key: cord-296075-8axbkyyz
authors: Castro, Raymond F.; Perlman, Stanley
title: Differential Antigen Recognition by T Cells from the Spleen and Central Nervous System of Coronavirus-Infected Mice
date: 1996-08-01
journal: Virology
DOI: 10.1006/viro.1996.0415
sha: 
doc_id: 296075
cord_uid: 8axbkyyz

Abstract CD8+cytotoxic T lymphocytes (CTLs) isolated from the central nervous system (CNS) of C57Bl/6 mice acutely infected with mouse hepatitis virus, strain JHM (MHV-JHM), and analyzed in a directex vivocytotoxicity assay recognize two epitopes (H-2Db- and H-2Kb-restricted encompassing amino acids 510–518 and 598–605, respectively) within the surface (S) glycoprotein. In contrast, CD8+T cells isolated from the spleens of mice inoculated intraperitoneally with MHV-JHM and restimulatedin vitroonly respond to the H-2Db-restricted epitope. In this report, the preferential recognition of the H-2Db-restricted epitope is confirmed using splenocytes stimulatedin vitrowith either MHV-JHM-infected MC57 cells or with a cell line expressing the S protein and analyzed in secondary CTL assays. To determine whether these results represent a difference in epitope recognition between the spleen and CNS, secondary CTL assays were performed using spleen cells coated with peptides encompassing the CTL epitopes as stimulators. Under these conditions, both epitopes sensitized cells for lysis by spleen-derived CTLs, suggesting that both epitopes were recognized by splenic CD8+T cells after infectionin vivo.Furthermore, limiting dilution analysis indicated that the precursor frequency of splenic CD8+T cells specific for both the H-2Kb- and H-2Db-restricted epitopes were not significantly different. Thus, the results suggest thatin vitrostimulation of splenocytes specific for the H-2Kb-restricted epitope is inefficient after endogenous processing but that this inefficiency can be corrected if peptide is provided exogenously at sufficiently high concentrations. As a consequence, the results also show that cells responsive to both of the previously identified CNS-derived CD8+T cell epitopes are present in the infected spleen at nearly the same frequency.

(26) or S-specific immunogenic peptides S510-518 or S598-605 (unpublished observations) in direct ex vivo chronic demyelinating encephalomyelitis in susceptible CTL assays using cells isolated from the spleens of mice mice and rats (3-8). The chronic disease results either with acute encephalitis. Bergmann et al., however, were if mice are infected with an attenuated variant of MHV able to isolate S protein-specific CD8 / T cell clones from or if they are protected from the acute encephalitis by a the spleens of B6 mice infected intraperitoneally with a passive infusion of anti-viral antibodies, CD4 / T cells or sublethal dose of virus. These clones only recognized CD8 / T cells (6, (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) . This chronic neurological disease one of the two epitopes which we identified in the direct has been studied as an animal model of human demyex vivo CTL assays. The same results were obtained elinating disease, including multiple sclerosis.

when spleen cells were analyzed in CTL assays after Both CD4 / and CD8 / T cells are required for virus secondary stimulation in vitro with MHV-infected spleen clearance (23-25). Target proteins for CD8 / T cell cytocells (2). toxicity have been identified in MHV-JHM-infected BALB/ In many cases, antigen-specific cytotoxicity by cells c and C57Bl/6 (B6) mice. In BALB/c mice, the majority of primed in vivo can only be demonstrated after secondary the CTL activity is directed against the nucleocapsid (N) in vitro stimulation with antigen. Conversely, there are protein, whereas in B6 mice, the S glycoprotein is the no reports identifying epitopes which are observed in predominant target (26, 27). Using cells isolated from the primary but not secondary cytotoxic T cell assays. The CNS of B6 mice with acute encephalitis and analyzed results described above, therefore, would be most condirectly ex vivo in cytotoxicity assays, we identified two sistent with different epitope recognition between efimmunogenic CD8 / T cell epitopes within the S protein fectors harvested from peripheral lymphoid tissue and (1). In contrast to these results, we could not identify a those harvested from the site of inflammation, the brain. significant amount of CTL activity against the S protein Alternatively, insufficient presentation of the H-2K b -restricted epitope in vitro and consequent inefficient expansion of lymphocytes primed in vivo to respond to this

To determine if inadequate presentation of the H-2K brestricted epitope could be overcome by using a different source of antigen for stimulation, two alternative approaches were tried. First, irradiated (9000 rad), MHVinfected MC57 cells (H-2 b ) were used as stimulators. MC57 cells in general are not readily infected by MHV-JHM (28), but a line which was partially susceptible to the virus (approximately 10-20% of the cells were positive for viral antigen in an immunofluorescence assay) was developed. When these cells were used to stimulate control (EL-4). The average spontaneous release for EL-4 targets corresponded to õ13% and for MC57 cells õ25%. Percentage of specific For this purpose, splenocytes were harvested 8 days release was calculated as described previously (26). after intraperitoneal inoculation and restimulated in vitro with EL-4-S35 cells. When analyzed in CTL assays, these cells recognized the H-2D b , but not the H-2K b -restricted stricted epitopes. Irradiated splenocytes were coated with both peptides at a concentration of 5 mM each, epitope (Fig. 1B) . The results from these experiments using two different types of cells expressing S protein washed, and added to bulk cultures of spleen cells from MHV-infected mice. As shown in Fig. 2 , splenocytes stim-as stimulators are the same as those obtained by Bergmann et al. using MHV-JHM-infected syngeneic spleen ulated in vitro with both peptides for 5 days were able to efficiently lyse targets coated with either peptide in cells. They showed that when S protein was processed endogenously, the H-2K b -restricted epitope was not rec-secondary cytotoxicity assays. These results suggested that both epitopes were rec-ognized by spleen-derived lymphocytes in secondary CTL assays.

