Role of De Novo Protein Synthesis in Target Cells Recognized by Cytotoxic T Lymphocytes Specific for Vesicular Stomatitis Virus

Transcription

1 JOURNAL OF VIROLOGY, Dec. 1991, p X/91/ $2./ Copyright 1991, American Society for Microbiology Vol. 65, No. 12 Role of De Novo Protein Synthesis in Target Cells Recognized by Cytotoxic T Lymphocytes Specific for Vesicular Stomatitis Virus DONNA M. ROSCOE,* KADZUE ISHIKAWA, AND DOUGLAS S. LYLES Department of Microbiology and Immunology, Bowman Gray School of Medicine of Wake Forest University, Winston-Salem, North Carolina 2713 Received 12 April 1991/Accepted 13 September 1991 The requirements for viral and host protein synthesis in the generation of target antigens for cytotoxic T lymphocytes (CTL) was evaluated by using vesicular stomatitis virus (VSV) inactivated by UV irradiation (UV-VSV). EL4 target cells incubated with UV-VSV were recognized and lysed by anti-vsv CTL, indicating that de novo synthesis of viral proteins was not required for the generation of antigens recognized by antiviral CTL. Anti-VSV CTL from H-2b mice primarily recognize determinants derived from the VSV N protein bound to the class I major histocompatibility complex (MHC) antigen H-2Kb. Comparison of a cloned CTL line representing this specificity and a heterogeneous population of anti-vsv CTL showed that determinants other than that recognized by the cloned CTL were generated more efficiently from UV-VSV. By using vaccinia virus recombinants that express deletion fragments of the N protein, it was shown that these additional determinants were probably derived from VSV proteins other than the N protein. The protein synthesis inhibitor emetine was used to determine whether newly synthesized host proteins were required for antigen generation. The addition of emetine to target cells prior to or at the time of the addition of UV-VSV inhibited lysis by anti-vsv CTL. This inhibition could be due to depletion of newly synthesized MHC molecules from intracellular membranes. This hypothesis was supported by using brefeldin A to delay membrane protein transport in target cells during the time of incubation with emetine and UV-VSV, which resulted in partial reversal of the effect of emetine. These results suggest that newly synthesized class I MHC molecules are required for the generation of antigens recognized by anti-vsv CTL. Cytotoxic T lymphocytes (CTL) recognize viral antigens complexed with host proteins of the major histocompatibility complex (MHC), usually class I MHC molecules, at the surface of infected cells. Peptide fragments derived from the proteolysis of viral proteins bind to a cleft in the class I MHC molecule formed by the highly polymorphic otl and (x2 domains (reviewed by Townsend and Bodmer [2]). The antigens recognized by class I MHC-restricted CTL are similar to antigens presented to helper T cells by class II MHC molecules in specialized antigen-presenting cells. As a general rule, antigens that bind to class II MHC molecules are derived from the uptake of exogenous proteins by endocytosis, while antigens associated with class I MHC molecules are primarily derived from intracellular sites such as the cytosol (reviewed by Yewdell and Bennink [27]). It is likely that many viral antigens recognized by class I MHCrestricted CTL are generated by de novo synthesis during virus infection, since in some cases CTL recognition of target cells can be prevented by the inhibition of viral protein synthesis (14). However, there are also examples of CTL recognition of antigens generated without protein biosynthesis. For instance, enveloped viruses inactivated by UV irradiation or heat treatment can sensitize target cells for CTL recognition, provided that the inactivated virions are capable of penetration into the cytosol by the fusion of the virus envelope with cellular membranes (9, 28). Cells have also been shown to process antigens from exogenously added proteins introduced into the cytosol by the osmotic lysis of pinosomes (13). We have previously shown that vesicular stomatitis virus (VSV) inactivated by UV irradiation (UV-VSV) can be * Corresponding author. efficiently used in the secondary elicitation of CTL in vitro (3). As in the case of CTL elicited by infectious virus (16), the anti-vsv