Review and Abstracts

Viruses and the Pathogenesis of Diabetes Mellitus Elliot J. Ray field, M.D., and Yoshiko Seto, Ph.D., New York INTRODUCTION

The etiology of juvenile-onset insulin-dependent (JIDM) diabetes mellitus in man, which represents about 20 per cent of all cases of diabetes, remains elusive. The purpose of this review is to present evidence supporting a viral pathogenesis of diabetes in man and experimental animal models and the possible relationship of viral mechanisms to genetic and immunologic factors. This topic has been reviewed in the virology literature in recent years,1*2 and a brief review appeared in this journal in 1974. 3 Data concerning the hypothetic or documented role of viruses in the etiology of diabetes are derived from (1) studies of pancreatic pathology (insulitis), (2) genetics (including twin studies), (3) anecdotal reports that followed several viral illnesses, (4) epidemiologic studies in which elevated specific viral antibody titers are correlated with an outbreak of a viral illness or significantly increased incidence of diabetes in a population, and (5) animal models of virus-induced diabetes and/or pancreatic disease. Table 1 is a summary of several viruses implicated in the pathogenesis of diabetes or pancreatic disease, the species and the specific parts of the pancreas affected, and the authors who reported the findings. Viral Infections in Man Mumps virus. In 1864, the Norwegian physician J. Stang reported that diabetes developed in one of his patients shortly after mumps infection and hypothesized a causal relation between the two events.4 Harris, in this country, described in 1899 "A Case of Diabetes Mellitus Quickly Following Mumps." 5 After these two early case reports, many others have appeared describing a temporal relationship between mumps virus infection and the deFrom the Department of Medicine, Mount Sinai School of Medicine, New York, NY 10029. Address reprint requests to Dr. Elliot J. Rayfield, Mount Sinai School of Medicine, New York, NY 10029. 1126

velopment of diabetes. 6 " 14 Indeed, mumps virus has been the virus most frequently reported to be associated with diabetes mellitus; much more common, however, is the association of mumps pancreatitis without diabetes mellitus. 15 " 19 Symptoms of diabetes after mumps virus infection have been reported to vary from a few days to several months, and the hyperglycemia and glycosuria may be evanescent or persistent. In this connection, it should be pointed out that viral infections 20 ' 21 and even fever22 promote nonspecific alterations in carbohydrate metabolism associated with elevation of insulin counter-regulatory hormones (cortisol, glucagon, and growth hormone). Two reports document the onset of diabetes after mumps in two siblings 9 ' 1 2 almost simultaneously. The few cases of mumps pancreatitis that have been examined pathologically have shown an acute inflammatory process with edema, mononuclear cell infiltration, and necrosis and degeneration of epithelial cells but may not be representative of all 18

cases. Using an epidemiologic approach, Sultz conducted a survey in Erie County, New York, having a defined childhood population of 300,000; 112 parents of diabetic children in the area (about one third of the diabetic cases) were interviewed. 23 Parents were questioned regarding the age at occurrence of mumps vaccination, exposure to mumps, or mumps in relation to age at the onset of diabetes mellitus. Interestingly, there was a mean lag time of 3.8 years between mumps infection or vaccination and the onset of diabetes mellitus. Koch's postulates remain to be demonstrated with regard to the relationship of mumps virus infection and subsequent diabetes mellitus- in man. Since sequential pancreatic biopsies during a mumps infection to determine whether mumps virus replicates in and injures beta cells is not possible in man, it would be important to ascertain whether mumps virus could replicate in vitro in cultured human pancreatic beta cells. In point of fact, Prince and co-workers recently showed that the ABC-strain of mumps virus DIABETES, VOL. 2 7 , NO. 11

ELLIOT J. RAYFIELD, M . D . , A N D YOSHIKO SETO, PH.D.

TABLE 1 Viruses implicated in the pathogenesis of diabetes mellitus or pancreatic disease in man and animal models Type of virus RNA

Virus Mumps

Species affected Man

Human (in vitro) Monkey (in vitro) Coxsackie B

Man

Mice

Pathologic changes Pancreatitisedema, mononuclear cell infiltration, and necrosis and degeneration of epithelial cells Establishment of infection in cultured pancreatic beta cells

References Craighead, 1975;1 Bostrb'm, 196818

Focal necrosis and inflammation of exocrine pancreas Necrotizing lesions of exocrine pancreas

Kibrick and Benirshke, 1958149

Occasional and fine structural alterations of beta cells Slight damage in both islet and acinar tissue Encephalomyocarditis, M variant

Foot-and-

Guinea mouth Disease

Mice

Alterations in the pigs Cattle

Mice Venezuelan equine encephalomyelitis

Hamsters

Monkey

Mice (C57KSJ db/db)

Degranulation and coagulation necrosis with subsequent shrinking of beta cells and architectural alteration of islets Functional abnormalities in both alpha and beta cells. Degeneration and necrosis of isolated acinar cells or acinar cells adjacent to the islets of Langerhans Platt, 1958153 pancreatic zymogen tissue Almost total absence of islets, with infiltration of round cells and some acinar and ductal necrosis Lesions of the pancreatic acinar tissue TC-83 strain—focal acinar necrosis, acutely No changes in convalescent phases of study No changes at 10 months postinoculation with Trinidad strain TC-83 strain—beta cell degranulation and changes in subcellular organelles, especially mitochondria

Prince et al., 197824

Pappenheimer et al., 1951;73 Vizoso and Sanders, 1964150 Robertson, 1954;155 Burchet al., 1971;78 Tsuietal., 1972;79 Harrison et al., 1972;80 Coleman et al., 1973*' and 197482

Boucher and Notkins, 1973;88 Craighead and McLane, 1968;86 Craighead and Steinke, 1971;87 Hayashiet al., 1974;90 Munterfering, 1972;151 Wellmannetal., 1972;1S2 Peterson et al., 197591

Craighead, 1975;1 Hayashiet al., 1974;90 Munterfering, 1972;1SI Wellmannetal. IS2

Barboni, 196667

Platt, 1956154 and 195970 Rayfield et al., 1 9 7 6 104

Rayfield and Bowen, 1977;IOS Bowen and Rayfield, 1977106 Goldberg et al., J 978 107

(continued) NOVEMBER, 1978

1127

VIRUSES AND THE PATHOGENESIS OF DIABETES MELUTUS

Table 1 (cont'd.) Type of Virus

Virus

Species affected

Rubella

Man

Rabbits

DNA

7

RNA

Cytomegalovirus

Man

Pathologic changes

References

Lymphocytic infiltration of interstitial periglandular tissue Beta cell degranulation and changes in subcellular organelles

Forrest et al., 1971;37 Bunnel and Monif,

Typical intranuclear inclusions in the pancreatic acinar, ductal, and insular cells Herpes virus-like particles in electron micrographs of alpha cells in the islets of Langerhans

Capped and McFarlane, 194750

Gepts, 1972"

Infectious mononucleosis

Man

Pancreatitis

Wislocki, 1966;51 Everett et al., 196952

Varicella

Man

Inclusions in acinar and endothelial cells in the pancreatic cells

Johnson, 1940;" Oppenheimer, 1944;54 Cheathamet al., 1956;55 Blattner, 195756

Spontaneous (transmissible agent) diabetes mellitus C-type virus particles induced by multiple subdiabetogenic injections of streptozotocin

Guinea Pig

Degranulation of beta cells, and cytoplasmic inclusions with sparing of A- and D-cells Insulitis (mononuclear infiltration of islets)

Lang and Munger, 1976;72 Munger and Lang, 197371 Like and Rossini, 1976;65 Rossini et al., 197766

Mice

was capable of infecting cultured human pancreatic beta cells (obtained at autopsy from seven subjects).24 A double-labeled antibody technique was used to reveal whether a beta cell or other pancreatic cell was infected.24 This technique used fluorescein-labeled anti-mumps antibody to localize mumps-infected cells and rhodamine-labeled anti-insulin antibody to localize insulin-containing beta cells. A doublestained cell, identified by screening sections first with the rhodamine filters and then with the fluorescein filters, represented a mumps-infected beta cell. The double-labeled antibody method provides an important technique for quantifying a specific viral infection in a specific polypeptide-containing pancreatic islet cell. While it is yet to be proved that mumps virus can produce diabetes in man, the ability of any virus to grow in human pancreas in culture may prove useful in screening viruses that are potentially diabetogenic. 1128

j 97244

Menseret al., 197840

Coxsackie virus. In 1969 Gamble and associates reported that within three months of onset of disease, insulin-dependent diabetics had a higher neutralizing antibody titer to Coxsackie-B4 virus than did either normal subjects or patients with diabetes of a duration longer than three months. 25 Types Bl and B3 demonstrated the same trend to a lesser extent. 25 At the same time, Gamble and Taylor described a seasonal incidence of insulin-dependent diabetes mellitus most pronounced in patients up to age 19, with one peak in July and a larger peak in October. 26 Furthermore, the pattern of variation showed a significant positive correlation with the annual prevalence data for Coxsackie virus, type B4, but not for other types of viral infection; 26 other workers have not been able to confirm these findings.27"31 In one of these studies, five years after an outbreak of Coxsackie B4 in the Pribolof Islands, glucose tolerance tests were performed.30 Of the 136 persons under age 25, none exhibited diabetic DIABETES, VOL. 2 7 , NO. 11

ELLIOT J. RAYFIELD, M.D., AND YOSHIKO SETO, PH.D.

