stage] increases greatly as you look further back in the earlier “spon- taneously” aborted material. Carr points out that Down’s syndrome cases occur in about one in six hundred births; the 17-18 trisomic is seen about once in two thousand, the D group anomalies occur about once in 4,000; and sex chromosome anomalies about once in 540. All this, according to Carr, adds up to about 1 in 240 of all births being of some aneuploid condition. This is a considerable rate and if we are looking for evidence of some cause or virus agent—this would be reasonably consistent. This is, of course, leading to the possible implication of viruses in part or all of this; but at present we do not know. When Carr examined the early aborted material he found a minimum of 22 percent of the abortuses showing some chromosomal anomaly and he uncovered, as did Szulman and others, trisomics and triploids of types which had never before been observed. These types were presumably lethal and such cases do not ordinarily come to term to be observed at all in the general population. In unselected newborn material his rate was zero in 500, that is, in material from birth to slightly older which is not out of line with a presumed rate of 103 1/240. All this only indicates the approaches; this does not directly relate to viruses. The relation of virus effects to cell genetics and cellular selection is in a primitive stage but we can start by making categories of cell-virus interactions. These classes are in the main: (A) the lytic type, (B) the carrier state and (C) the oncogenic type. An example of the first type, wherein chromosome damage is induced im- mediately by virus, would be Herpes simplex on the Chinese hamster cell. You get a variety of breaks and other damage, also rearrangement and non- disjunction (82). You also get cytopathic effects and cell death; so this is the lytic type of combination. If you put the same virus on human cells of similar type, in tissue culture, you obtain no damage whatsoever, for a time. So in the lytic type of system you have virus replication, cell destruction, and there is no stimulation of cell replication. The second type would be the carrier state, in which most of the cells sur- vive and continue to produce virus. In this cell-virus situation there is not necessarily any effect upon cell replication. An example of this would be rubella on human fibroblast-like cells. Studies at our institute (83) have shown that fibroblasts from certain types of tissues produce rubella virus in vitro when infected and there is no effect on cell replication. However, if you do the same infection experiments involving serial cultivation of fibro- blasts from other organs, as from fetal lung material, you do have a marked inhibition of cell replication. This effect doesn’t depend upon a direct CPE; there is a blockage of cell division, or limitation on the total cell cycles so that only three or four in vitro passages can be obtained. _ Now, the third cell-virus interaction would be the oncogenic type, as in SV40 with human cells and in polyoma virus with hamster cells. In these systems there is at first some CPE, and some virus production, but this is not the prime aspect of the situation. For SV40 a carrier state is established for some time* and then the virus assumes some cryptic form or relation with the human host fibroblast so that fully formed free virus is no longer produced. The classic situation, of course, is polyoma acting on hamster cells. There occur alterations of the cell, new properties, and whether these effects are mediated directly through the cell’s machinery or through the genetic changes evidenced by chromosome breakage and nondisjunction cannot yet be determined. A subclass would be the oncogenic RNA virus. Rous virus is a chicken virus but it can produce tumors in rats; however, it does not produce any chromo- some changes. Such tumors are the exception to the usual case in that you don’t have any evidence of possible cell genetic changes. All of this just relates to the nature of the interaction of the virus with the cell in the first place. We ourselves are interested in the system of SV40 with the human cell (84, 70, 85). After the initial interaction of the cell and the virus the cell acquires characteristics it didn’t have before, so the question *This period may be a few days or a few weeks for the human fibroblast with SV40 and may be negligible in the case of the Syrian hamster cell and polyoma virus. 104 arises, are these genetic only in a sense of the virus itself providing the DNA, or are they genetic in the sense of the virus causing permanent damage to the cell genome? By selection, certain cells are favored in the continued evolu- tion of the population, as in a tumor population or hyperplastic cell popula- tion. Is this really a model for neoplasia? This is our interest. In this situation SV40 is applied in vitro to human fibroblast-like cells from fetal material. (We have primarily used a particular strain. WI-38, although this is not im- portant. Any fetal material cultured in vitro for a time yields a fibroblast- like cell line and the generalizations I will mention apply to any human fibro- blast-like cells carried in vitro.) After the SV40 virus is introduced, there is a slight amount of lysis, which can hardly be measured, and then as you subculture the material, with virus being produced in a small amount, there is a period when a small number of cells begin to produce large amounts of the virus. Within about 12-14 days almost all of the cells are producing a very small amount of virus and after this point one begins to see the changes which comprise the transformation. We regard these as steps in a complex process which is selective and evolu- tionary in the in vitro situation. The initial characteristic step is the loss of contact inhibition of division. This is the earliest fundamental change in any growth characteristic of the cell population. In our view it represents a time at which the SV40 virus ge- nome has become established successfully in a majority of the population in whatever fashion it achieves this. Maturation of a complete virus is no longer occurring to any great extent. Maturation which occurs in some cells would presumably kill these and they are thus eliminated in competition with those cells in which the virus has assumed a new kind of relationship. This loss of contact inhibition of division represents an advantage in growth in tissue culture; these cells are capable of division under crowded conditions, where the normal cells, the uninfected cells, cannot divide so well. Within probably one or two cell divisions after this there are a large number of chro- mosome changes. Thus you have a very long delay between infection and the appearance of chromosome changes, but these coincide, or almost coincide, with this initial change in a growth property, acquired by the population. A second growth change occurs a few cell generations later. By this time you have tremendous variation of the cell genome by restructuring, tet- raploidy, breakage, and nondisjunction. So in terms of cytological categories of the damage that the virus can induce to cells you could establish four cate- gories: One would be simple breakage. The second would be nondisjunction, or numerical changes of whole chromosomes. The third would be polyploidi- zation, which accompanies all virus-cell interactions which produce breakage. The fourth would be a shattering or fragmentation of the chromosomes which (according to Warren Nichols) would probably represent a cell death situation. 105 I think the important point here is not that we have chromosome breakage; it’s what survives that matters. If you want to design a virus that is going to produce effects which might go on further, in terms of survival of the geneti- cally maimed cell, then you would copy the oncogenic ones such as SV40 and polyoma. This practically limits you to the oncogenic situation, since here you don’t get enough lysis or maturation of virus to kill all the cells. These damaged cells limp along and from this damage there evolves a population which is no longer subject to control. “Control” is used in the sense that the unin- fected cells are limited as to their total replication cycles in vitro, limited in the rate at which they grow and limited in terms of contact inhibition of division. (If you all have questions, I would appreciate interruptions.) Dr. Benirschke: Are there any categories of viruses that do one thing or the other more specifically? Dr. Moorhead: The RNA viruses don’t seem to affect chromosomes in general. However, a generalization is very dangerous, I think, because the number of viruses studied is still not large and they have been studied in different types of cell systems and by quite different approaches. For instance, in the human situation the leukemics may eventually be shown to be related to a virus but they don’t generally show a great amount of chromosome breakage. There is some evidence of specific types of chromo- some damage, and it’s not clear whether this could be cell selection or initial action of the virus, one being the Philadelphia chromosome occur- ring in one type of leukemia. You could argue that a certain type of virus has a propensity for a region in this chromosome which is then more likely to break. But this same break can be caused by radiation, too. Also you have aberrations in certain testicular tumors which are similar to each other. I think in the myelomas that someone has also found four out of five cases having similar karyotypes. To find similar karyotypes in situations in which you usually have a great variety of random changes is rather strong evidence for either selection for certain types of genomic rearrangements, or for a specific relation between virus and chromosome. This issue is not yet resolved. I realize I am getting far afield from congenital anomalies. Dr. Katz: Would you say that in general there is a correlation between the propensity to alter chromosomes and the viruses which primarily involve host cell nucleus? Dr. Moorhead: I think as a generalization the DNA viruses will involve chromosome breakage and, in general, are nucleus-associated. Dr. Miller: As you look at the preparation you see breaks; would you say they were caused by a virus—— Dr. Moorhead: You can’t. Nichols and his group have been able to say that mumps is, I think, more likely to cause shattering; but all of these over- 106 lap. You rarely see shattering with SV40 on human cells. You have very little assurance when you look at something, to say, that this is so and so virus or isn’t. You have few generalizations that you can make in this way. Dr. Eichenwald: Is it correct to say that every chromosome change which a virus can produce has also been found due to non-infectious causes? Dr. Moorhead: That is right. There are no clearly specific types of damage related to virus. That is to say, all categories of damage: breakage, nondisjunction, translocation resulting from breaks, etc., can also be pro- duced by various chemicals and by irradiation. Aside from general cate- gories of types of damage, there are only minor generalizations that can be made, implicating differential properties of certain chromosomal regions. Irradiation has been shown to induce breakage in a completely random way along all chromosomes whereas certain chemicals or viruses may induce statistically more breaks in the secondary constriction regions or at the centremere, etc. Although viruses induce true breaks which lead to rearrange- ment, viruses also tend to produce more “breaks” of the isolocus type than does X-ray. Also this relates to the suggestion made above regarding chronic granulocytic leukemia, and certain tumors, that some specific chromosomes or regions of chromosomes might respond differently or preferentially undergo breakage. Please keep in mind the excellent alternative that these are only selective phenomenon, not specific sites of initial virus interaction. Dr. Benirschke: Does an affinity of a virus to the nucleus enhance its ability to cause nondisjunction? Are there studies in this direction? Dr. Moorhead: We offered something like that as a conjecture in the early SV40 work, since Yerganian had noted it first with SV40 in human cells. He had found that a few passages after the growth changes appeared he had a very high proportion of elimination of chromosomes in the small acrocentric group of the human karyotype. In our material, we also had a very high proportion of the involvement of the acrocentrics. These par- ticular chromosomes are considered to be involved in the formation of the nucleolus and one obtains a very concentrated immunofluorescence around the nucleolus at a certain stage of infection, so this was interesting; but since then I think we have studies which pretty strongly argue that this is cell selection, even within a few cell divisions after transformation. I don’t believe you get any substantial preference for initial damage to those groups in human cells but within five or six cell generations chromosomal changes do preferentially involve these groups, so we think this is a selec- tion which operates very early in the SV40-human cell situation. Dr. Dancis: Has anybody investigated the frequency of chromosome anomalies as related to gestational age? Is this something more frequent in the younger age group? Dr. Moorhead: Those that fail, yes. 107 293-596 0—68——8 Dr. Dancis: The question is how to interpret the observation. Is this something that is directly related to death or not? Dr. Benirschke: You must know McFeely’s studies (86) Paul. Dr. Moorhead: Yes. Dr. Benirschke: He investigated the 10 day old blastocysts of seven swine uteri chromosomally. The chromosome number of the pig is 38 and the chromosomes are easily arranged. He found that in each uterus there were at least one or two blastocysts that had abnormal chromosome num- bers, very much like the human spontaneous abortions. And, of course, it’s well-known that in swine a large number of blastocysts fails to implant or, when implanted, undergo abortion. There were many types of errors, triploidy, tetraploidy, and mosaics, very much like in the human abortion material. So what holds for man holds also for swine. Dr. Dancis: This may be far out, but I don’t think it answers the question I am asking. The only way I could see that would be answered if you were to take a biopsy, grow it and permit pregnancy to continue to see whether a certain number of these peculiar chromosomal anomalies would permit continuation of pregnancy. Is there a possibility of acceptance of these anomalies and maybe rejection at a later date? Dr. Moorhead : Again, it’s the severity of the involvement. Aberrations of number for the larger chromosomes are only found in these early spon- taneous abortions. You involve more gene factors and increase the cer- tainty of a lethal imbalance. For the more minor ones like translocations, survival is quite common, of course; a balanced translocation is not neces- sarily detrimental and one might be among us in this room. From all the work in animals this condition is surprisingly frequent. Dr. Eichenwald: Are we assuming that in sows when cell division takes place, some cells may naturally acquire an abnormal number of chromo- somes? Thus there is no factor which is causing abnormal cell division other than pure random chance. Dr. Benirschke: For instance, bad sperms probably don’t make it to the tube so often as good sperms but we don’t know this. Certainly if there is an egg awaiting that has a chromosome too many or one missing, it will be fertilized and subsequently eliminated. Dr. Eichenwald: The reason the ovum or sperm is defective is then simply a matter of chance alone? Dr. Benirschke: That is an important basic biological question. Hayflick (87) is at least beginning to attempt to answer it. What is it that makes normal cell lines eventually die out? Is it because there are eventually mutations? Every cell has a chance of undergoing mitotic errors, and this is a frequent occurrence as far as we see it now. 108 Dr. Eichenwald: Do we know, how many of a population of 100 million sperm are defective? Dr. Benirschkle: Leuchtenberger (88) has examined this quantitatively, spectrophotometrically, on human and bull sperm DNA content. She finds a large proportion of sperm to have an abnormal amount of DNA. The accuracy of this methodology is now about 4 percent, at that time it was perhaps even higher (89, 90). Similarly, in meiosis studies one finds ab- normal meiotic division often which undoubtedly would end up with the production of abnormal sperm. Dr. Eichenwald: Is this level approximately the same from any one individual, or is there a great deal of variation? Dr. Benirschke: There are semi-sterile bulls, so-called, in whom the infertility is much higher and the DNA range is much wider, also in in- fertile men (88). An interesting finding is the Swedish report of transloca- tion in SRB cattle (90). Some bulls were used as donors for artificial insemination, and they happened to be translocation carriers, spreading translocation chromosomes (2n=>59) to many offspring. Dr. Miller: There are some viruses that cause chromosomal abnormalities, but do not cause congenital malformation. Are there other viruses that have no effect on the chromosomes but are teratogenic? Is there a correlation between Dr. Moorhead: I am not well versed on the viruses that are considered to be teratogenic in the first place. Dr. Miller: Take rubella. Dr. Moorhead: Chang (91) studied fibroblasts cultured from thera- peutically aborted material from rubella mothers that were serum-assayed and known to be infected. The cell material itself was shedding virus in cul- ture and this was studied in detail. Lumping together data from cells that were from such material (infected naturally in utero and then cultured for only a few passages) he was not able to show a very significant effect. There was a significant increase in chromosome breakage, but there was no nondisjunction, no polyploidization, no shattering to speak of and the amount of breakage was barely significant. In this study of nine human cell lines breakage was increased to 91% percent. The normal average is around 4% percent, for ordinary breaks in chromosomes, so it’s certainly not very en- couraging that this has anything to do with the rubella situation. Then he did a parallel study with about the same number of cell lines from material infected in vitro and found the same thing, about 9 percent break- age. He found no other particularly frightening chromosome effects. The more interesting thing is the work of Plotkin and Boué and others (81). In looking over all of their work over a period of more than a year they found fibroblasts that were derived from pharyngeal mucosa pituitary were more 109 likely to show breaks, some as high as 20 to 30 percent breakage, as compared to lung cell fibroblasts. They continued to grow and to shed virus for the full 40 or 50 passes that we normally obtain from such material, but fibroblasts from lung tissue are invariably stopped short in their growth in vitro. So there is some effect, entirely apart from the chromosomes, in rubella infection that seems implicated more than anything genetically transmitted by the cells. In all these cultures, aside from the significant but small increase in break- age, there was no transformation or enhancement of any of the growth prop- erties or changes in morphologic appearance. It has also been shown that there are no significant increases in chromo- some breakage or other effects in the peripheral blood leukocytes of newborns with Gregg’s syndrome. We only looked at fibroblasts in vitro and the complexities of virus effects on growth, I think, are much more rich and promising than anything con- cerned with Dr. Dancis: Chromosomal anomalies are a very gross effect. Dr. Moorhead: That is right. Dr. Dancis: Isn't it reasonable that you may expect any number of defects at the gene level ? Dr. Moorhead: Yes. Cytologically visible changes may just be the indi- cator, the top of the iceberg. The situation with importance may be under- neath. There may be actual gene mutation, but so far there is not a good system to demonstrate mutations, point mutations, by viruses. Dr. Eichenwald: One could say gene mutation could only occur if there is chromosome breakage, because otherwise the cell dies and that is the end of it. Dr. Moorhead: That is also true of transformation. You always get chromosomal changes. We have done repeated SV40-human cell transforma- tions and you always get chromosome changes, extensive ones. You wouldn’t think the cells could live. The changes seem as extensive as in the cells that die, so for this reason many people say the chromosome change is entirely extraneous, but I think they must have a role in this model of hyperplasia. Dr. White: Is it true that breaks in the chromosomes of fetal tissues can be induced by small amounts of X-ray? If the mother were X-rayed prior to delivery might this cause chromosomal breaks? Dr. Moorhead: Yes, if you didn’t check that out. Dr. White: If you look for viruses playing a role in congenital malforma- tions one obvious one is to search for seasonal variations. We were talking just a moment ago about gametes, and if we are thinking in terms of congenital malformations resulting from genetic imbalances in the gametes, is there any way we can use a seasonal approach to that question? When are the gametes of the mother most vulnerable? Are we more likely to 110 find the insult which results in non-disjunction occurring at ovulation, or some other time? Dr. Moorhead: It would have to be shortly after that, yes. I don’t know how long a time. Dr. White: If you are looking for trisomy 18, say, and it’s non-mosaic so that the initial insult occurred to the gamete or to the zygote and you want to determine if there is any variation in seasons, should you be looking at the time of conception? Or is the gamete just as vulnerable earlier? Dr. Benirschke: There is one recent study concerned with heavy maternal radiation of the gonads and fetal outcome (92). No deleterious effect could be ascertained. Numerous similar studies have been undertaken in experi- mental animals with variable results. Dr. Moorhead: Wasn’t this idea used to explain the age effect on the human egg of the mongoloid age effect? It doesn’t seem to fit. Dr. Benirschke: Preovulatory radiation damage of the potential gamete apparently is not always achieved. Dr. White: As I understand the process of meiosis, there is an initial pair- ing of chromosomes prior to segregation. If this were an incorrect pairing, might it result in nondisjunction ? Dr. Benirschke: No, I think there is good evidence it is unlikely to be incorrect pairing, rather we assume incorrect segregation. Dr. White: And this segregation occurs at about the time of ovulation? Dr. Benirschke: Which occurs immediately before ovulation in most mammals. The pairing in the female begins during embryonic life and remains until ovulation, when the first maturation division occurs. Dr. Dancis: There must be information from Hiroshima pregnancies taking place after radiation exposure. Dr. Miller: There were six measures of genetic damage, and they were all negative. With that size sample, as large as it was, no effect could be shown, although, on the basis of laboratory experimentation, there is no doubt that genetic damage must have been produced. It just can’t be demonstrated. Dr. Eichenwald : If we assume that a virus can cause damage to chromo- somes, shouldn’t we also be studying the father of a defective child to determine whether he had a viral illness just prior to the initiation of pregnancy? Perhaps the chromosomes in a significant proportion of his sperm were altered by the illness. Wouldn't it be just as logical to assume that it’s a paternal rather than maternal viral disease that is causing the problem? 