CTL elicited by UV-inactivated virus in H-2b mice primarily recognize determinants derived from the major viral nucleoprotein (N protein) and are primarily restricted by H-2Kb. The results presented here extend these observations to show that UV-VSV can be used to sensitize target cells for lysis by anti-vsv CTL. This is an important distinction, since CTL elicitation using heterogeneous spleen cell cultures could conceivably involve a function restricted to specialized antigen-presenting cells that process UV-inactivated virus, as shown recently for the presentation of ovalbumin to class I MHC-restricted T cells (17). The processing pathway by which antigen fragments derived from cytoplasmic proteins associate with class I MHC molecules is unknown. Recent evidence indicates that the inhibition of class I MHC transport from the endoplasmic reticulum (ER) prevents the CTL recognition of target cells, suggesting that newly synthesized MHC molecules are required for antigen presentation (8, 15, 26). Alternatively, it is possible that class I MHC molecules synthesized prior to antigen exposure can associate with antigenic peptides by internalization and recycling of MHC molecules to the ER. The use of UV-inactivated virus to sensitize targets for lysis by CTL in the presence of a protein synthesis inhibitor makes it possible to evaluate the role of newly synthesized MHC molecules in antigen recognition, since viral protein synthesis is not required. In the experiments presented here, the protein synthesis inhibitor emetine was used in CTL assays to prevent the synthesis of host proteins and thereby prevent the availability of MHC molecules for antigen binding. The results suggest that MHC molecules synthesized prior to exposure to UV-VSV play little if any role in antigen presentation. 6856

2 VOL. 65, 1991 PROTEIN SYNTHESIS REQUIREMENTS FOR ANTI-VSV CTL 6857 MATERIALS AND METHODS Materials. VSV (Indiana serotype) was grown in BHK cells and was purified and inactivated by UV irradiation as described previously (3). The peptide RGYVYQGL (singleletter code), corresponding to amino acids 52 to 59 of the VSV N protein, was synthesized on an Applied Biosystems model 43A peptide synthesizer and was purified by reversephase liquid chromatography. Oligonucleotides were synthesized on an Applied Biosystems model 38B DNA synthesizer. Anti-VSV CTL. The secondary elicitation of anti-vsv CTL from C57BL/6 mice in vitro by using UV-VSV was performed as described previously (3). Anti-VSV CTL clone 33, derived from a C57BL/6 mouse, was obtained from James Sheil (West Virginia University School of Medicine) and was maintained by restimulation with irradiated spleen cells from H-2Kbm8 mice as described previously (18). The conditions for the infection of target cells, labeling with 51Cr, and incubation with effector cells have been described previously (3). Briefly, EL4 cells were incubated with either infectious VSV (1 to 3 PFU per cell) or UV-VSV for 3 min to 1 h. MC57G cells were infected with vaccinia virus recombinants for 7 h. Cells were labeled with 51Cr for 1.5 h and then incubated with effectors in a 4-h 51Cr release assay. Generation of deletion fragments of the N gene. The plasmid puc18/n was constructed by subcloning a 1.4-kbp XhoI fragment from pjs223 (19) containing the full-length gene for the VSV N protein into the SalI site of puc18 oriented so that cleavage with PstI produced an approximately 6-bp fragment encoding the amino terminus and an 8-bp fragment encoding the carboxy terminus of the N protein. Two complementary oligonucleotides, 5'GTGTGATAA and 5'A GCTTTATCACACTGCA, were annealed and ligated to the 6-bp PstI fragment encoding amino acids 1 to 198 to introduce a new stop codon and HindIII cohesive ends into the fragment. Similarly, a new start codon and HindIlI cohesive ends were introduced into the 8-bp PstI fragment encoding amino acids 197 to 423 by ligation with an annealed mixture of the oligonucleotides 5'AGCTTGTCGACCACCA TGATTGCA and 5'ATCATGGTGGTCGACA. Finally, a new start codon and Hindlll cohesive ends were introduced into a 1-kbp HinPI fragment of puc18/n encoding amino acids 128 to 423 by ligation with an annealed mixture of the oligonucleotides 5'AGCTTGTCGACCACCATGTC and 5'C GGACATGGTGGTCGACA. The