key: cord-0005059-agiqcmru
authors: Löhr, M.; Bergstrome, B.; Maekawa, R.; Oldstone, M. B. A.; Klöppel, G.
title: Human cytomegalovirus in the pancreas of patients with type 2 diabetes: Is there a relation to clinical features, mRNA and protein expression of insulin, somatostatin, and MHC class II?
date: 1992
journal: Virchows Arch A Pathol Anat Histopathol
DOI: 10.1007/bf01606908
sha: 68e6271532854e6fc03e80a9f0e2442a36a6350e
doc_id: 5059
cord_uid: agiqcmru

Human cytomegalovirus (HCMV) was recently demonstrated in the pancreas of about half the patients with type 2 diabetes mellitus in the absence of mumps, rubella or Coxsackie B virus. The present study addresses the question as to whether type 2 diabetes with an HCMV-positive pancreas differs from those with HCMV-negative pancreases with respect to age, sex, treatment, duration of disease, volume densities of B-cells and D-cells, mRNA levels of insulin and somatostatin, islet amyloid peptide deposits and major histocompatibility complex (MHC) class I and class II gene transcription, and protein expression. HCMV-positive type 2 diabetic patients showed a tendency towards a shorter duration of disease and significantly increased levels of MHC class II on RNA. In addition, expression of MHC class II product (HLA-DR) was identified in duct epithelial cells and/or islet cells in 9 diabetic pancreases and in 2 non-diabetic glands. No MHC class I expression could be detected. No other clinical differences between HCMV-positive and HCMV-negative glands were found. All 10 HCMV-positive diabetics showed a strong expression of MHC class II mRN in the pancreas. By immunocytochemistry, 4 of 10 demonstrated expression on the islets; three of ten also expressed MHC DRβ on ductal cells. This finding might be related to the viral infection, as only 2 of the 9 HCMV-negative patients were HLA-DRβ positive and none of the non-diabetic controls showed increased levels of MHC class II mRNA. These data suggest that HCMV infection in the pancreas is associated with type 2 diabetes. However, no conclusions as to a role of this virus in the aetiopathology of type 2 diabetes can be drawn at present.

Diabetes mellitus is a syndrome with heterogeneous pathogenesis (K16ppel 1984; Lef6bvre 1988 ) and amongst the various aetiological factors that may be involved in the development of diabetes are viruses (Rayfield et al. 1976; Notkins et al. 1984; Leiter et al, 1989) . From experimental work and studies in children with fatal virus infections, it is known that the encephalomyocarditis virus, and reovirus or Coxsackie B4 viruses can cause necrosis of the endocrine cells of the pancreas (Craighead and Steinke 1971; Jenson et al. 1980; Ujevich and Jaffe 1980; Ahmad and Abraham 1982; Onodera et al. 1983; Yoon et al. 1986; Foulis et al. 1990; Szopa et al. 1990 ). Most pancreases from patients with diabetes, however, lack signs of lytic islet cell necrosis, but display changes that are either compatible with autoimmune destruction of the/3-cells, as in type 1 diabetes (K16ppel et al. 1991) , or B-cell dysfunction, as in type 2 diabetes (K16ppel et al. 1991 ). Yet viruses might still play a role in the pathogenesis of diabetes by either triggering auto-immune B-cell destruction via expression of major histocompatibility complex (MHC) proteins (Bottazzo et al. 1985; Jennings et al. 1985; Foulis and Farquharson 1986; Suzumura etal, 1986; Campbell etal. 1988) or, alternatively, by persisting in B-cells and impairing their specific function (Oldstone et al. 1984; Oldstone 1989) . The latter hypothesis has been investigated in depth in animal studies using lymphocytic choriomeningitis virus (LCMV) and Venezuelan equine encephalitis (VE) virus as models. LCMV infection has been shown to decrease insulin production (Oldstone etal. 1984; Rodriguez et al. 1985) without any evidence of cell injury or inflammation. This was also observed with VE virus (Rayfield et al. 1976 ) and Coxsackie B4 virus (Yoon (Preece et al. 1977) and diabetes (Ginsberg-Fellner et al. 1985) in the absence of any inflammatory or necrotic islet changes. Human cytomegalovirus (HCMV) has also been implicated in the aetiology and pathogenesis of diabetes. Diabetes developed in a child with congenital HCMV infection at the age of 13 months (Ward et al. 1979) . Nuclear inclusions in islet cells and insulitis were found in infants who died of disseminated HCMV infection (Hultquist et al. 1973; Jenson et al. 1980) . The HCMV genome was demonstrated in 15% of newly diagnosed type 1 diabetics who had islet cell antibodies as well (Pak et al. 1988 ). In a recent study we found HCMV nucleic acids in the islets of Langerhans in 44% of patients with type 2 diabetes mellitus, but not the nucleic acids of mumps, rubella or Coxsackie B (L6hr and Oldstone 1990) . The nucleic acid sequences were detected in fixed, paraffin-embedded pancreatic tissue using slot blot and in situ hybridization as well as polymerase chain reaction derived DNA sequencing. To analyse the significance of this finding for the development of type 2 diabetes further, two subgroups of diabetics were formed from the original study group in which detailed clinical data were available, one HCMV positive and one HCMV negative. These were compared with one another and the healthy controls with regard to several variables such as treatment and duration of diabetes, mRNA levels of insulin and somatostatin, volume density of B-cells and D-cells, transcription of class I and class II MHC genes, protein expression of HLA-DR, and islet amyloidosis.