two-hour values, while persistence of positiveneutralizing antibody titers was shown in 77 per cent of these subjects. In another study, concentrations of the antibody to Coxsackie B4 were assessed in 49 identical twins (22 concordant for diabetes and 27 discordant for diabetes). 31 If viruses are causal agents in diabetes, it would be anticipated that the diabetic twins in the discordant pairs would have either a higher incidence or a higher titer of antibody than the nondiabetic twins. The data showed that the diabetic twins in the discordant pairs, in fact, did not have higher antibody titers to several viruses, including Coxsackie B4, mumps, and rubella, than did the nondiabetic twins. 31 Further work will be necessary to evaluate fully the role, if any, Coxsackie-B virus plays in the pathogenesis of JIDM. Rubella virus. Since 1949, 32 documented cases of diabetes mellitus have been reported in patients with the congenital rubella syndrome. 32 ' 40 In 1971, Forrest, Menser, and Burgess37 reviewed their original 50 cases of congenital rubella, in which they had already described five diabetics by 1967. 33 They discovered that four additional patients had diabetic glucose tolerance curves and a further four had slight abnormalities in glucose tolerance together with abnormal insulin responses, suggestive of latent diabetes mellitus. Insulin concentrations were not measured in patients with normal glucose tolerance. Therefore, patients with decreased beta cell reserve who did not yet have decompensated glucose tolerance could have been overlooked. All patients had severe stigmata of the congenital rubella syndrome, including deafness, cataracts, pulmonary-valve stenosis, and chorioretinopathy, and four (of nine) had a family history of diabetes. The most recent report from the Australian workers cited that, in addition to the 9/45 of the congenital rubella patients from the 1967 follow-up series who had clinical diabetes (20 per cent incidence), there were eight other patients with diabetes mellitus and congenital rubella. 40 As part of these same studies, it was also shown that HLA-B8 was present in 50 per cent of the diabetics (both clinical diabetics and patients with abnormal glucose tolerance) in contrast to 24 per cent of a control population. 41 This antigen has been shown to be increased in frequency in insulin-dependent juvenile-onset diabetes. 42 These patients did not show an increase in HLA-BW15, which is also associated with JIDM. 4 2 It is noteworthy that, in children currently being followed in the Roosevelt Hospital (New York, NY) NOVEMBER, 1978

Rubella Project, the prevalence of insulin-dependent diabetes mellitus is now 20 per cent. 43 Since rubella virus multiplies and sometimes causes mononuclear pancreatic infiltrates, 44 the development of diabetes mellitus at a later date is not surprising. One could postulate that, since chronically infected cells cultured with rubella virus have a shortened life span 45.46 the same could be applicable to the pancreatic beta cell. Since growth hormone deficiency47 and hypothyroidism 48 have also been described in the congenital rubella syndrome, rubella virus may be tropic to a number of endocrine tissues in addition to the beta cell. Long-term prospective studies, clearly, are needed to follow patients with congenital rubella and their siblings for the subsequent development of diabetes mellitus (both subclinical and clinical). Encephalomyocarditis (EMC) virus (M-variant). Neutralizing antibody titers to EMC virus have been found in 12 per cent of 41 patients with juvenile diabetes in contrast to 6 per cent of 66 control subjects. 49 In some of the diabetics, serum was not obtained until several years after the onset of diabetes, when antibody titers could have returned to normal. While it is possible that certain cases of JIDM are induced by EMC virus, the data are too limited to establish definite conclusions. Other Viruses

Although cytomegalovirus,50 infectious mononucleosis, 51 - 52 and varicella53"56 can cause pancreatic lesions in man, and diabetes mellitus has been associated temporally with viral infections from cytomegalovirus, measles, polio, influenza, and tickborne encephalitis, 57 " 60 solid evidence is not available at present concerning other viruses that are diabetogenic in man. Insulitis Defined as infiltration of pancreatic islets by mononuclear cells, chiefly lymphocytes, insulitis has been recognized since 1902. 61 The lesion is characteristic in children or young adults with recent onset diabetes 62 - 63 who have come to autopsy, although it has been described in certain patients with late onset diabetes as well. 64 The presence of a.mononuclear infiltration in the pancreas of insulin-dependent diabetics is compatible with a previous viral infection, an autoimmune process, or a combination of the two. Such insulitis has also been demonstrated recently after multiple subdiabetogenic doses of streptozotocin in mice 65 ' 66 and in Wistar rats with a syndrome of spontaneous diabetes. 108 Thus, pancreatic pathologic changes occurring in recent onset, insulin-dependent 1129

VIRUSES A N D THE PATHOGENESIS OF DIABETES MELLITUS

diabetes in childhood are consistent with a viral and/or autoimmune process. Animal Models of Diabetes Mellitus with a Definite or Possible Viral Etiology Foot-and-mouth-disease virus (FMDV). In 1962 and 1963 several reports from Italy described a diabeteslike syndrome in cattle, with anorexia, polyuria, hyperglycemia, and glycosuria. 67 " 69 The pancreases of the affected animals contained a paucity of islets of Langerhans, which when found showed mononuclear cell infiltrates similar to the insulitis in pancreatic tissue of patients with juvenile-onset diabetes mellitus. 62 Platt noted that FMDV-infected mice showed acinar cell necrosis but no islet lesions.70 However, McVicar et al., using a different strain of FMDV (Ctype), observed no elevation of blood glucose concentrations or of glucose intolerance for more than a month. 2 These observations illustrate how specific subtypes of a virus may result in different tropism for cells or tissues (table 2). Spontaneous diabetes mellitus in guinea pigs. Munger and Lang 71 ' 72 reported a spontaneous diabetes mellitus in guinea pigs that is contagious and is possibly due to a viral infection. The syndrome is characterized by glycosuria and abnormal glucose tolerance tests, although fasting blood sugars are variable. Pathologic examination of the beta cells disclosed degranulation and cytoplasmic inclusions with sparing of A-cells and D-cells; necrosis and/or inflammation of islets was not

noted. To date, studies on the controlled transmission and isolation of an infectious agent have not been completed. Coxsackie B4. Much investigation has been focused, in recent years, on the possible role of the Coxsackie-B group of viruses in causing diabetes mellitus in animals as it has in man (see above). In 1950 Pappenheimer et al. reported that suckling mice infected with the Powers strain of group-B Coxsackie virus developed necrosis of exocrine pancreatic tissue, 73 a finding that was confirmed by other workers. 74 " 77 Next Burch et al. 78 described insular as well as exocrine lesions in suckling mice infected with Coxsackie-B4 virus. Patchy islet degeneration and necrosis as well as some degranulation of beta cells were found. Similar results were obtained from electron microscopic evaluation of islets in mice infected with Coxsackie viruses Bl and B 3 . 7 9 ' 8 0 Since neither plasma glucose concentrations nor the presence of glycosuria were stated, one cannot be certain that these lesions resulted in measurable alterations in carbohydrate metabolism. Coleman et al. 8 1 ' 8 2 reported that Coxsackie-B4 virus produced hyperglycemia in 20 to 30 per cent of adult CD1 mice 15 to 20 days after infection. Microscopic examination showed only slight damage in both islet and acinar tissue, and random serum insulin levels were slightly elevated in infected mice in comparison with control mice except for 30 days postinoculation mice. On the other hand,

TABLE 2 Factors that influence virus-induced diabetes mellitus in animal models'" Factor Characteristics of virus Type Strain Dose Virulence Passage

Characteristics of host Species Age Sex Hormonal factors Genetic background of host

Influence

Comments

Mostly RNA-type and a few DNA-type viruses B4 in Coxsackie viruses, M variant in EMC Must be large enough to cause a moderate infectious illness but not so large as to be fatal (M variant of EMC virus) Should be attenuated in its primary illness in host to allow expression of diabetic state (M variant of EMC virus) Passage in vivo or in vitro through host is usually necessary for pancreaticotropic effect (Coxsackie-B4 virus)

Depends on presence of specific surface virus receptor on beta cell (M variant of EMC virus) Young adults most susceptible (M variant of EMC virus) Male > female (M variant of EMC virus) Estrogen, testosterone, adrenal corticosteroids (M variant of EMC virus) SWR strain of mice are susceptible to M variant of EMC virus; C57 Bl/6 strain of mice are resistant to M variant of EMC virus

•Each factor in this table is not applicable to every animal model or to observations in man. Where possible, a specific virus that applies to each characteristic in the list is noted. 1130

DIABETES. VOL. 2 7 , NO. 11

ELLIOT J. RAYFIELD, M . D . , A N D YOSHIKO SETO, PH.D.

Ross et al., in an extensive series of experiments that used Coxsackie-Bl-5 viruses in as many as six inbred strains of mice, showed increased serum amylase levels and normal or transient hypoglycemic glucose levels. 83 ' 84 These findings correlated with pathologic evidence of the marked destruction of acinar cells with virtually no effect on pancreatic beta cells. Thus, discrepancy in data from various laboratories can be ascribed to a wide range of susceptibilities in different inbred strains of animals as well as the huge number of viral variants that exist, some of which may be more betacytotropic than others.