ligation mixtures were digested with HindIlI to remove multiple copies of the oligonucleotides, and the N gene fragments were cloned into the HindIII site of puc18. The correct insertion of the oligonucleotides was confirmed by dideoxy sequencing on supercoiled templates, as described previously (6). Vaccinia virus recombinants. The vaccinia virus expression vector psc11 (5) was modified to accept HindIII inserts by changing the original HindIII site to a BglII site and the SmaI site to a Hindlll site by restriction enzyme cleavage followed by ligation with the appropriate linkers (New England Biolabs, Beverly, Mass.). The N gene fragments were subcloned into the new HindIII site in the coding orientation under the control of the vaccinia virus 7.5K promoter. Transfection of CV1 cells infected with wild-type vaccinia virus and isolation of recombinants containing plasmid sequences was performed as described previously (5). The vaccinia virus recombinant that expresses the full-length N protein has been previously described (12, 29). CO)._a c-. Q - _o) OR E:T Rato FIG. 1. CTL recognition of EL4 cells sensitized with UV-VSV. EL4 cells were incubated for 3 min with infectious VSV ([1), no virus (), or UV-VSV at 4 (U), 1 (),.25 (A), or.6 (X) p.g/16 cells. Cells were labeled with 51Cr and incubated with secondary anti-vsv CTL at the indicated effector-to-target (E:T) ratios. The percentage of specific lysis above the level of spontaneous release of 51Cr in the absence of effectors was calculated. Spontaneous release from the different target cells was not more than 8%. The points represent the means of triplicate determinations. RESULTS UV-VSV was tested for its ability to sensitize target cells for lysis by anti-vsv CTL. Anti-VSV CTL effectors were elicited by the secondary in vitro stimulation of primed spleen cells from C57BL/6 (H-2b) mice with UV-VSV. EL4 target cells were incubated with various concentrations of UV-VSV for 3 min. Cells were then labeled with 51Cr and incubated with anti-vsv CTL at various effector-to-target ratios. The data in Fig. 1 show that anti-vsv CTL recognized and lysed EL4 targets sensitized with UV-VSV in a dose-dependent manner when UV-VSV was in the range of.6 to 4 p.g/16 cells. The maximum level of lysis obtained with UV-VSV was comparable to that obtained with infectious virus, although target cells incubated with infectious VSV were always lysed more efficiently than those incubated with UV-VSV. Residual viral protein synthesis in EL4 cells incubated with UV-VSV was not detectable by labeling with [35S]methionine and analysis by sodium dodecyl sulfate-gel electrophoresis and fluorography (data not shown). Furthermore, plaque assay of UV-VSV showed that the infectivity was reduced by a factor of >18 by UV inactivation. The target size for the UV inactivation of N protein synthesis is the smallest of any VSV protein and is about 1/1 the target size for the UV inactivation of viral infectivity (1). However, this still implies that N protein synthesis was reduced by a factor of >17, which is several orders of magnitude less than that required to generate significant CTL recognition. Anti-VSV CTL from H-2b mice primarily recognize determinants derived from the N protein in the context of H-2Kb (16). A CTL clone that is representative of this specificity, clone 33 (18), was tested for the recognition of EL4 targets sensitized with UV-VSV. Figure 2 demonstrates that EL4 targets generated with infectious VSV are much more efficiently killed by clone 33 CTL than by the heterogeneous

3 6858 ROSCOE ET AL. J. VIROL _na 32 3.Q CO 2 Ma 3 - cj C') E:T Ratio FIG. 2. Recognition by clone 33 CTL or by heterogeneous anti-vsv CTL of targets sensitized with UV-VSV. EL4 cells were incubated for 3 min with UV-VSV (4 i±g/16 cells, open symbols) or infectious virus (closed symbols) and then labeled with 51Cr and incubated with clone 33 CTL (squares) or secondary anti-vsv CTL (circles) at the indicated effector-to-target (E:T) ratios. anti-vsv CTL population. In contrast, cells sensitized with UV-VSV are killed as efficiently by the heterogeneous population as by the cloned CTL. Thus, targets sensitized with UV-VSV expressed the epitope recognized by clone 33 CTL; however, a significant percentage of the