111 Dr. Moorhead: I don’t know of any data on that. Dr. Eichenwald: I am talking about transfer of damaged chromosomes. Dr. Moorhead: Due to infection of the father? I don’t know, that is going a step farther back. Dr. Benirschke: Well, we rely quite heavily on the selection pressures in the ascent of the sperm from the cervix to the fallopian tubes. Not based on very good data. I know of no good comparative study of, say, the aneuploidy as measured by microspectrophotometry of sperm at the cervix, at the uterus, tube, and end of tube. This would be an important study to be done some time. Dr. Moorhead: 1 have one thing to say which will complete what I wanted to present. This intrigues me from the standpoint of Dr. Katz's interest in the dynamics of the virus infection itself, apart from anything on the chromosomes. I wanted to emphasize the fact that in the SV40~human cell you may wait some weeks after initial infection, subculture the material and so forth, before you see the series of changes which are quite reproducible. But Girardi (84) has shown that if he infects material when it’s dividing rapidly you obtain the same series of evenis but they are compacted. So the dividing cell is much more susceptible to perhaps the entrance of the virus or, more interestingly, the virus may be less likely to mature. If he uses material that is in a rapidly dividing state, he can achieve all the events described as well as the occurrence of induced complement fixing antigen but in a much more efficient proportion with the dividing cells. It is as if the cell were busy dividing so it can’t supply something the virus may need for maturation. Dr. Benirschke: I think, perhaps, we should have the epidemiologic presentation now. Bob, would you continue, please? 112 X—Epidemiologic Approaches to the Origins of Congenital Malformations by Robert W. Miller Dr. Miller: The bibliography that I prepared, for the most part, comes purposely from obscure journals. I think one can make a living at scaveng- ing good articles that have been overlooked. Epidemiology, however, is not quite the blank space that has been indi- cated on the agenda. (Laughter.) And, in fact, it really applies to everything that has been said to date, particularly to the rubella syndrome, which was recognized by a very simple epidemiologic study. Gregg (68), by looking backwards into the histories of mothers whose children had a combination of congenital defects, cataracts in particular, easily identified rubella infection during pregnancy as the cause. His dis- covery had a very large impact on human teratology and on chronic-disease epidemiology. Generally, epidemiology can be classified as descriptive or analytical. Descriptive epidemiology is concerned with the distribution of disease by such variables as age, sex, race and socioeconomic status. While these studies may seem tedious to nonepidemiologists, they can provide interesting clues to the origins of disease. Of particular value with regard to the infectious origin of congenital defects are seasonal patterns or case-clustering in other divisions of time or geography, and changing risks with maternal age or parity. Analytical epidemiology is often underestimated. It is just as much con- cerned with testing hypotheses as laboratory research is, but the epidemiol- ogist must take things as they are, not as he might want them ideally. He must try to recognize in natural occurrences the factors that influence devel- opment of disease. Let us consider an example of how one moves from descriptive to analyti- cal epidemiologic studies. Down’s syndrome was first recognized to be as- sociated with maternal age in 1909 (93), long before the era of big grants. In fact it did not require any money at all to recognize the relationship, since 20 case-histories which gave maternal age would be sufficient. The risk of having a child with Down’s syndrome has subsequently been shown to be about 20 times greater among pregnant women over 40 years of age than it is among women 20-24 years of age (94). Obstetricians have known for decades that the risk of miscarriage also in- creases with maternal age, but no one quantitated or documented the rela- tionship until 1962; and then the study was done not by an obstetrician, but by a statistician through the use of the records of the Hospital Insurance 113 Plan of Greater New York. Spontaneous abortion in women over 40 is about twice as common as it is among women 20 years younger. The risk of childhood leukemia is also related to maternal age, the prob- ability being 40 percent greater if the mother is over 40 years of age than if she is 20-24 years old at the time the child is born (95). The relationship of Down’s syndrome, miscarriage and childhood leukemia to maternal age suggests a hypothesis: this epidemiological characteristic in common has an explanation in common. One possibility pertains to cytoge- netic abnormality, for it has been observed that meiotic nondisjunction is re- lated to maternal age. In Down’s syndrome there is usually an extra chromo- some present from conception or soon thereafter. Among early spontaneous abortions, about 22 percent are characterized by chromosomal abnormalities (78). Childhood leukemia occurs excessively with Down’s syndrome, and there are other errors of meiosis that may well be associated with childhood leukemia (96). It is conceivable that errors in chromosomal division are the factor in common that relates these three disorders to maternal age. The next question is, what causes meiotic nondisjunction? Some people have suggested that infection does. In particular, Stoller and Collman (77) in Australia have claimed that peaks in infectious hepatitis are followed in 9 months by peaks in the births of children with Down’s syndrome. The num- ber of cases of Down’s syndrome in their study was small, the average each year being only 13. The range was from 6 to 20. With such a small sample size, there is a large random error. It is entirely possible that the seeming peaks in Down’s syndrome represent random fluctuations which by chance followed peaks in hepatitis occurrence. The hypothesis that the two diseases were related was tested further by Stark and Fraumeni (97) of the National Cancer Institute. They happened to have available to them information on a large sample of mongoloid children who were born in the lower peninsula of Michigan over the 15-year period, 1950-64. The sample, exhaustively ascertained, consisted of about 2,500 children with Down’s syndrome. In- formation on the occurrence of hepatitis in the same area was obtained from the Communicable Disease Center of the Public Health Service. The results showed that there were two peaks in the occurrence of hepatitis, but there was no sign of a peak in the occurrence of Down’s syndrome at any time during the 15-year interval. There were between 100 and 200 mongoloid children in each year of the study. With a sample of this considerable size, the results of Stoller and Collman in Australia could not be substantiated in the United States. Subsequent reports from Birmingham, England (98) ; Seattle (99) ; and Italy (100) similarly found no correlation between the two diseases. There is at the National Cancer Institute a mathematical genius named Mantel, who has devised a statistical technique to determine if there is a subtle clustering of rare events. The technique was developed for study of cancers, particularly leukemia, but applies as well to specific congenital malformations 114 such as Down’s syndrome. If one subdivided Michigan geographically over various intervals of time, he would expect a certain number of clusters, the subdivision by chance to contain clusters of the syndrome. By comparing this expected value with the observed number of clusters he could evaluate whether or not Down’s syndrome showed any subtle aggregation geographically or temporally. When the observed and expected numbers of clusters were com- pared through the use of a x* test, the value obtained was 0.26 (1.d.f.), a low figure which indicates no significant difference. When the same technique was applied to infectious hepatitis, a known contagious disease, the x* value was 6116, the largest in the history of the Biometry Branch of the National Cancer Institute. Obviously the technique works when the disease to which it is applied really occurs in aggregates. When applied to Down’s syndrome, the test was negative not only with regard to clusters by units of a year but also by months or seasons (101). Thus, even within small time-intervals there was no tendency toward aggregation. Retrospective studies, often maligned by the uninitiated, have made im- portant contributions to human teratology. Rubella has already been men- tioned in this respect. In a similar fashion, by looking back into the maternal histories of children with small head circumference and mental retardation, it was possible, 40 years ago, to implicate as a cause pelvic therapeutic X-ray exposure during pregnancy. Retrospective study also identified thalidomide as a teratogen. The gross findings of retrospective studies were subsequently refined through prospective studies which provided more accurate estimates of the risks involved. The prospective approach involves identification of a cohort of exposed persons as, for example, to rubella infection during preg- nancy. Determination can be made as to the true risk of (1) fetal wastage, (2) congenital defect in the liveborn, (3) the interval of greatest susceptibility during pregnancy, and (4) the more subtle manifestations of fetal injury. Thus, it has only recently been established by such techniques, as mentioned by Dr. White yesterday, that deafness occurs in the absence of other charac- teristics of the rubella syndrome. Mention has been made of retrospective studies in which one looks back- wards, and of prospective studies, in which one looks forward. It seems logi- cal that there should also be “laterospective” studies. There is no such term in textbooks of epidemiology, but there is such an approach. While it may not be popular in the Federal Government because it is cheap, it has been found to be extremely effective. It concerns looking sideways for associated diseases. In this way, leukemia was found to be excessively associated with Down’s syn- drome, the risk being about 20 times greater than usual (102). The question then arose, are other specific congenital malformations excessively associated with specific childhood neoplasms? The approach was simple. The medical charts of 440 children with Wilms’ tumor were reviewed, and all information pertaining to congenital defects was abstracted. The results showed that there 115 were six patients with aniridia (congenital absence of the iris of the eye). The normal frequency is about one in 50,000; in Wilms’ tumor it was one in 73. Also found in excess of normal expectation were congenital hemihyper- trophy (one side of the body, or a segment of it, several sizes larger than the other), genitourinary tract defects and hamartomas (103). The relationship of these 4 categories of congenital defects to one another and to Wilms’ tumor is shown schematically in figure 31. It will be noted that aniridia and congenital hemihypertrophy, though not associated with one another, are each associated with congenital defects of the genitourinary tract. The right-hand side of te diagram suggests that Wilms’ may occur with certain growth excesses—hemihypertrophy or hamartomas, whereas the left- hand side of the diagram indicates that the tumor may occur excessively with ocular (and cerebral) defects independent of growth excesses, the link between the two being congenital malformations of the genitourinary tract. The value of formulating a schema like this is that it indicates how an ad- vance in the understanding of any one of these diseases may advance under- standing of the several diseases with which it is associated. Thus, a virus found to induce pigmented nevi should also be evaluated for its capacity to induce the malformations or neoplasms with which this hamartoma has been linked; and the etiology of Wilms’ tumor can be examined not only in terms of the neoplasm itself, but also in terms of the four classes of malformations with which it occurs excessively. WILMS' TUMOR | 7 CN \ < ’ ANIRIDIA —_ |coneENITAL oeFecTs| _ | coneENnITAL | | PIGMENTED OR | OF G.u.TRACT | HEMIHYPERTROPHY | VASCULAR NEVI Figure 31 Wilms’ tumor and the congenital defects with which it is associated (Miller, R. W.: Yale J. Biol. Med. 37: 487-507, 1965). Another use of epidemiology, already illustrated with respect to Wilms’ tumor, is the definition of groups at exceptionally high risk of specific dis- eases. There are several advantages in the recognition of such groups, as ex- emplified by the study of leukemia. The usual frequency of the disease in the U.S. population is about one in 2,500 persons over a 10-year interval. Accord- ing to the medical literature to date (104) the groups at exceptionally high risk of leukemia and the magnitude of risk are as follows: 116 Category Risk Time Interval Identical twins of children with leukemia.. *209, Weeks or months after first twin falls ill. Polycythemia vera treated with X-ray or 179 15 years. P-32. Bloom’s syndrome. ..................... 1139, First 30 years of life. Japanese atomic-bomb survivors who were within 1000 meters of the hypocenter. . . . 29, 15 years. Down’s syndrome. ..............vnnnn 19; First 10 years of life. Radiation-treated ankylosing spondylitis... 0.49%, 15 years. “Among 22 identical twins with childhood leukemia, the co-twin was affected in 5 instances. 13 of 23 persons with Bloom’s syndrome developed leukemia. There are several features of interest in this listing. It should be noted that in polycythemia vera treated with radiation the risk of leukemia was much higher than it was among the more heavily exposed Japanese atomic- bomb survivors or among patients with radiation-treated ankylosing spon- dylitis. The implication is that in polycythemia vera the radiation treatment alone does not account for the increased risk of leukemia, but that the under- lying disease itself has something to do with the high probability of develop- ing the neoplasm. The listing also contains information of value to the virolo- gist, cytogeneticist, hematologist and immunologist, for it indicates that if they study twins of children who have developed leukemia, there is a high probability (one in five) that they will see the earliest manifestations of the disease as it develops in the co-twin, and so perhaps gain new under- standing of its genesis. Finally, the list provides an opportunity to seek a common denominator among the high-risk groups identified. In this particular example, they have in common a distinctive genetic feature (104). Identical twins have identical genes. Polycythemia vera, in advance of therapy, often exhibits aneuploidy, and after treatment has chromosome breaks as well. In Bloom’s syndrome (and Fanconi’s aplastic anemia) there is excessive chromosomal fragility, a special characteristic of these two genetically transmitted diseases. Long- lasting chromosomal breaks may also be acquired by exposure to ionizing radiation, which is known to be leukemogenic, or to benzene which prob- ably is. We come next to the question of serologic epidemiology, an approach recently pursued by Brown and Evans (50). From 9,000 women, beginning early in pregnancy, they acquired serial blood specimens which were stored until after the baby was born. If the initial physical examination of the child revealed no congenital defects, the blood was discarded. If the child had a congenital defect, the blood was examined for evidence of viral infec- tion during pregnancy. The tests performed concerned several strains each of Coxsackie virus, echovirus, influenza virus and adenovirus. There was serologic evidence of infection with only two of these, Coxsackie B3 and B4, significantly more often among the mothers of children with congenital heart 117 disease (22 cases) than among mothers in the matched control group (nine cases). The big question is, were the differences observed real or did they happen by chance? It is difficult to tell by statistical tests alone, because if one makes enough comparisons, he may very well find that significant dif- ferences occur in one or two of them purely by chance. To distinguish if the relationship is real, the results of this study should be consistent with observa- tions made in the animal laboratory or in other investigations of man. To my knowledge, no such studies have yet been made. The hypothesis can be rela- tively easily tested through the use of epidemiologic data already collected elsewhere. In the perinatal study of the Kaiser Foundation in the San Fran- cisco Bay Area, there are in storage serial blood samples from pregnancies of about 16,000 women, a substantial proportion of whom were first exam- ined in the early part of pregnancy. It should be a simple matter to determine if the blood from mothers of children with congenital heart disease shows evidence of Coxsackie virus infection more often during pregnancy than does a sample of similar women whose children were normal. Now, I had expected to be the first to mention veterinary studies but find I am the last. Fortunately, I can mention them in a different way than other speakers have. In 1961 the National Cancer Instiute established an Epizootiology Sec- tion, whose objective was to study cancers in domestic animals for clues to the origins of human cancer. The excellent veterinarian recruited to head the Section quickly found that he had no data with which to work. Veterinary hospital records, for example, are commonly filed under the owners’ names. Thus, one could not retrieve charts on the basis of the breed or species of an animal of interest or with respect to specific diseases. There are, of course, no vital certificates or special disease registries for animals as there are for human beings. To provide a data resource for epizootiologic re- search, a system was designed for making standardized observations on a large series of animals in a fashion which permitted machine-processing and easy retrieval (105). A brief form was designed on which the examin- ing veterinarian recorded identifying information, characteristics of the animal, procedures performed, diagnoses and the methods by which the diagnoses were made. To make possible machine-processing of the diagnoses, it was necessary to develop a standard nomenclature of diseases in domestic animals. The veterinary profession had long wanted such a nomenclature, but had not yet devised one. It was produced in about 6 months by the group which designed the data processing and retrieval system (106), and has since found wide application both here and abroad. The data system was pilot-tested at the College of Veterinary Medicine of Michigan State University where it was regarded with such enthusiasm that its use was soon adopted by 12 other veterinary medical centers in the United States and several overseas. In one year at Michigan State University the data 118 collection system revealed that of 6,156 diagnoses in all species, 6.3 percent were attributed to genetic or other prenatal influences (107). Included were such diagnoses as dislocated hips in German shepherd dogs, slipped patellas and microphthalmos in poodles, diaphragmatic hernia in a Basset hound, branchial cyst in a dachshund, and a substantial variety of other defects among other breeds and species. Not only is it possible to make epidemiologic studies on the basis of this resource, but through it one may also locate cases of special interest for laboratory research pertaining either to etiology or to therapy. In contrast to research in man, it is possible in veterinary medicine to manipulate breeding to determine the influence of genetic or environmental factors in the etiology of congenital defects. Finally, epizootiologists had to find a statistical procedure that avoided the complication of not knowing the population at risk. It is difficult to esti- mate how many dogs or cows or cats there are in a community. To determine whether or not certain breeds have a greater than usual predisposition to some disease, the relative risk method was adopted since it permits comparisons to be made when rates cannot be determined. The method has been exem- plified in a study of bone sarcomas in dogs (107). A collection was made of case histories concerning about 400 such neoplasms diagnosed at 14 veteri- nary medical centers. The relative risk among the smallest breed was regarded as 1.0; the other frequencies were measured against that standard. In the large breeds, bone sarcomas were seven times more frequent than in the small breeds and, most surprisingly, the relative frequency among giant breeds was more than 100 times greater than in the small breeds. This finding may, of course, have implications for the corresponding cancer in man. My purpose in this presentation has been to emphasize that epidemiologic research can present a variety of approaches to the origins of congenital de- fects; that it does not have to be boring or costly; that it can test hypotheses that are being developed in the laboratory or through clinical observations; that it can generate hypotheses; and that it can often perform these functions very quickly through the use of existing data if one knows where to look for suitable resources. It seems to me that what is really needed in teratology, whether related to viruses or not, is a research group prepared to use whatever epidemiologic method is most appropriate for generating new information. The unit should be diversified and fast moving, ready to use local, national or international resources to determine if ideas developed in the laboratory or from individ- ual patients have general applicability in man, or to develop notions that can be elucidated by other investigative approaches. Dr. Benirschke: Thank you very much, Bob. Dr. Mitchell: May I fit in a footnote in relation to Bob’s comments on the Brown and Evans study? (50) 119 As Dr. White mentioned yesterday, the Collaborative Study on Cerebral Palsy and Mental Retardation has serological samples from the pregnant women enrolled in this study. While only 20 percent of them were seen in the first trimester, 60,000 women have now been enrolled in the study. Two years ago Dr. Sever performed an antibody screen for some 30 viruses using maternal sera of 27 patients with congenital heart disease, and 53 matched controls. Most of the 27 patients had lethal forms of congenital heart dis- ease and so, undoubtedly, differ from the total group whose identification is only now being completed. All the mothers were first trimester registrants. Hence, these patients are very similar to the group studied by Brown and Evans. We did not find any significant correlations between maternal viral infection and infant congenital heart disease. This is not to say that there may not be an etiological relationship of this sort but to identify it two things are needed. First, we need to be able to group the cases in a meaningful morphogenetic way by aberrant precursor structure or cells. Secondly, we need to be able to eliminate antibody crossover be- tween viruses and to find some good way of identifying mothers with viral infection during pregnancy, so that we can work from two starting points, the mother with infection during pregnancy and the child with a specific mal- formation. It is not realistic to believe that a battery of 30 viral antigens can be assessed for 60,000 women. Some way must be found of focusing in upon the high risk groups. However, after the abnormal offspring have all been clearly identified and when etiologic hypotheses are formulated, full use can be made of this resource. Dr. Miller: This illustrates the use of an existing data resource to test ‘a notion or a question from some other study. Dr. Gruenwald: One advantage of a fast moving study would be a rapid monitor function, to prevent what happened with thalidomide. Dr. Eichenwald: It would seem to me that if an infectious agent which follows a strong seasonal pattern produces congenital malformations, one should expect to observe some temporal pattern to the incidence of those malformations. I would guess that in Michigan, Coxsackie viruses occur at a particular time of the year, much as they do in other parts of the United States, and since no seasonal variation was observed in the incidence of malformations, this would make me doubt that a teratologic effect of Cox- sackie viruses has been demonstrated. Dr. Miller: This is what I meant by consistency with other observations. Your comment refers to an inconsistency. I think that the congenital defects most appropriate for study would be those that do vary seasonally or by other intervals of time, as anencephaly does. And, if I remember correctly, so does congenital dislocation of the hip. Dr. Benirschke: Does anencephaly differ seasonally here too? Dr. Miller: In Scotland it did (108), but not in New York State (109). 