duced by EMC virus varied greatly among the mice but correlated well with the degree of beta cell damage as assessed by pancreatic histology. 90 Male mice were shown to be more susceptible to EMC virusinduced hyperglycemia than female mice. 92 Castration of male mice resulted in augmented resistance similar to that of female mice. 92 Furthermore, the administration of testosterone markedly enhanced susceptibility both in castrated male and female mice, in addition to intact females.92 The increased resistance in castrated males could be eliminated by adrenalectomy. 92 These experiments point to the importance of hormonal factors in modulating susceptibility to Encephalomyocarditis virus (M-variant) (EMC). The virus-induced disease, which may well be relevant to most convincing data to support a viral etiology of the situation in man (table 2). diabetes are derived from work with the M-variant of The effect of EMC vims—induced hyperglycemia in encephalomyocarditis (EMC) virus (a picornavirus, mice upon the microangiopathic complications of like Coxsackie-B and foot-and-mouth-disease virus) in diabetes is uncertain. Amorphous, electron-dense mice. This virus was found to produce a diabetes 85 88 material in glomerular mesangium has been observed mellitus—like syndrome, " characterized by 2Vi to AVi weeks after EMC virus inoculation in male symptoms of polydipsia and polyphagia. PathologiDBA/2 mice, 93 although these mice did not exhibit cally, the beta cells exhibit degranulation and coaguglomerular capillary basement membrane thickening. lation necrosis with subsequent shrinking and arAdditional studies will be of interest to delineate more chitectural alteration of the islets. Pancreatic alpha 89 fully the development of renal, other microcells appeared normal and necrosis of acinar cells 89 90 angiopathic, and macroangiopathic abnormalities in occurred rarely. However, Petersen et al. rethese mice. ported that, as early as six hours after infection with EMC virus, perifused mouse islets exhibited a reducThe next series of studies of EMC virus—induced tion in large glucagon immunoreactivity and proinsudiabetes dealt with genetically determined factors of lin, indicating functional abnormalities in both alpha the host that could influence the susceptibility to deand beta cells. 91 velop murine diabetes. An EMC virus—susceptible strain (DBA/2J) and -resistant strain (C3H/J) of mice Early studies by Craighead's laboratory employed were bred, and the Fl generation was infected with CD-I-strain mice, 85 " 87 ' 89 whereas Boucher and NotEMC as well as back crossed against the parental kins noted that DBA/2N mice were more susceptible strains (before infection).94 The data obtained from to EMC virus—induced diabetes. 88 EMC virus was these experiments were interpreted to be consistent shown to replicate in beta cells within one to two days with a polygenic inheritance of virus-induced diabetes after inoculation, as assessed by localization with in these mice. 94 Neel et al. had proposed a similar fluorescein-labeled anti-EMC antibody in beta cells scheme for the inheritance of diabetes mellitus in but not in acinar cells. 90 Virus particles were not man. 95 Moreover, studies by Ross et al. 96 indicated noted by electron microscopy.89 During the acute that the inheritance of EMC virus—induced murine phase of infection, starting at about two days after diabetes is autosomal recessive and polygenic. inoculation, release of insulin occurred from disrupted beta cells, leading to hypoglycemia and decreased Recent experiments have* attempted to explain pancreatic insulin content. 87 In the next phase of illmechanistically how genetic factors confer virus resisness, four to seven days after inoculation, hyperglytance or susceptibility. It has been suggested as a cemia and hypoinsulinemia developed. 86 " 88 Finally, generality that immune-response (Ir) genes are responthree general patterns of glucose response were noted sible for associations between HLA and disease97 and in EMC-infected DBA/2N mice: 60 to 80 per cent of that H-2—linked Ir genes influence the magnitude of the animals demonstrated a transient hyperglycemia, immune response of the host to a viral infection.98 10 to 15 per cent had persistent hyperglycemia for six However, in the case of EMC-virus infection in mice, or more months, and 15 to 30 per cent never became there were no differences in anti-EMC—neutralizing hyperglycemic. 88 The degree of hyperglycemia inantibody titers three to five days after infection in NOVEMBER,

1978

1131

VIRUSES AND THE PATHOGENESIS OF DIABETES MELLITUS

diabetes-susceptible versus diabetes-resistant strains of m i c e . " In addition, diabetes-susceptible strains of mice could not be grouped on the basis of a common H-2 t y p e . " Another mechanism by which genetic factors might be manifested is through controling the degree of viral replication in the beta cells of the host. Studies to explore this possibility demonstrated that EMC virus—inoculated, susceptible, murine strains had higher pancreatic viral titers 100 than resistant strains. Interestingly, the higher viral titer resulted from an increased number of beta cells being infected rather than from greater amounts of virus replicating per infected cell. 1 0 0 > l o o a Recent immunofluorescent studies indicate that some resistant strains of mice have several times more infected islet cells than other resistant strains, which raised the possibility that only certain subpopulations of beta cells contain EMCvirus receptors that would allow viral penetration into selected beta cells. 101 Furthermore, at least twice as many EMC viruses attached to the beta cells from susceptible (SJL/J) than from resistant (C57BL/6J) strains of mice. 102 Thus, these data are also compatible with the notion that genetic control of differences in viral receptors on the surface of the beta cells may confer susceptibility or resistance to EMC-induced murine diabetes. 102 In order to investigate a role of the immune system in EMC-induced murine diabetes, Jansen et al. studied the effect of immunosuppression with 500-R sublethal X-irradiation or 120 mg. per kilogram Asta 5122, a cyclophosphamide derivative. 103 X-irradiated virus—inoculated mice maintained normal glucose concentrations and histologically showed beta cell necrosis (no quantitation of the degree of necrosis is presented) with striking absence of insulitis. In contrast, Asta 5 122—immunosuppressed plus EMC virus—infected mice exhibited significantly greater plasma glucose concentrations than control mice that were infected with EMC alone. Severe insulitis was present histologically in all chemically immunosuppressed, EMC virus-infected mice. These workers postulated that the increased insulitis after druginduced immunosuppression might be the result of T-lymphocyte—mediated destruction of the beta cells that had escaped attack by EMC virus. 103 Venezuelan encephalitis virus. Golden Syrian hamsters, inoculated with the virulent (Trinidad) strain of Venezuelan encephalitis (VE) virus, were found to have mature virions and viral antigens in beta cells by electron microscopy and immunofluores1132

cence. 104 VE growth curves displayed viral replication in pancreas with both Trinidad strain and the less virulent, vaccine strain (TC-83). TC-83—infected hamster pancreas showed only focal areas of acinar necrosis, and neither viral particles nor viral antigens were present in beta cells. The rapid lethality of the Trinidad-infected group necessitated performing metabolic studies in TC-83-strain—infected hamsters. Glucose tolerance tests using 2 gm. or 5 gm. glucose per kilogram intraperitoneally were performed on TC-83-strain—infected hamsters two days, 24 days, and 90 days postvirus inoculation. Glucose intolerance occurred in hamsters in each of the infected groups given 5 gm. glucose per kilogram, except for the 90-day convalescent group. Severely diminished plasma insulin responses occurred in the 24-day and 90-day convalescent TC-83 hamsters after both 2 gm. and 5 gm. glucose per kilogram. Pancreatic insulin content was within the normal range in 24-day convalescent TC-83 hamsters, which suggested a defect in insulin release from the beta cells during this stage of convalescence. Because of the advantage of studying longitudinal changes that occur after infection with the Trinidad strain of VE virus, the investigations were extended to a primate species, in collaboration with Dr. G. S. Bowen, C.D.C., Fort Collins, Colorado. 105 Five young adult healthy male rhesus monkeys were given intravenous glucose tolerance tests (GTT) during a control, preinoculation period and for eight days, 2, 5, and 10 months after inoculation with Trinidad strain VE-virus. During each GTT, sequential samples were obtained for serum insulin, plasma glucose, cortisol, and glucagon. After the 10-month GTT, monkeys were killed to assess pancreatic pathology and extractable hormone content. After VE inoculation, there was a progressive impairment in glucose tolerance until five months, with 10-month values remaining in the same range but fasting hyperglycemia did not occur. Similar to the hamster data, there was a marked progressive diminution in both basal and glucose-stimulated serum insulin responses after VE inoculation, with 5-month and 10month values being comparable to each other. Mean basal plasma glucagon levels at 2, 5, and 10 months were significantly lower than corresponding values during the preinoculation study. Plasma cortisol concentrations did not change significantly during all the studies, making it unlikely that the observed hypoinsulinemia was induced by nonspecific stress. Light and electron microscopic evaluation of the pancreases DIABETES, VOL. 2 7 , NO. 11

ELLIOT J. RAYFIELD, M.D., AND YOSHIKO SETO, PH.D.

tions of alloxan, another beta cell toxin, did not produce insulitis. 66 Also, the administration of 3-0methyl-D-glucose, nicotinamide, or antilymphocyte serum before each of the multiple streptozotocin doses attenuated, but did not prevent, the insulitis and elevations of blood glucose noted with streptozotocin alone. 66 It is proposed that this model, in a susceptible host, is the consequence of three distinct factors—a direct beta cell cytotoxicity, viral induction within beta cells, and a cell-mediated autoimmune reaction—that interact with each other to varying degrees to produce the final metabolic and histologic picture. 66 Spontaneous diabetic syndrome in Wistar rats. A syndrome of spontaneous diabetes has recently been reported in both male (n = 3) and female (n = 15) nonobese Wistar rats. This syndrome is characterized by glycosuria, hyperglycemia, hyperketonemia, ketonuria, hypoinsulinemia, and hyperglucagonemia. 108 Pathologic examination of the pancreas in certain moderately ketotic and stable rats reveals beta cell destruction and infiltration of islets with lymphocytes and monocytes. In severely ketotic animals and in moderately ketotic rats that survived longer, an end-stage picture was apparent, with smaller islets in fewer numbers than normal consisting almost exclusively of nonbeta cells by aldehyde-fuchsin staining. Exogenous insulin treatment controls partly the metabolic abnormalities that return when insulin is discontinued. Genetic, possible viral, or immunologic factors have not as yet been established. Autoimmunity and Diabetes Several mechanisms exist by which a virus infection may trigger the subsequent development of immunologic events in the pancreas and other organs, progressing long after the acute infection. 108 These mechanisms include: (1) the formation of new antigenic sites in cellular proteins; (2) the production of cell-specific antibodies directed against pancreatic islets; (3) a nonspecific action of virus infection that 108a Thus, Streptozotocin model. Rossini and Like found that may lead to loss of immunologic tolerance. for example, it is conceivable that such loss of immultiple subdiabetogenic injections of streptozotocin munologic tolerance may place the pancreatic islets at in mice produced a pancreatic insulitis with progresrisk to additional pancreatotropic viruses or to an onsive destruction of beta cells and sustained going, smoldering, autoimmune process long after hyperglycemia.65 Streptozotocin in this model may the initial infection subsides, leading inexorably to have acted by initiating a cell-mediated immune reclinical diabetes mellitus. The following discussion sponse. It is most intriguing that electron microscopy brings together several pieces of information concerndemonstrated type-C viruses within the beta cells, ing immunologic events and disease states as either suggesting that streptozotocin may have activated causative factors or associations with insulinmurine leukemia virus in vivo in susceptible hosts. dependent diabetes mellitus. Further studies indicated that multiple, small injec-

