The natural mouse pathogen polyoma virus is highly oncogenic in H-2k mice carrying the endogenous superantigen encoded by the mouse mammary tumor provirus Mtv-7. This superantigen results in deletion of Vβ6 TCR-expressing polyoma-specific CD8+ CTL, which appear to be critical effectors against polyoma tumorigenesis. Here we have isolated cloned lines of CD8+ T cells from resistant (i.e., Mtv-7) H-2k mice that specifically lyse syngeneic polyoma virus-infected cells and polyoma tumor cells. Nearly all these CTL clones express Vβ6 and are restricted in their recognition of virus-infected cells by H-2Dk. Screening a panel of synthetic peptides predicted to bind to Dk, for which no consensus peptide binding motif is known, we identified a peptide corresponding to a nine-amino acid sequence in the carboxyl-terminus of the middle T (MT) protein (amino acids 389–397) that was recognized by all the Vβ6+CD8+ CTL clones. The inability of MT389–397-reactive CTL to recognize cells infected with a mutant polyoma virus encoding a MT truncated just proximal to this sequence indicates that MT389-397 is a naturally processed peptide. The frequencies of precursor CTL specific for polyoma virus and MT389–397 peptide were similar, indicating that MT389–397 is the immunodominant epitope in H-2k mice. In addition, polyoma-infected resistant mice possess a 10- to 20-fold higher MT389-397-specific precursor CTL frequency than susceptible mice. This highly focused CTL response to polyoma virus provides a valuable animal model to investigate the in vivo activity of CTL against virus-induced neoplasia.

Class I MHC-restricted CTL are critical elements of immune surveillance for virus-induced neoplasia (1). The importance of CTL in controlling viral oncogenesis is most clearly illustrated by the high frequency of EBV-positive lymphoproliferative disease in T cell-immunocompromised patients (2) and the ability of EBV-specific CTL to promote regression of EBV-associated B cell lymphomas (3). Much less is understood about immunologic control of tumors induced in natural hosts by viruses of the family Papovaviridae, whose most important human pathogen is human papillomavirus (HPV).3 Infection of cervical epithelial cells by certain genotypes of HPV is the antecedent of most cases of cervical carcinoma (4), and enhanced progression of HPV-associated cervical lesions is seen in immunosuppressed patients (5). That CD8+ CTL are involved in surveillance for HPV-induced cervical carcinoma is implicated by the high frequency of cervical cancers showing down-regulation of class I MHC molecules and TAP proteins (6) and evidence of specific mutations in a class I MHC CTL viral epitope in cervical cancers (7). A primary obstacle to dissecting the components of immune responsiveness to papovaviruses and their induced tumors is the absence of good animal models that mimic natural infection.

Polyoma virus is a mouse papovavirus capable of inducing a broad spectrum of epithelial and mesenchymal cell-derived tumors (8). The 5.3-kb circular polyoma DNA genome encodes three early region nonstructural proteins, termed small T (ST), middle T (MT), and large T (LT), and three late region virion capsid proteins. Cellular transformation, tumor induction, and virion assembly require constitutive expression of MT (9, 10, 11), the oncoprotein of polyoma virus. A number of studies using athymic mice and adoptive transfer of virus-immune T cells clearly document that anti-viral T lymphocytes are responsible for conferring resistance to polyoma tumors (12, 13). A protective anti-tumor effector function for anti-polyoma class I MHC-restricted CTL is suggested by a recent report showing that mice immunized with a synthetic peptide corresponding to an LT protein sequence that binds to a class I MHC molecule reject syngeneic polyoma tumors (14).

In immunocompetent mice, high susceptibility to polyoma virus-induced tumors is manifested only by neonatal inoculation of certain inbred strains of H-2k mice (15, 16). We recently presented evidence indicating that this susceptibility is conferred by the endogenous superantigen (SAG) encoded by the mouse mammary tumor provirus-7 (Mtv-7). By deleting thymocytes whose TCR use specific β-chain variable domains (Vβ) (17), Mtv-7SAG creates a hole in the peripheral T cell repertoire for T cells required for polyoma tumor immunosurveillance (18). Biased usage of the Mtv-7SAG-reactive Vβ domain, Vβ6, was found among CD8+ T cells infiltrating a polyoma tumor implanted in a virus-immune, resistant (i.e., Mtv-7-negative) H-2k mouse, and CD8+Vβ6+ T cells expressing virus-specific cytotoxicity dominated bulk cultures of T cells derived from neonatally infected, resistant mice (18). These findings strongly implicate polyoma virus-specific CD8+Vβ6+ CTL as essential anti-polyoma tumor effectors in vivo.

A long-standing difficulty in defining the effector T cell(s) that mediate anti-polyoma tumor activity and identifying T cell epitopes has been an inability to establish lines of polyoma-specific T cells. One early study reported induction of specific lytic activity against virus-infected and -transformed cells by CD8+ T cells from virus-infected mice after coculture with polyoma-transformed syngeneic cells (19). We recently showed that short term cultures of virus-immune T cells restimulated in vitro with infected syngeneic stimulator cells exhibited anti-viral and anti-tumor cytotoxicity (18). Here, we report the isolation of polyoma-specific CD8+ CTL cloned lines from virus-infected H-2k mice. As expected from the predominant usage of Vβ6 by anti-viral CTL in bulk cultures, most CTL clones express a Vβ6 TCR. Moreover, nearly all the clones are restricted by the H-2Dk molecule in their recognition of virus-infected cells. In the absence of a consensus motif for peptides bound to Dk, potential anchor residues for Dk binding peptides were predicted from crystallographically solved and molecularly modeled class I MHC:peptide structures. Using this deduced Dk peptide binding motif, we identified a nine-amino acid sequence in the MT protein as the immunodominant anti-viral CTL epitope. CTL directed to this epitope recognize both virus-infected cells and cells derived from polyoma-induced tumors. Polyoma tumor-susceptible mice were found to possess a 10- to 20-fold lower frequency of precursor CTL to this epitope than resistant mice. Possible explanations for the inability of a weak anti-viral CTL response to prevent the outgrowth of virus-transformed cells are discussed.

C57BR/cdJ and C3H/HeSnJ mice were purchased from The Jackson Laboratory (Bar Harbor, ME). C3H/HeNCr and C3H/BiDaCr mice were purchased from the Frederick Cancer Research and Development Center of the National Cancer Institute (Frederick, MD). Pregnant ICR outbred mice and female, 4-wk-old, Sprague-Dawley rats were purchased from Harlan Sprague-Dawley, Inc. (Indianapolis, IN).

The polyoma virus strain A2 was molecularly cloned and plaque purified. Virus stocks were prepared by inoculating primary kidney cells from 10-day-old ICR mice at a multiplicity of infection (moi) of 0.1 and carrying cultures in the presence of DMEM (Mediatech, Herndon, VA) containing 2% FBS (HyClone, Inc., Logan, UT) to complete the viral cytopathic effect. Virus stocks used for inoculating mice were similarly prepared, except that kidney cells were obtained from 10-day-old C57BR/cdJ mice, and virus infection was conducted in the absence of serum; these stocks are designated serum-free A2 virus. Cell lysates were titrated by plaque assay on UC1B cells (obtained from American Type Culture Collection, Rockville, MD) and typically contained 2 × 107 plaque-forming units/ml. Newborn mice (<18 h of age) were injected s.c. in hindfootpads with 20 to 30 μl of serum-free A2 virus.

The polyoma virus strain PTA1387T encodes a MT protein lacking the carboxyl-terminal 37 amino acids (11). The parental PTA virus, like the A2 virus, is a highly tumorigenic, large plaque, polyoma strain (8); the MT protein sequences of PTA and A2 are identical. PTA1387T virus stocks were prepared from molecularly cloned virus on primary baby mouse kidney cells and titrated as described above.

AG104A cells were derived from a spontaneous tumor of C3H/HeN (H-2k) origin (20) (provided by Dr. H. Schreiber, University of Chicago, Chicago, IL). SVCOL, SVB10.A, and SVD2 are SV40-transformed cell lines from C3H.OL (H-2KdDk), B10.A (H-2KkDd), and DBA/2 (H-2d) mice, respectively (21) (provided by Dr. L. Gooding, Emory University, Atlanta, GA). 6215 cells were derived from a polyoma virus-induced salivary tumor arising in a gamma-irradiated (900 rad), virus-infected adult C57BR/cdJ (H-2k) mouse (18). L929 cells were obtained from American Type Culture Collection. Cell lines were maintained in DMEM containing 10% FBS. P1.HTR.Dk and P1.HTR.Kk are Dk- and Kk-expressing transfectants, respectively, of P1.HTR (H-2d) cells (provided by Dr. A. Van Pel, Ludwig Institute for Cancer Research, Brussels, Belgium) and were maintained in Iscove’s modified Dulbecco’s medium (IMDM; Life Technologies, Gaithersburg, MD) supplemented with 10% FBS. L929, 6215, P1.HTR.Dk, and P1.HTR.Kk are nonpermissive, and AG104A, SVCOL, SVB10.A, and SVD2 are highly permissive for productive infection by polyoma virus (data not shown).