120 Dr. Gruenwald: It is amazing with the fantastic variation of instance of anencephaly from valley to valley in South Wales how little has actually been learned by doing this kind of work. It is apparently a matter of some- body—it happens by chance, or because he is very smart—associating this with the right kind of thing. As far as I can tell, this has not been done. The variations are so fantastic that one would think one ought to come upon something. And yet it has been so disappointing. Dr. Miller: I wonder if the next step isn’t up to the laboratory people: to study a series of women who have had such children to determine, if they can, what the virology or immunology or whatever -ology it is, might have been responsible for the defect. Dr. Gruenwald: Wasn't there something done with folic acid in this ? group’ Dr. Eichenwald: In Wales, does a mother give birth repeatedly to anencephalic infants? Is there a familial incidence ? Dr. Miller: In New York State, yes (109). Dr. Eichenwald: But is there in Wales? Dr. Miller: I don’t know whether it has been studied there. Dr. Gruenwald: There is apparently a relation to social class in Britain. Dr. Benirschke: Your reference to meiotic errors for these three cate- gories bothers me. It always does for maternal age when mongolism is brought up, and you even include leukemia and abortions. Maybe Dr. Moor- head can say something on this. I personally believe that data on mongolism, but certainly those on other chromosomal errors, are not as yet firmly convincing that they are secondary to meiotic nondisjunction. In fact, I think there is increasing concern that mitotic rather than meiotic errors are the base of many chromosomal disorders. You are lumping three conditions together, leukemia, mongolism and abortions. It makes me suspicious because in the abortions, perhaps even more so in some of the sex chromosomal disorders, there is a very high frequency of mosaics, which is good evidence that this is due to postzygotic, mitotic, nondisjunction. I am trying to say that this is important to keep in mind and that we have to come to grips with what can be distinguished as being meiotic and what is mitotic. Dr. Moorhead: I don’t remember that there was any clear distinction as to the ratio of mosaics in the studies of Carr (110) and in those on general newborns. Do you? For technical limitations I don’t think you can really define this difference very well. There is a certain amount of mosaicism in both, a rather large amount. 121 We haven’t any chance of studying anything like this in man, and if we are correct in making any inferences, the chances are that some of those lines would die out and would never be captured for study; you would be left with an X-O0 line. It was surprising two years ago when LaChapelle (111) and others found that when they studied a large number of Turner-like women, that half of these were mosaics; the more people studied all kinds of individuals, the more mosaicism was found in what was reckoned to be a rather clear-cut syndrome. Also 90 percent of your reported cases are based on blood cells alone. Dr. Gruenwald: In support of this, I think there are one or two reason- ably well-documented cases of monozygotic twins one of whom was a mongoloid. Dr. Benirschke: Five, and the implication is that this is a disturbance of early zygotic life, say, in the tube, which may be considerably altered by environment or old age. I was just told in the local (NIH) study of abortuses, triploidy is the most common condition. Carr (112) and others have had an excess of X-0 abortuses. Dr. Miller: Did you try to find the original reference? I couldn’t. It is said to be Mather in 1938. Dr. Moorhead: Don’t you have evidence in human populations? You don’t consider the ordinary age effect in relation to Down’s syndrome? Dr. Miller: All we know is that there is an age effect but we don’t know the reason. You have to tell us the reason. Dr. Moorhead: I don’t think we know either. Dr. Miller: The point I wanted to make is, whatever the reason, it may be common to all three. Dr. Benirschke: With respect to hepatitis it is very easy to talk about it now. If we had sat here four years ago, with the rubella virus isolation, a lot of worthless speculation might have gone on. Really the first order of business is to isolate the hepatitis virus and then life would be a lot easier. Dr. Katz: It might, and then it might not. This will depend on whether it is an hepatitis virus or a group of hepatitis viruses. Life might be more complicated than presently. It seems to be that there is a definite role in any epidemiological approach to congenital malformations to be played by the National Communicable Disease Center. Their Epidemic Intelligence Service (EIS) maintains close contact with all state and territorial health departments so that they are aware when epidemics or clusters of infectious disease occur in a given area. They maintain surveillance of the very same illnesses with which we are 122 concerned as possible etiologic factors in teratogenesis. In many states, they have officers attached for duty periods with the local health departments so that their antennae are particularly acute and almost ubiquitous. Somehow there should be a mechanism so that some fast-moving epidemiol- ogy group concerned with malformations could be alerted by the EIS of an outbreak of one of the infections which you select to follow. If there is hepatitis from Raritan Bay clams or ECHO 9 in Milwaukee, one can be certain that some first trimester ladies will have consumed infected clams or are entero- virus-susceptible. If EIS notified the malformation epidemiologists of such an episode, the latter could quickly move into the affected area and follow the child-bearing age group among whom they would undoubtedly find some who were both infected and pregnant. Dr. Miller: The Communicable Disease Center is doing just that for leukemia clusters. Its officers are in various parts of the country and as soon as a cluster is reported, the Communicable Disease Center makes an investiga- tion. There is no such plan at the moment in operation for congenital defects. It is of interest to recall that in the CDC’s own city, Atlanta, there was an “epidemic” of meningomyelocele. From August-October 1962, 16 cases were observed as compared with 3 expected (113). With all their laboratories and specialized personnel, CDC could not find the cause. Dr. Katz: By the time anencephalus has been noted, it is too late for the type of approach I was proposing. My thought was that at the time of an acute outbreak of an infectious disease serum and appropriate cultures would be obtained from women in the child-bearing age who were known to be pregnant and who were not but might be very early in a pregnancy. Careful follow-up over the ensuing 6 to 9 months would be conducted to ascertain the outcome of all pregnancies—abortions, stillbirths, live births, malformations. Correla- tion of pregnancy results with viral isolations and serology might then shed some light on the role of that particular virus on the embryo and fetus. Dr. Benirschke: They might never consider blue tongue virus as a possi- bility. Assume this had been hydranencephaly. Maybe this is as good a time as any to make some recommendations. One of the things one might recommend is a central reporting facility. Such a “clearinghouse” has been formally proposed before (114) in a meeting of scientists with interested industry shortly after the thalidomide event. A plan was even published, that there should be a center in this country to which 10, 15, 20 strategically located obstetrical hospitals would report monthly their statistics on births, deaths, abnormalities, premature rates, and whatever parameters one might include. Further, that someone in the center or a computer would screen these data continuously. If in Boston then there was a clustering of rare events, say anencephaly, an investigation would be started. Pretty soon this would shape up and, after a few trial balloons, it could be worked out just where the signal for alarm 123 293-596 0—68——9 should be set. In any event, such a system could be worked out and I am all for it. It is quite meaningful to the whole country because anticipation of any of these events involves great cost. It could screen teratogens, and infection trends and all kinds of things could be fed into this at very small cost. Perhaps this is a mammoth task of organization, but one in which people like Bob Miller might be interested enormously. Dr. Mitchell: Was there not a study started along these lines in New York? Dr. Benirschke: It folded up. It was in Philadelphia and was an entirely different thing. That study had much difficulty finding support and eventu- ally did not find enough money. It produced all kinds of data, and was not really as discriminating as we would like it to be. But the principal problem with that study was that it employed feed-in from numerous very small hospitals with very irregular and spurious reporting from local committees which may or may not have been very discriminatory. The study did not address itself to this same task but just tried to help the local community hospitals to pull themselves up by comparing their death rates with other death rates in the country, and so on. This was sponsored initially by the AMA, and the small community hos- pitals throughout the U.S. reported to this agency. Dr. Gruenwald: It was started by the AMA, their perinatal study, and the AMA finally refused them funds and they became an independent group. They produced all kinds of forms. But this was mostly obstetrical, perinatal. Dr. Miller: The Commission on Professional and Hospital Activities in Ann Arbor has about 250 member hospitals who submit summary reports on every discharge. The data are not the best in the world for research, but they might serve certain screening purposes. Dr. Mitchell: In newborns they would readily pick up skeletal or external abnormalities; kidney or cardiac malformation would rarely be identified. Dr. Gruenwald: They can only pick up what is fed into them, of course, and then it becomes a problem in how much detail can you try to do this. There is a limit to this, I suppose. Dr. Mitchell: It is not really just a question of detail. It is the acumen and interest of the diagnostician. Dr. Eichenwald: It would seem to me this type of study can be done, perhaps by some of the large prepaid medical groups such as the Permanente Foundation because these have large groups of women seen early during pregnancy. The women and their children are followed closely. If one com- bined information from all the Permanente plans in California, this alone would give one a sample of sufficient size. Doing what the Philadelphia group 124 tried to do, under circumstances where the accuracy of diagnoses is variable and background noise very high, would not result in clearly interpretable data. Dr. Miller: They would have to be adapted, because they can’t retrieve the cases by diagnosis at present. Dr. Gruenwald: There is, of course, one other point, and that is that this country is perhaps less suited for such investigations than many, say, Euro- pean countries, which on account of their health care systems, and also on account of more stable populations can do much better. To give you an example, the British perinatal mortality survey which not only managed to survey I think close to 99 percent of all births, regardless of where they took place—and half of them did not take place in hospitals— during a week in 1958. They decided six years later to find all these children, and they found almost all of them again without ever having previously arranged for follow-up. So the opportunities for doing some of this might be much better in other countries than in our country because of the different health system and also because of their more stable populations. Now, this is at the moment, for entirely technical or unscientific reasons, very difficult for us to support, I understand. But I think one should keep in mind that there may be better places to do things like this than in most of our populations here. Dr. Benirschke: Bob, your raising the Bloom’s syndrome brings to mind the many speculations we have undertaken at home how one might study this disease. Maybe Dr. Katz can chip in on this discussion. As you phrased it, the disease is a genetically transmitted disorder of chromosomes. If this were due to a virus that makes chromosome breaks, and I assume that you expect to find, or hope to elicit the presence of a virus that is vertically transmitted, how would you go about finding such a virus? Dr. Moorhead: Levitan (115) has an excellent parallel to this in Droso- phila; it is an instability which is transmitted only through the egg. Dr. Benirschke: How do you go about seeing if there is a virus? Dr. Katz: You could start by looking at ova from the mother of such a patient. If this were vertical transmission of an RNA virus, then both mother and infant might very well continue to shed the agent for long periods of time. How does the maternal ovum become infected in the first place? Pre- sumably she must have a viremia which seeds the ovary. The animal models of vertical RNA virus transmission are all marked by excretion of the agent both by parent and progeny. On the other hand, a DNA virus might well have achieved such a high degree of integration with the host cell genome that no complete virus can be found. The current techniques for ferreting out these agents center around the detection of new antigens in the host cell 125 which are virus-directed and the analysis of host cell RNA looking for mes- senger RNA which is complementary to viral DNA and is not found in normal cells. The first approach utilizes immunofluorescent methods and comple- ment-fixation testing while the latter depends upon biochemical separations and nucleic acid hybridization experiments. Both are fruitful, if one has some idea which virus he suspects as the original culprit. But with no knowl- edge of the possible agent for which one seeks clues, the possibility of nearly 300 different viruses to be screened makes the whole prospect rather unappeal- ing. Has anyone studied these patients in some such way? Dr. Benirschke: I don’t know. Dr. Katz: It’s easy to be a pessimist and to assume that you won't find anything. Dr. Miller: Do you mean that in Bloom’s syndrome the presence of the detrimental gene lowers resistance to an infectious agent transmitted in the ? ovum? Dr. Moorhead: You can just say that the agent causes a mild amount of breakage; the agent itself induces breakage as small numbers escape or ma- ture. But you may never find a mature virus in it. You also may have simply mutant genes on particular chromosomes which might just be running through this stock, genes that directly induce the cancer. By backcrossing plants you can shuffle the chromosomes around until one chromosome is obtained and that particular chromosome, with a particular background, can increase the plant’s tendency to pathologic growth (equated to cancer) sixty-or-eighty fold. You just don’t know. This is the same problem that you face with the Burkitt’s lymphoma. One can try interference with other viruses; the Henles (116) have some evidence for this. Dr. Katz: Another angle would be to examine the gametocytes under the electron microscope. This has been fruitful to some workers investigating the Burkitt lymphoma. With only five patients to study, this might not be too overwhelming a task. Given an electron photomicrograph with parti- cles that resemble a virus, what comes next? As one who does not do E.M. work himself, I remain a bit skeptical of shadows which are named or in- terpreted with no demonstration of their biological activities. However, the appearance of such particles may offer some direction for the others who are pursuing some of the techniques which would be facilitated by a clue as to the type agent suspected. Rather than serologic and virologic screening of hundreds of agents, it might then be possible to concentrate on 39 adeno- viruses or a dozen myxoviruses, if the particles had the structural character- istics of a particular group. Paul Moorhead seems to have the same reluctance as I in sending some- body off to find the needle of DNA in the haystack of chromosomes. 126 Dr. Eichenwald: In order for us to make significant advances in this area, we are going to have to await the development of new methodology and new approaches in superficially unrelated fields. For example, work in virol- ogy, and the development of new immunologic techniques has had a pro- found effect on cancer research. Similarly, new methods are developed in fields different from those we are talking about, these can be applied to try to solve the problems of teratology. The blind and repetitive approach of, say, looking for some antigens in eggs is going to be a fantastic waste of time. Dr. Katz: Were you to announce that you were studying this most extraordinary family in Hanover, N.H., and invited collaborative investiga- tions from other labs, it would seem reasonable and relatively easy to estab- lish some cell lines in culture from their tissues and to distribute these cul- tures to any competent research worker who thought he had some special gimmick with which to try to help solve the riddle. Dr. Eichenwald: That is literally trying to shoot a fly with a machine gun on the chance that one bullet is going to hit that fly. Dr. Benirschke: But if everybody knows about the existence of such a syndrome somebody might just happen to say, “My God, that was just what I needed to come around in my work.” Dr. Eichenwald: That is why we all have to stay alert as to what is going on in other fields. Dr. Miller: I think that applies not only to the work of different labora- tories but also to different research approaches: clinical, epidemiological and laboratory; the greater the interaction the greater the gain. Dr. Gruenwald: Could it be that the vertical transmission of virus toler- ance might make things more difficult? Dr. Katz: There are a number of models in the animal systems which exemplify the vertical transmission-tolerance relationship. These include LCM (lymphocytic choriomeningitis) of mice and the avian leucoses. Dr. Gruenwald: You might not find the antibodies then. Dr. Katz: That is correct, but in the known examples when antibody is not present, virus can be found with relative ease. With some of the avian lymphoma models, viral interference was the clue to the congenital infec- tion. Cell cultures from infected chicks could not be altered by Rous sarcoma virus (RSV) while those prepared from non-infected embryos were suscepti- ble to RSV. Thus, in the absence of antibody there was still a protective mechanism mediated by the congenital infection. Once again, I think that some of the biochemical approaches exploited by Green and Associates (117) in St. Louis will continue to contribute fundamental knowledge in the area of “latent” or “integrated” viral fractions. Although these are not adaptable to mass screening presently, it is conceivable 127 that they may someday be developed for such use if basic investigations have established their fruitfulness. Looking for the evidence of viral nucleic acid influence by examining the messenger RNA of suspect cells is an imaginative and elegant way to search for an oncogenic virus. This was Green’s original goal, but the success in that area may augur well for similar approaches to the malformation problem. This is a good example of the need to support competent investigators whose research may not seem initially to have developmental applicability but which eventually opens an entirely fresh avenue to our view. Dr. Axelrod: It seems that the real difficulty is to be able to spot the more subtle congenital difficulties that ensue due to viral or other infections. The subtleties in functional change may be reflected years postnatally, many years. I wonder if, instead of going from the top down, which is the epidemio- logical method, there also should not be a bottom up. I think that the vi- rologists have clinically been able to categorize a large number of viruses. Through the use of animal models whose susceptibility to human prototype virus is known, an in-depth study could be made of the pathology or the abnormal biochemical functions of these model animals which have been infected as fetuses or as ova or sperm. This sounds like a Herculean task but it really isn’t when you consider that it might lead to a predictability—after all, predictability is the thing we are looking for—which would not be possible by an epidemiological route. Although in the end I think they would fuse beautifully, because if you could have a chart of types of viruses and types of morphological and functional dysplasias which these viruses produce during the gestational period or before that to the egg and sperm, that then this type of chart could be extremely useful in prediction and presentation. Dr. Katz: While you were talking, I found no difficulty in drawing up a list of eight basic groups of viruses which include 250 members capable of infecting the human host. It would be a truly Herculean task to screen them all. Dr. Axelrod: You have got to weigh the gigantism of the job against what you are going to get out of it, not just now but in the next couple of generations. Dr. Katz: Although I completely agree that this would be wonderful were it done, who wants to do it? Dr. Axelrod: If you don’t do it, this kind of conference will exist many generations from now with all the pathologists running out trying to get specimens and identifying some unfortunately subtle functional change, like not being able to do arithmetic in the sixth grade because of a virus that hit the kid in utero some 9 to 12 years before. We are going to have to plan now to understand this functional change. 