upon sacrifice were normal. In a second group of rhesus monkeys studied, with six controls, six Trinidad VE-strain—infected, and six TC-83 VE— infected monkeys, glucose intolerance did not develop in either the Trinidad or the TC-83 group. 106 Thus it becomes clear from the previous animal models as well as the VE studies that multiple factors such as the dose and strain of virus, animal age and weight, and genetic background among other variables are operative in determining the extent of structural and functional beta cell damage in response to a viral infection. Table 2 summarizes many of the factors that influence virus-induced diabetes mellitus in animal models. Clearly more studies are necessary to assess the mechanisms by which VE virus promotes a metabolic picture of chemical diabetes with hypoinsulinemia. Rubella. Menser et al. 40 reported studies recently with rubella virus—infected virgin, white rabbits (Castle Hill Laboratory strain, University of Sydney) and their offspring delivered both spontaneously (N = 7) and by cesarean section (N = 37). Blood glucose concentrations in the test animals (mean, 52 mg. per deciliter; range, 18-113 mg. per deciliter) were lower than those in controls (mean, 66 mg. per deciliter; range, 30-146 mg. per deciliter), although serum insulin concentrations in the test animals (mean, 43 (i\J. per milliliter) were also lower than those in the controls (mean, 68 /xU. per milliliter). Electron microscopic examination of the pancreas showed beta cell degranulation and changes in subcellular organelles, but no evidence of necrosis was cited. The offspring, delivered spontaneously, were killed at nine months of age and, despite cortisone-acetate injections every six weeks (X three), did not develop diabetes and apparently had less pathologic alterations than the fetuses delivered by cesarean section. These studies are too limited to judge in terms of whether the virgin, white rabbit is a suitable animal model to use for investigating the pathogenesis of diabetes mellitus associated with the congenital rubella syndrome.

NOVEMBER, 1978

1133

VIRUSES A N D THE PATHOGENESIS OF DIABETES MELLTTUS

The pathologic basis for autoimmunity in the etiology of JIDM has been discussed in the section on insulitis. A multitude of clinical studies have documented an increased association of adrenal, thyroid (including thyroglobulin), and gastric parietal cell antibodies with JIDM. 1 0 9 " 1 1 6 In addition, an association of insulin-dependent diabetes with thyroid (hyperthyroidism, 117 primary hypothyroidism, 118 Hashimoto's thyroiditis 111 ), idiopathic hypoparathyroidism, 1 1 9 idiopathic hypoadrenocorticism, 1 2 0 Schmidt's syndrome, 1 1 0 pernicious anemia, 1 1 3 polyendocrine disorders, 121 and myasthenia gravis 122 have all been reported. Acute inflammation of the islets of Langerhans of rats and mice challenged with guinea-pig-insulin antiserum was noted by Lacy and co-workers123 as well as Logothetopoulos and Bell. 124 Glucose intolerance and overt diabetes have been produced in rabbits immunized with heterologous insulin. 1 2 5 The question remains unanswered as to whether tissue-directed autoantibodies in endocrine diseases represent a cause or an effect of the basic underlying process. Nerup et al. utilizing the leucocyte migration test (LMT) as an in vitro assessment of cellular hypersensitivity to fractions of atrophic porcine pancreas, demonstrated significantly lower values in diabetic leucocytes than in leucocytes from control patients. 114 MacCuish and co-workers measured cellular hypersensitivity to an extract of human pancreas using the LMT test and found that only the young insulin-dependent diabetic patients exhibited a significantly different response from normal; older diabetics, on insulin or oral hypoglycemic treatment, had responses within the normal range. 126 Finally, Huang and Maclaren have shown that lymphocytes from young patients with insulin-dependent diabetes mellitus induce specific cytotoxic changes in an insulinoma-derived cell line. 127 With regard to pancreatic islet-cell antibodies (ICAb), Lendrum et al. reported that 48 per cent of children with diabetes mellitus of recent onset manifested such antibodies. 128 ICAb are a pancreasspecific, species-nonspecific, IgG class of antibodies, many of which fix their complement, that react with all known types of endocrine cells within the islets. 129 - 130 The ICAb are directed against cytoplasmic components of islet cells. 129 At the time of diagnosis of insulin-dependent diabetes, ICAb are detectable in as high as 85 per cent of patients. The percentage drops to 50 per cent by one month and remains at this level for the first year; thereafter, the prevalence falls to a level of 10 to 20 per cent, with no further de1134

cline. 129 Lendrum and co-workers reported further that ICAb is not associated with insulin-independent diabetes mellitus. 129 The data of Irvine's laboratory in Edinburgh are of interest in this regard. 131 This group found that frequencies of ICAb in insulin-dependent diabetics with (62 per cent) and without (56 per cent) other clinical organ-specific autoimmune disorders were similar within a year of diagnosis of diabetes. 131 But at five years after diagnosis, 26 per cent of insulin-dependent diabetics with other autoimmune disease continued to have measurable serum ICAb titers in contrast to 7 per cent with insulin-dependent diabetes alone. 131 Also, diabetics who responded initially to oral hypoglycemic agents but who exhibited the presence of ICAb were more likely to have serum thyrogastric antibodies, and, subsequently, they required longterm insulin treatment. 132 These workers also found a sustained positive ICAb in a given insulin-dependent diabetic to be associated with HLA-B8. 133 Maclaren and associates documented circulating antibodies to human insulinoma cells in 34 of 39 insulin-dependent diabetics using an indirect immunofluorescent technique. 134 While Bottazzo et al. found pancreatic ICAb in two patients a year before symptomatic diabetes occurred, 135 little is known concerning the temporal appearance of these antibodies in relation to the development of clinical diabetes mellitus or their pathogenetic significance. 136 It has been postulated that a viral infection may act as a triggering event to pancreatic islet-cell damage with a latent period of variable duration until clinical diabetes ensues and that the findings of ICAb in patients who subsequently develop diabetes may act as a marker during this latent period. 128 Additional prospective studies of the role of ICAb as a marker in insulin-dependent diabetes mellitus are needed, and the standardization of existing techniques for the measurement of ICAb is essential. Genetic Considerations in Human Diabetes

Genetically determined factors in EMC-induced murine diabetes have been cited above. One of the strongest pieces of evidence for the importance of environmental factors (such as viral infection) in the genesis of diabetes comes from studies of twins. In juvenile-onset, insulin-dependent diabetes (under age 40 years), only 50 per cent of identical twins are concordant (both twins affected) for diabetes in contrast to maturity-onset diabetes mellitus (over age 40 years), in which 90 per cent of the second twins are concordant. 137 Such discordance (only one twin afDIABETES, VOL. 2 7 , NO. 1 1

ELLIOT J. RAYFIELD, M . D . , A N D YOSHIKO SETO, PH.D.

fected) in 50 per cent of the juvenile-onset diabetic twins is compatible with a nongenetic cause such as viral infection(s) or an environmental diabetogenic toxin in the diabetic twin. Several (eight) pairs of twins under the age of 45 have remained discordant for over 20 years. 138 Recent studies documenting associations between certain HLA antigens and juvenile-onset insulindependent diabetes strengthen the genetic basis of this disease. Specifically, HLA antigens B 8 , 4 2 - 1 3 9 B W 1 5 , 4 2 ' 1 3 9 D W 3 , 1 4 0 ' 1 4 1 and D W 4 1 4 0 - 1 4 1 are found in significantly higher frequencies in juvenileonset insulin-dependent diabetes, while no such associations have been found for insulin-independent maturity-onset diabetes. 42 ' 139 " 141 HLA-B7 appears to have a significantly decreased association with insulin-dependent diabetes. 140 ' 142 It is fascinating that HLA-B8, but not BW15, is associated with the presence of severe diabetic micrangiopathy (retinopathy and nephropathy). 143 Data from Rubinstein et al. indicate that the diabetes-predisposing gene is recessive, with 50 per cent penetrance. 144 The penetrance was estimated by the fact that only half the siblings who shared both haplotypes with the firstaffected sibling develops diabetes. 144 The report of Barbosa et al., in which HLA typing was performed in 24 families with two or more juvenile-onset insulindependent diabetic siblings, also showed that 55 per cent of the diabetic pairs were concordant for both HLA haplotypes. 145 Preliminary studies from Barbosa and co-workers revealed that five of eight nondiabetic siblings HLA-identical with their diabetic siblings had latent diabetes as assessed by abnormal cortisone-primed glucose tolerance. 145 These data are all compatible with an additional environmental diabetogenic factor superimposed on the diabetespredisposing gene (even when present in double dose). Evidence suggesting that HLA-products might play a role in the immune response to infectious agents has previously been confined to animal models 98 and could not be documented with respect to murine EMC-induced diabetes. 99 However, a recent study by de Vries and associates has demonstrated that HLACw3 or an HLA-product associated with Cw3 affects the cellular immune response to vaccinia virus in man. 146 Such a mechanism, while yet to be proved, could be operative in conferring either protection or vulnerability to diabetogenic viruses in man. Recent Human Studies

A retrospective epidemiologic study concerning the incidence, sex, seasonal, and geographic patterns of NOVEMBER,