Peptides were synthesized by the solid phase method on a Symphony/Multiplex Peptide Synthesizer (Rainin, Woburn, MA) with F-moc chemistries. HPLC analysis showed that peptides were >95% pure. Peptide stock solutions were prepared in water at a concentration of 3 mM and stored at −20°C. Peptides were diluted in IMDM and 10% FBS immediately before use in cytotoxicity assays.

Bulk cultures of polyoma-immune T cells were established as previously described (18). Briefly, 1 × 106 mononuclear cells from draining popliteal and inguinal lymph nodes of 12- to 14-day-old virus-infected mice were cocultured with 10 × 106 A2 virus-infected syngeneic splenocytes, gamma-irradiated (2000 rad) 24 h after infection, in 24-well cluster plates (Costar, Cambridge, MA) in 2 ml of IMDM complete medium (IMDM supplemented with 10% FBS, 8% medium of Con A-pulsed rat splenocytes (prepared as described in 22 , 4 mM glutamine, 5 mM α-methylmannoside, 50 μM 2-ME, penicillin, and streptomycin). In vitro tertiary cultures were established by restimulating 2 to 5 × 105 viable mononuclear cells from these day 7 cultures using the same protocol. All cultures were maintained in a humid atmosphere of 7% CO2 at 37°C.

T cells were cloned by limiting dilution from day 7 in vitro secondary or tertiary cultures. Viable cells were cocultured in 96-well flat-bottom microtiter plates (Costar) with 1 × 106 virus-infected, γ-irradiated (2000 rad) syngeneic splenocytes in 0.2 ml/well IMDM complete medium. T cells were cultured in replicates of 48 or 96 wells at dilutions ranging from 1 to 20 cells/well. Clusters of proliferating cells appeared at 5 to 7 days of incubation. To insure that expanded T cells were derived from individual precursor clones, proliferating cell clusters were selected for expansion when the frequency of positive wells at a given responder dilution was <15%, and individual wells contained only one proliferating cell cluster. Typically, 10 of 96 replicate wells from in vitro tertiary cultures plated at 10 cells/well were positive for proliferating cell clusters. T cells were initially expanded for 7 days in 48-well cluster plates containing 5 × 106 virus-infected, irradiated, syngeneic stimulators in 1 ml/well IMDM complete medium and subsequently expanded to 24-well dishes under the same conditions as those used to established the bulk cultures.

Polyoma virus-infected AG104A target cells were prepared as follows. AG104A monolayers harvested with trypsin-EDTA (Life Technologies) were plated at 5 × 105 cells/well into six-well cluster plates (Nunc, Naperville, IL) in 0.2 ml of DMEM containing polyoma virus at an moi of 5, incubated at 37°C for 1.5 to 2 h, overlaid with 3 ml of DMEM containing 10% FBS and murine rIFN-γ (100 U/ml), and incubated at 37°C for another 16 h. Uninfected cells were either treated with lysates of uninfected BMK cells or simply overlaid with medium. To radiolabel adherent target cells, medium was removed, and 150 to 200 μCi of Na251CrO4 (400–1200 Ci/g; New England Nuclear, Boston, MA) with 5% FBS was added to cells. After incubating at 37°C for 1.5 h, 51Cr was removed, then replaced with 1 ml of DMEM and 10% FBS, and incubated at 37°C for 1 h. For experiments using peptide-pulsed targets, peptides at the indicated concentration were prepared in DMEM and 10% FBS, and added at this second incubation step. Cells were then rinsed with DMEM three times, harvested with trypsin-EDTA, and washed twice with DMEM and 10% bovine calf serum (Summit Biotechnology, Ft. Collins, CO).

51Cr-labeled target cells were aliquoted at 5000 cells/well into either V-bottom or U-bottom 96-well microtiter plates (Costar). In certain experiments, 50 μl of peptides were added at 3 times their final concentration to wells containing 50 μl of 51Cr-labeled cells, then after 1-h incubation at 37°C, 50 μl of T cells were added. For experiments using virus-infected or peptide-pulsed targets, 100 μl of target cells and 100 μl of T cells were cocultured in each well. The assay medium was IMDM and 10% FBS. After a 4- to 6-h incubation at 37°C, half the volume of each well was removed and counted in a Beckman Gamma-4000 counter (Beckman, Fullerton, CA). Spontaneous 51Cr release from target cells in all assays was 10 to 20% of the total lysis. The percent specific lysis was calculated as follows: ((51Cr released with effector cells) − (spontaneous 51Cr release)/((total 51Cr released with 1% Triton X-100) − (spontaneous 51Cr release)) × 100. The percent specific lysis values represent the means of four replicate wells. SEMs were always <5% of the mean values and are omitted. Each 51Cr release assay was performed a minimum of three times with similar results; representative experiments are shown.

Viable, nucleated spleen cells from polyoma virus-immune mice were titrated in replicates of 16 wells into U-bottom 96-well microtiter plates (Costar) containing 5 × 105 virus-infected, irradiated (2000 rad), C3H/HeN spleen cells in a total volume of 0.2 ml/well IMDM complete medium. Seven days later, each well received 5 × 105 infected, irradiated C3H/HeN spleen cells to further expand Ag-specific T cells and increase sensitivity in detecting anti-polyoma CTL. Five days later, murine rIL-12 (PharMingen, San Diego, CA) was added at a final concentration of 10 U/ml. IL-12 boosts the Ag-specific cytotoxicity of mature, anti-polyoma CD8+ CTL (A. Lukacher, manuscript in preparation). After an additional 2 days of incubation, each well was split into three wells of U-bottom 96-well microtiter plates, to which were added 2500 51Cr-labeled, murine rIFN-γ (PharMingen)-treated (100 U/ml for 24 h) AG104A target cells that were untreated, pulsed with 10 μM MT389–397 peptide, or infected with polyoma virus. Positive wells were defined as wells for which 51Cr release values exceeded the mean release values of uninfected target cells by >3 times the SD. CTL precursor frequencies were determined according to the method of Quintans and Lefkovits (23).

Viable T cells were isolated on LSM (Organon Teknika Corp., Durham, NC) step gradients and indirectly stained with FITC-conjugated goat anti-hamster IgG or FITC-conjugated goat anti-rat IgG F(ab′)2 secondary Abs (Caltag Laboratories, South San Francisco, CA) and culture supernatants of the following hybridomas: H57-597 (anti-TCR-αβ), B20.6 (anti-Vβ2), 44-22-1 (anti-Vβ6), and KJ16 (anti-Vβ8.1/8.2). Phycoerythrin-conjugated anti-CD4, anti-CD8α, and IgG2a isotype control Abs were also purchased from Caltag Laboratories. Flow cytometry was performed on a Becton Dickinson (Mountain View, CA) FACSort using LYSIS II software, and data were analyzed using CellQuest software.

Virus-specific CTL from two polyoma tumor-resistant H-2k inbred mouse strains, C57BR/cdJ and C3H/HeSnJ (18), were cloned by limiting dilution from in vitro tertiary cultures of T cells harvested from draining lymph nodes of approximately 2-wk-old mice inoculated with polyoma virus at birth. Each round of in vitro restimulation consisted of weekly coculture of T cells with syngeneic spleen cells gamma-irradiated 24 h after polyoma virus infection in the presence of conditioned medium of Con A-pulsed rat spleen cells. Since polyoma DNA replication requires G1 to S cell cycle transition of the host cell (24), and gamma-irradiation induces G1 arrest in primary mammalian cells (25), the spleen cell stimulators were irradiated only after the 72-h viral lytic life cycle was engaged. Approximately 40 T cell clones were isolated from nine virus-primed mice, and all were found to express CD8 by flow cytometry (data not shown). The strongly biased usage of Vβ6 by CD8+ anti-polyoma CTL in bulk cultures (18) (Fig. 6) was reflected in the expression of Vβ6 by roughly 90% of these T cell clones. To exclude the possibility that Vβ6+ T cell clones from the same virus-primed mouse were progeny of a common CTL precursor, Vβ6+ T cell clones from eight different mice were selected for further analysis as were the only two non-Vβ6 clones isolated in this limiting dilution series. All Vβ phenotyping was confirmed and the Vβ8.1 subfamily was determined by RT-PCR analysis using Vβ family consensus sense primers and a Cβ antisense primer (26). The designations, mouse strain origin, and Vβ phenotype of these 10 CD8+ T cell clones are summarized in Table I.

FIGURE 6.

Viral specificity, Vβ6 and coreceptor usage, and MT389–397 epitope recognition by polyoma virus-immune T cell bulk cultures from resistant H-2k mice. In vitro tertiary cultures of virus-immune T cells from four neonatally inoculated C3H/HeSnJ mice were assayed for cytolytic activity against 51Cr-labeled uninfected, polyoma-infected, MT389–397 peptide-pulsed (10 μM), and gag88–96 peptide-pulsed (10 μM) AG104A target cells at the indicated E:T cell ratios for 4 h.

FIGURE 6.

Viral specificity, Vβ6 and coreceptor usage, and MT389–397 epitope recognition by polyoma virus-immune T cell bulk cultures from resistant H-2k mice. In vitro tertiary cultures of virus-immune T cells from four neonatally inoculated C3H/HeSnJ mice were assayed for cytolytic activity against 51Cr-labeled uninfected, polyoma-infected, MT389–397 peptide-pulsed (10 μM), and gag88–96 peptide-pulsed (10 μM) AG104A target cells at the indicated E:T cell ratios for 4 h.