heterogeneous anti-vsv CTL population must have recognized additional determinants in targets generated with UV-VSV to account for its relative efficiency of killing compared with that of the clone 33 CTL. The epitope recognized by clone 33 CTL consists of amino acids 52 to 59 of the N protein (23). This was shown by using a set of overlapping chemically synthesized peptides corresponding to the amino terminal one-third of the N protein (amino acids 1 to 161) and by analysis of an endogenous peptide bound to H-2Kb isolated from VSV-infected cells. The additional determinants recognized by the heterogeneous anti-vsv CTL population could be derived either from sequences of the N protein other than the atnino terminal one-third or else from VSV proteins other than the N protein. Anti-VSV CTL were tested for their ability to lyse target cells infected with vaccinia virus recombinants expressing deletion fragments of the N protein to determine whether other determinants derived from the N protein were recognized, as shown in Fig. 3. Target cells infected with vaccinia virus recombinants expressing the full-length N protein (423 amino acids) or the amino terminal 198 amino acids of the N protein were lysed by anti-vsv CTL. However, two other viruses, one expressing amino acids 128 to 423 and the other expressing amino acids 197 to 423, were not recognized by anti-vsv CTL, indicating that additional epitopes do not map to the carboxy terminus of the N protein. Thus, the additional determinants recognized by anti-vsv CTL from H-2b mice are probably derived from VSV proteins other than the N protein. The existence of CTL specific for VSV proteins other than N protein has been previously demonstrated. Secondary anti-vsv CTL show lower reactivity to EL4 cells transfected with the N gene than to VSV-infected EL4 cells (16). In addition, clonal analysis of anti-vsv CTL has demonstrated the existence of o, lo E:T Ratio FIG. 3. CTL recognition of cells infected with vaccinia virus recombinants that express deletion fragments of the N protein. MC57G cells were infected with VSV (l), no virus (), wild-type vaccinia virus (A), or recombinant vaccinia virus expressing the full-length N protein (amino acids 1 to 423) (-), amino acids 1 to 198 (), amino acids 128 to 423 (A), or amino acids 197 to 423 (x). Cells were labeled with 51Cr and incubated with secondary anti-vsv CTL at the indicated effector-to-target (E:T) ratios. CTL that do not recognize EL4 cells transfected with the N gene (18). These determinants are not likely to be derived from the VSV glycoprotein (G protein), since cells transfected with the G gene or infected with vaccinia virus recombinants that express the G protein are not recognized by H_2b CTL (3, 16). Target cells sensitized with UV-VSV were used to determine whether viral antigens associate only with newly synthesized class I MHC molecules by incubating the cells with the protein synthesis inhibitor emetine. EL4 cells were incubated with emetine for various times before and after the addition of UV-VSV. Cells were then labeled with 51Cr and incubated with anti-vsv effector cells. Since emetine is an irreversible inhibitor of protein synthesis, it was not necessary to maintain the cells in medium with emetine throughout the incubation with effectors. In contrast, reversible inhibitors of protein synthesis, such as cycloheximide, were found to be toxic to effector cell function when included throughout the experiment (data not shown), as reported previously (4). The data in Fig. 4 indicate that when emetine was added 1 h after the addition of virus, there was little or no inhibition of CTL recognition and lysis. When emetine was added at the time of the addition or 3 min after the addition of virus, there was partial inhibition, and when emetine was added 1 or 2 h before the addition of virus, there was significant inhibition of CTL recognition. These results suggest that newly synthesized MHC molecules are required for antigen recognition by anti-vsv CTL. Alternatively, the emetine could be inhibiting other processes in target cells required either for antigen presentation or for lysis by CTL. To rule out the latter possibility, EL4 cells were incubated with emetine either before or after the addition of the chemically synthesized N peptide corresponding to amino acids 52 to 59. Exogenous peptides appear to be able to bind to surface MHC molecules, bypassing the intracellular processing pathway (8, 13, 15, 26). Figure 5 demonstrates that targets incubated with the peptide either before or after the