1978

JIDM was reported from Denmark 147 in 1977 and included one third of the country's population for the period of 1970 to 1974. The incidence of the disease remained at about 13.2 per 100,000 per year; 27 per cent more boys than girls were affected. Similar to the British data, 26 a seasonal variation in onset of diabetes mellitus was noted with a reduction of incidence during May, June, and July. Of note, the incidence of JIDM was related to socioeconomic status, being greatest in areas with the lowest status. A prospective study of 110 people was conducted in England from 1974 to 1976 with respect to interrelationships among HLA type, viral antibody titers, ICAb, presence of antecedent viral illness, and seasonal variation in the genesis of JIDM. 1 4 8 The findings demonstrated a clustering of HLA BW-15— positive subjects in the winter peak of 1976 who had increased neutralizing antibody titer to Coxsackie viruses (Bl-4); such clustering did not occur in the autumn peak of 1975. The presence of ICAb did not correlate with HLA type, viral antibody titers, or history of antecedent infection. Similar studies in other countries but of longer duration than the present study are required to further delineate the interrelationships among specific viral infections, HLA-linked factors, and ICAb. Conclusions

The threads of the tapestry depicting a viral etiology of JIDM are coming closer together, but conclusive proof of such in man is lacking. Animal models, most notably EMC-induced murine diabetes mellitus, have demonstrated convincingly a juvenile-type diabetic state in the genetically susceptible host. There is evidence that one site of genetic control in mice relates to differences in viral receptors on the surface of the beta cells that may confer susceptibility or resistance to EMC-virus— induced diabetes. The availability of beta cell tissue culture techniques, assuming that such cultivation does not render beta cells nonspecifically susceptible to any virus, has potential in screening for possible diabetogenic viruses in man. Animal models of virus—induced diabetes in primates could offer the possibility to study carefully the evolution of diabetic complications. Certain animal models and human studies are consistent with an autoimmune component playing a part in the genesis of JIDM. A viral infection and an autoimmune process are not incompatible with each other: A virus can trigger an autoimmune event and 1135

VIRUSES A N D THE PATHOGENESIS OF DIABETES MELLITUS

an autoimmune phenomenon can heighten susceptibility to a diabetogenic virus. The pathologic entity of insulitis in JIDM is compatible with a viral and/or autoimmune process being operative etiologically. Genetic studies of twins concordant and discordant for JIDM suggest an environmental component, such as a viral infection, in the etiology of this type of diabetes. HLA associations have been established with JIDM, but the nature of the gene product(s) of specific HLA haplotypes is not known. Pancreatic islet cell antibodies (ICAb) are associated with JIDM. Whether ICAb are formed in response to a viral or autoimmune process, or, whether they actually play a role in the etiology of JIDM is unclear. Epidemiologic investigations reveal a relationship of JIDM with seasonal patterns and, in certain studies, an association with increased titers of neutralizing antibodies to Coxsackie-B viruses. While a seasonal variation in JIDM can be explained by a concomitant seasonal variation in viral infections, other interpretations (such as dietary and exercise alterations) are also possible. Although these findings, taken together, suggest a role for certain viruses and other environmental factors (drugs, toxins, foods) in the pathogenesis of JIDM, further studies will be necessary to establish unequivocally such relationships and to define the specific mechanisms that are operative. ADDENDUM Recently, Yoon et al. demonstrated that Coxsackie-B3 virus (Nancy strain), passaged 10 times in secondary mouse embryo cells, was capable of infecting human beta cells in an in vitro culture system. 1 Although less than 20 per cent of the cultured cells were beta cells, proof that such beta cells were infected was derived by the double-labeled antibody technique. Also a decreased content of radioimmunoassayable insulin was noted in these CoxsackieB3 virus—infected human pancreatic cultures. 1 Studies performed in the same laboratory have shown that reovirus type 3, passaged at least seven times in pancreatic beta cell cultures from SJL/J mice, produced pancreatic islet necrosis and mononuclear leukocyte infiltration after inoculation into one- to two-weekold mice. 2 A double-labeled antibody technique demonstrated the presence of reovirus antigens in beta cells from frozen sections of mouse pancreas. These reovirus-infected mice did not exhibit overt hyperglycemia up to 20 days after infection but did show abnormal glucose tolerance. 1136

References 1

Yoon, J-W., Onodera, T., Jenson, A. B., and Notkins, A. L.: Virus-induced diabetes mellitus. XI. Replication of Coxsackie B3 virus in human pancreatic beta cell cultures. Diabetes 27:778-81, 1978. 2 Onodera, T., Jenson, A. B., Yoon, J-W., and Notkins, A. L.: Virus-induced diabetes mellitus: Reovirus infection of pancreatic B cells in mice. Science 201:529-31, 1978. ACKNOWLEDGMENTS The work was supported in part by grants from NIH (AM 18522), New York Diabetes Association, Irving Graef Medical Research Foundation, Helena Rubinstein Foundation, Herman Goldman Foundation, Mr. Harold Weiss, and an NIH Research Career Development Award KO4-AM 00089 to Dr. Rayfield. Thanks are due to Doctor Henry Dolger for his continued support and encouragement, to D. Gary Langer for assisting with the bibliography, and to Dr. J-W. Yoon for providing the authors with manuscripts that hadn't been published yet. REFERENCES (The prefix "A" indicates an abstract following this section.) 'Craighead, J. E.: The role of viruses in the pathogenesis of pancreatic disease and diabetes mellitus. Prog. Med. Virol. 29:161-214, 1975. 2 Notkins, A. L.: Virus-induced diabetes mellitus: Brief review. Arch. Virol. .54:1-17, 1977. 3 Steinke, J., and Taylor, K. W.: Viruses and the etiology of diabetes. Diabetes 23:631-33, 1974. 4 Maugh, T. H.: Diabetes: Epidemiology suggests a viral connection. Science 788:347-51, 1975. 5 Harris, H. F.: A case of diabetes mellitus quickly following mumps. Boston Med. Surg. J. CXL:465-69, 1899. 6 Patrick, A.: Acute diabetes following mumps. Br. Med. J. «:802, 1924. 7 Gundersen, E.: Is diabetes of infectious origin? J. Infect. Dis. 42:197-202, 1927. 8 Kremer, H. V.: Juvenile diabetes as a sequel to mumps. Am. J. Med. 3:257-58, 1947. 9 King, R. C : Mumps followed by diabetes. Lancet 2:1055, 1962. 10 Hinden, E.: Mumps followed by diabetes. Lancet 7:1381, 1962. n McCrae, Wm.: Diabetes mellitus following mumps. Lancet /: 1300-01, 1963. 12 Messaritakis, J., Karabula, C , Kattamis, C , and Matsaniotis, N.: Diabetes following mumps in sibs. Arch. Dis. Child. 46:561-62, 1971. 13 Block, M. B., Berk, J. E., Fridhandler, L. S., Steiner, D. F., and Rubenstein, A. H.: Diabetic ketoacidosis associated with mumps virus infection: Occurrence in a patient with macroamylasemia. Ann. Intern. Med. 78:663-67, 197314 Dacou-Voutetakis, C , Constantinidis, M., Moschos, A., DIABETES, VOL. 2 7 , NO.

11

ELLIOT J. RAYFIELD, M . D . , A N D YOSHIKO SETO, PH.D.

Vlachou, C , and Matsaniotis, N.: Diabetes mellitus following mumps: Insulin reserve. Am. J. Dis. Child. 727:890-91, 1974. 15 Farnam, L. W.: Pancreatitis following mumps: Report of a case with operation. Am. J. Med. Sci. 763:859-70, 1922. 16 Brahdy, M. B., and Scheffer, I. H.: Pancreatitis complicating mumps. Am. J. Med. Sci. 181:255-6O, 1931. 17 Zelman, S.: Blood diastase values in mumps and mumps pancreatitis. Am. J. Med. Sci. 207:461-64, 1944. 18 Bostrom, K.: Patho-anatomical findings in a case of mumps, with pancreatitis, myocarditis, orchitis, epididymitis and seminal vesiculitis. Virchows Arch. 344:111-17, 1968. 19 Witte, C. L., and Schanzer, B.: Pancreatitis due to mumps. JAMA 203:1068-69, 1968. 20 Rayfield, E. J., Curnow, R. T., George, D. T , and Beisel, W. R.: Impaired carbohydrate metabolism during a mild viral illness. N. Engl. J. Med. 289:618-21, 1973. 21 Aiyathurai, J. E. J.: Pathogenesis of viral ketoacidosis in children. Proc., 8th Singapore-Malaysia Congr. Med. 8:26-29, 1973. 22 Kirstein, M. B., and Bromberg, L.: The effect of fever therapy upon carbohydrate metabolism. J. Lab. Clin. Med. 25:7-10, 193323 Sultz, H. A., Hart, B. A., Zielezny, M., and Schlesinger, E. R.: Is mumps virus an etiologic factor in juvenile diabetes mellitus? J. Pediatr. 86:654-56, 1975. A24Prince, G. A., Jenson, A. B., Billups, L. C , and Notkins, A. L.: Infection of human pancreatic beta cell cultures with mumps virus. Nature (London) 277:158-61, 1978. "Gamble, D. R., Kinsley, M.L., FitzGerald, M. G., Bolton, R., and Taylor, K. W.: Viral antibodies in diabetes mellitus. Br. Med. J. 3:627-30, 1969. 26 Gamble, D. R., and Taylor, K. W.: Seasonal incidence of diabetes mellitus. Br. Med. J. 3:631-33, 1969. "Huff, J. C , Hierholzer, J. C , and Farris, W. A.: An "outbreak" of juvenile diabetes mellitus. Consideration of a viral etiology. Am. J. Epidemiol. 700:277-87, 1974. 28 Hierholzer, J. C , and Farris, W. A.: Follow-up of children infected in a Coxsackie-virus B-3 and B-4 outbreak: No evidence of diabetes. J. Infect. Dis. 729:741-46, 1974. 29 Dippe, S. E., Bennett, P. H., and Miller, M.: Coxsackie B virus and diabetes. Br. Med. J. 2:443-44, 1974. Letter. 30 Dippe, S. E., Bennett, P. H., Miller, M., Maynard, J. E., and Berquist, K. R.: Lack of causal association between Coxsackie B4 virus infection and diabetes. Lancet 7:1314-17, 1975. 31 Nelson, P. G., Pyke, D. A., and Gamble, D. R.: Viruses and the aetiology of diabetes: A study in identical twins. Br. Med. J. 4:249-51, 1975. 32 Hay, D. R.: Studies in preventive hygiene from the Otago Medical School. I. The relation of maternal rubella to congenital deafness and other abnormalities in New Zealand. N. Z. Med. J. 48:604-08, 194933 Menser, M. A., Dods, L., and Harley, J. D.: A twenty-five year follow-up of congenital rubella. Lancet 2:1347-50, 1967. 34 Forrest, J. M., Menser, M. A., and Harley, S. S.: Diabetes mellitus and congenital rubella. Pediatrics 44:445-46, 1969. 35 Plotkin, S. A., and Kaye, R.: Diabetes mellitus and congenital rubella. Pediatrics 46:650-51, 1970. 36 Johnson, G. M., and Tudor, R. B.: Diabetes mellitus and congenital rubella infection. Am. J. Dis. Child. 720:453-55, 1970. 37 Forrest, J. M., Menser, M. A., and Burgess, J. A.: High frequency of diabetes mellitus. in young adults with congenital rubella. Lancet 2:332-34, 1971. NOVEMBER,