Close modal
Table I.

Polyoma virus-specific CD8+ T cell clonesa

MouseCloneMouse Strainb
6-6 C57BR/cdJ 
8-1 C57BR/cdJ 
9-12 C57BR/cdJ 
10-11 C57BR/cdJ 
11-1 C57BR/cdJ 
13-2 C3H/HeSnJ 
14-1 C3H/HeSnJ 
16-3 C3H/HeSnJ 
16-5 C3H/H3SnJ 8.1 
15-5 C3H/HeSnJ 
MouseCloneMouse Strainb
6-6 C57BR/cdJ 
8-1 C57BR/cdJ 
9-12 C57BR/cdJ 
10-11 C57BR/cdJ 
11-1 C57BR/cdJ 
13-2 C3H/HeSnJ 
14-1 C3H/HeSnJ 
16-3 C3H/HeSnJ 
16-5 C3H/H3SnJ 8.1 
15-5 C3H/HeSnJ 
a

T cell clones were isolated by limiting dilution from in vitro tertiary cultures of lymph node cells of C57BR/cdJ or C3H/HeSnJ mice neonatally inoculated with polyoma virus.

b

T cell clones were stained directly with phycoerythrin-conjugated anti-CD8α or anti-CD4 mAbs or stained indirectly with culture supernatants of hybridomas 44-22-1 (anti-Vβ6), KJ16 (anti-Vβ8.1/8.2), and B20.6 (anti-Vβ2) and FITC-conjugated anti-rat IgG(Fab′)2 secondary Abs. All Vβ phenotyping was confirmed by RT-PCR using a Cβ antisense primer and sense primers specific for Vβ6, Vβ8.1, and Vβ2 genes (26).

Each of these T cell clones lysed polyoma virus-infected, but not uninfected, syngeneic target cells (Fig. 1). Identification of the class I MHC molecule that restricts viral Ag presentation to these anti-polyoma CTL clones was determined using as targets SV40-transformed fibroblasts from B10.A (KkDd), C3H.OL (KdDk), and B10.D2 (KdDd) mice. These target cells are sensitive to CTL-mediated lysis (21) and are highly permissive for productive polyoma virus infection (A. Lukacher, unpublished observations). Because H-2L genes are absent in H-2k mice (27), only H-2K and D molecules are available to present viral peptides to anti-polyoma CTL. As shown in Figure 2, all 10 CTL clones specifically recognized virus-infected cells bearing H-2Dk.

FIGURE 1.

Specific lysis of polyoma-infected cells by virus-immune T cell clones. T cell clones were assayed for cytolytic activity against uninfected or polyoma virus-infected 51Cr-labeled AG104A (H-2k) target cells for 4 h at an E:T cell ratio of 10:1.

FIGURE 1.

Specific lysis of polyoma-infected cells by virus-immune T cell clones. T cell clones were assayed for cytolytic activity against uninfected or polyoma virus-infected 51Cr-labeled AG104A (H-2k) target cells for 4 h at an E:T cell ratio of 10:1.

Close modal
FIGURE 2.

MHC restriction mapping of polyoma virus-specific CTL clones. CTL clones were assayed for cytolytic activity against uninfected or virus-infected SV40-transformed target cells of DBA/2 (KdDd), C3H.COL (KdDk), and B10.A (KkDd) origin for 4 h at an E:T cell ratio of 10:1.

FIGURE 2.

MHC restriction mapping of polyoma virus-specific CTL clones. CTL clones were assayed for cytolytic activity against uninfected or virus-infected SV40-transformed target cells of DBA/2 (KdDd), C3H.COL (KdDk), and B10.A (KkDd) origin for 4 h at an E:T cell ratio of 10:1.

Close modal

The highly biased usage of Vβ6 and common Dk restriction pattern of the anti-polyoma CTL clones led us to hypothesize that polyoma virus-specific CTL recognize very few epitopes. To determine whether these CTL are directed to an epitope derived from an early region viral protein, a cell line derived from a syngeneic polyoma virus-induced tumor, designated 6215, was assayed for recognition by anti-polyoma CTL. 6215 cells express full-length ST and MT, a carboxyl-truncated LT, and no viral capsid proteins (18). We had previously shown a preferential infiltration of CD8+Vβ6+ T cells into 6215 tumors implanted in virus-immune C57BR/cdJ mice (18). As shown in Table II, representative Vβ6+ CTL clones 11-1 and 13-2 as well as an in vitro tertiary culture from virus-primed C3H/HeSnJ mice composed nearly entirely of CD8+Vβ6+ T cells lysed 6215 cells but not the syngeneic, spontaneous tumor cell line AG104A. The Vβ8.1+ CTL clone, 16-5, also specifically lysed this polyoma tumor cell target. The ability of anti-polyoma CTL clones to lyse 6215 cells shows that CTL induced in polyoma-infected mice and passaged in vitro with infected syngeneic stimulators recognize Ags shared by syngeneic virus-infected and -transformed cells, and that these epitopes are derived from an early region viral protein(s).

Table II.

Anti-polyoma virus CTL specifically lyse a polyoma tumor cell line

CTLb% Specific Lysisa
AG104A6215c
11-1 49 
13-2 16 
16-5 8.1 29 
Bulk cultured 26 
CTLb% Specific Lysisa
AG104A6215c
11-1 49 
13-2 16 
16-5 8.1 29 
Bulk cultured 26 
a

The AG104A and 6215 cell lines were used as targets in a 4-h 51Cr-release assay at an E:T ratio of 10:1.

b

Vβ phenotyping was performed as described in Table I.

c

The 6215 cell line is derived from a polyoma virus-induced tumor in an irradiated C57BR/cdJ mouse (see Materials and Methods).

d

The bulk culture is a >98% CD8+Vβ6+ T cell in vitro tertiary culture of lymph node cells from a polyoma virus-immune C3H/HeSnJ mice.

No consensus motif has been described for peptides bound to Dk to permit prediction of CTL epitopes from polyoma early region proteins presented by this class I MHC molecule. Only one Dk-bound peptide has been reported (28). This nonapeptide, gag88–96, is encoded by a transposed element within the gag gene of an endogenous defective retrovirus selectively expressed in a spontaneous leukemia. To deduce a plausible Dk peptide binding motif, we examined the polymorphic residues of the α1/α2 domains of the Dk heavy chain (29) predicted by MHC class I:peptide structures to occupy peptide binding pockets and to interact with specific peptide side chains (reviewed in 30 . Prediction of Dk peptide anchor residues was gauged against the sequence of gag88–96. Among class I MHC molecules, the collection of polymorphic residues whose side chains point into the B pocket of HLA-B27 (H9, T24, E45, C67, Y99) most closely approximates the corresponding residues in Dk (E9, S24, D45, A67, S99) in charge and/or polarity. Because threonine 24 and glutamic acid 45 in HLA-B27 are particularly important in stabilizing binding to peptides with arginine at position 2 (31), we predict a basic residue at this position (arginine in gag88–96) for peptides binding to Dk. Peptides bound to murine class I molecules invariably have a hydrophobic carboxyl-terminal anchor residue (32), with pocket F residue 116 primarily determining the identity of this peptide anchor (33). Like Kb, which accommodates peptides with a carboxyl-terminal leucine, Dk also has a phenylalanine 116; leucine occupies the carboxyl-terminal position of gag88–96. Finally, a basic anchor residue at position 5 (lysine in gag88–96) is predicted based on molecular modelling of peptides bound to Dd, where aspartic acid 156 in the D pocket, a polymorphism shared by Dd and Dk, interacts with a position 5 arginine in Dd binding peptides (34). Table III lists the sequences of four peptides, three nonamers and an octamer, from polyoma virus early region proteins that satisfy this three-residue anchor motif for Dk binding peptides.

Table III.

Polyoma virus early region amino acid sequences that satisfy a predicted Dk peptide-binding motifa

1 X2 Basic3 X4 X5 Basic6 X7 X8 X9 L
gag88–96b 
T6–14c 
MT102–109  
MT236–244 
MT389–397 
1 X2 Basic3 X4 X5 Basic6 X7 X8 X9 L
gag88–96b 
T6–14c 
MT102–109  
MT236–244 
MT389–397 
a

The predicted motif for peptides binding to H-2Dk molecules is a basic amino acid at positions 2 and 5 and a carboxyl-terminal leucine; X designates nonanchor residues. Peptide residue positions are numbered starting at the amino terminus.

b

gag88–96 is derived from an endogenous murine retrovirus and is the only reported Dk bound peptide (28).

c

T6–14 refers to amino acids 6–14 in the common amino terminal sequence of all three polyoma T proteins. The other three peptides correspond to the indicated amino acid sequences in the MT protein.

Synthetic peptides corresponding to each of these four potential Dk polyoma epitopes were screened for their ability to sensitize syngeneic target cells to lysis by the 10 Dk-restricted polyoma virus-specific CTL clones. As shown in Figure 3, the MT389–397 peptide was specifically recognized by all eight Vβ6+ CTL clones and by the Vβ8.1+ CTL clone 16-5. Clone 15-5 failed to recognize target cells presenting any of these peptides. Of note, none of the MT389–397-reactive clones lysed targets sensitized with gag88–96, which, in addition to its three deduced anchor residues, shares with MT389–397 an arginine at position 1 and a glycine at position 4. Eleven of twelve additional CD8+ polyoma virus-specific CTL clones isolated from neonatally infected C3H/HeSnJ mice also recognized MT389–397; of these, 10 expressed Vβ6 and one expressed Vβ8.1 (data not shown). An MT389–397-nonreactive CTL clone, 26-2 (Fig. 4 C), was restricted by Kk in its recognition of virus-infected cells (data not shown). The MHC and peptide specificities of the MT389–397 peptide-specific CTL clones were confirmed by their capacity to specifically lyse MT389–397-pulsed allogeneic target cells expressing transfected Dk, but not Kk, molecules (data not shown).

FIGURE 3.

Identification of a Dk-bound peptide by polyoma virus-specific CTL clones. Anti-polyoma virus CTL clones described in Table I were assayed for cytolytic activity against 51Cr-labeled L929 targets at an E:T cell ratio of 5:1 for 4 h. Peptides were added to the assay at a final concentration of 1 μM.

FIGURE 3.

Identification of a Dk-bound peptide by polyoma virus-specific CTL clones. Anti-polyoma virus CTL clones described in Table I were assayed for cytolytic activity against 51Cr-labeled L929 targets at an E:T cell ratio of 5:1 for 4 h. Peptides were added to the assay at a final concentration of 1 μM.

Close modal
FIGURE 4.

Cross-reactive recognition of MT236–244 by MT389–397-reactive CTL clones. A, CTL clones 8-1 and 13-2 were assayed for 4 h at an E:T cell ratio of 5:1 for cytolytic activity against 51Cr-labeled L929 target cells in the presence of the indicated concentration of peptide. The flu NP50–57 (SDYEGRLI) peptide binds to H-2Kk (72). B, CTL clones 11-1 and 14-1 were assayed for 4 h at the indicated E:T cell ratios for cytolytic activity against 51Cr-labeled L929 target cells pulsed with MT389–397 (1 μM), MT236–244 (100 μM), or gag88–96 (100 μM) peptides. C, Anti-polyoma CTL clones 15-5 and 26-2 (MT389–397-nonreactive), and clones 11-1 and 13-2 (MT389–397-reactive) were assayed for 7 h at an E:T cell ratio of 10:1 for their capacity to lyse 51Cr-labeled AG104A cells infected with the wild-type virus, A2 strain, or the mutant polyoma virus, PTA1387T, whose MT ends at amino acid 384.

FIGURE 4.

Cross-reactive recognition of MT236–244 by MT389–397-reactive CTL clones. A, CTL clones 8-1 and 13-2 were assayed for 4 h at an E:T cell ratio of 5:1 for cytolytic activity against 51Cr-labeled L929 target cells in the presence of the indicated concentration of peptide. The flu NP50–57 (SDYEGRLI) peptide binds to H-2Kk (72). B, CTL clones 11-1 and 14-1 were assayed for 4 h at the indicated E:T cell ratios for cytolytic activity against 51Cr-labeled L929 target cells pulsed with MT389–397 (1 μM), MT236–244 (100 μM), or gag88–96 (100 μM) peptides. C, Anti-polyoma CTL clones 15-5 and 26-2 (MT389–397-nonreactive), and clones 11-1 and 13-2 (MT389–397-reactive) were assayed for 7 h at an E:T cell ratio of 10:1 for their capacity to lyse 51Cr-labeled AG104A cells infected with the wild-type virus, A2 strain, or the mutant polyoma virus, PTA1387T, whose MT ends at amino acid 384.

Close modal

Titration of the early region peptides revealed cross-reactivity by MT389–397-reactive CTL clones with the MT236–244 peptide at high peptide concentrations. MT389–397 peptide concentrations as low as 10−9 to 10−8 M were able to sensitize L929 (H-2k) targets for CTL lysis, with maximal levels of lysis seen at 0.1 to 1 μM (Fig. 4,A); this concentration range is similar to that reported for CTL recognition of peptide-pulsed L929 targets by Kk-restricted CTL clones (35). At concentrations 3 to 4 logs greater than that required for MT389–397 target cell sensitization (100 μM), the MT236–244 peptide rendered target cells sensitive to CTL-mediated lysis (Fig. 4,A). No target cell lysis was seen with the four other peptides tested in this concentration range. The Dk-restricted presentation of MT236–244 was confirmed using peptide-pulsed P1.HTR.Dk and P1.HTR.Kk cells as CTL targets (data not shown). Of the other MT389–397-reactive CTL clones, only clone 14-1 lysed target cells pulsed with MT389–397 but not MT236–244 (Fig. 4,B), indicating that recognition of MT389–397 is sufficient for killing polyoma-infected cells. To determine whether cross-reactive recognition of MT236–244 contributes to the overall lysis of infected cells by MT389–397-reactive CTL, CTL clones were tested for their ability to recognize target cells infected with the mutant polyoma virus PTA1387T, whose MT is truncated at amino acid 384. As shown in Figure 4 C, the MT389–397-nonreactive CTL clones 15-5 and 26-2 lysed cells infected with either wild-type virus or PTA1387T virus, while the MT389–397-reactive CTL clones 11-1 and 13-2 efficiently lysed only wild-type virus-infected targets. The low level lysis of PTA1387T-infected targets by clone 11-1, but not by clone 13-2, may indicate a minor contribution of the MT236–244 peptide to recognition of polyoma-infected cells by some MT389–397-reactive CTL. Taken together, these results demonstrate that the MT389–397 epitope is a naturally processed peptide in polyoma-infected cells, and that CTL recognition of MT389–397 is necessary and sufficient for lysis of infected cells.

Because the MT389–397 peptide has repetitive amino acids at both its amino and carboxyl termini, we asked whether truncations at either end affect CTL recognition. Peptides lacking the arginine at position 1, lacking either one or both carboxyl-terminal leucines, or lacking the first and last amino acids were assayed over a 5-log concentration range for their efficiency in sensitizing syngeneic cells for lysis by MT389–397-specific CTL clones. None of these peptides was recognized as well as the full-length peptide (Fig. 5), indicating that the truncated peptides do not readily shift in register in the Dk peptide binding groove. MT388–397 (leucine at position 388) and MT389–398 (valine at position 398) decamer peptides sensitized target cells over similar concentration ranges as the MT389–397 nonapeptide (data not shown).

FIGURE 5.

Determination of the minimum epitope optimally recognized by MT389–397-reactive CTL. The ability of MT389–397-reactive CTL clones 11-1 and 13-2 to recognize titrated concentrations of peptides lacking amino- and/or carboxyl-terminal residues of the MT389–397 sequence were tested in a 4-h 51Cr release assay against L929 target cells at an E:T cell ratio of 5:1.

FIGURE 5.

Determination of the minimum epitope optimally recognized by MT389–397-reactive CTL. The ability of MT389–397-reactive CTL clones 11-1 and 13-2 to recognize titrated concentrations of peptides lacking amino- and/or carboxyl-terminal residues of the MT389–397 sequence were tested in a 4-h 51Cr release assay against L929 target cells at an E:T cell ratio of 5:1.

Close modal

While reduced affinity of the truncated peptides for Dk could account for their weaker recognition compared with the MT389–397 peptide, each CTL clone exhibited a distinct profile of concentration ranges over which the truncated peptides triggered target cell lysis. This suggests that clonotypic differences in TCR interactions with Dk:MT389–397 complexes are responsible for these clone-specific reactivity patterns.

Since MT389–397 is recognized by the majority of the anti-polyoma CTL clones, we asked whether this epitope is immunodominant in polyoma tumor-resistant H-2k mice. Short term cultures of polyoma virus-immune T cells established from neonatally infected C3H/HeSnJ mice were evaluated for cytolytic activity against infected or MT389–397 peptide-pulsed syngeneic target cells. As shown in Figure 6, virus-immune T cells from four individual mice lysed MT389–397-pulsed and virus-infected targets, but not gag88–96 peptide-coated and uninfected targets. The level of killing of MT389–397-coated targets tended to exceed that of infected targets, suggesting that this epitope may be inefficiently processed and/or presented by cells productively infected with polyoma, as described in other anti-viral CTL systems (36, 37, 38).

In our previous report (18), the resistant H-2k mouse strain C57BR/cdJ was used to assess the anti-polyoma CTL response. C57BR/cdJ is a Vβa genotype mouse strain in which approximately 40% of the germ-line Vβ gene segments are deleted (39). We wanted to determine whether the highly biased usage of Vβ6 by anti-polyoma CTL in C57BR/cdJ mice also extended to the CTL response elicited by polyoma in C3H/HeSnJ mice, a polyoma tumor-resistant H-2k strain possessing a complete repository of Vβ genes (i.e., Vβb genotype). Single-color flow cytometric analysis for Vβ6, CD8, and CD4 was therefore conducted on each of the C3H/HeSnJ anti-viral CTL cultures shown in Figure 6. As expected based on the impact of Mtv-7SAG deletion of Vβ6-expressing thymocytes on polyoma tumor susceptibility (18), T cells in each of these in vitro tertiary cultures predominantly expressed Vβ6 and CD8 (Fig. 6). Moreover, in contrast to T cell Ag unresponsiveness incurred by the Vβa-truncated TCR repertoire in other systems (40, 41), C57BR/cdJ and C3H/HeSnJ mice both mounted an immunodominant CTL response to the MT389–397 epitope.

