key: cord-0698998-pv4sjsbp authors: Chairez, Ruben; Yoon, Ji-Won; Notkins, Abner Louis title: Virus-induced diabetes mellitus. X. Attachment of encephalomyocarditis virus and permissiveness of cultured pancreatic β cells to infection date: 1978-04-30 journal: Virology DOI: 10.1016/0042-6822(78)90465-8 sha: 35986d93702911ced2b8f81686d04909ac40c5b5 doc_id: 698998 cord_uid: pv4sjsbp Abstract Monolayers of pancreatic β-cells from strains of mice susceptible (SJL/J) and resistant (C57BL/6J) to the development of virus-induced diabetes mellitus were inoculated with the M variant of encephalomyocarditis (EMC) virus. Immunofluorescence showed that viral antigens appeared in up to 10 times more β cells from susceptible SJL/J mice than from resistant C57BL/6J mice. Infectious center assays revealed that 10–30 times more SJL/J β cells contained infectious virus than C57BL/6J β cells. Viral attachment experiments showed no difference in the binding of EMC virus when embryonic fibroblasts, pancreatic fibroblasts, and kidney cells from SJL/J and C57BL/6J mice were compared. However, at least twice as much virus attached to the pancreatic β cells from susceptible than from resistant strains of mice. Our data suggest that genetically determined differences in viral receptors on the surface of β cells may be one of the factors controlling susceptibility to EMC-induced diabetes mellitus. The M variant of encephalomyocarditis (EM0 virus produces a diabetes-like syndrome in mice by infecting and destroying pancreatic /3 cells (1) (2) (3) (4) (5) . Only certain strains of mice develop this syndrome and susceptibility to EMC-induced diabetes is inherited as an autosomal recessive trait (6) . The genetic factors controlling susceptibility operate at the level of the 8 cells, and whether a particular strain of mouse develops diabetes appears to be related to differences in the permissiveness of the 8 cells to infection with EMC virus (7, 8) . Recently pancreatic p cells from strains of mice that develop EMC-induced diabetes (susceptible) and pancreatic 8 cells from strains of mice that do not develop EMC-induced diabetes (resistant) were grown in culture and infected with EMC virus (8) . Examination of these cultures revealed that /3 cells from susceptible mice produced up to 50 times more virus than did 6 cells from resistant mice. However, it had not been determined whether the I Author to whom requests for reprints should be addressed. higher viral yield in cultures from susceptible mice was due to more virus being produced per cell or more cells being infected. The present investigation was initiated to resolve this question and, in addition, to see whether there was any difference in the attachment of EMC virus to 8 cells from susceptible as compared to resistant strains of mice. The M variant of EMC virus was grown and assayed on CAF-1 mouse embryo fibroblasts (MEF) (8) . Monolayers of MEF, kidney cells, and pancreatic fibroblasts were prepared by routine methods. Monolayer cultures enriched for pancreatic 8 cells were prepared as described previously (8) . Based on staining with fluorescein-labeled anti-insulin antibody, between 50 and 85% of the cells in these cultures were 8 cells. 3H-labeled EMC virus was prepared by infecting MEF monolayers with 10 plaqueforming units (PFU) of virus/cell. After a 1-hr incubation, Eagle's minimal essential medium (MEM) with 1% calf serum and 10 $X/ml of [3Hluridine was added . When approximately 80% of the cells showed percentage of virus bound to cells at any cytopathology, the medium containing the given time was determined by calculating labeled virus was harvested, and the virus the difference between the amount of virus was purified essentially by the methods recovered in the medium and the amount used by Ziola and Scraba for Mengo virus of virus inoculated. Petri dishes incubated (9) . The virus banded at a bouyant density with MEM but not containing cells served of 1.33 g/cm3 in CsCl and contained 1.6 x as controls. lo5 cpm/ml with an infectious titer of 8.4 The permissiveness of p-cell cultures x 10' PFU/ml. from strains of mice that develop diabetes Infectious center assays were performed (SJL/J, NIH-Swiss) and from strains of by inoculating monolayers of /3 cells (35 mice that do not develop diabetes (C57BL/ mm petri dishes) with EMC virus at a 6J) to EMC infection is illustrated in Table multiplicity of infection (m.o.i.) of 100. At 1. Immunofluorescence showed that up to the end of 60 min, the unattached