128 Dr. Benirschke: If you do away with studying poliomyelitis, we will never understand it completely, but then maybe it is not necessary to under- stand what really goes on. Dr. Axelrod: That is a big “if.” Because there are viruses that we don’t know about yet. Dr. Benirschke: But there are not many people who are going to study poliomyelitis anymore. Dr. Axelrod: That will cut off a large segment of knowledge. Dr. Benirschke: That is why we are continuously rediscovering old articles. Dr. Silverstein : Let me make one comment with respect to the magnitude of the task that Dr. Axelrod suggests. In our interest in understanding something about the pathogenesis of two intrauterine congenital diseases, the blue tongue and the congenital syphilis models, our interest has been primarily in understanding the fetal immunology and pathogenesis of the disease process. I have roughly calculated that it would take me a hundred animals to begin to get an understanding—hope- fully a more or less satisfactory understanding—of either of these two entities, with respect to the development of the disease as it is affected by age, by dose, and by a number of other parameters. The point to be made here is that these 100 animals are 100 successfully infected animals, allowing for a little bit of wastage, not 100 animals where we put material into the mother and get perhaps 20 percent transmission to the fetus. These are instances in which we can successfully and directly infect the fetus. This is obviously a figure pulled out of a hat, but this is certainly the order of magnitude of animals that would be required. Dr. Katz: You raised the same point and I would agree that from your perspective it is also important. When we still remain uncertain of the pathogenesis of an intrauterine infection with rubella, it is misleading to assume that injecting virus directly into the fetus is the same as putting virus into the mother. Dr. Silverstein: This is one of the points I think that one should bring out. Dr. Axelrod: I think it is the level at which you are working that we are discussing here. You are working on a more fundamental level of under- standing as to how a disease occurs or immunological responses, how they occur. What I want to know is, do other viruses cause malformations, and if so which ones? Dr. Margolis: While this is all very important, we must remember that what we have done so far is take an intense look at but one side of the moon; now we must also look at the other side. I think we ought to keep alive the 129 contrary hypothesis that none of these things we are talking about are of viral origin. We must consider a genetic approach. I have been impressed with how easy it is to bring out genetic abnormalities which are quite parallel or identical with those in man by, simply, brother-sister, mother-son matings among animals. Maybe Bloom’s syndrome would be much better approached that way. Dr. Miller: I have two comments now instead of one. Yes, you could do that in the veterinary field when you recognize congenital defects. You can’t do it with Bloom syndrome so well, because you can’t mate the people according to your will. Dr. Margolis: My contention is that you would find Bloom’s syndrome eventually if you did this in animals. Dr. Miller: Then you can mate them at will can’t you? But I wanted to ask the virologists, can they detect from the experience to date, say with rubella virus, what other teratogenic viruses are apt to have their characteristics? So one does not have to test all the viruses in the world, but can make a good guess as to which are most probably teratogenic? Dr. Katz: Unfortunately, it would only be a wild guess and this may reflect the ignorance of the field today. There are some ways of using present knowledge of offending agents to venture a prediction. Rubella fits somewhere in the myxovirus family—RNA nucleoprotein with a lipoprotein membrane, helical symmetry, medium size, hemagglutinating antigen, etc. There are a number of other viruses which share these same properties but on the basis of present clinical and epidemiological experience we would not select them as likely candidates to implicate in teratogenesis. They include the influenzas, the parainfluenzas, mumps, measles, Newcastle disease and respi- ratory synctial virus. Similarly, a survey of the properties of cytomegaloviruses would lead one to investigate all the members of the herpes virus family because they share most of the same characteristics. Dr. White: I think it is not quite as grim as that. If you are looking for a candidate virus, it can’t be one virus which is going to destroy the cells. Dr. Katz: This depends almost completely upon which cells you choose to discuss. Dr. White: If you have a virus and put it in tissue culture and the cells are totally destroyed, that virus is probably a less likely candidate than an agent like rubella which does not have so devastating an effect. Dr. Eichenwald: You are using ad hoc post facto reasoning. You are saying that rubella is teratogenic because it has the right properties in tissue culture. There is no assurance that another teratogenic virus might not kill every cell in tissue culture. 130 Dr. Katz: A tissue culture tube is in some ways an artificial system, or at best an incomplete one. Destruction of the cell sheet by a virus would exclude Herpes simplex, group B Coxsackies and cytomegalovirus as agents for con- sideration. And yet, aside from rubella, they are the only ones where any sort of established relationship to intrauterine damage seems clear. We all enjoy extrapolating our test tube model but there are too many other factors operative in the maternal-fetal-placental system which can not be duplicated in vitro as yet. Dr. White: But you need indicators, and you do the best with what you have. Dr. Eichenwald: One is obviously attracted to those facts which best fit one’s theory. For example, the rubella virus inhibits cellular multiplication in tissue culture, but even if it didn’t, it wouldn’t make any difference in terms of the demonstrated effect it has on the fetus. It may just be an accident that these two things happen in parallel. There are other viruses that act similarly in tissue culture, but they do not produce teratogenic effects. I agree completely with Sam, that there is no way that we can sit down at the present moment and select from the long list of viruses an agent which we could predict on the basis of experience with hog cholera vaccine that measles and yellow fever vaccines might be teratogenic. Yet there is not a shred of evidence that this is so. There is a whole series of events that has to occur before a virus is capable of producing anomalies. Some of these events may be related to adult immunity or to placental infection or to whether the virus persists sufficiently long in the placenta. Rubella terato- genesis perhaps occurs because, by chance, all the various requirements are met. Dr. Benirschke: Let’s have the summary. 131 XI—Summary of the Conference Presentations by Heinz F. Eichenwald Dr. Eichenwald: I will not attempt to summarize everything that has been said, but will restrict my remarks to those aspects of the discussion which bear most closely on the central problem of this conference; viruses as a cause of congenital malformations. Infection with certain viruses during pregnancy may result in five separate but related deleterious effects in the unborn child: (1) abortion, (2) fetal death and later abortion, (3) a teratogenic action sufficient to kill the fetus and thus observable only after delivery, (4) intrauterine infection with complete recovery prior to birth, and (5) intrauterine infection with disease recognizable at birth or shortly thereafter. Infection of the fetus can theoretically occur by several different routes: by extension of a disease process involving uterine tissue, the amnion and eventually the fetus, by an ascending infection from the vagina, and the cervix into the uterine cavity, by contact of the fetus with infected tissues of the cervix or labia during the birth process or by transplacental trans- mission of the pathogen. It would appear that from the standpoint of abor- tions and teratogenic infections, transplacental transfer is most important, but the mechanisms involved are far from clear. For example, even though viremic infections are common in adults, the fetus is rarely affected. Is this protection due to a placental barrier operating at various levels of efficacy? Is it due to the fact that it is necessary for a virus to be able to set up a nidus of disease in the placenta which may mechanically rupture across this theoretical barrier? Is it related to the particular cellular component of the blood with which the virus is physically associated, such as the white cells, or the platelets? Is it due to a combination of factors which must operate concurrently ? Basically, it would appear that we know very little about these quantitative and qualitative functions of the placenta which would permit it to act so selectively. One fascinating area for study would consist of an examination of the local defense mechanisms within the placenta, perhaps antibody- related, or of a cellular nature. Plasma cell-like cells do appear in the villi. Do they form in response to antigenic stress? Do they produce anti- bodies locally? Is their presence or production totally unrelated to immuno- logic function? We do know that some viruses as well as other infectious agents may pass across the placental barrier with ease. This ability is not necessarily related to particle size. In this connection, it should not be assumed that a deleterious effect on the fetus can only be produced if the pathogenic agent manages to invade the conceptus directly. It appears reasonably well-established that abortion and fetal death may occur in the absence of direct fetal involvement since a 132 number of products of conception have been examined virologically without agents being isolated from fetal tissue, even when the fetus was expelled at the height of an acute infectious maternal illness. Abortion, premature birth, and miscarriages are most commonly asso- ciated with those viral diseases that are accompanied by severe toxemia and febrile responses. The best recognized examples are variola, measles, polio, and influenza. In these conditions, the rate of fetal deaths is generally propor- tional to the severity of maternal illness. Furthermore, we must consider another possible mechanism. A viral infec- tion, an infection of any type, may produce a change in placental function which, for a variable period of time, deprives the fetus of his more or less optimal environment. Is it possible that such a disturbance of relatively short duration which does not affect the fetus directly could produce permanent developmental retardation in the embryo if the stimulus is applied during some crucial time in intra-uterine development? Recent studies on growth in infants, malnourished for relatively limited periods of time would suggest that such a possibility should be investigated. One of the most striking, theoretically disturbing, but practically fortunate features of viral teratology is the fact that teratogenic action in human beings seems to be limited to the single virus rubella, an agent which, in the child, or adult, rarely causes severe illness. Is this due to the fact that this agent is able to surmount various obstacles offered by the early placenta in the human, whatever they might consist of, singly or in combination? This does not appear to be a likely explanation. Rather, a more logical tentative explanation would be that this disease is associated with congenital malformations for the very reason that it is so mild—the virus, when acting in the fetus, only damages some cells engaged in organogenesis, rather than killing the fetus outright. Other viruses, of greater disease-causing potential when affecting the early embryo, might well kill it and thus cause abortion. This point could be clarified by a more systematic study of early human abortuses, or through the use of well- selected animal models. Some support for the rubella hypothesis just outlined does exist in the finding that hog cholera vaccine, when injected into pregnant sows during the first month of pregnancy results in a high incidence of damaged piglets. The potential capacity to cause fetal damage is apparently a direct result of the attenuation of the virus. A consideration of the hog cholera model could lead to the conclusion that attenuated vaccines can be hazardous to pregnant women if given during the first trimester. There is no support for this view, except in the isolated report of MacArthur (124) who found that of 34 pregnant women vacci- nated against smallpox during the first to third month of pregnancy, only 18 had normal infants, while there were 4 abnormal babies in 169 mothers vaccinated after the fourth month. So far no similar examples have been presented for attenuated polio, measles, or yellow fever viruses, 133 The mechanism of rubella action in producing a teratogenic effect is unknown. It is possible that the virus affects only those cells most rapidly multiplying at the time the embryo is affected. Rather than a specific tissue affinity, the virus may possess a more general affinity for cells while they are in the particular biochemical state required for rapid proliferation. It is of interest, therefore, that in some tissue culture systems, rubella virus does not destroy the cells, but inhibits their multiplication, which would nicely—perhaps too nicely—explain the observation that rubella babies have fewer cells than normal infants in some of the organs where cells have been counted. There is no doubt, however, that there is evidence of virus multiplication within the embryo. This may indicate that the anomaly results from direct infection and destruction of rapidly differentiating cells, but other possible modes of action may be considered. For example, disturbed growth might result from disease at the terminal branches of the capillaries nourishing these organs, or perhaps, as Dr. Silverstein pointed out, by a destructive, inflammatory manifestation of immunity, an immunopathologic event affecting some tissues more than others. Perhaps the virus, upon entering the cell, alters its genetic material, the physical expression of which might be an alteration of the chromosomal structure or arrangement. This latter possibility appears to me to be remote. There is no evidence that rubella babies have abnormal chromosomal patterns. Furthermore, there are no data which would suggest that the high proportion of dis- ordered chromosomes found in spontaneous abortion material is related to infection. It would seem useful, however, to suggest that the next person studying chromosomes in abortuses do virologic studies of the material as well as scrutinize the placenta. The concept formerly existed that the embryo, when infected, was unable to react or to mount a defense—this view has been thoroughly dispelled. With suitable stimuli, the embryo appears to be able to respond remarkably early, both with antibody and cellular components. It is, in fact, possible that in some cases, in order for the disease to develop, an immunologic or cellular response is required, as in syphilis, in LCM and perhaps in other diseases. The role of immunity in the pathogenesis of fetal lesions and disease is a fascinating and challenging aspect of the problem which obviously requires considerably increased attention. A more difficult aspect for study is pre- sented by the possibility that fetal susceptibility to infection may well be governed by as many factors as that of the mature individual. Another major avenue of approach has been opened by the investigations with the rat viruses. One must obviously consider the possibility that agents of this type can play a role in human disease, perhaps producing specific lesions in the fetus which might fall into the classification of congenital 134 malformation, or perhaps “degenerative” diseases, such as progressive ataxia, or Schilder’s disease, etc. We should determine promptly whether the known rat agents cause human infections with any degree of frequency and whether the human is burdened with similar or related viruses. Obviously, the possibility of the existence of totally unrelated agents as a cause of congenital malformation in man must be kept in mind and diligently explored, using the most advanced virologic techniques. Some of the agents involved may be viruses encountered among lower animals but which may also be capable of infecting human beings, such as blue tongue, etc. While discussing infections of the fetus, one should mention some of the true congenital diseases such as syphilis, toxoplasmosis, cytomegalo- virus disease, herpes, Coxsackie, and so on, not excluding that strange puzzle, hepatitis. To make a few brief points: it would appear at this point in our knowledge that herpes and Coxsackie are generally acquired by the fetus late in gesta- tion, perhaps most commonly during the birth process or postpartum, while toxoplasmosis or perhaps syphilis and CID can be acquired at almost any time during pregnancy. With toxoplasmosis it is known with reasonable certainty that infection in the first trimester usually produces an abortion; while in the second trimester, diffuse disease occurs in the fetus, which may be more or less healed by the time of birth, leaving only central nervous system residua. If the organism is acquired late in pregnancy, a generalized illness will result affecting all organ systems. CID seems to work similarly. With syphilis, lesions occur perhaps only when antibodies have been produced by the fetus. The pattern of disease with these organisms thus depends on the time in utero when the fetus is infected. A clear distinction should be made between the true teratogenic action of rubella virus and the congenital anomalies produced by toxoplasma, treponema, and cytomegaloviruses. The defects in the latter instances are due to specific destructive pathologic lesions, such as inflammation and necrosis in and around the aqueduct of Sylvius, rather than to an effect on organogenesis. It is also possible that congenital infections with viruses, both known and unknown, exist which are asymptomatic for months or years, but eventu- ally result in some degenerative disease. Practicality, ethics and other factors prevent many of the problems presented by fetal-viral interaction from being adequately investigated in the ultimate source of our concern, the human. Suitable animal models must thus be sought, but our discussion has suggested that some of the models presently used were employed not because they seemed to reflect closely the human situation, but rather because they were accessible, and cheap. Models, per se, have only a limited usefulness. Nevertheless, it would appear essential 135 to attempt to define some model or models which most closely resemble the various features of reproductive and fetal physiology of the human. The baboon, in many ways, seems to be preferable to the rhesus, who in turn is preferable to the rat, or the chick embryo. Do we know for each purpose which animal is best? Once this point is defined, experiments can be designed with some probable relevance to human problems. In summary, it would appear that as far as can be determined at the present moment, viral infections contribute only a small proportion of congenitally malformed children, that the majority of the other etiologic factors remain unknown, but that the most likely way some of these can be defined is by the application of epidemiologic methods, with many different disciplines contributing bits and scraps of information. It would seem that “break-throughs” are going to come as improved methodology is developed in seemingly unrelated fields, such as tumor virol- ogy, immunology, cytogenetics, etc., with a subsequent application of these new methods to the problem at hand. (Applause.) Dr. Benirschke: I think this is the first summary I have ever heard that really was of great value. Dr. Miller: I thought he must have written it before the conference. (Laughter. ) Dr. Benirschke: I think the greater problem is for us to do a similarly admirable job to make suitable recommendations to either of the Institutes or to ourselves on what we would like to see done, either by us collectively or individually at home. Dr. Silverstein: It would be worthwhile, before we consider that aspect to hear some additional comment on what really is the significance of viral infection in fetal malformation and embryopathy. Dr. Benirschke: You mean quantitatively. Dr. Silverstein: How big is the problem? How significant is it considered of itself, and in relation to other mechanisms of malformation and embryopathy. The impression I have had during this meeting is that it is only important with respect to rubella, which is not going to be a problem very much longer. Otherwise it may or may not be big. We don’t really know. Dr. Benirschke: Right. Let’s take that narrow portion. Dr. Margolis: Would you add smallpox to this list, tentatively? The recent smallpox epidemic and the mass immunization program in Kuwait offers an opportunity for a field study of this question. Dr. Benirschke: Well, fetal disease and fetal malformation, are you willing *o grant that, if you have a sick fetus with encephalitis and with subsequent aqueduct stenosis, these are malformations? 