Well-preserved tissue obtained at autopsy within 24 h of death from the pancreas of 19 patients with type 2 (non-insulin-dependent) diabetes mellitus and 18 normoglycaemic patients without evidence of pancreatic disease, was retrieved as blocks of formalin-fixed or Bouin-fixed, paraffin-embedded material from the files of the Departments of Pathology at the University of Hamburg, FRG; the University of Essen, FRG; the Free University of Brussels, Belgium; and the University of California at San Diego. The blocks of all 18 controls and the diabetics were subjected to RNA analysis (see below); in 15 of the 18 normoglycaemic patients detailed clinical data were available and blocks of those patients were therefore used for further studies. The clinical criteria for type 2 diabetes mellitus were abnormal glucose tolerance and elevated fasting blood glucose levels. No HLA typing was performed in these patients. Information on clinical data (age, sex, treatment with diet and/or oral hypoglycaemics or with insulin, and duration of diabetes) was extracted from the patients' records and summarized in Table 1 . Ten patients with a duration of the disease between 1 and 24 years received insulin (mean, 5 years). All other patients were treated with diet and/or oral hypoglycaemic drugs.

Preparation of RNA from paraffin-embedded tissue was carried out as described in previous studies (L6hr and Nerenberg 1990; L6hr and Oldstone 1990) . In brief, tissue was cut, digested with proteinase K in the presence of a chaotropic agent, sodium dodecyl sulphate (SDS), mixed with GTC, subjected to ultracentrifugation with caesium chloride, extracted with phenol/chloroform, precipitated and finally assessed for quality and quantity by mini-gel electrophoresis and spectrophotometrical readings.

For slot-blot analysis, 25 gg RNA aliquots were denatured at 65~ for 15 min in 6 x SSC (1 x SSC=0.15 M sodium chloride 0.015 M trisodium citrate), 7.4% formaldehyde, then serially diluted in 15 x SSC. Samples were applied to nitrocellulose membranes using a 72-slot minifold blot apparatus. Nitrocellulose membranes were baked 2 h at 80 ~ C, prehybridized 4 h at 37 ~ C in 50% deionized formamide, 5 x SSC, 2.5 x Denhardt's solution, 100 gg/ml boiled, sonicated salmon sperm DNA and then hybridized for 24 h at 37 ~ C with the respective labelled probes (see below). After hybridization, membranes were washed in 2xSSC, 0.1% SDS at 37~ for 30 rain, in 0.1 x SSC, 0.1% SDS at 55~ or 65~ for 30 min, then exposed against Kodak XAR-5 film at -70 ~ C with Cronex lighting plus intensifying screen. Following exposure to film, membranes were washed in 0.1 xSSC, 0.1% SDS at 85~ for 2 h to remove cDNA hybrids. After re-exposure to film to ensure efficient loss of hybridization signal, membranes were reused in hybridization experiments.

For densitometry, films were scanned with an LKB Ultrascan XL laser densitometer (LKB/Pharmacia, Uppsala, Sweden). In order to maintain an internal standard for these experiments and as an additional control to insure that mRNA levels were measured for equivalent amounts of total cellular RNA, insulin and somatostatin mRNA levels were related to the individual 28S ribosomal RNA levels (Bowles et al. 1986; Lipkin et al. 1988 ). For each sample, the ratio of insulin and somatostatin hybridization intensity to the ribosomal hybridization signal was calculated. These ratios were used to compare the different diabetic groups among each other and with the controls. Data were entered in an Apple Macintosh computer and analysed with StatView (Abacus Concepts, Berkeley, Calif., USA).