virus 10 times more SJL/J cells contained viral was removed by washing the cells three antigens than did C57BL/6J cells. At 24 hr times. The monolayers then were briefly after infection (m.o.i., lo), 5% of the trypsinized (5 min at 37"), and the cells C57BLKiJ cells were positive for viral anwere resuspended and incubated at 37" for tigens, while 46% of the NIH-Swiss cells 40 min in MEM containing antibody to and 54% of the SJL/J cells contained viral EMC virus (neutralization titer, ~300) to antigens. When a higher m.o.i. (100) was neutralize any infectious virus in the me-employed, immunofluorescence was seen dium or on the surface of the cells. The in 19% of the C57BL/6J cells and in 91% of cells were then washed three times, resus-the SJL/J cells. In contrast, no difference pended, and counted, and the appropriate was observed in the number of cells connumbers were added to confluent MEF taining viral antigens when MEF cultures monolayers. The monolayers then were were infected with EMC virus; approxioverlaid with methylcellulose and the mately 95% of the cells in C57BL/6J and number of cells that produced plaques (infectious centers) was determined 72 hr SJL/J monolayers contained viral antigens at 18 hr after infection. Viral antilater. The number of cells containing viral gens were not detected in uninfected P-cell monolayers. antigens was determined by immunofluo- The difference in susceptibility was even rescence. p cells grown on coverslips were stained with fluorescein isothiocyanate-more apparent by infectious center assay. The data in Fig. 1 show that at each of the (FITC) labeled anti-EMC antibody as de-cell concentrations tested, lo-30 times scribed previously (8) . more SJL/J than C57BL/6J cells contained Viral attachment assays were performed on monolayers (60~mm petri which EMC virus attached to SJL/J as At various times after viral inoculation, 0.85 ml of cold 0.1 M phosphate-buffered compared to C57BL/6J cells, 3H-labeled saline (PBS), pH 7.5, was added. The me-virus was added to monolayers and at different times thereafter the amount of dium then was removed and assayed for virus that attached was determined. The infectivity (8) and radioactivity (10) . The attachment of "H-labeled EMC virus to Fig. 2 . Viral attachment ranged from 9 to 22%, depending on the cell type. No significant difference in binding of 3H-labeled virus was observed when cells from susceptible and resistant strains of mice were compared. The attachment of virus as measured by radioactivity leveled off at 16 to 32 min after inoculation. In contrast, the attachment of virus as measured by infectivity continued for at least 64 min. Although the total amount of infectious virus which attached varied depending on the cell type, there was no significant difference in attachment when MEF, pancreatic fibroblasts, and kidney cells from susceptible and resistant strains were compared. However, in the case of pancreatic p cells, more infectious virus attached to cells from the susceptible than resistant strain. The data in Fig. 2h difference at the 95% confidence level (P < 0.05). In another experiment in which a different virus pool was used (data not shown), 2.5 times more virus attached to the &IL/J monolayers than to the C57BLW monolayers. The immunofluorescence and infectious center data on cultured /3 cells reported here is consistent with earlier in uiuo findings (8) which showed that SWR/J pcells were more susceptible to infection than C57BLKJ p-cells. Moreover, the present findings indicate that the difference in susceptibility of /3 cells observed in uiuo is retained under in vitro cultivation conditions. Blocks at any one of several sites in the viral cycle (e.g., attachment, uncoating, or replication) could account for the observed differences in susceptibility. In the case of the picornaviruses, it is known that only certain cell types have receptors for these viruses (U-131, and differences in the concentration of these receptors could influence the susceptibility of the various organs within the host. Our experiments with the M variant of EMC virus showed that more of this virus attached to /3 cells from mice that developed diabetes @IL/J) than to p cells from mice that did not develop diabetes (C57BLKJ). In contrast, no difference in attachment was observed when other cell types (e.g., kidney, embryonic fibroblast, pancreatic fibroblast) from these two strains of mice were tested. These findings are compatible with earlier observations