4 VOL. 65, ro PROTEIN SYNTHESIS REQUIREMENTS FOR ANTI-VSV CTL 6859 a) CD.~2 U ~ E:T Ratio FIG. 4. Emetine inhibition of CTL recognition of target cells sensitized with UV-VSV. EL4 cells were incubated in medium without emetine (El) or in medium with 2,uM emetine at 37 C for 2 h before (-), 1 h before (), h before (A), 3 min after (A), or 1 h after () the addition of UV-VSV (4 p.g/16 cells). Cells were washed to remove the emetine, labeled with 51Cr, and incubated with secondary anti-vsv CTL at the indicated effector-to-target (E:T) ratios. addition of emetine were lysed by anti-vsv CTL. Thus, emetine does not inhibit the ability of targets to be lysed by anti-vsv CTL, provided the cell expresses viral determinants. Further support for the idea that the inhibition of CTL recognition by emetine was due to depletion of intracellular 5 4 >" 3.ES co 2 1 >% 3- -J 1./ antigens (14). We provide evidence here that de novo synthesis of viral proteins is not a requirement for recognition by anti-vsv CTL, since EL4 cells incubated with E:T Rato FIG. 5. Emetine does not inhibit CTL recogniition of target cells UV-VSV readily become targets for lysis. This result was sensitized with a chemically synthesized pepti[de fragment of N also obtained with other target cell types (unpublished protein. EL4 cells were incubated in medium with 2 p.m emetine results). Similar results have been obtained with UV- and for 2 h before () or 1 h after (A) the additiorn of 5,uM peptide heat-inactivated influenza virus (11, 28) and Sendai virus (9). corresponding to amino acids 52 to 59 of the N include uninfected cells (), VSV-infected cells ( p), anduninfected In these cases, the generation of class I MHC-associated cells incubated with the peptide in the absence of. emetine(). Cells antigens requires virus penetration into the cytosol. Al- 5"Cr, and incu- though we did not test this point, it is also likely that the were washed to remove the emetine, labeled with bated with secondary anti-vsv CTL at the inclicated effector-to- recognition of UV-VSV also requires virus penetration. target (E:T) ratios. The ability of UV-VSV to sensitize targets for lysis by E:T Rato FIG. 6. CTL recognition of cells sensitized with UV-VSV in the presence of BFA and emetine. EL4 cells were incubated in medium with 5,uM BFA (open symbols). After 1 h, emetine was added to a final concentration of 2,uM (O and LI). Thirty minutes later, cells were incubated with UV-VSV (4 p.g/16 cells) for 1 h. The cells were washed and resuspended in medium without BFA (A and ) or with 2.5,uM BFA (O) and then labeled with 51Cr and incubated with secondary anti-vsv CTL at the indicated effector-to-target (E:T) ratios. Controls shown are uninfected () and infected (A) EL4 cells without treatment with BFA or emetine. H-2Kb was obtained by using brefeldin A (BFA) to delay membrane protein transport. BFA reversibly inhibits the transport of membrane proteins, such as MHC molecules (8, 15), out of the ER. EL4 cells were incubated with BFA for 1 h before the addition of emetine and then of UV-VSV. The cells were incubated in the presence of BFA for an additional 1 h, and the BFA was then removed to allow MHC protein transport to the cell surface prior to incubation with effectors. As shown in Fig. 6, incubation with BFA partially reversed the inhibition of CTL recognition caused by emetine. These results suggest that delaying the transport of newly synthesized proteins from the ER can partially compensate for the inhibitory effects of emetine. DISCUSSION Antigens associated with class I MHC molecules are generated by an as yet unidentified proteolytic degradation process. There is evidence to support the idea that the processing pathways for these antigens use a cytosolic route (reviewed by Yewdell and Bennink [27]). Previously, it was suggested that endogenous synthesis of viral proteins was :, = v necessary for the generation of class I MHC-associated