1978

38 Lundstr6m, R., Ahnsjo, S., Berczy, J., Blomqvist, B., and Eklund, G.: Pediatria XIV (Proc., XIV Int. Congr. Pediatr., Vol. 4, p. 61.) Buenos Aires, 1974. 39 Cooper, L. Z.: Congenital rubella in the United States. In Infections of the Fetus and Newborn Infant. Krugman, S., and Gershon, A. A., Editors. Alan R. Liss, Inc., New York, 1975, p. 1. A40Menser, M. A., Forrest, J. M., and Bransby, R. D.: Rubella infection and diabetes mellitus. Lancet 7:57-60, 1978. 41 Menser, M. A., Forrest, J. M., and Honeyman, M. C : Diabetes, HL-A antigens and congenital rubella. Lancet 2:150809, 1974. 42 Cudworth, A. G., and Woodrow, J. C : HL-A system and diabetes mellitus. Diabetes 24:345-49, 1975. 43 Ziering, P.: Personal communication, 1978. 44 Bunnell, C. E., and Monif, G. R. G.: Interstitial pancreatitis in the congenital rubella syndrome. J. Pediatr. 80:465-66, 1972. 45 Rawls, W. E., and Melnick, J. L.: Rubella virus carrier cultures derived from congenitally infected infants. J. Exp. Med. 723:795-816, 1966. 46 Plotkin, S. A., and Vaheri, A.: Human fibroblasts infected with rubella virus produce a growth inhibitor. Science 756:65961, 1967. 47 Preece, M. A., Kearney, P. J., and Marshall, W. C : Growth hormone deficiency in congenital rubella. Lancet 2:842-44, 1977. 48 Hanid, T. K.: Hypothyroidism in congenital rubella. Lancet 2:854, 1976. 49 Yoon, J-W., Huang, S. W., Maclaren, N. K., Wheeler, C. J., Selvaggio, S. S., and Notkins, A. L.: Antibody to encephalomyocarditis virus in juvenile diabetes. N. Engl. J. Med. 297:1235-36, 1977. Letter. 50 Cappell, D. F., and McFarlane, M. N.: Inclusion bodies (protozoan-like cells) in the organs of infants. J. Pathol. Bacteriol. 59:385-98, 1947. 51 Wislocki, L. C : Acute pancreatitis in infectious mononucleosis. N. Engl. J. Med. 275:322-23, 1966. 52 Everett, E. D., Volpe, J. A., and Bergin, J. J.: Pancreatitis in infectious mononucleosis. South. Med. J. 62:359-60, 196953 Johnson, H. N.: Visceral lesions associated with varicella. Arch. Pathol. 30:292-307, 1940. 54 Oppenheimer, E. H.: Congenital chickenpox with disseminated visceral lesions. Johns Hopkins Hosp. Bull. 74:240-49, 1944. 55 Cheatham, W. J., Weller, T. H., Dolan, T. F., and Dower, J. C : Varicella: Report of two fatal cases with necropsy, virus isolation, and serologic studies. Am. J. Pathol. 32:1015-36, 1956. 56 Blattner, R. J.: Serious complications of varicella including fatalities. J. Pediatr. 50:515-17, 1957. 57 John, H. J.: The diabetic child. Etiologic factors. Ann. Intern. Med. 8:198-213, 1934. 58 Grishaw, W. H., West, H. F., and Smith, B.: Juvenile diabetes mellitus. Arch. Intern. Med. 64:787-99, 1939. 59 Warfield, L. M.: Acute pancreatitis followed by diabetes. JAMA 89:654-58, 1927. 60 Vizen, E. M.: On the atypical forms of tick-borne encephalitis, [in Russian] Zh. Nevropatol. Psikhiatr. im S. S. Korsakova 63:1462-66, 1963. 61 Schmidt, M. B.: Ueber die Beziehung der Langerhanssche Inseln des Pankreas zum Diabetes Mellitus. Muench. Med. Wochenschr. 49:51, 1902. 62 Gepts, W.: Pathologic anatomy of the pancreas in juvenile diabetes mellitus. Diabetes 74:619-33, 1965.

1137

VIRUSES A N D THE PATHOGENESIS OF DIABETES MELLITUS 63 Le Compte, P. M.: "Insulins" in early juvenile diabetes. Arch. Pathol. 66:450-57, 1958. 64 Le Compte, P. M., and Legg, M. A.: Insulitis (lymphocytic infiltration of pancreatic islets) in late-onset diabetes. Diabetes 27:762-69, 1972. 65 Like, A. A., and Rossini, A. A.: Streptozotocin-induced pancreatic insulitis: New model of diabetes mellitus. Science 793:415-17, 1976. A 66 Rossini, A. A., Like, A. A., Chick, W. L., Appel, M. C , and Cahill, G. F., Jr.: Studies of streptozotocin-induced insulitis and diabetes. Proc. Natl. Acad. Sci. USA 74:2485-89, 1977. 67 Barboni, E., and Manocchio, I.: Alterazionia pancreatiche in bovini con diabete mellito post-aftoso. Arch. Vet. Ital. 73:47789, 1962. 68 Pauluzzi, L.: Sindrome diabetica post-aftosa nel bovino e nel caprino. Clin. Vet. (Milano) 86:113-32, 196369 Pedini, B., Avellini, G., Morettini, B., and Comodo, N.: Diabete mellito post-aftoso nei bovini. Atti Soc. Ital. Sci. Vet. 76:443-50, 1962. 70 Platt, H.: The occurrence of pancreatic lesions in adult mice infected with the virus of FMD. Virology 9:484-86, 1959. 71 Munger, B. L., and Lang, C. M.: Spontaneous diabetes mellitus in guinea pigs: The acute cytopathology of the islets of Langerhans. Lab. Invest. 29:685-702, 197372 Lang, C. M., and Munger, B. L.: Diabetes mellitus in the guinea pig. Diabetes 2.5:434-43, 1976. 73 Pappenheimer, A. M., Kunz, L. J., and Richardson, S.: Passage of Coxsackie virus (Connecticut -S strain) in adult mice with production of pancreatic disease. J. Exp. Med. 94:45-64, 1951. 74 Melnick, J. L., and Godman, G. C : Pathogenesis of Coxsackie virus infection. Multiplication of virus and evolution of the muscle lesion in mice. J. Exp. Med. 93:247-66, 1951. 75 Godman, G. C , Bunting, H., and Melnick, J. L.: The histopathology of Coxsackie virus infection in mice. I. Morphologic observations with four different viral types. Am. J. Pathol. 28:223-57, 1952. 76 Pappenheimer, A. M.: Fuchsinophile granules in the tissues of mice infected with the Connecticut -S strain of Coxsackie virus. J. Exp. Med. 93:251-58, 1952. 77 Woodruff, J. F.: The influence of quantitated post-weanling undernutrition on Coxsackie B3 infection of adult mice. II. Alteration of host defense mechanisms. J. Infect. Dis. 727:164-81, 1970. 78 Burch, G. E., Tsui, C-Y., Harb, J. M., and Colcolough, H. L.: Pathologic findings in the pancreas of mice infected with Coxsackie B4. Arch. Intern. Med. 728:40-47, 1971. 79 Tsui, C-Y., Burch, G. E., and Harb, J. M.: Pancreatitis in mice infected with Coxsackie B l . Arch. Pathol. 93:379-89, 1972. 80 Harrison, A. K., Bauer, S. P., and Murphy, F. A.: Viral pancreatitis: Ultra-structural pathological effects of Coxsackie virus B3 infection in newborn mouse pancreas. Exp. Mol. Pathol. 77:206-19, 1972. 81 Coleman, T. J., Gamble, D. R., and Taylor, K. W.: Diabetes in mice after Coxsackie B4 virus infection. Br. Med. J. 3:25-27, 1973. 82 Coleman, T. J., Taylor, K. W . , and Gamble, D. R.: The development of diabetes following Coxsackie B virus infection in mice. Diabetologia 70:755-59, 1974. 83 Ross, M. E., and Notkins, A. L.: Coxsackie B viruses and diabetes mellitus. Br. Med. J. 2:226, 1974. 84 Ross, M. E., Hayashi, K., and Notkins, A. L.: Virus-