136 Dr. Silverstein: Yes. Dr. Benirschke: Fetal disease, growth retardation and what we consider classical congenital abnormalities. Dr. Silverstein: In a certain sense I think we are doing our job, if we have a job to do, by saying that this is not an extremely important field. This is probably a valid recommendation, if it should seem that this is reasonable. Dr. Benirschke: What do the pediatricians consider the magnitude of the problem to be? Dr. Dancis: We can only say how large does the problem seem to be to us at this point. You would accept that? Dr. Silverstein: Sure. Dr. Benirschke: What then was the size of the rubella problem in the country in 1966? How many rubella babies, defective, were born. Dr. Miller: I have no idea. Dr. Benirschke: One per 5,000? Dr. Miller: I have no idea. I think it would be lower than that. Dr. Dancis: Why did you pick 1966? Dr. Benirschke: Well, any year, where there was no major epidemic All right. 1940 to 1970. Dr. White: The estimated result of the rubella epidemic of 1964 was 20,000 to 30,000 defective babies. We used the Perinatal Research Study data to look at the incidence of clinical rubella during pregnancy prior to 1964, and in 30,000 preg- nancies there were 25 total cases of maternal illness due to rubella. We would guess, I suppose, a ratio of subclinical to clinical infection somewhere around one to one. So perhaps between 40 and 60 cases occurred in 30,000 pregnancies during the non-epidemic years of 1959 through 1963. Dr. Benirschke: So this is one in a thousand. Dr. Miller: But the babies were not affected. Dr. White: This was the incidence of maternal rubella. If that is evenly divided up between trimesters and knowing what we know now about damage, I think we can estimate how many affected babies there there were. Dr. Miller: I heard Dr. Sever say in 1965 that as a result of the 1964 epidemic, there were five definite rubella-syndrome babies born in the peri- natal study and five probable. Is that incorrect? 137 Dr. Mitchell: There must be more than that. There are at least 10 young- sters with congenital heart disease whose mothers had rubella during pregnancy. Dr. Benirschke: I think this information could be obtained, with Lon’s help. Dr. White: I may have that report here. Dr. Benirschke: If I am not mistaken, Dr. Sever has checked the fre- quency of rises of antibody titers to cytomegalovirus and, if I am not mis- taken, 6 percent of pregnancies in the NINDB study had a rise of titers, is that right? Dr. White: I can check that too. I have here the report entitled “Rubella Epidemic, 1964: Its effect on 6,000 Pregnancies,” (118). In the summary it is stated that 10 percent of the pregnant women studied reported exposure. Two percent of all women studied developed clinical rubella and 40 percent of those were in the first trimester. Approximately 10 percent of women with clinical rubella in the first trimester had a child with congenital rubella syndrome which was recognized within the first month of birth. Dr. Mitchell: That is out of 6,000. There are more than that who had rubella in the perinatal study. Dr. White: They chose women who were under study during the peak of the epidemic. Dr. Benirschke: Does this satisfy you? Dr. Katz: These 5 babies represent only those which were picked up in the first month of life. You may certainly have passed as normal, at that time, a number who will turn up later with small heads, cardiac lesions, hearing deficits or combinations of these which were inapparent in the first month. Dr. Silverstein: And based on 1966 knowledge of rubella and its implica- tions, is it safe to say that, if there is another epidemic or potential epidemic in 1970, it won’t be a serious problem ? Dr. Katz: This is a matter we haven't discussed but I don’t believe present information permits a firm conclusion. Most of us feel that all evidence to date suggests there is only one type of rubella virus which is antigenically of the same identity throughout the world and has not varied over the years. On the other hand, there are observers who feel that the incidence of damaged infants subsequent to the 1964-65 epidemic far exceeds past experience and they would question whether the 1964-65 strain was not more “embryo-tropic” than previous strains. Since this was the first major out- break to follow the development of techniques for studying virus and anti- bodies to rubella, comparison with past data is not valid because it was not 138 ever based on more than clinical impressions. The detailed investigations that have focused on the congenital rubella babies and their follow-up in social clinics were just not part of the picture in the past. Perhaps Dr. Miller can cast an epidemiologist’s eye on this question. Dr. Miller: There was a review of the literature made by Swann in 1949 of 538 case reports of children with rubella, and only 9 percent were reported to have mental retardation. (69) None were reported to have motor impairment, though I suppose some did. In the 1964. epidemic there was a very high percentage of mental retarda- tion and a high percentage of cerebral palsy. I don’t think the doctors 20 years ago would have missed the mental retardation or the thrombocytopenia. Dr. Katz: This is the sort of thing that makes the question so difficult to answer. There is no data presently to support the expectation of a 1970 epidemic which would differ sharply in character because its etiologic agent was a mutant of marked variation from past rubella virus. Strains of the agent from laboratories throughout the world display identical biological, immunological and biochemical properties. If the baboon proves to be a good model for embryopathic effects of the rubella virus, this might be a marker which could be compared and contrasted with strains isolated in future years. Dr. Benirschke: But do you have that virus for 1944? Dr. Katz: No. I meant that even in the absence of antigenic shifts of rubella virus, such as those displayed regularly by the influenza group, one might look for naturally-occurring alterations of virulence. In that context, I was thinking of teratogenesis as the expression of virulence. This is a continuing process beginning with the initial strains grown in 1962 but no answers will be forthcoming until the passage of a sufficient number of years to offer some degree of confidence that the virus of 1962 is identical with that of 1972. Dr. Benirschke: It is done? Dr. Katz: Yes, for the few years that have elapsed, virus from all over the world has been exchanged among many investigators. Dr. Axelrod: I think we are dwelling very heavily on the one virus in- fection which shows some promise of being really related to teratogenesis, that is rubella. But what about all the spontaneous abortions that occur? What about all the fertilized eggs that never get implanted. As a rule, a viral infection is of a general type, even the common cold. Dr. Benirschke: I think in the summary it was said that anyone now studying causes of abortion, should look for possible virus diseases. Dr. Axelrod: Can anyone really define what an abortion is? 139 293-596 0—68——10 Dr. Benirschke: I think the World Health Organization definition is quite useful for this purpose. Dr. Axelrod: Is that what they pass out? Dr. Benirschke: What you would like to see studied very logically is fairly well circumscribed. Dr. Axelrod: I would like to include absorptions and failures of implantations. Dr. Benirschke: You can’t open women to fine those embryos. Dr. Axelrod: That is why I am using animals. Dr. Benirschke: But such studies are in progress, people are attempting, with limited methodology, to correlate the tissue culture growth rate of abortions with chromosomal structure, with serologic abnormalities in the mother and with virus recoverability. They are extremely expensive and cumbersome, but it is hoped that a larger number of people will undertake such study. The WHO recommendation for a model system of chromosome study of abortions has been set up with exactly this in mind, to obtain the maximum data. Dr. Axelrod: From the anthropological standpoint, the obstetrician and psychologists have given progesterone for many many years for threatened abortions. This was considered “therapy”. Then, about 3 years ago, a nice double blind study was done and they found that with administered pro- gesterone the same number aborted as with the placebo. Dr. Benirschke: Does this satisfy your question with regard to the magni- tude of the problem? Dr. Silverstein: I think it is very useful to be able to admit, as I think we have done, that we don’t really know what the significance of the problem is, and that in a sense we are dealing with it on a speculative basis. I have to confess that I came here thinking that viruses were a much more impor- tant cause of malformation and embryopathy than I now believe on the basis of this discussion. Dr. Benirschke: But from a single epidemic, 30,000 babies, 20,000 of whom were significantly damaged, that is a substantial number of people, isn’tit? Dr. White: In Dr. Sever’s (119) paper he said: “Cytomegalovirus in- fection in normal pregnant women has been studied in several recent investi- gations. Serological studies of pregnant women in Boston and Philadelphia revealed that 30-60 percent had detectable complement fixing antibody 4.3 to 7.7 percent showed a significant increase in titer during the 3145—4-month period of observation. This is in agreement with the data shown in table 18. 140 “In our recent investigation of 200 pregnant women, virus was isolated at delivery from the urine of 7 (3.5 percent) and the child of one of the women with viruria also had virus present in the first voided urine after birth.” Dr. Katz: The detection of cytomegaloviruria on the first day of life bespeaks a very highly developed approach to what is known about the behavior of this virus. But if we are to depart on bolder paths, we must consider the data of Hanshaw (120) on the correlation of microcephaly at age 2 years with evidence of cytomegalovirus infection and the reports from Finland (121) suggesting viremia as a fairly common occurrence in normal blood donors. If the 4.3 to 7.7 percent seroconversion rate in pregnancies studied by the Sever group was accompanied by a similar incidence of viremia, a great many infants without viruria, at birth, may nonetheless have been exposed to cytomegalovirus in utero. Dr. Margolis: Just to give one other totally different viewpoint and one which sounds really wild, consider what Sir McFarland Burnet said a year or so ago (122). He stated that there are possibly being nurtured in experimental labora- tories viruses to which the human has never been exposed and which pose a tremendous potential hazard for man. Should they escape from a laboratory they could destroy mankind by invasion of a virgin soil in epidemic form. So we must think about preventive medicine too. Dr. Benirschke: Do you think you want to approach that point, Arthur? Dr. Silverstein: No. I think this is quite adequate. Dr. Benirschke: I think it is a substantial number. Dr. Dancis: I think maybe we can compromise in this way. From the presentations that we have heard in the last day, there is little doubt that virus infections have the potential for producing severe disease in the infant. Rubella has been the most striking example to date. Recent advances in technology have permitted us to learn much more about the disease than we have ever known and we have much more respect for it, as a result. It would be naive to assume this will not happen again with some other disease in the future. On the other hand the sum total of information that we have now precludes in my mind anything that would imply, to use the common expression of today, a crash program. A reasonable program would be to explore the fundamentals involved in injury to the fetus by viral infections in the mother—how viruses pene- trate the placenta, how they infect the placenta, how they interfere with normal development, etc. Techniques are now available to make sizeable increases in our information in all these areas. And we can be confident 141 that one day this information will be useful—possibly in the face of some new epidemic, or in re-evaluating currently unrecognized threats. Dr. Gruenwald: I think we ought to try and put this into perspective. I will not be able to answer the question myself which I am going to pose now. Maybe Dr. Miller can. How does damaging virus disease in pregnancy compare numerically with things we all accept as something we should worry about, such as severe asphyxia, prematurity and all these things that we all accept are important problems that we must worry about, and some others—how do they all numerically compare in our population? It would be nice, I think, if some day somebody could put this all together and tell us that there are as many babies damaged by this as by something else. We have at times, I think, been inconsistent about what we have worried about and what we have not worried about. For instance, just to give you an example, I had an argument once at a conference whether one should do blood sugar determinations in growth retarded newborn babies because hypoglycemia might affect their brain, and somebody said, “Oh, this would cost too much.” I said, “Do you think that phenylketonuria should be tested for?” “Oh, yes.” My answer was: “Don’t you think more are damaged by hypoglycemia than by phenyl- ketonuria abnormalities?” So I think we ought to try—and I hope our epidemiologists can help with this—to put this all in perspective so that we will all know how morbidity and mortality from virus diseases in pregnancies compares with a lot of other things which we have long accepted as pressing and important problems. Dr. Benirschke: I have just estimated frequencies of significantly dam- aged babies, through rubella, in the non-epidemic year, they are 1.3 to 3.3 per 10,000 pregnancies. 1.3 to 3.3 per 10,000. In epidemic years 10 to 20 per 10,000. Dr. Dancis: In trying to decide what types of effort should be supported, I am afraid it is an over-simplification to judge the desirability for research by the incidence of the disease. Dr. Gruenwald: That was not my intention. It was meant for our information. Dr. Dancis: I have heard it said that maximal effort should be put into the prevention of brain damage in the newborn period because statistically it is extremely important. And, similarly, the Congressman is attracted to supporting research in coronary artery disease because so many of his colleagues die from it. Of course, this is a useful measure, but another, and probably more important consideration, is the timeliness of major support for research. As all of us know, it commonly takes many years of patient work by individual investigators to make those essential discoveries that makes a field ripe for rapid advances. And it often takes great judgment to recognize when a field is ripe. 142 Dr. Gruenwald: I didn’t mean this in the sense of judging what merits support. What I said was meant only for our own information, to put it into perspective for us, not necessarily as a criterion to judge the need for research. Dr. Benirschke: I personally feel that it is a shame that we don’t know more about the precise nature of the thalidomide embryopathy. We know it has damaged some 10,000 German children and that it can be reproduced in some animals. We don’t know anything about the mode of action but are we going to find out the mode of action now that the problem is licked? The problem of virus embryopathy is of greatest concern here. We dis- cussed yesterday Parkman’s studies on 8 rhesus monkeys with rubella virus. Dr. Miller presented dose response curves in radiation because we dropped an atom bomb over Hiroshima, and we have some samples to study. Dr. Axelrod injects an enormous dose of rubella intravenously into the baboon but cannot cause abortions. We have a suggestion of the possibility of a dose response curve with this virus. We know that one can come up with all kinds of dreams why there are differences. There are different virus types, there is different tropism of one virus or there is a different infectious dose, or a different mode of infection— all kinds of possibilities. The intriguing paper of Parkman’s needs verfica- tion, and study of many other parameters in some detail. We have been talking about the only virus causing an appreciable number of malformations in man, and one can have a model in which to study this. So it costs $500 per pregnant monkey. I think it ought to be studied, not in rats, rabbits or hamsters. We ought to find out in such an animal model the dose response, different types of viruses, placental pathology, etc., all the things that were discussed here. That is a succinct study which can be undertaken. I hope all agree that, so far as thalidomide is concerned, we wish now we had a decent paper to which we could refer and say, this is exactly how thalidomide works. It goes into the gut, it goes into the circula- tion, so many milligrams go across into the baby and on day 27 it does this or that, chemically or morphologically, etc. So my plea would be that, with this one specific example, a really good study be made in depth. Dr. Katz: You raised as a model the experiments of Paul Parkman and his colleagues (66) in which they studied rubella virus infection of the pregnant monkey. They came up with some data which showed that infection of the pregnant rhesus early in gestation was in some ways similar to the human system. Virus was transmitted to the conceptus in 50 percent of their animals, a figure strikingly close to that reported by Alford and Weller (10) in human pregnancies complicated by first trimester rubella. There have been several other attempts to study in a ferret and in a rabbit model the pathogenesis of rubella in pregnancy. However, the great pres- sures in rubella research have focused on the development of an immunizing agent to prevent the disease. Pregnant monkeys have been utilized to compare 143 virulent and attenuated variants in their capacity to cross the placenta; but I have seen no reports of investigations to elaborate how virus crosses when it is successful in reaching the fetus. This may prove an academic matter for the rubella virus, as Dr. Silverstein has suggested, for the advent of a vac- cine may obviate the need to answer the question. But the problem will remain with us until we are satisfied that no other viruses use the same mechanism. To overlook opportunities to explore these problems in our haste to produce a vaccine is, to my mind, wasteful and short-sighted. To eliminate congenital heart disease induced by rubella without ever learning how the interaction of virus and fetus produced the malformation would be a small victory, for it is hard to believe that other agents do not exploit similar pathways. This is not intended as a polemic against vaccine development, only as a plea for the encouragement of studies along both lines, not one to the exclusion of the other. Dr. Eichenwald: I would like to comment further on this point. Thalidomide and rubella virus are interesting tools for the study of experi- mental embryology. Teratology perhaps might be considered a branch of the pathology of embryology. The study of the events which produce abnormal consequences could eventually lead to an understanding of the normal processes. I am surprised at the lack of interest among embryologists; their studies sometimes consists of cutting the tails of salamanders, and then drawing conclusions in terms of mammalian embryology. Some of the answers that we are seeking, can be gained by studying in detail abnormal situations which can be induced at will. It seems likely that eventually we will discover a final common pathway, that a number of differ- ent stimuli produce similar lesions. Dr. Margolis: I would like to extend that concept a little further. We recognize that certain potentially teratogenic agents are often given to preg- nant women, even late in pregnancy—therapeutic agents against infectious organisms, for example. In our own studies, we have seen, for example an antimitotic do the same thing to the developing cerebellum that a mitolytic virus does. Another thing which has come out from this study is the resolution of the argument in the literature about the histogenesis of the cerebellar cortex. A choice between two conflicting opinions has been made possible by studying the viral cerebellar lesion (24). And thirdly, in these laboratory animals born so much more immature than the human infant, we have, in effect, a “fourth trimester” to study and compare with the human third trimester in man. In other words, if we bypass the question of transplacental infection, we can get some idea of the conse- quences of retardation of growth in late gestation by studying postnatal lesions in animals. 144, Dr. Axelrod: What Dr. Margolis says is very interesting, and we can bear that out in recent experiments. Quite by accident—I like to think otherwise—it was found that the postnatal rat brain can acetylate hydrocor- tisone. As we went back in the rat, we found that indeed there are certain aspects of the growth of the brain which we can correlate with this acetylation phenomenon. This is a tool for studying other parameters. But within a few months postnatally this rat brain can no longer actively acetylate the hydrocortisone. So we went over to the baboon and found that postnatally the baboon cannot acetylate the hydrocortisone very well. But going back, we found that in the third, and particularly the second trimester when certain modifications were taking place in the brain, acetylation occurred to at least 10 percent of the substrate. So this concept of Dr. Margolis and also Dr. Eichenwald, that these models can serve as premature human models, is a good one because of the sequence of the development. The baboon is pretty well developed by the time it is born, but it still lacks some of the developmental processes that the human has at birth. Dr. Benirschke: This kind of interchange reminds me of another aspect which I would like to recommend to the Institutes. This is to suggest that an important endeavor is an interdisciplinary conference. It is so trivial and everybody talks about it but every time one goes to another conference of your own field one speaks the same jargon and really does not learn any more. We ought to go to the veterinary Congresses and vice versa. A florist would say, “Go to the botanist and find out about the tulip viruses there”. Dr. Katz started this meeting by reminding us that Blue Tongue and Scrapie have been studied without any one of us paying the least bit of heed to it. And to the Institutes this ought to be a lesson, that they should not call meetings of virologists if they want to lick scrapie, but that geneticists might be the most important participants in such a conference. It is very difficult, not only to organize but also, to find support for such conferences and to make them really meaningful. Dr. Gruenwald: I would like to make another commonplace suggestion and recommendation, and that is if we are going to invest a considerable amount of effort in doing animal experiments with viruses or with anything else, we should, more than has been done in the past, follow these animals through their entire life span. There is basically no difference between a malformation that develops before birth or a characteristic kidney change that develops in certain mice when they get to be 4 years old. I would think that more of this should be done. There is another aspect, and that this, I think we will save money in the long run if we do as much work as possible, or get as much information as possible from the same animal, or the same group of animals. 