5 686 ROSCOE ET AL. anti-vsv CTL does not rule out the possibility that class I MHC-associated antigens are more aptly derived from de novo synthesized viral proteins. In fact, this may be the case, since cells treated with infectious VSV were better targets for anti-vsv CTL than those treated with UV-VSV. In addition, the relative sensitivity to lysis by different CTL populations of cells sensitized with UV-VSV was different from that of cells sensitized with infectious virus. Killing by clone 33 CTL of cells sensitized with infectious virus was greater than killing by heterogeneous anti-vsv CTL, while killing by clone 33 CTL of cells sensitized with UV-VSV was relatively less efficient than killing by the heterogeneous population. This is probably because the epitope recognized by clone 33 CTL is generated in abundance in infected cells but is present in limited amounts on cells sensitized with UV-VSV. Although the epitope recognized by clone 33 CTL represents the dominant specificity of anti-vsv CTL (23), the heterogeneous CTL population also contains a variety of CTL with receptors specific for other VSV epitopes (18). CTL recognition of vaccinia virus recombinants expressing deletion fragments of the N protein was tested to determine whether the additional determinants recognized by anti-vsv CTL include other epitopes derived from the N protein. Two recombinants that express carboxy terminal N protein fragments were not recognized, while a recombinant that expresses an amino terminal fragment that includes the epitope previously mapped was recognized. These results do not completely rule out the existence of other N protein epitopes, since the carboxy terminal fragments may lack conformational properties necessary for the generation of a putative epitope. Nonetheless, they provide strong evidence that the additional epitopes reside on VSV proteins other than the N protein. Anti-VSV CTL from H-2b mice do not recognize cells infected with vaccinia virus recombinants that express the G protein (3) or are transfected with the G gene (16). Therefore, the additional epitopes probably reside on the other internal viral proteins, the L, NS, or M protein. Previous experiments have suggested that newly synthesized MHC molecules are required for antigen presentation. For example, inhibition of the transport of class I MHC molecules to the cell surface prevents the recognition of viral antigens by CTL (8, 15, 26). In addition, some class I MHC molecules require antigenic peptides for proper folding and association with 12 microglobulin, which are necessary for transport out of the ER (21). Since class I MHC molecules on the cell surface have been shown to undergo recycling to intracellular membranes (22), these results do not rule out the possibility that previously synthesized MHC molecules also present antigens after being recycled to intracellular sites of antigen-mhc association. To test this possibility, we used the protein synthesis inhibitor emetine to prevent the synthesis of MHC molecules in cells incubated with UV- VSV. The logic is that the addition of emetine for a sufficient time prior to the addition of virus should result in the depletion of MHC molecules from intracellular membranes due to the transport of the MHC molecules to the cell surface. The addition of emetine 1 to 2 h prior to the addition of antigen inhibits CTL recognition with cells treated with UV-VSV but not with cells treated with exogenously added antigenic peptide, which bypasses the intracellular pathway of antigen generation. These data support the idea that newly synthesized MHC molecules are required for antigen presentation, although they do not rule out the possibility that synthesis of host components other than MHC molecules are required for antigen processing. When emetine was added to target cells at the same time J. VIROL. that UV-VSV was added, there was partial inhibition of CTL recognition, and when emetine was added 3 min to 1 h after the virus was added, there was little or no inhibition. These results are consistent with the kinetics of intracellular transport of viral and MHC antigens. Inhibition of cellular protein synthesis occurs within minutes of the addition of emetine (1). Transport of H-2Kb to the cell surface has a half-time of about 3 min (24). Since