All probes were obtained as plasmids and transformed in competent bacteria (E. coli, DH5e) (Maniatis et al. 1982) . The HCMV DNA probes (pCM3, pJN201) were a gift from Dr. J. Nelson (Schrier et al. 1985) . cDNA probes for insulin (gift from Dr. P. Southern, Minneapolis, Minn., USA; Cordell et al. 1979) , somatostatin (gift from Dr. W. Rutter, San Francisco, Calif., USA; Lipkin et al. 1988 ), non-polymorphic region of human class I and class II MHC (pdpl and pDRfl, gift from Dr. S. Lawrence, La Jolla, Calif., USA (Curtsinger et al. 1987) ) were used as well as a probe to the 28S ribosomal RNA (S 138, gift from Dr. P. Southern; Lipkin et al. 1988 ). Probes were prepared by the random hexoprimer method with 32p (Feinberg and Vogelstein 1983) and specific activity of greater than 5 x 108 cpm/gg. For each hybridization, 1 2 x 10 6 cpm probe/ml hybridization mix were used.

Immunocytochemistry was performed (L6hr and Kl6ppel 1987) on 3 gm serial sections using a monoclonal antibody against insulin (1 : 5000; Biogenex via Camon, Wiesbaden, FRG) and polyclonal antibodies against glucagon, somatostatin (1 : 2000 and 1 : 3000; both INC, Stillwater, Minn., USA), pancreatic polypeptide (1:1000; gift from Dr. R.E. Chance, Indianapolis, Ind., USA) and islet amyloid polypeptide (1:2000; rat IAPPtl 2_ 37t, kindly donated by Dr. O. Madsen, Hagedorn Res. Laboratories, Gentofte, Denmark). Class II MHC expression was identified by a monoclonal antibody A II (gift from Dr. A. Foulis, Glasgow, UK). This antibody is known to react with paraffin-embedded tissue (Epenetos et al. 1985; L6hr and K16ppel 1987) and raised against the purified human class II antigens, DR and DQ (Neefjes et al. 1986 ). After deparaffinizing, the endogenous peroxidase was blocked with 3% hydrogen peroxide in methanol. After further rehydration and washes in phosphate-buffered saline (PBS), the slides were incubated with 10% normal porcine serum in a humid chamber at room temperature for 30 min. After shaking off the blocking serum, the monoclonal antibodies were applied diluted in PBS containing 0.1% bovine serum albumin. Incubation took place at room temperature for 18 h. After several washes in PBS, the secondary biotinylated antibody (rat anti-mouse IgG, Boehringer Mannheim) was applied at a 1:50 dilution for 30 min at room temperature. After several washes and a 30 min incubation with the avidin-biotin-complex (ABC, Vectastain) at room temperature, the slides were rewashed in PBS, reacted with 2,2 diaminobenzidine, rewashed, dehydrated and mounted for microscopic evaluation. Specificity of the antibodies was demonstrated by systematically including positive and negative controls. Human tonsillar tissue was used as positive control for the HLA-DR antibody. For identification of islet amyloid, sections adjacent to those immunostained for IAPP were stained with Congo red. The percentage of islets containing extracellular IAPP deposits was estimated and graded according to a semi-quantitative score ranging from grade 0 (no extracellular IAPP present in the islets), grade 2 (10 50% of all islets positive for extracellular IAPP) to grade 3 (more than 50% of the islets contain extracellular IAPP).

Determination of the volume density of the immunoreactive area of the insulin-containing B-cells and the somatostatin containing D-cells was performed as described (K16ppel et al. 1985; L6hr et al. 1989 ). Briefly, H & E-stained sections were evaluated at a final magnification of x 22 for the volume densities (volume fractions) of the mesenchymal tissue and parenchyma expressed relative to the whole pancreas according to established principles (Oberholzer 1983) . Volumes of mesenchymal tissue and parenchyma of the pancreas, including the endocrine tissue, were calculated as described (K16ppel et al. 1985; L6hr et al. 1989) . In a second step, volume densities of the immunochemically stained endocrine cells were measured at a total magnification of x 400 by a semiautomatic image analyser (L6hr et al. 1989) . In this analysis we report only data on B-cells and D-cells, because we could relate these data to the mRNA values of insulin and somatostatin. Statistical evaluation was performed with the Mann-Whitney test. Two-sidedp-values <0.05 were considered statistically significant.