145 Dr. Katz: I must add my voice to the plea for better communication between those working primarily with animal problems and those engaged in human investigations. One of the major frustrations of most serological studies stems from the unavailability of an initial, or pre-infection, specimen of blood. Since a great many children are born to women who have married, and premarital serology for evidence of syphilis is a fairly widespread requirement, couldn’t some effort be directed toward preservation in some retrievable fashion of these pre-marital serum specimens to serve as the base line for later investi- gations of abnormal pregnancies and infants. The Institute of Laboratories of the state health department in Massachusetts has made judicious use of such materials to survey antibody levels against diphtheria and tetanus toxins. To carry this one step further, Lon White and I discussed, outside the conference, the feasibility of using the first pregnancy of any woman as her base line observations and continuing to keep her under study throughout the next pregnancy also. There must be statistics on which to base an estimate of how many of the recently delivered women will be pregnant again within any given elapsed time period. The classic 6-week post-partum checkup can be the first visit to the next prenatal course, so to speak. Stable populations, such as the New York H.LP. group or the California Kaiser participants, might prove ideal for this longer term surveillance. One more entreaty for the exploitation of the E.I.S. of the National Com- municable Disease Center as a watchdog for those who want to do broad studies of infectious disease and its effects on pregnancy does not seem too redundant. Although rubella virus justifiably has dominated our discussions, I would call attention once more to the cytomegaloviruses. Because we know of no overt clinical syndrome which calls to our attention the pregnant mother undergoing infection, the potentially affected infants have not been identified except when they presented a full-blown picture of generalized infection. The high rate of antibody conversion during pregnancy and the emerging picture of what might be late effects of a congenital infection demand our thoughtful attention. This may be as prevalent, in some years, as rubella in 1964-65. Our own limited experience suggests a cyclic epidemic pattern. We seem to find our newborns with overt cytomegalovirus infection clustered in bunches every 2 or 3 years as if there were some rhythmic epidemiologic pattern in some ways reminiscent of rubella cycles. Actually two members of the DNA herpes virus family merit more strenuous investi- gations by the experimental teratologists, simplex and cytomegalo. They are especially intriguing from an evolutionary viewpoint also, since every species seems to possess a representative of each type. With this note I've covered all the additional points I had listed. Dr. Silverstein: On the level of basic science, and of what I still continue to feel is our social obligation to general health, I think my main argument 146 would be in favor of more pathogenetic, pathophysiologic studies of the mechanism of embryopathy and fetal malformation, and of fetal disease on all levels. If I had been more virologically oriented and a little more intelligent I think I might very well have made a better choice of cytomegalovirus in the rhesus instead of blue tongue in the lamb. We ought to improve our ability to pick things up sooner in the general population, to pick up earlier than perhaps we did the last time, the next epidemic that is going to be important for embryopathy; to pick up a little sooner perhaps than we are doing now on an epidemiologic level, the relation- ships of a given epidemic to a certain kind of embryopathy that follows it, perhaps much later. Kurt’s suggestion of getting good sampling from 20 or 30 or 40 of the main hospitals in the country is really an excellent one. I don’t think this can be overstated. The value of this would, dollar for dollar, give us ever so much more important information than we are likely to get by any other mechanism. Dr. Benirschke: Did you notice anything in Téndury’s (7) book? Dr. Silverstein: Yes. We were discussing earlier the plasma cells and the placenta and Heinz’s observations. Tondury mentions that rubella in the 11 week fetus is a necrobiotic process, and that there is no competence in a fetus of this age to engage in an inflammatory reaction. In discussing the direct cytopathogenic effects of rubella in the development of heart lesions, he goes on to say: “A reaction in the sense of inflammatory process is absent in this case. Fetuses of 3 or 4 months are not capable of giving this type of reaction.” Presumably this includes not only chronic inflammation, but also the formation of plasma cells. Dr. Gruenwald: He is an anatomist, and I think somebody ought to go over his material. It is a common observation that a pathologist and an anatomist look at this with different eyes. Dr. Frank: There are two points that have come to my mind, and one is that I am very impressed with the amount of materials that are already available, as Art said, and the fact that, as Dr. Gruenwald said, we haven’t really been using them. I am thinking in terms of things like the closed Indian populations where people have gone in and obtained serum samples and done major studies, looking essentially at every individual in the popula- tion. They have very good methods for taking care of their immediate prob- lems. They know all the babies that are being born. They have serum sam- ples and know all the epidemics that are going through the population, and yet no one has ever looked at this kind of material for just the sort of things we are interested in. There is an awful lot available, and part of the problem is that people don’t know it is available. I think this is a major problem. The other thing that came to mind from the discussion that Dr. Silverstein and Dr. Dancis were having is that when Lord Rutherford, at the height of 147 his career, was asked what he though the importance of the atomic particle study was going to be to society, he said he felt strongly it would never have any importance but it was of interest to the physicist in knowing how the world works. I thought this was an important comment. Dr. Benirschke: You raise this problem, but you don’t answer what should be done about it. Dr. Frank: I think that one of the ways is by having conferences of this sort. Part of the problem is that the left hand doesn’t know what the right hand is doing. Part of NIH is very involved in collecting these serum sam- ples and studying these Indian populations. I think that it is only by some sort of trade of information of the sort that we are having here that interested research people are going to become aware of the possibilities in Indian populations and of the banks of frozen serum samples at the Kaiser Foundation. It would be nice if there were some sort of clearing house where all this information was available. I don’t know of one. Maybe one has to be developed. Dr. Gruenwald: What about Alaska? Would that be similar? Dr. Frank: I don’t know. Dr. McCracken: They are not yet collecting samples, but that is one of the things we are looking at. Dr. Miller: As a staff member of the NIH, who has never been asked before for his opinion as to what an institute should do (Laughter.) My own feeling is that the Institutes should engage particularly in activi- ties which university-based scientists cannot pursue as easily. They should tap resources and provide resources, demonstrate by example, and cross dis- ciplines by symposia such as this, including not only people outside NIH but also inside. I have never before met most of the NIH people who are here now. Dr. Benirschke: Heinz, would you have anything more to aid? Dr. Eichenwald: No. I think the expression is I have shot my wad. And everything that I will say others have already said or would say more elegantly so I will just pass. Dr. Benirschke: I doubt very much that any one of us can say anything any more elegantly than you have summarized. Dr. McCracken? Dr. McCracken: You brought up something that we had been doing at the Institute and really no one knows too much about it. We got interested in the Indian population from many different standpoints. There have been many studies done with the Indian population but they have all been buried 148 in very obscure literature and in health statistics that sit in Indian reserva- tions. No one ever knows about what happens to them yet some of this has now been brought to the clinical horizon and is very interesting. There have been large studies or analysis of the congenital malformations in Indians. In certain populations there is clustering of certain malforma- tions. This study actually originated from here, in the Dental Institute. While it had many shortcomings the mothers in this study could be studied for prenatal infections. This has not been done at all. The malnutrition alteration of young infants with infectious disease and problems of growth and development particularly in the very young could also be studied here. What we are now starting to do, which I think will be helpful in the long run, is to organize an international conference, bringing in people from various areas, including Guatemala and certain areas in Africa where they are dealing with approximately the same problems, to point out two differ- ent populations in the United States, the Indian and the Harlem or Ap- palachia populations. They will go into some of these problems of growth and development from the newborn period through childhood. The cytomegalovirus infection does show that congenital infection really is a spectrum of disease, no longer is the diagnosis made only at post mortem. In a large proportion of babies this infection probably does nothing which we can recognize at birth but does give mental retardation which may be very subtle. So one must really take the length of an animal or human life, and certainly look at the infant through childhood and the preschool age into school, where maybe his inability to add or subtract or multiply or whatever, is his manifestation of something that happened very early in his experiences. These are the only two aspects that I would bring up. Dr. Silverstein: I have been impressed, and reimpressed last night by reading Tondury’s book by the fact that almost all of the viral diseases leading to congenital anomalies have ocular complications. I think that this is important, and something that one doesn’t pay enough attention to. We can see through the cornea into the body, in a sense, and often visualize an otherwise occult disease. The point was made that if we have an amelia, it is likely to be reported, but some heart or other congenital disorder may escape us. My point is that we have in the embryology of the eye, during one of the most critical periods of development, a step by step developmental sequence, anomalies of which we can see later by looking into the eye. We can see anomalies of corneal development, of the lens development, of retinal development, and other anomalies at the optic disc. We have in the eye the ability to see these anomalies which may reflect other problems that are more occult elsewhere in the body. At later stages of development, when we are dealing with congenital inflammatory disease a very significant number of occular disorders can be seen clinically in cases where the disease 149 might not otherwise be recognized. I think we are missing a tremendous amount of extremely valuable information by not looking carefully in the eyes of these newborns. Just as an example, a resident (and now fellow) at Hopkins is now going through his second thousand consecutive newborns at Baltimore City Hos- pital. He happens to be looking for cupping of the optic disc, and for retinal and optic disc hemorrhage and finds a surprising number of hemorrhages at the optic disc, in addition to some interesting findings on the color of the optic disc, which relates to its vascularization and to the amount of blood the eye is getting. If he is successful in this and if he can hang on for the five or ten years that this study will take, he may be able to relate these hemorrhages with the vague suspicion that has been raised in the literature that there may be a relationship between fundus hemorrhage and later developmental brain abnormality or later behavioral or learning problem. So the point here is that we may be missing a very good opportunity to get some extremely valuable data. Dr. Benirschke: Amen. I think the same applies to us pathologists. You can’t find many pathologists who know how to look at or even cut an eye really adequately. Most diagnoses pathologists make of a placental specimen are: mature placenta. Peter will bear me out. The same applies to the eye and the inner ear. That is why we wrote this book (62) because there is nothing that the pathologist has to hold on to. He is not trained to do it and so he says it’s a mature placenta. Dr. Margolis: The eye is easier to understand than the placenta. Dr. Silverstein: The eye is a microcosm most of which we can visualize, and there is very little in physiology or in pathophysiology that has not got its ocular analogue. Dr. Miller: Following that rationale, we examined 7,000 children in Japan, did a screening examination of visual acuity, looked at their eyes with ophthalmoscopes and wrote a paper. Not only do pediatricians perhaps not look enough at the eyes but they don’t look at publications on eyes. Out 0f 400 reprints we have 388 left. Dr. Benirschke: I think this is very true. You can say the same about the placenta, all the information is in literature. As you know, I am very much interested in the process of the so-called placental membranitis, the amniotic sac infection syndrome. Is it infection or just inflammation? There appeared last month a fascinating paper by McLellan et al (123). A pediatrician took an otoscope and looked at 800 newborn ears to find a surprising number of cases of otitis. It pays to be attentive, this is all the message there is to this conference. 150 Dr. Katz: At our own hospital the residents are alert to the middle ear as a site from which infection may extend in the newborn to produce menin- gitis and sepsis. Just last week we had a small infant with sepsis accompanied by the skin lesions typical of pseudomonas infection. His ear was cultured because of slight drainage therefrom and when it produced a pure growth of pseudomonas we alerted the otolaryngology resident whose attitude was one of studied disdain. Dr. Miller: One person with aniridia in our study was described by an intern as having widely dilated pupils that did not react to light. Dr. Katz: We have been fortunate in the last 2 years to establish a full- time eye service within the hospital, whereas in years past sporadic consulta- tions were the only available assistance. The new service has come at a fortuitous time since many of the rubella infants are in need of their minis- trations. Dr. Mitchell: May I ask Art, if you had a choice between making it mandatory to take a Wasserman reaction for the newborn or to look in his eyes, which would you choose? Dr. Silverstein: I can’t answer that question. Dr. Gruenwald: You would wait about 6 weeks. Dr. Mitchell: Is it not still mandatory in a large number of states to drop silver nitrate into the eye of newborns? Dr. Gruenwald: Yes. You see the black spots on the cheek where the silver nitrate went. Dr. Dancis: Let me make a comment about that. About three weeks ago in this same room we were discussing retrolental fibroplasia, which I have always considered an example of teratology, post-partum. It was stressed that the ophthalmologists have completely lost interest in the newborn eye because they believe that retrolental fibroplasia has been solved for the time being. I am delighted that you are having more success than most of us. The art of looking at the newborn eye is an art which is not shared by most ophthalmologists. Arnold Patz then discussed this question of eye pathology and the absolute necessity of providing simple instructions to pathologists across the country merely as to how to fix the eye and where to send it so that it will reach a pathol- ogist who knows how to examine eyes. In brief, accepting the desirability— of examining eyes in the newborn proved we have the problem of finding the people who will do it. Dr. Benirschke: You really have to go out proselytizing as I have done with the placenta. You have to push this as a personal hobby and make a laughing stock of yourself to the extent that people eventually come and do a 151 study just to disprove you. This is eventually how this happens. And then something is being learned. I think you or your resident would write such a book on eye disease of newborns, there is no other way. Dr. Silverstein: I don’t mean to imply the ophthalmologists were beating down the doors of the lying-in hospitals looking for this information. Ob- viously they are not. But if one is going to make a recommendation, then it should be in some way to find a mechanism to stimulate the departments of ophthalmology to send their residents through the newborn wards and to stimulate the people who run the newborn wards to invite the ophthalmolo- gists to come and take a look. Dr. Gruenwald: The same thing goes for studying the placenta. In hospitals where they begin to ask the pathologist what was found in the pla- centa he will perhaps hopefully some day begin to look at placentas. But there must be motivation. He must know there is something to be gained by doing this and not just be asked to do it. One has to make it clear to one’s colleagues that they can really see something, learn something that is worthwhile. Dr. Mitchell: In other words, one must motivate them. One way, which I personally deplore, is through legislation as with PKU. Are there other better ways of motivating ? Dr. Silverstein: Sure. There are any number of ways. The first is to write a book and get it to department chairmen. The other is to make avail- able to a dozen of the significant departments of ophthalmology a little bit of fellowship support so that he can send a resident part-time into the new- born wards to make this type of study. You must buy the service. Dr. Axelrod: As a member of the conference, and at the risk of platitudi- nous utterances | would plead for open minds on the part of the reviewing committees and the people involved in distribution of funds so that studies on all levels may be supported whether or not they are involved with the specific topic of viruses. I firmly believe that we must understand the normal processes of growth and development before we can further interpret patho- logical development. Although we recognize this was a conference on virus etiology and congenital malformation, it should be recognized that from different orientations and disciplines, the understanding of clinical entities still come from studies of fundamental biochemical, physiological and molecular biological lesions. In a recent conference held on the use of drug testing in primates, members of the FDA were convinced that drugs at least should be primarily tested in certain animal species and brought rapidly into the human, then taken down into a primate, or species showing parallel effects to the human, for studies in depth At the epidemiological level, I think better detective work is in order. Rapid isolation of the infective agent should be sought and then 152 studied in some primates so that a rapid understanding of potential terato- genic effect of the isolated culprit can be achieved I think that thalidomide itself is uninteresting. But it represents a unique and rather select chemical poison of an organogenic event. Many of the viruses yet to be understood may represent unique or fairly unique poisons, interruptors of molecular events at the histogenic level. I think an under- standing of normal organogenesis and growth can be helped by an under- standing of how these drugs and infectious agents act to interrupt organogenesis and histogenesis. Therefore, support systems at the level of operation in which normal organogenesis and histogenesis are investigated are important. Dr. Moorhead: I want to emphasize one point that Dr. Katz reminded me of. I am not a virologist, but it is important to emphasize, for people who are not close to virologists, the nature of the host cell. You may take one virus, call it a “human virus” or an “oncogenic virus,” but it does quite dif- ferent things in different cells. For example SV40 lyses green monkey cells, it does nothing to mule cells, it makes hamster cells malignant, it transforms human cells. So this area of “detection of the culprit” in our environment as you mentioned is, I think, completely wide open. For example, I think it is extremely surprising that the Rous virus of fowl causes tumors in mammals! This still excites me to think of this un- known sea of viruses from all sorts of animals, as detected and associated with that animal, and then named cat or rat virus. I think the area should be encouraged and this is, of course, a basic area of work in virus exploration. But this comes back to what Dr. Eichenwald said in summarizing about the genetic background. Let’s say in human abortion that there is no evidence yet that there is any virus associated with the very high proportion of abortuses which seem to have chromosomal anomalies, at least 22 percent in the ones studied. But we know viruses can enjoy a transient infection and set these events off. It is the very cells that are not damaged much that may be important. We know of the various things viruses can do; we have this variety of viruses available and we don’t know if there are any species bar- riers. Then you couple this with the tremendous variety of effects that any one virus could have. For example, the replication of virus is not necessary for the induction of cell-fusion, certainly not a minor sort of effect. Dr. Benirschke: We must consider getting disciplines into our vista which are not represented here and which we don’t ever consider as making a contribution. For instance, as you talked I was thinking that we really ought to have someone with a broad knowledge of evolution and how this fits in. Dr. Katz: This is an area in which we found veterinary virologists most receptive. With the government research laboratory on Plum Island, off the tip of Long Island, we conducted a collaborative study of the relationships of human measles virus, canine distemper virus and rinderpest (cattle plague) 153 virus. Since rinderpest is not found in the United States and the Agriculture Department is anxious to maintain this freedom from so devastating an illness of cattle, it is permissible to work with the virus only at their Plum Island establishment where elaborate precautions are maintained. All three viruses must have had some common point of origin and the veterinary people were as anxious to explore this problem as were we. Dr. Margolis: My contribution to this conference has been to talk about a concept of virus which produces a neurologic deficit, but a virus which is not neurotropic. I have been functioning as a neuropathologist. However, I am not a virologist, nor am I versed in teratology. I suppose these hiatuses force me to maintain an open mind, at least one not cluttered by preconceptions. You may have noticed among my references a paper with the title, “In Pursuit of an Ataxic Hamster” (26). This paper states: “Research has been likened to the pursuit of a pregnant fox through a maze, with the fox repre- senting a provocative observation, a research approach, or hypothesis and the maze, the vagaries inherent in original investigations. Somewhere during the chase the fox whelps, and a choice must be made as to which pup to pursue. Further along the chase this new target fox matures, becomes impregnated, and whelps. Hence another choice, another chase, and so ad infinitum. More- over, as foxes are prone to do, there is an unpredictable doubling back and criss-crossing of old trails, and often target foxes which were once elusive are encountered and captured.” The studies I have related illustrate this parallel. Kilham was searching for a tumor virus and instead he encountered ataxia in golden hamsters, and this observation led to the discovery of a class of mitolytic viruses, or the recogni- tion of a new type of virus-cell interaction. I was fortunate in joining this chase in its most exciting phase. The progressive uncovering of new viruses of this class and the continuing expansion of the host spectrum leads me to a question. Is there any way for us to predict the potential for interspecies transmissibility of viruses? Must this remain simply an empirical approach? Must we merely wait until, as has so often happened in the past, one of our laboratory people contracts a new virus disease? Should we actually resort to an “Operation Alert” and try to probe in advance whether the prediction of Burnet is likely to materialize? Should we support broadside or wholesale tests of these potentially teratogenic or pathogenic viruses in baboons so we know what could happen should they infect man? Dr. Benirschke: Conversely, should you study the antibodies of defective man against a host of mammalian viruses such as Blue Tongue? I am left with a lack of knowledge whether there are antibodies in man against Blue Tongue. I think this is a pertinent question which needs to be answered. I am sure Art would be delighted to know whether shepherds have antibodies and if, say, Dr. Miller could answer the question whether in the area in which Blue Tongue is endemic hydranencephaly occurs in man more frequently. And if so, is it due to the virus. Conversely, if a pregnant woman, 154 a shepherdess has been exposed to the virus, does she make antibodies during this time and then have a known affected offspring? Dr. Silverstein: I may say that the reason we chose Blue Tongue was that we were assured by the veterinarians that it wouldn’t bother us. Dr. Katz: It is certainly impressive to the pediatrician-virologist that many of the apparent species barriers are crossed with ease in the test tube. Even when the intact virion is unsuccessful in establishing an infection in some resistant cell, nucleic acid from that virus, in those instances where it has been available, will induce a growth cycle of virus in the previously inhos- pitable cell. This leads to the question: Does free, infectious viral nucleic acid turn up very often in our internal or external milieu? The answer, if affirma- tive, still does not establish the importance of any such DNA or RNA in the transmission of virus infection. However, as Paul Moorhead has emphasized and I would underline, all that may be required to induce extreme behavioral alterations in a cell is one cycle of viral replication without any further production of the intact virus. Dr. White: The perinatal Research Study and projects like it suffer from certain definable shortcomings and seldom yield their full potential for several years. Epidemiologic studies which should be encouraged in the next few years should really be aimed at complementing rather than aug- menting projects like the Perinatal Study. An alternate approach is through identification and intensive study of the high risk pregnancy and high risk newborn. One mechanism we have used involves screening cord sera for high IgM levels. Others which have been defined here are looking in the eyes very closely and looking at the placenta. These should obviously, I think, be utilized and should probably be integrated if at all possible. Finally, perhaps we should be searching for better means of recognizing the high risk infant. Dr. Gruenwald: May I say a word to this. I wonder whether it would be worthwhile if the people concerned with this collaborative study, perhaps with the help of some other consultants, will sit down some day and put down on paper just what you suggested. What is it that the study has not done that other studies should do? What can one learn in this way; I think this might be very helpful, because I am sure there are many people con- sidering new studies and it is very difficult for an outsider to know just exactly what this particular study has and has not accomplished. It might be very, very helpful if some form could be found to do this. Dr. Miller: It is being done, but not at the invitation of the perinatal study. An expert committee is evaluating the quality of the data and its usefulness. Dr. Singer: I would just really second what Lon had said in terms of the complementation of the type of study, the perinatal study, or the type of approach the perinatal study has. 155 293-596 0—68——11 I would also like to emphasize again the need for better communication between disciplines and between different geographic areas. Dr. Friedman: I personally have found this to be an extremely enlight- ening conference. As a pediatric cardiologist, I get this distinct feeling that if I were able to wish away all viral disease, I would not in any way be put out of business. As far as lack of information concerning fundamental mechanisms, I am impressed with the complexity of genetic and environmental factors. This sort of conference and interplay has been very profitable. It is a shame in a way, that rather than getting together once in awhile under one roof, that these sorts of representations can be under one roof working on one specific problem at a time. I must echo Dr. Miller’s feelings about the NIH. There is probably opportunity here for just this sort of thing. Dr. Benirschke: That is the NIH’s problem, not ours. Dr. Overall: To me, it has been a very stimulating conference. The answers have not been provided but it certainly has opened a whole new spectrum of avenues and approaches to questions that were raised and needed to be answered. Dr. Benirschke: Dr. Woodside, will you ask us some searching questions now, if it is not to late? Dr. Woodside: Well, I don’t know that I can ask searching questions. When Dr. Mitchell first mentioned the possibility of our looking together at this problem, the two Institutes, it struck me as an extremely important thing. And I do believe that one of the major things we hoped to learn has been learned, namely, the state of the art at the moment. For example, we were aware of the fact that rubella was really the only proven criminal, and yet we wanted to have experts here to talk about whether there are other im- portant possibilities that should be examined. Now, as I am sure all of you know, our Institute is probably unique at NIH in having as its mandate the study of normal growth and develop- ment. The whole history of NIH has been an attack on categorical problems, and I think everybody knows why this is so. You start with Cancer which was the first Institute. Then you add Heart, which was the next most serious killer, and so on right down the line. The historical development of the NIH finally got to an Institute which could study normal growth and development right through the whole life span. Well, of course one of the best ways to do this is by means of research grants which are aimed at fundamental problems in basic research. And we don’t always have to prove that if we do supply money for this it is going to cure this or that disease. And this is an important thing, as many of you have mentioned during this day and a half. It enables us, for example, to spend a lot of money on animal models. I am particularly interested in 156 learning the baboon study, and we intend to learn it at firsthand in the near future. But also I think Dr. Mitchell and I were interested in knowing whether this expert group would think that the field was ripe for a crash program. And 1 was particularly grateful to have this mentioned by a number of you. We now have the impression that you don’t feel it is ready for that, that there has to be a certain development of the science with enough basic information available so that it would be worth the money that we would have to put into a crash program. And we are able to do that if we think the time is ripe; but we need advice such as yours before we can make that kind of decision. An example where we did feel this was the case is in population. And this is one reason that Phil Corfman hasnt been able to spend any more time at this conference. He is involved in getting started just such a crash program in research on population. He has been with us now for three years, during which time we have been disappointed in the number of good applications for research in population that have been coming in through the regular grant mechanism. As a consequence of this disappointment, we were able to get an additional million dollars for contracts this year. Out of these we hope will come some of the answers that we just don’t have yet, for example, on the effects of the oral contraceptives, physiological and biochemical effects, the way the IUD works and similar problems. So we conceivably could have launched a crash program, either alone or in cooperation with the Heart Institute, which I think was one of the things Dr. Mitchell may have had in mind when she first came to us with the sug- gestion that we talk about this problem. As to the benefit of the conference itself, it has been tremendous as far as I personally am concerned. And I think it will be of value to the scientific world especially if the conference is published. In this connection, our Institute is actively engaged in developing scien- tific information centers, hopefully with the idea that the sort of thing which happened to Bob’s paper on eyes of Hiroshima infants won’t happen as often in the future. We are going to try to set it up so that the information can be widely disseminated to investigators and so coded that, for example, pedia- tricians will be able to learn what was published in a journal that they don’t have the time to read ordinarily. This is still a hope. We don’t have any of these really functioning yet. But there will be four of these science informa- tion centers in operation in the very near future. This has the blessing of the National Library of Medicine; it is not in con- flict with them; it won’t duplicate what they are doing, but will focus very definitely on the problems that we feel are important. Dr. Benirschke: Tell us a little more about the science information centers. Dr. Woodside: Well, they are being set up in our Institute and will func- tion right here in terms of the specific areas that we are charged with. One 157 will be in reproduction and population studies; one will be in perinatal bi- ology and infant mortality; a third will be in aging, and a fourth in growth and development. We don’t have a special one for mental retardation, although that is one of our programs. The reason we don’t is that the Mental Health Institute, now a bureau, has its own information center, with which we are cooperating. Dr. Gruenwald: Will this be by distribution to a list of subscribers or interested people, or will it be only upon request? Dr. Woodside: It will include the publication of abstracts and we are hoping to be able to distribute them to investigators without charge. Much of this has yet to be worked out, but this is what we are hoping to do. Dr. Benirschke: Something like the National Foundation does with the chromosome abnormalities. Dr. Woodside: Very much like it. There will also be English abstracts of foreign articles. The whole idea goes back to Bob Aldrich, the first di- rector of the Institute. Dr. Benirschke: Shiela, may I ask you if you have any charges that we have missed or questions that we should answer? Dr. Mitchell: I am afraid I do not have anything to offer such as Gil does. However, I would like to raise the question of publication from this con- ference. It is my impression both from the formal and informal discussions which have taken place here, that you feel much of the information which has been presented, while it is knowledge that you have, is not available to the scientific community even to those who are reasonably knowledgeable in the field. Publication of the highlights of these proceedings in, say Science would alert the academic community to the fact that this conference has been held and direct them to the source of further information. Such a report could, and hopefully would, present the questions which have been raised, the “now” problems, rather than try to give all the factual details. Then to follow-up we would publish an edited transcript. That is to say each of you would receive a personally marked copy of the entire proceedings, to correct and delete his contribution and comments and send it back to Gil and me for final editing. Is this the consensus of the group? 158 XII—Summary of the Conference Recommendations by Kurt Benirschke Dr. Benirschke: I would agree with that certainly. Is that the general wish? Fine. Before we disperse, it has been suggested to me that I might try to recapitulate, not the data and ideas presented, Heinz did that superbly, but rather our collective recommendations to the Institutes. We would unani- mously agree with Joe that a “crash” program in maternal viral infections and fetal malformations is not warranted at this time. However, the many opportunities which this field offers in understanding basic pathology, immune response development, viral action and end results should be ex- ploited to the full. To do this, cross fertilization should be actively supported either by holding or sponsoring conferences such as this one with unusual but involved people being invited; that is evolutionists and veterinarian virologists to this type of conference, geneticists and botanists to virology conferences and so forth. While the immediate gain might be small the long term import could be enormous. Secondly, we should encourage the NIH to tap and make available overlooked national resources. As Sam has mentioned, some states do and many more could, save sera from premarital blood tests cross linking these through their vital statistics records with epidemics and pregnancy. An extremely valuable source of viral and epidemiologic data could thus be made available to virologists, pediatricians and epidemiologists. Without any additional expense early and prepregnancy sera could be collected. As Art's discussion of ophthalmology brought out this morning, a great deal could be gained by establishing very special residencies or fellowships for cross-disciplinary study. That is to say, for a pediatrician interested in congental malformations or developmental pathology to spend 6 months or a year in the department of ophthalmology, neurologists in otorhinolaryn- gology and pathologists to travel to the very few centers where they can really learn in detail the pathology of the placenta, the eye or the ear. Career development awards might also be considered for such multifaceted careers. Encouragement should be given to detailed studies of thalidomide and viral embryopathy, to dose response and to placental pathology in rubella not only in rats and hamsters but in primates and, where possible, in man. These problems currently are not “glamour” issues but they can be attacked and from them we could learn a very great deal about fundamental, and hence generally applicable, events. Peter has suggested that serious thought should be given to encouraging epidemiologic studies in countries with more stable and more manageable populations, countries with health systems and records that make population studies effective, complete and relatively inexpensive. 159 At the risk of riding my own hobbyhorse, I would like to reemphasize the serious inadequacies in our knowledge of the placenta. Dr. Driscoll and I have tried to remedy this in part with our book, but either through lec- tureships, student fellowships, summer workshops or post-graduate seminars, studies of all aspects of the placenta, ils anatomy, phyiology, immunology and pathology must be encouraged if we are ever to elucidate fully the mechanisms and the causes of congenital malformations. Finally although perhaps this should have come first as Heinz suggested, and despite the recurring criticism which has been levelled at it, we should recommend epidemiologic studies which would complement the Collabora- tive Perinatal Study. To confirm or develop hypotheses which it has raised, the most appropriate of the many epidemiologic methods available should be utilized that the situation in man may be correlated with studies in animals and in test tubes. Let me thank you again for coming and you for inviting us. 160 1. 10. 11. Cited Bibliography Sommers, Donald F. Biochemistry of animal virus replication. New Eng. J. Med. 276: 1016-1023 and 1076-1081, 1967. Nahmias, A. J., Naib, Z. M., Highsmith, A. K., and Josey, W. E. Experimental genital herpes simplex infection in the mouse. Abstract, Society for Pediatric Research, April 1967, p. 25. Alford, C. A. Jr. Studies on antibody in congenital rubella infections. Amer. J. Dis. Child. 110: 455-463, 1965. Rawls, W. E. and Melnick, J. L. Rubella virus carrier cultures derived from congenitally infected infants. J. Exp. Med. 123: 795-816, 1966. Boué, A., Hannoun, C., Boué, J. G., and Plotkin, S. A. Cytological, virological and chromosomal studies of cell strains from aborted human fetuses. Proc. Soc. Exp. Biol. Med. 122: 11-16, 1966. Klein, S. W. and Huang, W. C. Adenovirus recovered from tissue cultures of human kidney cells. Abstracts, Society for Pediatric Research, May 1963, p. 89. Tondury, G. Embryopathien. Springer-Verlag, Berlin, 1962. Van Furth, R., Schutt, H. R. E. and Hijmans, W. The immunological development of the human foetus. J. of Exp. Med. 122: 1173, 1965. Osborn, J. J., Dancis, J. and Julia, J. F. Studies of the immunology of the newborn infant. Pediatrics 9: 736, 1952. Alford, C. A. Jr., Neva, F. A. and Weller, T. H. Virologic and serologic studies on human products of conception after maternal rubella. New Eng. J. Med. 271: 1275-1281, 1964. Medearis, D. N., Jr. Comparative aspects of reproductive failure induced in mammals by viruses. Comparative Aspects of Reproductive Failure, K. Benirschke (ed.), p. 333, Springer-Verlag, New York, 1967. 161 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 162 Miller, W. C. Virus infections affecting the fetus in animals and man. Proceedings of the Royal Society of Medicine, April 20, 1966. Emerson, J. L. and Delez, A. L. Cerebellar hypoplasia, hypomyelinogenesis, and congenital tremors of pigs, associated with prenatal hog cholera vaccination of sows. J. Am. Vet. Med. Assoc. 147: 47-54, 1965. McKercher, D. G. and Wada, W. J. The virus of infectious bovine rhinotracheitis as a cause of abortion in cattle. J. Am. Vet. Med. Assoc. 144: 136-142, 1964. McKeown, T. and Record, R. G. The influence of placental size on foetal growth in man, with special reference to multiple pregnancy. J. Endocrin. 9: 418-426, 1953. Russell, L. B. X-ray induced developmental abnormalities in the mouse and their use in the analysis of embrological patterns. I. External and gross visceral changes. J. Exper. Zool. 114: 545X602, 1950. Dagg, C. P. and Karnofsky, D. A. Teratogenic effects of azaserine on the chick embryo. J. Exper. Zool. 130: 555-572, 1955. Karnofsky, D. A., Patterson, P. A. and Ridgway, L. P. Effect of folic acid, “4-amino” folic acids and related substances on growth of chick embryo. Proc. Soc. Exper. Biol. Med. 71: 447-452, 1949. Schwalbe, E. Die Morphologie der Missbildungen des Menschen und der Tiere. Fischer, Jena, 1906. Schutt, W. Foetal factors in intrauterine growth retardation. Clinics in Developmental Medicine 19: 1, 1965. Naeye, R. L., Benirschke, K., Hagstrom, J. W. C. and Marcus, C.C. Intrauterine growth of twins as estimated from liveborn birthweight data. Pediatrics 37: 409, 1966. 22, 23. 24. 25. 26. 27. 28. 29. 30. Harlow, H. F. and Harlow, M. K. The affectional systems. Behavior of Non-Human Primates, A. M. Schrier, H. F. Harlow, and F. Stollnitz (eds.) Chapter 8, Volume II, Modern Research Trends, Acad. Press, New York, 1965, pp. 287-334. Margolis, G., Kilham, L. and Davenport, J. A model for viral induced reproductive failure: Theory, observations and speculations. Comparative Aspects of Reproductive Failure pp. 350-360, Springer- Verlag, New York, Inc., 1967. Margolis, G. and Kilham, L. Viral induced cerebellar hypoplasia. Infections of the Nervous System. Proceedings of the 44th Annual Meeting of the Association for Re- search in Nervous & Mental Disease, New York City, 1964. Margolis, G. and Kilham, L. Rat virus, an agent with an affinity for the dividing cell. NINDB Monograph #2, Slow, Latent and Temperate Virus Infec- tions, D. C. Gajdusek, C. J. Gibbs and M. Alpers (eds.), 1966, pp. 361-367. Margolis, G. and Kilham, L. In pursuit of an ataxic hamster, or virus induced cerebellar hypoplasia. International Academy of Pathology Monograph #8, The Central Nervous System. Toolan, H. W. Susceptibility of the rhesus monkey (Macaca Mulatta) to H-1 virus. Nature 209: 833, 1966. Ferm, V. H. and Kilham, L. Histopathologic basis of the teratogenic affects of H-1 virus and ham- ster embryos. J. Embrol. Exp. Morph. 13: 151-158, 1965. Kilham, L. Viruses of laboratory and wild rats. National Cancer Institute Monograph #20, Viruses of Laboratory Rodents, pp. 117-140, 1965. Kilham, L. and Margolis, G. Viral etiology of spontaneous ataxia of cats. Am. J. Path. 48: 991-1011, 1966. 