virus attachment and penetration requires about 3 min, it would be expected that the addition of emetine at the time of the addition of virus would result in the significant depletion of intracellular H-2Kb prior to the availability of viral antigenic peptides for binding to MHC molecules. This depletion of intracellular H-2Kb could be prevented by treating cells with BFA (8, 15), which partially reversed the inhibition of CTL recognition by emetine, provided the BSA was subsequently removed to allow transport of H-2Kb to the cell surface (Fig. 6). The observation that newly synthesized host proteins are required for the presentation of viral antigens has interesting implications for CTL recognition of cells infected with cytopathic viruses, such as VSV, which inhibit host protein synthesis. It can be predicted that only those viral proteins expressed early in infection, prior to the inhibition of host protein synthesis, can serve to generate antigens recognized by CTL. In contrast, cells infected with viruses which are minimally cytopathic or which establish persistent infections might be more easily lysed by CTL because of their higher level of antigen expression. It has been shown previously that CTL specific for the VSV N protein recognize transfected cell lines which permanently express the N protein gene more efficiently than cells infected with virus in a cold target inhibition experiment (16). This occurs despite the much higher level of N protein synthesis in VSV-infected cells than in the transfectants. In light of the results presented here, this could be explained by the fact that VSV effectively shuts off host protein synthesis within the first 2 h postinfection and thereby limits the availability of class I MHC molecules for antigen presentation. It has also been shown that cells are poor targets for CTL recognition when the influenza virus hemagglutinin is expressed under the control of a vaccinia virus late promoter in recombinant viruses (7). Since vaccinia virus inhibits host protein synthesis late in infection, it seems reasonable that the lack of recognition is due to the lack of newly synthesized MHC molecules available for antigen binding. Class I MHC molecules vary widely in their transport times from their site of synthesis in the ER to the cell surface. For example, within the same murine haplotype, class I MHC molecules may differ in their transport times from the ER to the Golgi membranes by as much as 4 to 6 h (2, 25). In the case of antigens associated with slowly transported MHC molecules, it would be expected that the inhibition of protein synthesis would be relatively ineffective at preventing antigen generation for CTL recognition. The presence of such slowly transported class I MHC molecules would appear to offer an advantage in the CTL recognition of antigens derived from viruses that are rapidly cytopathic. ACKNOWLEDGMENTS We thank Bernard Moss for providing us with vaccinia virus vectors, James Sheil for the clone 33 CTL, and Grada M. Van Bleek and Stanley Nathenson for communicating their results prior to publication. This work was supported by NIH grant A12778 and by North Carolina Biotechnology Center grant 86-G-224. Peptide synthesis was performed in the Protein Analysis Core Laboratory, and

6 VOL. 65, 1991 PROTEIN SYNTHESIS REQUIREMENTS FOR ANTI-VSV CTL 6861 oligonucleotide synthesis was performed in the DNA Synthesis Core Laboratory of the Cancer Center of Wake Forest University and was supported in part by NIH grants CA12197 and RR4869 and by a grant from the North Carolina Biotechnology Center. REFERENCES 1. Ball, L. A., and C. N. White Order of transcription of genes of vesicular stomatitis virus. Proc. Natl. Acad. Sci. USA 73: Beck, J. C., T. H. Hansen, S. E. Culien, and D. R. Lee Slower processing, weaker P2-M association, and lower surface expression of H-2Ld are influenced by its amino terminus. J. Immunol. 137: Bowman, M. R., D. S. Lyles, and J. W. Parce Possible mechanisms by which the H-2Kb13 mutation may decrease cytotoxic T lymphocyte recognition of vesicular stomatitis virus nucleoprotein antigen. J. Virol. 61: Brunner, K. T., J. Mauel, J.