As part of the former study group of 32 pancreases from patients with type 2 diabetes, 44% of which we reported to contain HCMV sequences (L6hr and Oldstone 1990), we selected the samples from 19 patients from whom we had further information on duration of disease and treatment. Of these, 10 were positive for HCMV by slotblot filter hybridization to pancreatic RNA, whereas none of 18 age-and sex-matched controls yielded a positive signal (Fig. 1) . There was no difference in age, sex, or treatment of diabetes (diet and/or oral hypoglycaemic versus insulin) between the patients containing HCMV nucleic acids in the pancreas and those who did not (Table 1) . However, patients expressing HCMV tended to have a shorter duration of diabetes when compared to the HCMV-negative type 2 diabetics (two-sided P< 0.06). The volume densities of B-cells and D-cells in HCMV-positive diabetic patients did not differ significantly from those of the HCMV-negative diabetics or from the non-diabetic patients (Table 1 ). In diabetics the mean volume density of B-cells tended to be lower than in non-diabetics; no such difference was noted between the mean volume densities of D-cells in diabetics and non-diabetics (Table 1) .

Insulin and somatostatin mRNA was measured by quantitation of autoradiographs. Although the densitometrically read value of the hybridization signal for insulin was slightly lower in the HCMV-positive type 2 diabetics (Table 1 : insulin mRNA/28S RNA), a statistically significant level was not reached, due to interindividual variation. Similar results were obtained on the protein level (Table 1) . No changes could be observed for somatostatin, on either the RNA or protein level. Extracellular IAPP was found in the islets of 15 of 19 (79%) type 2 diabetics and 0 of 15 non-diabetic controls (Fig. 2) . The number of HCMV-positive diabetics and HCMV-negative diabetics with extracellular deposition of IAPP in the islets did not differ significantly (Table 2) . Likewise, there was no obvious difference within the IAPP-negative group. The amount of extracellular islet deposits of IAPP varied from islet to islet and from case to case. There was no difference in the grades of IAPP positivity between the HCMV-positive Grading of percentage of islets containing extracellular IAPP : 1 < 10%; 2 10-50%; 3>50% of islets and HCMV-negative diabetics (Table 2 ). In islets without extracellular IAPP all or some of the B-cells also stained for IAPP. This intracellular IAPP immunoreactivity disappeared from the B-cells with the occurrence of significant extracellular IAPP deposits.

None of the diabetic nor any of the non-diabetic control pancreases showed a R N A signal for hybridization with the M H C class I probe (data not shown). Twelve of 19 diabetic patients expressed various levels of class II M H C R N A in the pancreas, and 10 of these were H C M V positive (Fig. 1, Table 3 ). None of the 15 ageand sex-matched control patients was positive for class II M H C RNA. The DR/~ probe did not cross-hybridize with H C M V (Fig. ] ).

Expression of class II M H C antigen ( H L A -D R ) was identified on endothelial cells as well as scattered mononuclear cells in all specimens from the diabetic patients as well as the non-diabetic control patients (Fig. 3a) . In 7 of the diabetic and in 2 of the non-diabetic patients there was positive staining on duct cells lining small ducts, and/or islet ceils (Table 3) . When adjacent serial sections were stained with insulin antiserum, the results suggested that the endocrine cells expressing class II M H C belonged to the group of insulin-producing cells (Fig. 3b, c) . There was no clear relationship between Fig. 2 . Immunoreactivity to islet amyloid polypeptide (IAPP) in islets of a type 2 diabetic. Note intensive extracellular staining in islet with amyloid (long arrows) and labelling of islet cells in an islet containing no extracellular IAPP (short arrows), x 125 the class II MHC RNA expression level and the degree and distribution pattern of class II MHC antigens in the diabetic and non-diabetic pancreases (Table 3) . Expression of class II MHC antigens on ducts and/or islets was found in 5 of 10 HCMV-positive diabetic pancreases and 2 of 9 HCMV-negative diabetic pancreases. It is noteworthy that one of the pancreases with the highest expression level of class II MHC RNA, which exhibited no class II MHC positivity on ducts or islets but was HCMV positive, revealed a discrete lymphocytic infiltration throughout the pancreatic tissue.