Limiting dilution analysis was performed to quantitate the numbers of CTL precursors (pCTL) in neonatally infected C3H/HeSnJ that were directed toward polyoma virus-infected and MT389–397 peptide-pulsed cells. pCTL analysis was conducted separately on two virus-primed mice. As shown in Figure 7 A, the precursor frequency for CTL recognizing polyoma-infected syngeneic targets is approximately 1/40,000 spleen cells, and that for pCTL recognizing MT389–397-pulsed targets is approximately 1/20,000 spleen cells. This virus-specific pCTL frequency is similar to that reported in other virus-mouse systems (42, 43). Given that each well in the limiting dilution assay containing cytotoxic activity against MT389–397-pulsed targets also exhibited specific killing, although generally at lower levels, of virus-infected targets (data not shown), the higher precursor frequency for MT389–397-reactive CTL than for anti-viral CTL probably reflects higher sensitivity in detecting specific CTL recognition using peptide-pulsed targets.

FIGURE 7.

Precursor frequency analysis for anti-polyoma CTL. Spleen cells from neonatally inoculated mice were titrated in replicates of 16 in 96-well microtiter plates and stimulated twice with polyoma-infected syngeneic spleen cells in the presence of conditioned medium from Con A-stimulated rat spleen cells. Individual wells were split and assayed for cytolytic activity against uninfected, virus-infected, and MT389–397 peptide-pulsed (10 μM) 51Cr-labeled AG104A target cells. Positive wells were defined as those wells for which 51Cr release values exceeded those of uninfected targets by >3 SD. A, Virus-specific and MT389–397 peptide-specific pCTL frequency analysis for two C3H/HeSnJ mice. B, MT389–397 peptide-specific pCTL frequency analysis for two C3H/HeSnJ and two C3H/BiDa mice.

FIGURE 7.

Precursor frequency analysis for anti-polyoma CTL. Spleen cells from neonatally inoculated mice were titrated in replicates of 16 in 96-well microtiter plates and stimulated twice with polyoma-infected syngeneic spleen cells in the presence of conditioned medium from Con A-stimulated rat spleen cells. Individual wells were split and assayed for cytolytic activity against uninfected, virus-infected, and MT389–397 peptide-pulsed (10 μM) 51Cr-labeled AG104A target cells. Positive wells were defined as those wells for which 51Cr release values exceeded those of uninfected targets by >3 SD. A, Virus-specific and MT389–397 peptide-specific pCTL frequency analysis for two C3H/HeSnJ mice. B, MT389–397 peptide-specific pCTL frequency analysis for two C3H/HeSnJ and two C3H/BiDa mice.

Close modal

Since the MT389–397 peptide is the immunodominant epitope for anti-polyoma CTL in resistant (i.e., Mtv-7) mice, we asked whether the precursor frequencies for MT389–397-reactive CTL differed between resistant and susceptible (i.e., Mtv-7+) mice. Spleen cells from neonatally immunized C3H/HeSnJ (resistant) and C3H/BiDa (susceptible) mice were plated in limiting dilution cocultures with virus-infected, irradiated syngeneic stimulators. After 1 wk, each well was restimulated with virus-infected, irradiated syngeneic spleen cells and assayed 7 days later for cytolytic activity against virus-infected and MT389–397-pulsed AG104A target cells. While virus-primed C3H/HeSnJ mice exhibited a CTL precursor frequency of 1/40,000, no specific lysis of virus-infected targets was detected by splenic T cells from infected C3H/BiDa mice even at dilutions as low as 5 × 105 spleen cells/well (data not shown). Given the apparently higher sensitivity for detecting anti-polyoma CTL with MT389–397-pulsed targets than virus-infected targets, we sought to determine whether MT389–397-reactive CTL were elicited by polyoma infection in C3H/BiDa mice. As shown in Figure 7 B, CTL that recognize MT389–397-pulsed targets were indeed generated in these tumor-susceptible mice, albeit at a considerably lower precursor frequency (1/150,000 to 1/460,000) than the resistant mice.

The presence of MT389–397-reactive CTL in the susceptible mice prompted us to attempt to isolate polyoma-specific CTL from neonatally infected C3H/BiDa mice. Multiple rounds of restimulation of T cells from the draining lymph nodes of polyoma-infected C3H/BiDa mice with virus-infected syngeneic stimulators succeeded in generating αβ+CD8+ T cell lines that specifically lysed both virus-infected and MT389–397-coated AG104A targets (Fig. 8). As expected given their derivation from a Mtv-7 provirus-positive mouse strain, these CTL lines do not express Vβ6 or Vβ8.1 TCRs. Analysis of the TCR sequences of MT389–397-reactive CTL from C3H/BiDa mice is in progress. Note that line 2 lysed MT389–397-pulsed, but not infected, target cells, again suggesting inefficient processing of this epitope by virus-infected cells. Thus, despite marked differences in their polyoma-specific pCTL frequency, the anti-polyoma CTL response in the tumor-susceptible and -resistant H-2k mice is directed to the same immunodominant epitope in MT.

FIGURE 8.

Virus- and MT389–397 peptide-specific lysis by virus-immune T cell lines from C3H/BiDa mice. T cell lines were established from the draining lymph nodes of three neonatally inoculated C3H/BiDa mice by multiple weekly restimulations with virus-infected, irradiated syngeneic spleen cells. These lines were assayed for cytolytic activity against 51Cr-labeled uninfected, polyoma-infected, MT389–397 peptide-pulsed (10 μM), and gag88–96 peptide-pulsed (10 μM) AG104A target cells at the indicated E:T cell ratios for 4 h.

FIGURE 8.

Virus- and MT389–397 peptide-specific lysis by virus-immune T cell lines from C3H/BiDa mice. T cell lines were established from the draining lymph nodes of three neonatally inoculated C3H/BiDa mice by multiple weekly restimulations with virus-infected, irradiated syngeneic spleen cells. These lines were assayed for cytolytic activity against 51Cr-labeled uninfected, polyoma-infected, MT389–397 peptide-pulsed (10 μM), and gag88–96 peptide-pulsed (10 μM) AG104A target cells at the indicated E:T cell ratios for 4 h.

Close modal

In this report we provide the first identification of an epitope for polyoma virus-specific CTL. Derived from the viral oncoprotein, MT, the MT389–397 nonapeptide was predicted to bind to H-2Dk based on a three-residue MHC anchor motif deduced from crystallographic and molecular modelling analyses of class I MHC:peptide complexes. Comparison of this CTL epitope with the only reported Dk-associated CTL epitope (28) lends support to the prediction that peptides with basic amino acids at positions 2 and 5 and a carboxyl-terminal leucine interact with pockets B, D, and F, respectively, in the peptide binding groove of Dk. Additional support for this Dk peptide consensus motif comes from the finding that two other polyoma early region peptides (T6–14 and MT102–109), selected on the basis of this putative motif, compete with MT389–397 in sensitizing cells for lysis by anti-polyoma CTL clones, while a Kk-binding peptide (flu NP50–57) did not compete (data not shown). A third peptide from the polyoma early region sequence that satisfies the Dk binding motif, MT236–244, also associates with Dk, as evidenced by its weak agonist activity for MT389–397-reactive CTL clones (Fig. 4 A). Studies are in progress to directly demonstrate that peptide positions 2, 5, and 9 are Dk anchor residues and to determine their relative contributions to the overall binding of peptides to Dk.

A central question is whether CTL mediate protection from polyoma virus tumorigenesis by eradicating virus-infected cells, virus-transformed cells, or both. Adoptive transfer of virus-immune spleen cells into infected nude mice has been shown to eliminate replicating viral DNA, which, in turn, is associated with the absence of polyoma virus-induced tumors (44). That polyoma-transformed cells may also serve as in vivo targets for virus-immune T cells is illustrated by the inability of CD4+ and/or CD8+ T cell-depleted, virus-infected mice to reject a polyoma tumor cell challenge (45). The MT389–397 sequence spans the region of MT immediately amino-terminal to and extending into the hydrophobic membrane-anchoring region of this type II-integral membrane, 421-amino acid protein. Because localization of MT to membranes is essential for its capacity to transform cells and induce tumors (11), and full-length MT is constitutively expressed in polyoma-induced tumors of all histotypes (46) (A. Lukacher, unpublished observations), MT389–397 is an ideal target epitope for a host CTL response against polyoma tumors. Furthermore, as an early region viral protein, MT is expressed by cells that are nonpermissive for viral DNA replication (24) in addition to its expression by productively infected cells. CTL surveillance for Dk:MT389–397 complexes, then, should control polyoma tumors not only by destroying virus-transformed cells but also by eliminating permissively and nonpermissively infected cells, both of which have the potential to undergo neoplastic transformation in vivo.