1138

induced pancreatic disease: Alterations in concentration of glucose and amylase in blood. J. Infect. Dis. 729:669-76, 1974. 85 From, G. L. A., Craighead, J. E., McLane, M. F., and Steinke, J.: Virus-induced diabetes in mice. Metabolism 77:1154-58, 1968. 86 Craighead, J. E., and McLane, M. F.: Diabetes mellitus: Induction in mice by encephalomyocarditis virus. Science 762:913-14, 1968. 87 Craighead, J. E., and Steinke, J.: Diabetes mellitus-like syndrome in mice infected with encephalomyocarditis virus. Am. J. Pathol. 63:119-30, 1971. 88 Boucher, D. W . , and Notkins, A. L.: Virus-induced diabetes mellitus: I. Hyperglycemia and hypoinsulinemia in mice infected with encephalomyocarditis virus. J. Exp. Med. 737:1226-39, 1973. 89 Craighead, J. E., Kanich, R. E., and Kessler, J. B.: Lesions of the islets of Langerhans in encephalomyocarditis virus-infected mice with diabetes-like disease. I. Acute lesions. Am. J. Pathol. 74:287-300, 1974. 90 Hayashi, K., Boucher, D. W . , and Notkins, A. L.: Virusinduced diabetes mellitus. II. Relationship between beta cell damage and hyperglycemia in mice infected with encephalomyocarditis virus. Am. J. Pathol. 7.5:91-102, 1974. 91 Petersen, K.-G., Heilmeyer, P., and Kerp, L.: Synthesis of proinsulin and large glucagon immunoreactivity in isolated Langerhans islets from EMC-virus infected mice. Diabetologia 77:21-25, 1975. 92 Friedman, S. F., Grota, L. J., and Glasgow, L. A.: Differential susceptibility of male and female mice to encephalomyocarditis virus: Effects of castration, adrenalectomy, and the administration of sex hormones. Infect. Immun. 3:637-44, 1972. 93 Kanich, R. E., Craighead, J. E., and Kessler, J. B.: Lesions of renal glomeruli in mice with virus-induced diabetes mellituslike disease. Diabetologia 9:203-09, 1973. 94 Craighead, J. E., and Higgins, D. A.: Genetic influences affecting the occurrence of a diabetes mellitus-like disease in mice infected with the encephalomyocarditis virus. J. Exp. Med. 739:414-26, 1974. 95 Neel, J. V., Fajans, S. S., Conn, J. W . , and Davidson, R. T.: Diabetes mellitus. In Genetics and the Epidemiology of Chronic Diseases. Washington, D.C., U. S. Govt. Printing Office, 1965, pp. 102-32. 96 Ross, M. E., Onodera, T., Brown, K. S., and Notkins, A. L.: Virus-induced diabetes mellitus. IV. Genetic and environmental factors influencing the development of diabetes after infection with the M variant of encephalomyocarditis virus. Diabetes 23:190-97, 1976. 97 Svejgaard, A., Platz, P., Ryder, L. P., Nielsen, L. S., and Thomsen, M.: HL-A and disease associations—A survey. Transplant. Rev. 22:1-43, 1975. 98 Benacerraf, B., and McDevitt, H. O.: Histocompatibilitylinked immune response genes. Science 773:273-79, 1972. "Boucher, D. W . , Hayashi, K., Rosenthal, J., and Notkins, A. L.: Virus-induced diabetes mellitus. III. Influence of the sex and strain of the host. J. Infect. Dis. 737:462-66, 1975. 100 Yoon, J - W . , and Notkins, A. L.: Virus-induced diabetes mellitus. VI. Genetically determined host differences in the replication of encephalomyocarditis virus in pancreatic beta cells. J. Exp. Med. 743:1170-85, 1976. A 10 ° a Yoon, J. W . , Lesniak, M. A., Fussganger, R., and Notkins, A. B.: Genetic differences in susceptibility of pancreatic B cells to virus-induced diabetes mellitus. Nature (London) 264:178-80, 1976. DIABETES, VOL. 2 7 , NO. 11

ELLIOT J. RAYFIELD, M . D . , A N D YOSHIKO SETO, PH.D. 101

A Yoon, J-W., Onodera, T., and Notkins, A. L.: Virusinduced diabetes mellitus. VIII. Passage of encephalomyocarditis virus and severity of diabetes in susceptible and resistant strains of mice. J. Gen. Virol. 37:225-32, 1977. A102Chairez, R., Yoon, J-W., and Notkins, A. L.: Virusinduced diabetes mellitus. X. Attachment of encephalomyocarditis virus and permissiveness of cultured pancreatic beta cells to infection. Virology. In press, 1978. 103 Jansen, F. K., Munterfering, H., and Schmidt, W. A. K.: Virus induced diabetes and the immune system. I. Suggestion that appearance of diabetes depends on immune reactions. Diabetologia 73:545-49, 1977. A104Rayfield, E. J., Gorelkin, L., Curnow, R. T., and Jahrling, P. B.: Virus-induced pancreatic disease by Venezuelan encephalitis virus: Alterations in glucose tolerance and insulin release. Diabetes 2.5:623-31, 1976. 105 Rayfield, E. J., and Bowen, G. S.: The evolution of Venezuelan encephalitis virus-induced carbohydrate abnormalities in rhesus monkeys. Clin. Res. 2.5:398A, 1977. Abstract. 106 Bowen, G. S., and Rayfield, E. J.: Unpublished observations. 107 Goldberg, S. L., Kochicheril, N. M., Schulman, R., Walker, G. F., and Rayfield, E. J.: Venezuelan encephalitis virus—induced defects in carbohydrate metabolism in genetically diabetic mice. Diabetes 27 (Suppl. 2):477, 1978. Abstract. 108 Nakhooda, A. F., Like, A. A., Chappel, C. I., Murray, F. T., and Marliss, E. B.: The spontaneously diabetic Wistar rat: Metabolic and morphologic studies. Diabetes 26:100-12, 1977. 108a Freytag, G.: Do viruses serve as mediators of immunological reactions? In Immunity and Autoimmunity in Diabetes Mellitus. Bastenie, P. A., and Gepts, W., Editors. Amsterdam, Excerpta Medica, 1974, pp. 247-51. 109 Landing, B. H., Petit, M. D., Weins, R. L , Knowles, M., and Guest, G. M.: Antithyroid antibodies and chronic thyroiditis in diabetes mellitus. J. Clin. Endocrinol. Metab. 23:119-20, 1963. 110 Solomon, N., Carpenter, C. J., Bennett, I. D., Jr., and Harvey, A. M.: Schmidt's syndrome (thyroid and adrenal insufficiency) and coexistent diabetes mellitus. Diabetes 24:300-04, 1965. ul Crome, L., Erdohazy, M., and Rivers, R. P. S.: Fulminating diabetes with lymphocytic thyroiditis. Arch. Dis. Child. 42:677-81, 1967. 112 Simkins, S.: Antithyroglobulin antibodies in diabetes mellitus. Diabetes 77:136-40, 1968. u3 Irvine, W. J., Clarke, S. F., Scarth, L., and Cullen, D. R.: Thyroid and gastric autoimmunity in patients with diabetes mellitus. Lancet 2:163-68, 1970. U4 Nerup, J., Ortved Anderson, O., Bendixen, G., Egeberg, J., and Poulsen, J. E.: Antipancreatic cellular hypersensitivity in diabetes mellitus. Diabetes 20:424-27, 1971. ll5 Nerup, J., and Binder, C : Thyroid, gastric, and adrenal autoimmunity in diabetes mellitus. Acta Endocrinol. (Copenhagen) 72:279-86, 1973. U6 Goldstein, D. E., Drash, A., Gibbs, J., and Blizzard, R. M.: Diabetes mellitus: The incidence of circulating antibodies against thyroid, gastric, and adrenal tissue. J. Pediatr. 304-06, 1970. 117 Perlman, L. V.: Familial incidence of diabetes in hyperthyroidism. Ann. Intern. Med. .5.5:796-99, 1961. U8 Ganz, K., and Kozak, G. P.: Diabetes mellitus and primary hypothyroidism. Arch. Intern. Med. 734:430-32, 1974. U9 Blizzard, R. M.: Idiopathic hypoparathyroidism: A probable NOVEMBER,