ognized efficiently in the mouse and that stimulation in vitro was inefficient in the case of the H-2K b -restricted The results presented thus far are consistent with inefficient presentation of the H-2K b -restricted epitope either epitope. It was not possible from these secondary CTL assays using bulk cultures to determine if there were in vivo during the primary response or later in vitro. Stimulation with exogenous peptide should circumvent diffi-quantitative differences in the number of CD8 / T cells responding in vivo to the two epitopes. Such quantitative culties associated with presentation of a suboptimal concentration of the H-2K b -restricted epitope in vitro after differences could contribute to the results shown in Figs. 1 and 2. If in vivo priming after intraperitoneal inoculation endogenous processing. To determine if this in fact occurred, splenocytes from intraperitoneally infected mice with MHV-JHM occurred more efficiently with the H-2D brestricted epitope, cells responding to the two epitopes were stimulated in vitro with splenocytes coated with peptides corresponding to the H-2D b -and H-2K b -re-should be present in the spleen at different concentra-of cultures that contained stimulators alone without responders. These assays using peptide-coated splenocytes as stimulators revealed comparable CTL precursor/effector frequencies for the H-2K b -(average 1/1291, range 1/927-1/1542 spleen cells) and H-2D b -restricted (average 1/987, range 1/380-1/1323 spleen cells) epitopes in these mice. As expected, the CTL precursor/effector frequencies obtained from infected B6 mice were significantly higher than those obtained from naive B6 mice (average 1/171, 872 spleen cells and 1/660,803 spleen cells for the H-2K b -and H-2D b -restricted epitopes, respectively). Therefore, in vivo priming was equally effec- shown in Fig. 4 , similar amounts of the two H-2 molecules were present on EL-4-S35 cells, suggesting that preferential down-regulation of the H-2K b molecule was not the tions. On the other hand, if inefficient expansion occurring in vitro is the sole explanation for the different explanation for the differences in recognition between the H-2D b -and H-2K b -restricted epitopes. results presented above, precursors to the two epitopes should be present at similar concentrations in the spleen.

Three independent studies showed that spleen-derived CD8 / cytotoxic T cells from intraperitoneally in-To compare the CTL precursor frequency for the two epitopes, limiting dilution analyses (LDA) using spleen-fected B6 mice recognize the H-2D b -but not the H-2K brestricted epitope in secondary CTL assays. First, Berg-derived lymphocytes from B6 mice intraperitoneally infected with MHV-JHM 8 days previously and from naive B6 mice were performed (Fig. 3) . Both virus-specific memory and effector cells should be present in the spleen at this time (29), although only memory cells may be measured in this secondary CTL assay (30). In these assays, threefold serial dilutions of responder spleen cells from individual mice were plated in a volume of 200 ml in the wells of 96-well round bottom tissue culture plates (Costar, Cambridge, MA) in complete RPMI media supplemented with 5% rat concanavalin A supernatant containing 50 mM methyl-a-D-mannopyranoside. In addition to immune or naive responder splenocytes, individual wells received 5 1 10 5 irradiated (3000 rad) syngeneic spleen cells which had been coated with peptide (10 was greater than 3 standard deviations above the mean suggested that in vivo priming in the spleen was similarly efficient for both epitopes. These data are most consistent with inefficient in vitro stimulation of the H-2K b -restricted CD8 / T cells. Antigen is likely to be processed by different antigen presenting cells (APCs) in vivo and in vitro and at different sites in vivo. The microenvironment may allow adequate presentation of the H-2K b -restricted epitope in vivo but this may not occur in vitro. Although some H-2K b -restricted epitope is likely expressed on the cell surface of APCs in vitro, this amount may be inadequate to stimulate in vivo primed cells, in fected syngeneic spleen cells, a step which our study 24, 76-85 (1973) .

indicates inadequately stimulates the H-2K b -restricted 5. Nagashima, K., Wege, H., Meyermann, R., and ter Meulen, V., Acta precursors, leading to a skewed result. Third, the data H-2D b -restricted epitope in cytotoxicity assays. 9. Yamaguchi, K., Goto, N., Kyuwa, S., Hayami, M., and Toyoda, Y., J.

Neuroimmunol. 32, 1-9 (1991).

In contrast to these data, use of peptide-coated sple- frequencies were measured for the two epitopes and

Proc. Natl. Acad. Sci. USA