The marked immunodominance of the MT389–397 epitope in the anti-polyoma CTL response is also notable given that this CTL reactivity is directed toward a nonstructural target protein in a system permissive for polyoma replication. Class I MHC-restricted CTL epitopes from proteins of oncogenic viruses are typically identified using CTL isolated from virus-inoculated nonnatural hosts (i.e., nonpermissive for viral replication) and/or by in vitro culture with virus-transformed cells (47, 48, 49). Under these experimental conditions, potential CTL epitopes from viral proteins only expressed in productively infected cells are excluded. In the present study, polyoma-specific CTLs are isolated from virus-infected mice by in vitro restimulation with virus-infected spleen cells, which support productive polyoma virus replication (D. Drake and A. Lukacher, unpublished observations).

While high MHC binding affinity is likely to be a prerequisite for the immunodominance of MT389–397 as for other class I MHC epitopes (50), virus-host cell interactions may also guide selection of MT389–397 as the immunodominant CTL epitope. The MT389–397 epitope would be expected to be presented by productively infected, nonpermissive cells and transformed cells. Such widespread Ag presentation as well as persistent presentation by nonlytically infected cells may preferentially drive expansion of CTL directed to MT389–397 over, for example, CTL recognizing epitopes from viral capsid proteins, which are expressed in lytically infected cells and variably expressed in polyoma tumors (46). Also, since the anti-polyoma CTL described here are derived from mice 2 wk postinfection, the dominant anti-polyoma virus CTL response to MT389–397 may reflect a narrowing of the repertoire of anti-viral CTL specificities elicited during acute infection to those found in the memory CTL population (43, 51).

A number of scenarios can be invoked to account for the failure of the weak anti-polyoma CTL response to control polyoma virus tumorigenesis. A low precursor frequency anti-polyoma CTL response may provide sufficient time for transformed cells to acquire mutations that enable them to evade CTL surveillance by any of a number of mechanisms (52). Although rare immune escape polyoma tumor variants emerge in polyoma tumor-resistant mice (18), in vivo selection of such tumor cell variants by anti-polyoma CTL is unlikely to serve as a general mechanism for neoplasia in susceptible mice given their high incidence of tumors, multiplicity of tumors per animal, and wide diversity of tumor types (8, 15). We, instead, favor the hypothesis that Mtv-7SAG-mediated deletion of Vβ6+ CTL in neonatally infected mice, which include the predominant polyoma-specific CTL response in Mtv-7 (resistant) H-2k mice, leads to a persistently high viral Ag state that interferes with the expansion and/or the function of the residual non-Vβ6+ anti-polyoma CTL. In tumor-susceptible mice, high levels of unintegrated, often defective, replicating viral DNA are present in tissues destined to develop tumors as well as in the tumors themselves, and only low amounts of infectious virus can be recovered from tumors (46, 53, 54). In contrast, little replicating viral DNA is detectable in virus-infected immunocompetent adult mice (53), which are uniformly resistant to polyoma virus tumorigenesis (15). In this regard, it is interesting to note that polyoma-infected adult Mtv-7+H-2k mice mount a vigorous anti-viral CTL response directed primarily to the MT389–397 epitope (D. Drake and A. Lukacher, unpublished observation). Experiments are underway to compare the levels of viral DNA and viral protein expression in tumor-prone tissues in neonatally infected susceptible mice to those sites in neonatally infected resistant mice. Selective loss of anti-viral CTL in the setting of high systemic viral loads is seen in mice neonatally infected with lymphocytic choriomeningitis virus (55, 56), and high doses of antigenic peptides can induce apoptosis of CD8+ CTL in vitro (57, 58). In light of the regulatory effect of Ag dose on Th1 vs Th2 differentiation (59), it is also conceivable that a persistently elevated Ag state could indirectly limit differentiation and/or clonal expansion of anti-polyoma CD8+ CTL by promoting Th2 over Th1 differentiation of polyoma-specific CD4+ T cells. Studies are in progress to investigate these possibilities.

A striking feature of the anti-polyoma CTL response in resistant H-2k mice is its strongly biased usage of Vβ6 (Fig. 6) (18). The recently solved structures of two TCRs with class I MHC:peptide complexes show the Vβ CDR1 domain positioned over the peptide’s carboxyl end (60, 61). Because all Vβ6+ polyoma virus-specific CTL clones examined to date (n = 20) recognize the Dk:MT389–397 ligand, it is conceivable that the Vβ6 CDR1 may directly contact common residues in this peptide. The distinct profiles of recognition of truncated MT389–397 peptides by Vβ6+ CTL clones (Fig. 5) further suggest a level of diversity in their CDR3s to accommodate interactions with different residues in the peptide. This conclusion is supported by preliminary TCR sequence analysis of the anti-polyoma Vβ6+ CTL clones, indicating considerable length and sequence heterogeneity in their CDR3β regions and diverse Vα gene usage (C. Wilson and A. Lukacher, manuscript in preparation). Diverse TCR repertoires to single immunodominant class I MHC-restricted CTL epitopes have been described in other viral systems (62, 63, 64). Although strongly preferred, the Vβ6 domain is not essential for TCR interaction with Dk:MT389–397 complexes, as rare MT389–397-reactive Vβ8.1+ CTL clones have been isolated from resistant mice, and Mtv-7+ susceptible mice (which lack Vβ6- and Vβ8.1-bearing T cells; data not shown and 65 also recognize this MHC:peptide complex. Whether MT389–397-reactive thymocytes expressing the Vβ6 domain are preferentially positively selected by polyoma-infected/-transformed thymic epithelial cells (8) or an endogenous Dk peptide ligand, or naive Vβ6+ MT389–397-specific T cells are preferentially activated upon contact with virus-infected cells remains to be determined.

The in vivo relevance of MT236–244 cross-reactivity by MT389–397-reactive CTL is unclear. The expression of two functional TCRs is unlikely given the high frequency of MT236–244 cross-reactivity among the MT389–397-reactive CTL clones (66, 67). In light of recent studies showing that a given TCR can interact with several unrelated peptides in the context of the same MHC molecule (68, 69, 70), we favor the concept that MT389–397-reactive CTL express a single TCR that can recognize the MT236–244 peptide. In this regard, it is interesting to note that MT236–244 shares only the three putative Dk anchor residues with MT389–397, with the corresponding residues at the other positions having nonconservative side chains (Table III). Studies are in progress to distinguish between single and dual TCR expression by these cross-reactive CTL. In vivo CTL recognition of the MT236–244 peptide is also suspect given the high concentrations of MT236–244 peptide required to sensitize target cells for lysis by MT389–397-specific CTL clones. In addition, attempts to expand T cells from virus-immune mice with MT236–244-pulsed splenic stimulators have been unsuccessful (data not shown). This dissociation between target cell sensitization and proliferation is in line with a report that higher ligand concentrations are required to drive CD8+ CTL proliferation than to trigger target cell lysis (71). In contrast, MT389–397-pulsed splenic stimulators efficiently promoted the outgrowth of T cells from polyoma-immune mice; these T cells specifically killed not only virus-infected and MT389–397-pulsed target cells, but, notably, MT236–244-pulsed targets as well (data not shown). We were also unable to detect TCR antagonism of target cell lysis by MT236–244 or by the nonagonist peptides T6–14 and MT102–109 of MT389–397-specific CTL clones (data not shown). Another possibility is that Dk binding peptides from polyoma proteins compete intracellularly with MT389–397 for binding to Dk; the consequent reduction in cell surface MT389–397 epitope density could contribute to the apparent weaker CTL recognition of virus-infected than MT389–397-pulsed targets.

In conclusion, we have identified a Dk-bound nonamer peptide derived from the polyoma virus oncoprotein, MT, as the immunodominant anti-polyoma CTL epitope in H-2k mice. To the best of our knowledge, this is the first identification of an epitope for CTL specific for polyoma virus-infected/-transformed cells. Importantly, CTL directed to the MT389–397 epitope were detected in mice susceptible to developing polyoma virus-induced tumors, but their precursor frequency was at least 10-fold lower than that of syngeneic resistant mice. This raises interesting questions concerning the inability of these CTL to control outgrowth of polyoma tumors. Rare polyoma virus-specific CTL clones that do not recognize MT389–397 have also been isolated, and studies to map their epitopes are in progress. Understanding the contributions of CTL directed to the dominant MT389–397 epitope and to subdominant viral epitopes in immunity to polyoma tumorigenesis has general importance in the development of vaccination strategies against virus-induced tumors.

We thank Hans Schreiber, Linda Gooding, and Aline Van Pel for providing cell lines; David Margulies for valuable discussions in predicting a peptide binding motif for Dk; Thomas Benjamin for providing A2 and PTA1387T genomic DNA; and Peter Jensen, Brian Evavold and Charles Parkos for critically reviewing this manuscript. The excellent technical assistance of Annette Hadley is appreciated.

1

This work was supported by National Institutes of Health Grant CA71971 and in part by grants from the University Research Committee of Emory University, the American Cancer Society, Inc., and the Winship Cancer Center of Emory University.

3

Abbreviations used in this paper: HPV, human papillomavirus; ST, small T protein; MT, middle T protein; LT, large T protein; SAG, superantigen; Mtv, mouse mammary tumor provirus; moi, multiplicity of infection; BMK, primary baby mouse kidney cell cultures; IMDM, Iscove’s modified Dulbecco’s medium; pCTL, CTL precursor.