which showed differences in viral replication in P-cell cultures but not kidney or embryo cultures, when SWFVJ and C57BL&I mice were compared (8) . In this connection, it is known that some viruses can grow in monolayer cultures derived from organs duplicate or triplicate determinations were made. Panels a, c, e, and g: Attachment of 3H-labeled virus to SJL/J cells which in the host are ordinarily resistant to these viruses. The development of susceptibility in culture presumably is due to induction of receptors (11, 12, 14, 15 ). It appears, however, that at least in the case of our p-cell system, differences among strains in susceptibility to the M variant of EMC virus persist in culture. A significant difference in the attachment of EMC virus was observed only when attachment was assayed by infectivity. The failure to demonstrate any difference when radioactivity was measured may in part be explained by the fact that the assay for 3H-labeled virus detects both defective particles and virus particles which have eluted from the cell and have lost their infectivity (10, (16) (17) (18) . Other investigators have reported results similar to our findings: Viral attachment (e.g., poliovirus type 1, rhinovirus-14, and human coronavirus) when measured by radioactivity may give lower values than when measured by infectivity (10,18; K. Holmes, personal communication). The actual attachment of the M variant of EMC virus to pancreatic /3 cells in fact may be greater than the apparent twofold difference observed here. Recent immunofluorescence studies on sections of pancreas from several strains of mice inoculated with EMC virus but resistant to the development of diabetes (i.e., C57BL/6J, CBA/J, AKR, and A/J) revealed a considerable difference in the number of p cells that became infected (19) . Although neither C57BL/6J nor CBA/J mice developed diabetes when inoculated with EMC virus, approximately 12% of the /3 cells from C57BL/6J mice contained viral antigens at 72 hr after inoculation, as compared to 3% of the p cells from CBA/J mice (19) . The fact that so few CBA/J p cells became infected suggested that it might be possible to detect even greater differences in viral attachment if the /3 cells from this strain were compared with /3 cells from the susceptible SJL/J strain. Recent experiments using these two strains showed that three to four times more virus attached to SJL/J cells than to CBA/J /3 cells (unpublished data). Another factor that might contribute to differences in viral attachment is the source of the virus. Recently, it was found that EMC virus which had been passaged in tissue culture several times was less diabetogenic than EMC virus which had been passaged in mice (19) . Whether this was due to a reduced tropism for pancreatic /3 cells is not known, but the possibility that virus passaged in mice might show better attachment than the tissue culture-passaged t3H-labeled) virus used in the present experiments merits investigation. Moreover, it is known that different strains of virus show very different rates of attachment. For Still another factor that could influence viral attachment is the actual percentage of p cells in the enriched cultures. Our cultures usually contained about 65% /3 cells, but it is still not technically possible to obtain large quantities of /3 cells which are absolutely free of other pancreatic cells (e.g., fibroblasts, ductal cells, or acinar cells). Moreover, certain cell types (e.g., Fig. 2 , pancreatic fibroblasts) may bind more virus than p cells. Thus, the greater binding of EMC virus to non-p cells that contaminate the enriched cultures may have obscured an even larger difference in the specific binding of EMC virus to p cells than the apparent twofold difference observed in our studies. In conclusion, differences favoring the attachment of EMC virus to /3 cells from strains of mice that develop diabetes now have been observed in seven out of seven experiments performed. Whether restriction at levels other than attachment (e.g., transcription and/or translation) also contributes to the degree of permissiveness of p cells from different strains of mice to EMC replication is not known. However, based on present information, it appears that genetically determined differences in viral receptors may at the least be one of the factors controlling susceptibility to EMC-induced diabetes mellitus. Progress in Medical Virology Proc. Sot The authors thank Dr. Kenneth S. Brown for the statistical analysis.