Recently, we reported that HCMV nucleic acids are present in the endocrine pancreas in 44% of patients with type 2 diabetes mellitus (L6hr and Oldstone 1990 ). In the present study we examined whether any of seven variables, age, sex, treatment of diabetes, duration of disease, volume density of B-cells and D-cells, expression of insulin and somatostatin RNA levels, HLA class II (DR/~) mRNA or islet amyloid, correlated with the HCMV positivity of the diabetic pancreases. The only findings which showed a relation with the HCMV status of the patients were a tendency towards shorter duration of diabetes and a more frequent expression of HLA-DR mRNA in the HCMV-positive patients.

From experimental studies in mice and hamsters infected with the LCM or VE virus it is known that these animals may develop a type 2-like diabetes without any obvious morphological damage to the islets. In these animals' diabetes, the possibility has been considered that the virus selectively suppresses the specific function of the pancreatic B-cells but leaves their cellular structure unaffected (Oldstone et al. 1984; Oldstone 1989) .

The only islet change associated with LCM infection was islet hypertrophy in animals with diabetes of short duration. This was interpreted as an attempt of the endocrine pancreas to increase insulin production during the early stage of disease. In our series of HCMV-positive type 2 diabetics with a mean duration of 11 years, we found neither B-cell hypertrophy nor any increase in volume density of B-cells. Instead, the B-cells showed a tendency towards a decrease in volume density as well as insulin RNA/ribosomal RNA ratio, which rather suggests a reduction in their number. However, even these B-cell changes appear to be unrelated to any specific damage caused or mediated by HCMV, since they were also found in HCMV-negative patients.

The tendency of the HCMV-positive patients to have a shorter mean duration of diabetes could imply that HCMV infection aggravates diabetes by further impairing B-cell function. However, as the length of the disease varied considerably from patient to patient, this finding lacked statistical significance and did not allow us to draw any firm conclusions with respect to an additive effect of HCMV in the development of type 2 diabetes.

In view of the results obtained with transgenic mice expressing a class II MHC molecule in the pancreas and developing diabetes mellitus in the absence of cellular injury or infiltration of the islets (Lo et al. 1988 ), we tested our samples with a probe to the non-polymorphous region of the human class II MHC DR/~ gene. Nine of 10 type 2 pancreases positive for HCMV expressed various levels of MHC class II RNA, but none of 18 controls and only 2 of 9 pancreases of HCMVnegative diabetics. At the protein level we found an expression of class II MHC (HLA-DR/~) on endothelial cells, some mononuclear cells and duct cells in the nondiabetic as well as in the diabetic pancreases. These findings confirm the results from from other studies (Daar pos. 

a Grading of percentage of endothelial cells/macrophages (E), ducts (D) and islets (I) expressing HLA-DR: 1 < 10%; 2 10-30%; 3 30 50%; 4>50% b RNA-DR//expression score: 1, weak; 2, moderate; 3, strong et al. 1984a, b; Foulis and Farquharson 1986; L6hr and K16ppel 1987; Dib et al. 1988 ). However, we also observed expression on islet cells and in particular B-cells in 1 control and 7 diabetic pancreases, findings which have so far been considered to be specific for type 1 diabetes of recent onset (Foulis and Farquharson 1986) . Why there was no clear relationship between class II MHC mRNA and its product is poorly understood. Technical problems or yet unknown variations in the interplay between transcription and translation may be important. However, the fact that diabetic HCMV-positive pancreases expressed class II MNC mRNA more frequently than HCMV-negative diabetic glands may be related to the fact that viruses are known to induce class I and class II expression (Jennings et al. 1985; Rosenthal et al. 1985; Suzumura et al. 1986; Gaulton et al. 1989) . Islet amyloidosis, the deposition of extracellular IAPP as amyloid within the islets is a morphological indicator lesion for type 2 diabetes in the pancreas (Westermark et al. 1987; Clark et al. 1990; K16ppel et al. 1991 ). Yet its significance for the pathogenesis of the disease is unknown. We found insular IAPP deposits in about 80% of the diabetic pancreases confirming the data of other studies (Westermark et al. 1987; Clark et al. 1990 ), but failed to demonstrate any relationship with the HCMV status in the patients.

In summary, this study revealed no association between HCMV infection and clinical or morphological variables characterizing patients with type-2 diabetes. The only finding that might be related to the presence of HCMV is positivity for MHC class II RNA. The possible significance of the HCMV infection for the function of the endocrine pancreas in man remains unclear. Preliminary results in transgenic mice using the major promoter of the HCMV immediate early gene hooked to a/?-galactosidase reporter gene demonstrated targeting of this gene to the islets of Langerhans (Jay A. Nelson, personal communication). Using this model, it should be possible to study the biology of HCMV infection in mammalian islets in more detail.