1
Greenberg, P. D..
1991
. Adoptive T cell therapy of tumors: mechanisms operative in the recognition and elimination of tumors.
Adv. Immunol.
49
:
281
2
Craig, F. E., M. L. Gulley, P. M. Banks.
1993
. Post-transplantation lymphoproliferative disorders.
Am. J. Clin. Pathol.
99
:
265
3
Rooney, C. M., C. A. Smith, C. Y. C. Ng, S. Loftin, C. Li, R. A. Krance, M. K. Brenner, H. E. Heslop.
1995
. Use of gene-modified virus-specific T lymphocytes to control Epstein-Barr virus-related lymphoproliferation.
Lancet
345
:
9
4
Kjaer, S. K., A. J. C. Van Den Brule, J. E. Bock, P. A. Poll, G. Engholm, M. E. Sherman, J. M. M. Walboomer, C. J. L. M. Meijer.
1996
. Human papillomavirus: the most significant risk determinant of cervical intraepithelial neoplasia.
Int. J. Cancer
65
:
601
5
Petry, K. U., D. Scheffel, U. Bode, T. Gabrysiak, H. Kochel, E. Kupsch, M. Glaubitz, S. Niesert, H. Kuhnle, I. Schedel.
1994
. Cellular immunodeficiency enhances the progression of human papillomavirus-associated cervical lesions.
Int. J. Cancer
57
:
836
6
Keating, P. J., F. V. Cromme, M. Duggan-Keen, P. J. F. Snijders, J. M. M. Walboomers, R. D. Hunter, P. A. Dyer, P. L. Stern.
1995
. Frequency of down-regulation of individual HLA-A and -B alleles in cervical carcinomas in relation to TAP-1 expression.
Br. J. Cancer
72
:
405
7
Ellis, J. R. M., P. J. Keating, B. J. E. F. Hounsell, D. V. Renouf, M. Rowe, D. Hopkins, M. F. Dugan-Keen, P. L. Stern.
1995
. The association of an HPV 16 oncogene variant with HLA-B7 has implications for vaccine design in cervical cancer.
Nat. Med.
1
:
565
8
Dawe, C. J., R. Freund, G. Mandel, K. Ballmer-Hofer, D. A. Talmage, T. L. Benjamin.
1987
. Variations in polyoma virus genotype in relation to tumor induction in mice: characterization of wild type strains with widely differing tumor profiles.
Am. J. Pathol.
127
:
243
9
Raptis, L., H. Lamfrom, T. L. Benjamin.
1985
. Regulation of cellular phenotype and expression of polyomavirus middle T antigen in rat fibroblasts.
Mol. Cell. Biol.
5
:
2476
10
Garcea, R. L., D. A. Talmage, A. Harmatz, R. Freund, T. L. Benjamin.
1989
. Separation of host range from transformation functions of the hr-t gene of polyomavirus.
Virology
168
:
312
11
Freund, R., A. Sotnikov, R. T. Bronson, T. L. Benjamin.
1992
. Polyoma virus middle T is essential for virus replication and persistence as well as for tumor induction in mice.
Virology
191
:
716
12
Law, L. W., R. C. Ting, E. Leckband.
1967
. Prevention of virus-induced neoplasms in mice through passive transfer of immunity by sensitized syngeneic lymphoid cells.
Proc. Natl. Acad. Sci. USA
57
:
1068
13
Allison, A. C., J. N. Monga, V. Hammond.
1974
. Increased susceptibility to virus oncogenesis of congenitally thymus-deprived nude mice.
Nature
252
:
746
14
Berke, Z., S. Palmer, T. Bergman, D. Wester, J. Svedmyr, S. Linder, H. Jornvall, T. Dalianis.
1996
. A short peptide eluted from the H-2Kb molecule of a polyomavirus-positive tumor corresponds to polyomavirus large T antigen peptide at amino acids 578 to 585 and induces polyomavirus-specific immunity.
J. Virol.
70
:
3093
15
Gross, L. G..
1983
.
Oncogenic Viruses
737
Pergamon Press, Oxford.
16
Freund, R., T. Dubensky, R. Bronson, A. Sotkinov, J. Carroll, T. Benjamin.
1992
. Polyoma tumorigenesis in mice: evidence for dominant resistance and dominant susceptibility genes of the host.
Virology
191
:
724
17
Acha-Orbea, H., H. R. MacDonald.
1995
. Superantigens of mouse mammary tumor virus.
Annu. Rev. Immunol.
13
:
459
18
Lukacher, A. E., Y. Ma, J. P. Carroll, S. R. Abromson-Leeman, J. C. Laning, M. E. Dorf, T. L. Benjamin.
1995
. Susceptibility to tumors induced by polyoma virus is conferred by an endogenous mouse mammary tumor virus superantigen.
J. Exp. Med.
181
:
1683
19
Green, M. I., L. L. Perry, E. Kinney-Thomas, T. L. Benjamin.
1982
. Specific thymus-derived (T) cell recognition of papova virus-transformed cells.
J. Immunol.
128
:
732
20
Ward, P. L., H. Koeppen, T. Hurteau, H. Schreiber.
1989
. Tumor antigens defined by cloned immunological probes are highly polymorphic and are not detected on autologous normal cells.
J. Exp. Med.
170
:
217
21
O’Connell, K. A., L. R. Gooding.
1984
. Cloned cytotoxic T lymphocytes recognize cells expressing discrete fragments of the SV40 tumor antigen.
J. Immunol.
132
:
953
22
Braciale, T. J., M. E. Andrew, V. L. Braciale.
1981
. Heterogeneity and specificity of cloned lines of influenza-virus-specific cytotoxic T lymphocytes.
J. Exp. Med.
153
:
910
23
Quintans, J., I. Lefkovits.
1973
. Precursor cells specific to sheep red cells in nude mice. Estimation of frequency in the microculture system.
Eur. J. Immunol.
3
:
392
24
Shadan, F. F., L. M. Cowsert, L. P. Villarreal.
1994
. n-Butyrate, a cell cycle blocker, inhibits the replication of polyomaviruses and papillomaviruses but not that of adenoviruses and herpesviruses.
J. Virol.
68
:
4785
25
Kuerbitz, S. J., B. S. Plunkett, W. V. Walsh, M. B. Kastan.
1992
. Wild-type p53 is a cell cycle checkpoint determinant following irradiation.
Proc. Natl. Acad. Sci. USA
89
:
7491
26
Casanova, J.-L., P. Romero, C. Widmann, P. Kourilsky, J. L. Maryanski.
1991
. T cell receptor genes in a series of class I major histocompatibility complex-restricted cytotoxic T lymphocyte clones specific for a Plasmodiumburghei nonapeptide: implications for T cell allelic exclusion and antigenic-specific repertoire.
J. Exp. Med.
174
:
1371
27
Hood, L., M. Steinmetz, B. Malissen.
1983
. Genes of the major histocompatibility complex of the mouse.
Annu. Rev. Immunol.
1
:
529
28
de Bergeyck, V., E. De Plaen, P. Chomez, T. Boon, A. Van Pel.
1994
. An intracisternal A-particle sequence codes for an antigen recognized by syngeneic cytolytic T lymphocytes on a mouse spontaneous leukemia.
Eur. J. Immunol.
24
:
2203
29
Watts, S., J. M. Vogel, W. D. Harriman, T. Itoh, H. J. Stauss, R. S. Goodenow.
1987
. DNA sequence analysis of the C3H H-2Kk and H-2Dk loci: evolutionary relationships to H-2 genes from four other mouse strains.
J. Immunol.
139
:
3878
30
Young, A. C. M., S. G. Nathenson, J. C. Sacchettini.
1995
. Structural studies of class I major histocompatibility complex proteins: insights into antigen presentation.
FASEB J.
9
:
26
31
Madden, D. R., J. C. Gorga, J. L. Strominger, D. C. Wiley.
1992
. The three-dimensional structure of HLA-B27 at 2.1 A resolution suggests a general mechanism for tight peptide binding to MHC.
Cell
70
:
1035
32
Falk, K., O. Rotzchke, K. Deres, J. Metzger, G. Jung, H.-G. Rammensee.
1991
. Allele-specific motifs revealed by sequencing of self-peptides eluted from MHC molecules.
Nature
351
:
290
33
Young, A. C. M., W. Zhang, J. C. Sacchettini, S. G. Nathenson.
1994
. The three dimensional structure of H-2Db at 2.4A resolution: implications for antigen determinant selection.
Cell
76
:
39
34
Corr, M., L. F. Boyd, E. A. Padlan, D. H. Margulies.
1993
. H-2Dd exploits a four residue peptide binding motif.
J. Exp. Med.
178
:
1877
35
Cossins, J., K. G. Gould, M. Smith, P. Driscoll, G. G. Brownlee.
1993
. Precise prediction of a Kk-restricted cytotoxic T cell epitope in the NS1 protein of influenza virus using an MHC allele-specific motif.
Virology
193
:
289
36
van der Most, R. G., A. Sette, C. Oseroff, J. Alexander, K. Marali-Krishna, L. L. Lau, S. Southjwood, J. Sidney, R. W. Chesnut, M. Matloubian, R. Ahmed.
1996
. Analysis of cytotoxic T cell responses to dominant and subdominant epitopes during acute and chronic lymphocytic choriomeningitis virus infection.
J. Immunol.
157
:
5543
37
Sijts, A. J. A. M., A. Neisig, J. Neefjes, E. G. Pamer.
1996
. Two Listeria monocytogenes CTL epitopes are processed from the same antigen with different efficiencies.
J. Immunol.
156
:
685
38
Selin, L. K., R. M. Welsh.
1997
. Cytolytically active memory CTL present in lymphocytic choriomeningitis virus-immune mice after clearance of virus infection.
J. Immunol.
158
:
5366
39
Behlke, M. A., H. S. Chou, K. Huppi, D. Y. Loh.
1986
. Murine T-cell receptor mutants with deletions of β-chain variable region genes.
Proc. Natl. Acad. Sci. USA
83
:
767
40
Nanda, N. K., R. Apple, E. E. Sercarz.
1991
. Limitations in plasticity of the T cell receptor repertoire.
Proc. Natl. Acad. Sci. USA
88
:
9503
41
Nanda, N. K., E. Sercarz.
1996
. A truncated T cell receptor repertoire reveals underlying immunogenicity of an antigenic determinant.
J. Exp. Med.
184
:
1037
42
Cao, W., B. A. Myers-Powell, T. J. Braciale.
1996
. The weak CD8+ CTL response to an influenza hemagglutinin epitope reflects limited T cell availability.
J. Immunol.
157
:
505
43
Cole, G. A., T. L. Hogg, M. A. Coppola, D. L. Woodland.
1997
. Efficient priming of CD8+ memory T cells specific for a subdominant epitope following Sendai virus infection.
J. Immunol.
158
:
4301
44
Wirth, J. J., M. M. Fluck.
1991
. Immunological elimination of infected cells as the candidate mechanism for tumor protection in polyomavirus-infected mice.
J. Virol.
65
:
6985
45
Ljunggren, G., H.-G. Ljunggren, T. Dalianis.
1994
. T cell subsets involved in immunity against polyoma virus-induced tumors.
Virology
198
:
714
46
Talmage, D. A., R. Freund, T. Dubensky, M. Salcedo, P. Gariglio, L. M. Rangel, C. J. Dawe, T. L. Benjamin.
1992
. Heterogeneity in state and expression of viral DNA in polyoma virus-induced tumors of the mouse.
Virology
187
:
734
47
Tanaka, Y., M. J. Tevethia, D. Kalderon, A. E. Smith, S. S. Tevethia.
1988
. Clustering of antigenic sites recognized by cytotoxic T lymphocyte clones in the amino terminal half of SV40 T antigen.
Virology
162
:
427
48
Kast, W. M., R. Offringa, P. J. Peters, A. C. Voordouw, R. H. Meloen, A. J. van der Erb, C. J. M. Melief.
1989
. Eradication of adenovirus E1-induced tumors by E1A-specific cytotoxic T lymphocytes.
Cell
59
:
603
49
Murray, R. J., M. G. Kurilla, J. M. Brooks, W. A. Thomas, M. Rowe, E. Kieff, A. B. Rickinson.
1992
. Identification of target antigens for the human cytotoxic T cell response to Epstein-Barr virus (EBV): implications for the immune control of EBV-positive malignancies.
J. Exp. Med.
176
:
157
50
Sette, A., A. Vitiello, B. Reherman, P. Fowler, R. Nayersina, W. M. Kast, C. J. M. Melief, C. Oseroff, L. Yuan, J. Ruppert, J. Sidney, M.-F. Del Guercio, S. Southwood, R. T. Kubo, R. W. Chesnut, H. M. Grey, F. V. Chisari.
1994
. The relationship between class I binding affinity and immunogenicity of potential cytotoxic T cell epitopes.
J. Immunol.
153
:
5586
51
Bachmann, M. F., D. E. Speiser, P. S. Ohashi.
1997
. Functional maturation of an antiviral cytotoxic T-cell response.
J. Virol.
71
:
5764
52
Toes, R. E., M. C. Feltkamp, M. E. Ressing, M. P. Vierboom, R. J. Blom, R. M. Brandt, M. Hartman, R. Offringa, C. J. Melief, W. M. Kast.
1995
. Cellular immunity against DNA tumour viruses: possibilities for peptide-based vaccines and immune escape.
Biochem. Soc. Trans.
23
:
692
53
Wirth, J. L., A. Amalfitano, R. Gross, M. B. A. Oldstone, M. M. Fluck.
1992
. Organ- and age-specific replication of polyomavirus in mice.
J. Virol.
66
:
3278
54
Wirth, J. J., L. G. Martin, M. M. Fluck.
1997
. Oncogenesis of mammary glands, skin, and bones by polyomavirus correlates with viral persistence and prolonged genome replication potential.
J. Virol.
71
:
1072
55
Ahmed, R., A. Salmi, L. D. Butler, J. Chiller, M. B. A. Oldstone.
1984
. Selection of genetic variants of lymphocytic choriomeningitis virus in spleens of persistently infected mice: role of suppression of cytotoxic T lymphocyte response and viral persistence.
J. Exp. Med.
60
:
521
56
Moskophidis, D., F. Lechner, H. P. Pircher, R. M. Zinkernagel.
1993
. Virus persistence in acutely infected immunocompetent mice by exhaustion of antiviral cytotoxic effector T cells.
Nature
362
:
758
57
Moss, D. J., S. R. Burrows, G. D. Baxter, M. F. Lavin.
1991
. T cell-T cell killing is induced by specific epitopes: evidence for an apoptotic mechanism.
J. Exp. Med.
173
:
681
58
Alexander-Miller, M. A., G. R. Leggatt, A. Sarin, J. A. Berzofsky.
1996
. Role of antigen, CD8, and cytotoxic T lymphocyte (CTL) avidity in high dose antigen induction of apoptosis of effector CTL.
J. Exp. Med.
184
:
485
59
Hosken, N. A., K. Shibuya, A. W. Heath, K. M. Murphy, A. O’Garra.
1995
. The effect of antigen dose on CD4+ T helper cell phenotype development in a T cell receptor-αβ-transgenic model.
J. Exp. Med.
182
:
1579
60
Garboczi, D. N., P. Ghosh, U. Utz, Q. R. Fan, W. E. Biddison, D. C. Wiley.
1996
. Structure of the complex between human T-cell receptor, viral peptide and HLA-A2.
Nature
384
:
134
61
Garcia, K. C., M. Degano, R. L. Stanfield, A. Brunmark, M. R. Jackson, P. A. Peterson, L. Teyton, I. A. Wilson.
1996
. An αβ T cell receptor structure at 2.5A and its orientation in the TCR-MHC complex.
Science
274
:
209
62
Cole, G. A., T. L. Hogg, D. L. Woodland.
1994
. The MHC class I-restricted T cell response to Sendai virus infection in C57BL/6 mice: a single immunodominant epitope elicits an extremely diverse repertoire of T cells.
Int. Immunol.
6
:
1767
63
Horwitz, M. S., Y. Yanagi, M. B. A. Oldstone.
1994
. T-cell receptors from virus-specific cytotoxic T lymphocytes recognizing a single immunodominant nine-amino-acid viral epitope show marked diversity.
J. Virol.
68
:
352
64
Imarai, M., E. C. Goyarts, G. M. van Bleek, S. G. Nathenson.
1995
. Diversity of T cell receptors specific for the VSV antigenic peptide (N52–59) bound by the H-2Kb class I molecule.
Cell. Immunol.
160
:
33
65
MacDonald, H. R., R. Schneider, R. K. Lees, R. C. Howe, H. Acha-Orbea, H. Festenstein, R. M. Zinkernagel, H. Hengartner.
1988
. T-cell receptor Vβ use predicts reactivity and tolerance to Mlsa-encoded antigens.
Nature
332
:
40
66
Padovan, E., G. Casorati, P. Dellabona, S. Meyer, M. Brockhaus, and A. Lanzavecchia. 1993. Expression of two T cell receptor α chains: dual receptor T cells. Science 262:.
67
Hardardottir, F., J. L. Baron, C. A. Janeway.
1995
. T cells with two functional antigen-specific receptors.
Proc. Natl. Acad. Sci. USA
92
:
354
68
Evavold, B. D., J. Sloan-Lancaster, K. J. Wilson, J. B. Rothbard, P. M. Allen.
1995
. Specific T cell recognition of minimally homologous peptides: evidence for multiple endogenous ligands.
Immunity
2
:
655
69
Nanda, N. K., K. K. Arzoo, H. M. Geysen, A. Sette, E. E. Sercarz.
1995
. Recognition of multiple peptide cores by a single T cell receptor.
J. Exp. Med.
182
:
531
70
Tallquist, M. D., T. J. Yun, L. R. Pease.
1996
. A single T cell receptor recognizes structurally distinct MHC/peptide complexes with high specificity.
J. Exp. Med.
184
:
1017
71
Valitutti, S., S. Muller, M. Dessing, A. Lanzavecchia.
1996
. Different responses are elicited in cytotoxic T lymphocytes by different levels of T cell receptor occupancy.
J. Exp. Med.
183
:
1917
72
Gould, K. G., H. Scotney, G. G. Brownlee.
1991
. Characterization of two distinct major histocompatibility complex class I Kk-restricted T-cell epitopes within the influenza A/PR/8/34 virus hemagglutinin.
J. Virol.
65
:
5401