1978

autoimmune disease. In Textbook of Immunopathology, Vol. 2. Miescher, P. A., and Mueller-Eberhard, H. J., Editors. New York, Grune and Stratton, 1969, pp. 547-50. 120 Irvine, W. J., and Barnes, E. W.: Adrenal insufficiency. Clin. Endocrinol. Metab. 7:549-94, 1972. 121 Whittingham, S., Mathews, J. D., Mackay, I. R., Stocks, A. E., Ungar, B., and Martin, F. I. R.: Diabetes mellitus, autoimmunity and aging. Lancet 7:763-67, 1971. 122 Osserman, K. E.: Muscles (myasthenia gravis). In Textbook of Immunopathology, Vol. 2. Miescher, P. A., and MuellerEberhard, M., Editors. New York, Grune and Stratton, 1969, pp. 607-23123 Lacy, P. E., Wright, P. H., and Silverman, J. L.: Eosinophilic infiltration in the pancreas of rats infected with antiinsulin serum. Fed. Proc. 22:604, 1963. (Abstract.) 124 Logothetopoulos, J., and Bell, E. G.: Histologic and autoradiographic studies of the islets of mice infected with insulin antibody. Diabetes 7.5:205-11, 1968. 125 Grodsky, G. M., Feldman, R., Toreson, W. E., and Lee, J. C : Diabetes mellitus in rabbits immunized with insulin. Diabetes 7.5:579-85, 1966. 126 MacCuish, A. C , Jordan, J., Campbell, C. J., Duncan, L. J. P., and Irvine, W. J.: Cell-mediated immunity to human pancreas in diabetes mellitus. Diabetes 23:693-97, 1974. 127 Huang, S-W., and Maclaren, N. K.: Insulin-dependent diabetes: A disease of auto-aggression. Science 792:64-66, 1976. 128 Lendrum, R., Walker, G., and Gamble, D. R.: Islet cell antibodies in juvenile diabetes mellitus of recent onset. Lancet 7:880-83, 1975. A129Lendrum, R., Walker, G., Cudworth, A. G., Theophanides, C , Pyke, D. A., Bloom, A., and Gamble, D. R.: Islet-cell antibodies in diabetes mellitus. Lancet it: 1273-76, 1976. 130 Bottazzo, G. F., Doneach, D., and Pouplard, A.: Humoral autoimmunity in diabetes mellitus. Acta Endocrinol. (Copenhagen), Suppl. 20.5:55, 1976. 131 Irvine, W. J.: Classification of idiopathic diabetes. Lancet 7:638-43, 1977. 132 Irvine, W. J., McCallum, C. J., Gray, R. S., and Duncan, L. J. P.: Clinical and pathogenic significance of pancreatic-isletcell antibodies in diabetics treated with oral hypoglycemic agents. Lancet 7:1025-27, 1977. 133 Morris, P. J., Vaughan, H., Irvine, W. J., McCallum, F. J., Gray, R. S., Campbell, C. J., Duncan, L. J. P., and Farquhar, J. W.: HLA and pancreatic islet cell antibodies in diabetes. Lancet 2:652-53, 1976. 134 Maclaren, N. K., Huang, S-W., and Fogh, J.: Antibody to cultured human insulinoma cells in insulin-dependent diabetes. Lancet 7:997-99, 1975. 135 Bottazzo, G. F., Florin-Christensen, A., and Doniach, D.: Islet-cell antibodies in diabetes mellitus with autoimmune polyendocrine deficiencies. Lancet 2:1279-82, 1974. 136 Irvine, W. J., Gray, R. S., and McCallum, C. J.: Pancreatic islet-cell antibody as a marker for asymptomatic and latent diabetes and prediabetes. Lancet 2:1097-1102, 1976. 137 Tattersall, R. B., and Pyke, D. A.: Diabetes in identical twins. Lancet 2:1120-25, 1972. 138 Report of the Workshop on Etiology and Pathogenesis of Insulin-Dependent Diabetes Mellitus. Philadelphia, Pa., September 8-10, 1977. 139 Nerup, J., Platz, P., Andersen, O. O., Christy, M., Lyngs0e, J., Poulsen, J. E., Ryder, L. P., Nielsen, L. S., Thom1139

VIRUSES AND THE PATHOGENESIS OF DIABETES MELLITUS

sen, M., and Svejgaard, A.: HL-A antigens and diabetes mellitus. Lancet 2:864-66, 1974. 140 Creutzfeldt, W . , Kobberling, J., and Neel, J. V., Editors: The Genetics of Diabetes Mellitus. New York, Springer Verlag, 1976. 141 Thomsen, M., Platz, P., Anderson, O. O., Christy, M., Lyngs0e, J., Nerup, J., Rasmussen, K., Ryder, L. P., Nielsen, L. S., and Svejgaard, A.: MLC typing in juvenile diabetes mellitus and idiopathic Addison's disease. Transplant. Rev. 22:125-47, 1975. 142 Ludwig, H . , Schernthaner, G., and Mayr, W . R.: Is HLA-B7 a marker associated with a protective gene in juvenileonset diabetes mellitus? N . Engl. J. Med. 294:1066, 1976. 143 Barbosa, J., Noreen, H . , Emme, L., Goetz, F., Simmons, R., deLeiva, A., Najarian, J., and Yunis, E. J.: Histocompatibility (HLA) antigens and diabetic microangiopathy. Tissue Antigens 7:233-37, 1976. A 144 Rubinstein, P., Suciu-Foca, N . , and Nicholson, J. F.: Genetics of juvenile diabetes mellitus: A recessive gene closely linked to HLA D and with 50 per cent penetrance. N . Engl. J. Med. 297:1036-40, 1977. 145 Barbosa, J., King, R., Noreen, H . , and Yunis, E. J.: The histocompatibility system in juvenile, insulin-dependent diabetics multiplex kindreds. J. Clin. Invest. 60:989-98, 1977. 14(! deVries, R. R. P., Kreeftenberg, H. G., Loggen, H. G., and van Rood, J. J.: In vitro immune responsiveness to vaccinia virus and HLA. N . Engl. J. Med. 297:692-97, 1977. A 147 Christau, B., Kromann, H . , Andersen, O. O., Christy, M., Buschard, K., Arnung, K., Kristensen, I. H . , Peitersen, B., Steinrud, J., and Nerup, J.: Incidence, seasonal and geographical patterns of juvenile-onset insulin-dependent diabetes mellitus in Denmark. Diabetologia 23:281-84, 1977. A 148 Cudworth, A. G., Gamble, D. R., White, G. B. B., Lendrum, R., Woodrow, J. C , and Bloom, A.: Etiology of juvenile-onset diabetes: A prospective study. Lancet 7:385-88, 1977. 149 Kibrick, S., and Benirschke, K.: Severe generalized disease (encephalo-hepatomyocarditis) occurring in the newborn period and due to infection with Coxsackie virus, group B. Pediatrics 22:857-75, 1958. 150 Vizso, A. D . , and Sanders, F. K.: Alteration of the pathogenicity of some group B Coxsackie viruses under different conditions of passage. I. Virus type 4. Acta Virol. (Praho) 8:3845, 1964. 151 Munterfering, H.: Zur Pathologie des Diabetes Mellitus der weissen Maus bei der EMC-Virusinfektion: Histologishe, elektron-mikroskopische und quantitativ morphologische Befunde an den Langerhanssacken Inseln. Virchows Arch. A 3.56:207-34, 1972. 152 Wellmann, K. F., Amsterdam, D., Brancato, P., and Volk, P. W.: Fine structure of pancreatic islets of mice infected with the M variant of the encephalomyocarditis virus. Diabetologia 8:349-57, 1972. 153

Platt, H.: Observations on the pathology of experimental foot-and-mouth disease in the adult guinea pig. J. Pathol. Bacteriol. 76:119-31, 1958. 154

Platt, H.: A study of the pathological changes produced in young mice by the virus of foot-and-mouth disease. J. Pathol. Bacteriol. 72:299-312, 1956. 155 Robertson, J. S.: The pancreatic lesion in adult mice infected with a strain of pleurodynia virus. I. Electron microscopical observations. Aust. J. Exp. Biol. Med. Sci. 32:393-409, 1954.

1140

ABSTRACTS (All are verbatim summaries, except those from Nature) Prince, G. A.; Jenson, A. B.; Billups, L. C; and Notkins, A. L. (Lab. of Oral Med., Natl. Inst. of Dental Res., N I H , Bethesda, Md.): INFECTION OF HUMAN PANCREATIC BETA CELL CULTURES

WITH MUMPS VIRUS. Nature (London) 272:158-61, 1978. Human pancreatic cultures from seven autopsy subjects (both sexes) ranging in age from 3 days to 46 years were examined for their susceptibility to mumps virus. The proportion of beta cells in these cultures, determined by staining with rhodamine-labelled anti-insulin antibody, ranged from 1-5%. The cultures were infected with the ABC strain of mumps virus and at various times thereafter stained by the double-label antibody technique using rhodamine-labelled anti-insulin antibody and fluorescein-labelled anti-mumps antibody. 60-90% of the beta cells in these infected cultures contained mumps antigen in the cytoplasm. Similarly, 60-95% of noninsulin containing cells in the same cultures also contained mumps antigens. The effect of mumps virus replication on pancreatic cell survival was also examined by following these infected cultures daily for 6 days. Virus titers increased by 72 hours after infection, and then declined. At 6 days after infection, the ratio of the number of beta cells in infected cultures to the number of beta cells in uninfected cultures decreased from 1 to 0 . 1 . The highly lytic nature of the mumps infection points to beta cell death as the most likely explanation for this decrease rather than virus-induced degranulation. Thus, by use of the double-label antibody technique, it is shown that human beta cells in vitro are susceptible to infection by mumps virus.

Menser, M. A.; Forrest, J. M.; andBransby, R. D. (Children's Med. Res. Found., Royal Alexandra Hosp. for Children, Camperdown, New South Wales, Australia): RUBELLA INFECTION AND DIABETES MELLITUS. Lancet 2:57-60, 1978. The incidence of diabetes mellitus was increased in patients with congenital rubella. Experimental congenital rubella infection in rabbits caused histological changes in the /3-cells of the pancreatic islets similar to those found in mice made diabetic by the M variant of the encephalomyocarditis virus. It is concluded that the diabetes seen in congenital rubella is due to viral infection of the pancreatic islet cells.

Rossini, A. A.; Like, A. A.; Chick, W. L.; Appel, M. C; and Cahill, G. F.,Jr, (Joslin Res. Lab., Harvard Med. Sch., and Peter Bent Brigham Hosp., Boston, and Dept. of Pathol., Univ. of Mass. Med. Sch., Worcester, MA): STUDIES OF STREPTOZOTOCIN-INDUCED INSULITIS AND DIABETES. Proc. Natl. Acad. Sci. USA 74:2485-89, 1977. Multiple small injections of streptozotocin produce a delayed, progressive increase in plasma glucose in mice within 5-6 days after the injections, in association with pronounced insulitis and induction of type C viruses within beta cells. Multiple subdiabetogenic doses of streptozotocin in rats and multiple injections of another beta cell toxin, alloxan, in mice did not induce insulitis although hypoglycemia followed the injection of larger quantities of both agents. In mice, the prior injection of 30-methyl-D-glucose (3-OMG) or nicotinamide attenuated the diabetic syndrome produced by streptozotocin; however, 3-OMG was more protective. Rabbit antimouse lymphocyte serum, alone, DIABETES, VOL. 2 7 , NO. 11