Pancreatic isleitis with coxsackie virus B5 infection

In situ characterization of autoimmune phenomena and expression of HLA molecules in the pancreas in diabetes mellitus

Detection of Coxsackie B virus specific RNA sequences in myocardial biopsy samples from patients with myocarditis and dilated cardiomyopathy

Reovirus infection enhances expression of class I MHC proteins on human/~-cell and rat RIN m5F cell

Islet amyloid polypeptide in diabetic and non-diabetic Pima indians

Isolation and characterization of a cloned rat insulin gene

Diabetes mellitus-like syndrome in mice infected with encephalomyocarditis virus

Evolutionary and genetic implications of sequence variation in two nonallelic HLA-DR fl-chain cDNA sequences

The detailed distribution of HLA-A, B, C antigens in normal human organs

The detailed distribution of MHC class II antigens in normal human organs

Selective localization of factor VIII antigenicity to islet endothelial cells and expression of class II antigens by normal human pancreatic ductal epithelium

A monoclonal antibody that detects HLA-D region antigen in routinely fixed, wax embedded sections of normal and neoplastic lymphoid tissues

A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity

Aberrant expression of HLA-DR antigens by insulin-containing fl-cells in recent-onset type 1 diabetes mellitus

A search for the presence of the enteroviral capsid protein VP1 in pancreases of patients with type 1 (insulin-dependent) diabetes and pancreases and hearts of infants who died of coxsackie viral myocarditis

Direct induction of Ia antigen on murine thyroid-derived epithelial cells by reovirus

Diabetes mellitus and autoimmunity in patients with the congenital rubella syndrome

Insulitis in cytomegalovirus infection in a newborn infant

Effect of herpes simplex virus types 1 and 2 on surface expression of class I major histocompatibility complex antigens on infected cells

Virus-induced diabetes mellitus XVII. Pancreatic islet cell damage in children with fatal viral infections

Islet histopathology in diabetes mellitus

Islet pathology and the pathogenesis of type 1 and type 2 diabetes revisited

The endocrine pancreas

Clinical forms of diabetes mellitus

Viral interactions with pancreatic B-cells

Viral infection of neurons can depress neurotransmitter mRNA levels without histologic injury

Diabetes and tolerance in transgenic mice expressing class II MHC molecules in pancreatic beta cells

Residual insulin positivity and pancreatic atrophy in relation to duration of chronic type 1 (insulindependent) diabetes mellitus and microangiopathy. Diabetologia

Nucleic acid extraction and detection from paraffin-embedded tissues

Detection of cytomegalovirus nucleic acid sequences in pancreas in type 2 diabetes

Insulitis and diabetes are preceeded by reduced endocrine and exocrine pancreatic volumes in diabetes prone BB rats

Ploegh HL (1986) A biochemical characterization of feline MHC products: unusually high expression of class II antigens on peripheral blood lymphocytes

Virus-induced diabetes mellitus

Morphometrie in der klinischen Pathologie. Allgemeine Grundlagen

Virus persists in beta cells of islets of Langerhans and is associated with chemical manifestations of diabetes

Virus-induced diabetes : cytomegalovirus and multiple enviromental insults

Association of cytomegalovirus infection with autoimmune type 1 diabetes

Growth hormone deficiency in congenital rubella

Virusinduced pancreatic disease by Venezuelan encephalitis virus. Alterations in glucose tolerance and insulin resistance

Virus persist in beta cells of islets of Langerhans and infection is associated with chemical manifestations of diabetes

Increased MHC H-2K gene transcription in cultured mouse embryo cells after adenovirus infection

Detection of human cytomegalovirus in peripheral blood lymphocytes in a natural infection

Coronavirus infection induces H-2 antigen expression on oligodendrocytes and astrocytes

Coxsackie B4 viruses with the potential to damage beta cells of the islets are present in clinical isolates

Pancreatic islet cell damage. Its occurrence in neonatal coxsackievirus encephalomyocarditis

Congenital cytomegalovirus infection and diabetes

Islet amyloid polypeptide-like immunoreactivity in the islet B cells of type 2 (non-insulin-dependent) diabetic and non-diabetic individuals

Coxsackie virus B4 produces transient diabetes in nonhuman primates