163 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 164 Margolis, G., Kilham, L. and Ruffolo, P. R. Rat Virus Disease: An experimental model of neonatal hepatitis. Experimental and Molecular Pathology. Engler, W. O., Baer, P. N. and Kilham, L. Effects of rat virus on healing osseous wounds. Arch. Path. 82: 93-98, 1966. Ruffolo, P. R., Margolis, G. and Kilham, L. The induction of hepatitis by prior partial hepatectomy in resistant adult rats injected with H-1 virus. Am. J. Path. 49: 795, 1966. Margolis, G. and Kilham, L. Unpublished Observations. Monif, G. R. G., Sever, J. L. and Cochran, W. D. The H-1 and the RV viruses and pregnancy: Serological study of certain groups of pregnant women. J. Pediatrics 67: 253, 1965. Herringham, W. P. and Andrewes, F. W. Two cases of cerebellar disease in cats, with staggering. St. Barth’s. Hosp. Rep. (London) 24: 241-248, 1888. Kilham, L. and Margolis, G. Spontaneous hepatitis and cerebellar “hypoplasia” in suckling rats due to congenital infections with rat virus. Am. J. Path. 49: 457-475, 1966. Johnson, R. H., Margolis, G. and Kilham, L. Identity of feline ataxia virus (FAV) with that of feline panleukopenia (FV). Nature 214: 175-177, 1967. Kilham, L. and Margolis, G. Congenital infections of cats due to feline panleukopenia virus (PLV) and manifested by cerebellar ataxia. Laboratory Investigation. Fischer, D. S. and Jonas, A. M. Cerebellar hypoplasia resulting from cytosine arabinoside treatment in the neonatal hamster. Clin. Res. 13: 540, 1965. Hicks, S. P. and D’Amato, C. J. How to design and build abnormal brains using radiation during development. 42. 43. 44. 45. 46. 47. 48. 49. 50. Disorders of the Developing Nervous System. Compiled and edited by W. S. Fields, and W. M. Desmond. Charles C. Thomas Co., 1961. Davison, A. N. and Dobbing, J. Myelination as a vulnerable period in brain development. Brit. Med. Bull. 22: 40-44, 1966. Javett, S. N. et al. Myocarditis in the newborn infant. A study of an outbreak associated with Coxsackie Group B virus infection in a maternity home in Johannesburg. J. Pediat. 48: 1, 1956. Gear, J., Measroch, V. and Prinsloo, F. R. The medical and public health importance of the Coxsackie viruses. South Afr. Med. J. 30: 806, 1956. Bierman, H. R., Crile, D. M., Dod, K. S., Kelly, K. H., Petrakis, N. L., White, L. P. and Shimkin, M. B. Remissions in leukemia of childhood following acute infectious disease, staphylococcus and streptococcus, varicella and feline panleukopenia. Cancer 6: 591-605, 1953. Lucas, A. M. and Riser, W. H. Intranuclear inclusions in panleukopenia of cats. A correlation with the pathogenesis of the disease and comparison with inclusions of Herpes, B-virus, yellow fever and burns. Am J. Path. 21: 435-465, 1945. Verlinde, J. E. Congenitale cerebellaire ataxie bij katten in samenhang met een ver- moedelijke virusinfectie bij de moeder gedurande de graviditeit. Tijdschrift voor Diergeneeskunde 74: 659-661, 1949. Jones, T. C. Personal Communication. Toolan, H. W. Experimental production of mongoloid hamsters. Science 131: 1446, 1960. Brown, G. C. and Evans, T. N. Serologic evidence of cocksackie virus etiology of congenital heart disease. J.AM.A. 199: 183, 1967. 165 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 166 Sandberg, A. A., Sofini, T. and Moore, G. E. Chronology and pattern of human chromosome replications. IV. Auto- radiographic studies of binucleate cells. Proc. Nat. Acad. Sci. 56: 105, 1966. Benirschke, K. (ed.) Comparative Aspects of Reproductive Failure. Springer-Verlag, New York, 1967. Horstmann, D. M., Banatvala, J. E., Riordan, J. T., Payne, M. C., Whitemore, R., Opton, F. M., DuVeflorey, C. Maternal rubella and a rubella syndrome infants. Am. J. Dis. Child. 110: 408, 1965. Stewart, H. L., Jr. Hormone secretion by human placenta grown in the eyes of rabbits. Am. J. Obst. Gynec. 61: 990, 1951. Thiede, H. A. and Rudolph, J. H. A method for obtaining monolayer cultures of human fetal cells from term placentas. Proc. Soc. Exper. Biol. Med. 107: 565, 1961. Tao, T. W. and Hertig, A. T. Viability and differentiation of human trophoblast in organ culture. Am. J. Anatomy 116: 315, 1965. Alvarez, H. Diagnosis of hydatidiform mole by transabdominal placental biopsy. Amer. J. Obstet. Gynec. 95: 538, 1966. Marine, W. M., Freeman, M. G. and McLennan, J. G. Localization of rubella virus in therapeutic abortions. Cl. Res. 14: 59, 1966. Selzer, G. Rubella in pregnancy. South Afr. J. Obst. Gynaec. 2: 5, 1964. Scothorn, M. W., Koutz, F. R. and Groves, H. F. Prenatal Toxocara canis infection in pups. J. Am. Vet. M. Ass. 146: 45, 1965. Koutz, F. R., Groves, H. F. and Scothorn, M. W. The prenatal migration of Toxocara canis larvae and their relation- ship to infection in pregnant bitches and in pups. Am. J. Vet. Res. 27: 789, 1966. 62. 63. 65. 67. 68. 69. 70. 71. 72. 73. Benirschke, K. and Driscoll, S. G. The Pathology of the Human Placenta. Springer-Verlag, New York, 1967. Farber, S. and Craig, J. M. Clinical pathological conference. J. Pediat. 49: 752, 1956. Kline, B. S. Microscopic observations of development of human placenta. Am. J. Obstet. Gynec. 61: 1065, 1951. Cohen, F. and Zuelzer, W. W. The transplacental passage of maternal erythrocytes into the fetus. Am. J. Obstet. & Gyn. 93: 566, 1965. Parkman, P. D., Phillipps, P. E. and Meyer, H. M. Experimental rubella virus infection in pregnant monkeys. Am. J. Dis. Child. 110: 390, 1965. Hendricks, A. G., Axelrod, L. R. and Claborn, L. C. Thalidomide syndrome in the baboon. Nature 210: 958, 1966. Gregg, N. M. Congenital cataract following German measles in the mother. Tr. Ophth. Soc. Australia 3: 35-46, 1941. Swan, C. Rubella in pregnancy as an aetiological factor in congenital malfor- mation, stillbirth, miscarriage and abortion. J. Obstet, Gynaec. Brit. Empire 56: 341-363, 1949. Hampar, B. and Ellison, S. Chromosomal aberrations induced by an animal virus. Nature (London) 192: 145-147, 1961. Koprowski, H., Ponten, J., Jensen, F., Ravdin, R., Moorhead, P. and Saksela, E. Transformation of cultures of human tissue infected with simian virus SV40. J. Cell. and Compar. Physiol. 59: 281-292, 1962. Robinson, A. and Puck, T. Studies on chromosomal nondisjunction in man. II. Amer. J. of Human Genetics 19: 2, 112-129, 1967. Nichols, W., Levan, A. and Kihlman, B. Chromosome breakage associated with viruses and DNA inhibitors. Cytogenetics of Cells in Culture, R. J. C. Harris (ed.) Academic Press, Inc., New York, pp. 255-271, 1964. 167 74. 75. Aula, P. Virus-associated chromosome breakage. A sytogenic studyof chicken- pox, measles and mumps patients and of cell cultures infected with measles virus. Ann. Acad. Sci. Fenn. Series A, IV Biologica 89: 1-75, 1965. El-Alfi, Omar S., Smith, P. M. and Biesele, J. Chromosomal breaks in human leucocyte cultures induced by an agent in the plasma of infectious hepatitis patients. Separat ur Hereditas 52: 285-294, 1964. 76. Harnden, D. G. 77. 78. 79. 80. Cytogenetic studies in patients with virus infections and subjects vaccinated against yellow fever. Amer. J. Hum. Genet. 16: 204-213, 1964. Stoller, A. and Collmann, R. Virus Aetiology for Down’s Syndrome (mongolism). Nature 208: 903-904, 1965. Carr, D. H. Chromosome anomalies as a cause of spontaneous abortion. Am. J. Obstet. Gynec. 97: 3, 283-293, 1967. Szulman, A. Chromosomal aberrations in spontaneous human abortions. New Eng. J. Med. 272: 811-818, 1965. Geneva Conference: Standardization of procedures for chro- mosome studies in abortion. Cytogenetics 5: 361-393, 1966. 81. Plotkin, S. A., Boué, A. and Boué, J. 82. 83. 168 The in vitro growth of rubella virus in human embryo cells. Am. J. Epid. 81: 71, 1965. Schlegel, R., Neu, R., Carneiro Leao, J., Farias, E., Aspillage, M. and Gardner, L. Observations on the chromosomal, cytological and anatomical char- acteristics of 75 human conceptuses including euploid, triploid XXX, triploid XYY and mosaic triploid XXY/diploid XY cases. Cytogenetics 5: 430-446, 1966. Nichols, W., Levan, A., Aula, P. and Norrby, E. Chromosome damage associated with the measles virus in virto. Hereditas 54: 101-118, 1965. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. Girardi, A., Weinstein, D. and Moorhead, P. SV40 transformation of human diploid cells: Parallel studies of viral and karyological parameters. Ann. Med. Exp. Fenn. 44: 242-254, 1966. Martineau, M. and Oxon, B. A. A similar marker chromosome in testicular tumours. Lancet 1: 839-842, 1966. McFeely, R. A. A direct method for the display of chromosomes from early pig embryos. J. Reprod. Fertil. 11: 161, 1966. Hayflick, L. The limited in vitro lifetime of human diploid cell strains. Exp. Cell Res. 37: 614, 1965. Leuchtenberger, C. and Leuchtenberger, R. Quantitative cytochemical studies on the relation of deoxyribonucleic acid of cells to various pathological conditions. Biochem. Pharm. 4: 128, 1960. Bahr, G. F. and Wied, G. L. Cytochemical determinations of DNA and basic protein in bull spermatozoa. Ultraviolet spectrophotometry, cytophotometry and microfluorometry. Acta Cytol. 10: 393, 1966. Gustavsson, I. Chromosome abnormality in cattle. Nature 211: 865, 1966. Chang, T. H., Moorhead, P., Boué, J. Plotkin, S. and Hoskins, J. M. Chromosome studies of human cells infected in utero and in vitro with rubella virus. Proc. Soc. Biol. & Med. 122: 236-243, 1966. Whelton, J. A., Ingall, D., Bamford, S., Smedal, M. I. and Wright, K. Evaluation for genetic damage in offspring of women with high- level gonadal irradiation. New Eng. J. Med. 273: 1266, 1965. See also editorial comment in Obstet. Gynec. Survey 22: 329, 1967. Shuttleworth, G. E. Mongolism Imbecility. Brit. Med. J. 2: 661-665, 1909. 169 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 170 Stark, C. R. and Mantel, N. Effects of maternal age and birth order on the risk of mongolism and leukemia. J. Nat. Cancel Inst. 37: 687-698, 1966. MacMahon, B. and Newill, V. A. Birth characteristics of children dying of malignant neoplasms. J. Nat. Cancel Inst. 28: 231-244, 1962. Fraumeni, J. F., Jr. and Miller, R. W. Epidemiology of human leukemia: recent observations. J. Nat. Cancer Inst. 38: 593-605, 1967. Stark, C. R. and Fraumeni, J. F., Jr. Viral hepatitis and Down’s syndrome. Lancet 1: 1036-1037, 1966. Leck, I. Incidence and epidemicity of Down’s syndrome. Lancet 2: 457-460, 1966. Kogen, A., Kronmal, R. and Peterson, D. R. Viral hepatitis and Down’s syndrome. Lancet 1: 615, 1967. Ceccarelli, G. and Torbidoni, L. Viral hepatitis and Down’s syndrome. Lancet 1: 438, 1967. Stark, C. R. and Mantel, N. Lack of seasonal- or temporal-spatial clustering of Down’s syndrome birth in Michigan. Amer. J. Epid. 86: 199-213, 1967. Stewart, A., Webb, J. and Hewitt, D. A survey of childhood malignancies. Brit. Med. J. 1: 1495-1508, 1958. Miller, R. W. Relation between cancer and congenital defects in man. New Eng. J. Med. 275: 87-93, 1966. Miller, R. W. Persons at exceptionally high risk of leukemia. Cancer Res. (in press). Peters, J. A. Standard Nomenclature and data retrieval in veterinary medicine. Amer. J. Vet. Res. 28: 531-537, 1967. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. Priester, W. A., Adelstein, E. H. and Peters, J. A. Standard Nomenclature of Veterinary Diseases and Operations. Washington, D.C., Government Printing Office, 1966, p. 622. Priester, W. A., Schirmer, R. G. and Rines, M. P. A one-year analysis of veterinary patients, diagnoses, and operations at Michigan State University. J. Amer. Vet. Med. Ass. 148: 666-671, 1966. Edwards, J. H. Congenital malformations of the central nervous system in Scotland. Brit. J. Prev. & Soc. Med. 12: 115-130, 1958. Milham, S., Jr. Increase incidence of anencephalus and spina bifida in siblings of affected cases. Science 138: 593-594, 1962. Carr, D. H. Chromosome studies in spontaneous abortions. Obstet. & Gynec. 26: 3, 308-326, 1965. LaChapelle, A. Cytogenetical and clinical observations in female gonadal dysgenesis. Acta Endocrinologica 40: Supple. 65, 1962. Carr, D. H. Cytogenetics of abortions. Comparative Aspects of Reproductive Failure, K. Benirschke (ed.). Springer-Verlag, New York, 1967, p. 96. Boris, M., Blumberg, R., Feldman, D. B. and Sellers, T. F., Jr. Increased incidence of meningomyeloceles. J.A M.A. 184: 766-767, 1963. Report: Conference on Prenatal Effects of Drugs. Commission on Drug Safety, Chicago, Illinois, March 29-30, 1963. Levitan, M. and Schiller, R. Further evidence that the chromosome breakage factor in drosophilia robusta involves a maternal effect. Genetics 48: 1231-1238, 1963. Henle, G. and Henle, W. Evidence for a persistent viral infection in a cell line derived from Burkitts lymphoma. J. Bacteriology 89: 252, 1965. 171 293-596 0—68——12 117. 118. 119. 120. 121. 122. 123. 124. 172 Fujinaga, K. and Green, M. Mechanisms of viral carcinogenesis by deoxyribonucleic acid mam- malian viruses. J. Virology 1: 576, 1967. Sever, J. L., Nelson, K. B. and Gilkeson, M. R. Rubella epidemic, 1964: Effect on 6,000 pregnancies. Diseases of Children 110: 395, 1965. Sever, J. L., Huebner, R. J., Castellano, G. A. and Bell, J. A. Serologic diagnosis “en masse” with multiple antigens. Am. Rev. Resp. Dis. 88: 342, 1963. Hanshaw, J. B. Cytomegalovirus complement fixing antibody in microcephaly. New Eng. J. Med. 275: 476, 1966. Robbins, F. C. and Oker-blom, G. Rubella studies in Finland. Personal Communication. Burnet, F. M. Men or molecules? A tilt at molecular biology. Lancet 1: 37-39, 1966. McLellan, M. D., Strong, J. P., Vautier, T. and Blatt, I. M. Otitis media in the newborn. Relationship to duration of rupture of amniotic membrane. Arch. Otolaryng. 85: 380, 1967. MacArthur, P. Congenital Vaccinia and Vaccinia Gravidarum. Lancet 2: 1104, 1952. Reference Bibliography VIRAL REFERENCES Alford, C., Snider, M. and Stubbs, G. Studies on the virulence of herpes simplex viruses isolated from different clinical entities. (Abs.) Soc. Ped. Res., Atlantic City, N.J., April 28-29, 1967, p. 26. Ashley, C. A. Study of the human placenta with the E. M. Arch. Path. 80: 377, 1965. Banatvala, J. E., Horstman, D. M., Payne, M. C. and Gluck, L. Rubella syndrome and thrombocytopenic purpura in newborn infants. New Eng. J. Med. 273: 474-478, 1965. Benyesh-Melnick, M., Rosenberg, H. S. and Watson, B. Viruses in cell cultures of kidneys of children with congenital heart mal- formations and other diseases. Proc. Soc. Exp. Biol. Med. 117: 451-459, 1964. Brown, G. C. Recent advances in the viral aetiology of congenital anomalies. Advances in Teratology, D. H. M. Woollam (ed.) Academic Press, London, 1966, p. 55. Burch, G. E., DePasquale, N. P., Sun, S. C., Hale, A. R. and Mogabgab, W. J. Experimental Coxsackievirus endocarditis. J.A.M.A. 196: 349-353, 1966. Defendi, V. Defendi, V. Cytopathology of virus infection. Federation Proceedings 21: 1113-1117, 1962. Eichenwald, H. F. and Shinefield, H. R. Viral infections of the fetus and of the premature and newborn infant. Advances in Pediatrics 12: 249-305, 1962, S. Z. Levine, ed. Ferm, V. H. Developmental malformations as manifestations of reproductive failure and Experimental viral infection of the placenta. Comparative Aspects of Reproductive Failure, K. Benirschke (ed.) Springer-Verlag, New York, 1967. Ferm, V. H. and Kilham, L. Congenital anomalies induced in hamster embryos with H-1 virus. Science 145: 510-511, 1964. 173 Ferm, V.H. and Low, R. J. Herpes simplex virus infection in the pregnant hamster. J. Path. Bact. 89: 295, 1965. Gersony, W. B., Katz, S. L. and Nadas, A. S. Endocardial fibroelastosis and the mumps virus. Pediatrics 37: 430-434, 1966. Gwatkin, R. B. L. Effect of viruses on early mammalian development. I. Action of mengo encephalitis virus on mouse ova cultivated in vitro. Proc. Nat. Acad. Sci. 50: 576-581, 1963. Gwatkin, R. B. L. Effect of viruses on early mammalian development. III. Further studies con- cerning the interaction of mengo encephalitis virus with mouse ova. Fertility and Sterility 17: 411-420, 1966. Gwatkin, R. B. L. and Auerbach, S. Synthesis of a ribonucleic acid virus by the mammalian ovum. Nature 209: 993-994, 1966. McAllister, R. M., Straw, R. M., Filbert, J. E. and Goodheart, C. R. Human cytomegalovirus. Cytochemical observations of intracellular lesion development correlated with viral synthesis and release. Virology 19: 521-531, 1963. McFeely, R. A. Cromosome abnormalities in early embryos of the pig. J. Reprod. Fertil. 13: 579, 1967. Mims, C. A. Immunofluorescence study of the carrier state and mechanism of infection in mice. J. Path. Bact. 91: 395-402, 1966. Mims, C. A. and Subrahmanyan, T. P. Immunofluorescence study of the mechanism of resistance to superinfection in mice carrying the lymphocytic choriomeningitis virus. J. Path. Bact. 91: 403-415, 1966. Naeye, R. L. and Blanc, W. Pathogenesis of congenital rubella. JLAM.A. 194: 1277-1283, 1965. Naib, Z. M. Exfoliative cytology of viral cervicovaginitis. Acta cytol. 10: 126, 1966. 174 Plotkin, S. A. and Vaheri, A. Human fibroblasts infected with rubella virus produce a growth inhibitor. Science 156: 659-661, 1967. Rubella Symposium, S. Krugman, ed. Amer. J. Dis. Child. 110: 345-476, 1965. Selzer, G. Rubella in pregnancy: Virus isolation and inclusion bodies. South African J. of Obs. & Gyn. 2: 5, 1964. Shultz, G. and DeLay, P. D. Losses in newborn lambs associated with blue tongue vaccination of preg- nant ewes. J. Am. Vet. M. A. 127: 224, 1955. Toolan, H. W., Dalldorf, G., Barclay, M., Chandra, S. and Moore, A.C. An unidentified filterable agent isolated from transplanted human tumors. Proc. Nat. Acad. Sci. U.S. 46(9) : 1256, 1960. Weller, T. H. and Hanshaw, J. B. Virologic and clinical observations on cytomegalic inclusion disease. New Eng. J. Med. 266: 1233-1244, 1962. Young, G. A., Kitchell, R. L., Luedke, A. J. and Sautter, J. H. Effect of viral and other infections of the dam on fetal development in swine: Modified live hog-cholera viruses. J. Am. Vet. M. A. 126: 165, 1955. IMMUNOLOGIC REFERENCES Billingham, R. E. Transplantation immunity and the maternal-fetal relation. New Eng. J. Med. 270: 667, 720, 1964. Galton, M. Immunologic interaction between mother and fetus. Comparative Aspects of Reproductive Failure, K. Benirschke, ed. Springer-Verlag, New York, 1967. Hasek, M., Lengerova, A. and Hraba, T. Immunologic tolerance. 1st 2 chapters in Advances Immunol. 1: 1, 67, 1961. Hotchin, J. The biology of lymphocytic choriomeningitis infection: virus-induced im- mune disease. Cold Spring Harbor Symp., Quant. Biol. 27: 475, 1962. 175 Ontogeny of Immunity, R. F. Smith, et al., eds. University of Florida Press, Gainesville, 1967. Osborn, J. J., Dancis, J. and Julia, J. F. Studies of the immunology of the newborn infant. Pediatrics 9: 736, 1952. Richards, W. P. C. and Cordy, D. R. Blue tongue virus infection: Pathologic responses of nervous systems in sheep and mice. Science 156: 530, 1967. Silverstein, A. M. Congenital syphilis and the timing of immunogenesis in the human fetus. Nature 194: 196, 1962. Silverstein, A. M. Ontogeny of the immune response. Science 144: 1423, 1964. Silverstein, A. M., Prendergast, R. A. and Kraner, K. L. Fetal response to antigenic stimulus. IV. The rejection of skin homografts by the fetal lamb. J. Exp. Med. 119: 955, 1964. Sterzl, J. and Silverstein, A. M. Developmental aspects of immunity. Advances Immunol. 6: 337-459, 1967. FETAL REFERENCES Gruenwald, P. Chronic fetal distress and placental insufficiency. Biol. Neonat. 5: 215-265, 1963. Gruenwald, P. Growth retardation of the embryo and fetus. Official J. Cong. Anom. Res. Ass. Japan 6: 9-10, 1966. Gruenwald, P. Growth of the human fetus. I. Normal growth and its variation. Am. J. Obstet. Gynec. 94: 1112-1119, 1966. Gruenwald, P. Malformations caused by necrosis in the embryo, illustrated by the effect of selenium compounds on chick embryos. Am. J. Path. 34: 77-103, 1958. Gruenwald, P., Dawkins, M. and Hepner, R. Panel discussion: chronic deprivation of the fetus. Sinai Hosp. J. 11: 51-80, 1963. 176 Gruenwald, P. Funakawa, H., Mitani, S., Nishimura, T. and Takeuchi, S. Influence of environmental factors on foetal growth in man. Lancet 1: 1027-1029, 1967. Naeye, R. L. and Blanc, W. Pathogenesis of congenital rubella. J.AM.A. 194: 1277-1283, 1965. Siegel, M. and Fuerst, H. T. Low birth weight and maternal virus diseases: a prospective study of rubella, measles, mumps, chickenpox, and hepatitis. J.A.M.A. 197: 680-684, 1966. PLACENTAL REFERENCES Adamsons, K., Jr. and Joelsson, I. The effects of pharmacologic agents upon fetus and newborn. Ah. J. Obst. Gynec. 96: 437, 1966. Dancis, J. The role of the placenta in fetal survival. Ped. Clin. N. Amer. 12: 477, 1965. Dancis, J., Douglas, G. W. and Fierer, J. Immunologic competence of mouse placental cells in irradiated hosts. Am. J. Obst. Gynec. 94: 50, 1966. Dancis, J. and Shafran, M. The origin of plasma proteins in the guinea pig fetus. J. Clin. Invest. 37: 1093, 1958. Garby, L. Studies on transfer of matter across membranes with special reference to isolated human amniotic membrane. Acta Physiol. Scandinav. 40: suppl. 137, 1957. Le Placenta Humain, Jean Snoeck, ed. Masson et Cie, Paris, 1958. Potter, E. L. The effect on the fetus of viral disease in the mother. Clin. Obstet. Gynec. 4: 327, 1961. Symposium on the Placenta, A. A. Plentl, ed. Am. J. Obstet. Gynec. 84: #11, part 2, 1962. The Placenta and Fetal Membranes, C. Villee, ed. Williams and Wilkins, 1960. Uhr, J. W., Dancis, J. and Finkelstein, M. S. Passage of bacteriophage ¢X174 across the placenta in guinea pigs. Proc. Soc. Exper. Biol. Med. 113: 391, 1963. 177 CYTOGENETIC REFERENCES Diaz, M. and Pavan, C. Changes in chromosomes induced by microorganism infection. Proc. Nat. Acad. Sci. 54: 1321-1327, 1965. Moorhead, P. and Weinstein, D. Cytogenetic alterations during malignant transformation. Recent Results in Cancer Research VI; Malignant Transformation by Viruses, W. H. Kirsten, ed. Springer-Verlag, New York, 1966, pp. 104-111. Shein, H. and Enders, J. Multiplication and cytopathogenicity of simian vacuolating virus 40 in cul- tures of human tissues. Proc. Soc. Exp. Biol. and Med. 109: 495-500, 1962. Stich, H., Hsu, T. and Rapp, F. Viruses and mammalian chromosomes 1. Localization of chromosome aber- rations after infection with herpes simplex virus. Virology 22: 439-445, 1964. EPIDEMIOLOGIC REFERENCES Editorial: Viruses and congenital heart disease. Lancet 1: 608-609, 1967. Hardy, J. B. Viral infection in pregnancy: a review. Amer. J. Obstet. & Gynec. 93: 1052-1065, 1965. MacMahon, B. and Sowa, J. M. Physical damage to the fetus. Milbank Mem. Fund. Quart. 39: 14-73, 1961. Manson, M. M., Logan, W. P. D. and Loy, R. M. Rubella and other virus infections during pregnancy. Her Majesty’s Stationery Office, Ministry of Health Publication #101, London, 1960. Shapiro, S., Jones, E. W. and Densen, P. M. A life table of pregnancy terminations and correlates of fetal loss. Milbank Mem. Fund. Quart. 40: 7-45, 1962. Shapiro, S., Ross, L. J. and Levine, H. S. Relationship of selected prenatal factors to pregnancy outcome and con- genital anomalies. Amer. J. Public Health 55: 268-282, 1965. Tjalma, R. A. Canine bone sarcoma: estimation or relative risk as a function of body size. J. Nat. Cancer Inst. 36: 1137-1150, 1966. 178 U.S. GOVERNMENT PRINTING OFFICE: 1968 O—293-596 U.C. BERKELEY LIBRARIES UMAR co223uuak3