-C. Cerottini, and B. Chapuis Quantitative assay for the lytic action of immune lymphoid cells on 5"Cr-labelled allogeneic target cells in vitro. Inhibition by isoantibody and by drugs. Immunology 14: Chakrabarti, S., K. Brechling, and B. Moss Vaccinia virus expression vector: coexpression of,3-galactosidase provides visual screening of recombinant virus plaques. Mol. Cell. Biol. 5: Chen, E. J., and P. H. Seeburg Supercoil sequencing: a fast simple method for sequencing plasmid DNA. DNA 4: Coupar, B. E. H., M. E. Andrew, G. W. Both, and D. B. Boyle Temporal regulation of influenza hemagglutinin expression in vaccinia virus recombinants and effects on the immune response. Eur. J. Immunol. 16: Cox, J. H., J. W. Yewdeli, L. C. Eisenlohr, P. R. Johnson, and J. R. Bennink Antigen presentation requires transport of MHC class I molecules from the endoplasmic reticulum. Science 247: Gething, M.-J., U. Koszinowski, and M. Waterfield Fusion of Sendai virus with the target cell membrane is required for T cell cytotoxicity. Nature (London) 274: Goldberg, I. H., and K. Mitsugi Sparsomycin, an inhibitor of aminoacyl transfer to polypeptide. Biochem. Biophys. Res. Commun. 23: Hosaka, Y., F. Sasao, K. Yamanaka, J. R. Bennink, and J. W. Yewdell Recognition of noninfectious influenza virus by class I-restricted murine cytotoxic T lymphocytes. J. Immunol. 14: Mackett, M., T. Yilma, J. K. Rose, and B. Moss Vaccinia virus recombinants: expression of VSV genes and protective immunization of mice and cattle. Science 227: Moore, M. W., F. R. Carbone, and M. J. Bevan Introduction of soluble protein into the class I pathway of antigen processing and presentation. Cell 54: Morrison, L. A., A. E. Lukacher, V. L. Braciale, D. P. Fan, and T. J. Braciale Differences in antigen presentation to MHC class I- and class II-restricted influenza virus-specific cytolytic T lymphocyte clones. J. Exp. Med. 163: Nuchtern, J. G., J. S. Bonafacino, W. E. Biddison, and R. D. Klausner Brefeldin A implicates egress from endoplasmic reticulum in class I restricted antigen presentation. Nature (London) 339: Puddington, L., M. J. Bevan, J. K. Rose, and L. Lefranqois N protein is the predominant antigen recognized by vesicular stomatitis virus-specific cytotoxic T cells. J. Virol. 6: Rock, K. L., S. Gamble, and L. Rothstein Presentation of exogenous antigen with class I major histocompatibility complex molecules. Science 249: Sheil, J. M., M. J. Bevan, and L. Lefrancois Characterization of dual reactive H-2Kb-restricted anti-vesicular stomatitis virus and alloreactive cytotoxic T cells. J. Immunol. 138: Sprague, J., J. H. Condra, H. Arnheiter, and R. A. Lazzarini Expression of a recombinant DNA gene coding for the vesicular stomatitis virus nucleocapsid protein. J. Virol. 45: Townsend, A., and H. Bodmer Antigen recognition by class I-restricted T lymphocytes. Annu. Rev. Immunol. 7: Townsend, A., C. Ohlen, J. Bastin, H.-G. Ljunggren, L. Foster, and K. Kairre Association of class I major histocompatibility heavy and light chains induced by viral peptides. Nature (London) 34: Tse, D. B., and B. Pernis Spontaneous internalization of class I major histocompatibility complex molecules in T lymphoid cells. J. Exp. Med. 159: Van Bleek, G. M., and S. G. Nathenson Isolation of an endogenously processed immunodominant viral peptide from the class I H-2Kb molecule. Nature (London) 348: Williams, D. B., F. Boriello, R. A. Zeff, and S. G. Nathenson Intracellular transport of class I histocompatibility molecules: influence of protein folding on transport to the cell surface. J. Biol. Chem. 263: Williams, D. B., S. J. Swiedler, and G. W. Hart Intracellular transport of membrane glycoproteins: two closely related histocompatibility antigens differ in their rates of transit to the cell surface. J. Cell Biol. 11: Yewdell, J. W., and J. R. Bennink Brefeldin A specifically inhibits presentation of protein antigens to cytotoxic T lymphocytes. Science 244: Yewdell, J. W., and J. R. Bennink The binary logic of antigen processing and presentation to T cells. Cell 62: Yewdeil, J. W., J. R. Bennink, and Y. Hosaka Cells process exogenous proteins for recognition by cytotoxic T lymphocytes. Science 239: Yewdell, J. W., J. R. Bennink, M. Mackett, L. Lefrancois, D. S. Lyles, and B. Moss Recognition of cloned vesicular stomatitis virus internal and external gene products by cytotoxic T lymphocytes. J. Exp. Med. 163: