- Email: [email protected]
Human T cell responses to the Epstein-Barr nuclear antigen-1 (EBNA-1) as evaluated by synthetic peptides
CELLULAR
IMMUNOLOGY
123,325-333 (1989)
Human T Cell Responses to the Epstein-Barr Nuclear Antigen-l (EBNA-1) as Evaluated by Synthetic Peptides’ J~RGEN PETERSEN, GARY RHODES, KEVIN PATRICK,* JEAN ROUDIER, AND JOHN H. VAUGHAN Research Institute ofScripps Clinic, Department ofBasic and Clinical Research, 10666 North Torrey Pines Road, La Jolla, California 92037, and *San Diego State University, Student Health Services, 5300 Campanile Street, San Diego, California 92182 Received April 24, 1989; accepted July 6, 1989
A panelof synthetic peptides derived from Epstein-Barr virus (EBV) nuclear antigen 1 (EBNA-1) was used to examine human T cell responses to this antigen. In six of seven normal persons with past EBV infection, T cell precursors specific for five peptides (P27, amino acid residues 83-101; P62, 148-166; E31, 353-367; E41, 368-381; and El 1,461-474) were detectable. The precursor frequencies were in the range of 1:20,000 to < 1:100,000 peripheral blood mononuclear cells as determined by limiting dilution analyses. Only two of these peptides were predicted as a-helices; all peptides were glycine-rich. Four other peptides were not reactive in the seven individuals tested. T cell responseswere not detectable in donors without prior EBV infection. Infectious mononucleosis patients investigated 4-6 weeks after diagnosis had likewise no detectable peptide-specific T cell precursors. Thus, it appears that T cells recognizing peptides from EBNA-I arise and persist in people with past EBV infection. o 1989 Academic press, IX. INTRODUCTION
Epstein-Barr virus (EBV)2 is a ubiquitous human herpes virus that infects the majority of the population. The infection persists for life, B cells serving as a reservoir. EBV infection is the cause of infectious mononucleosis (IM) and has been associated with other clinical disorders, such as Burkitt lymphoma, nasopharyngeal carcinoma, and immunodeficiency syndromes. Human B lymphocytes latently infected with EBV express at least five nuclear antigens (EBNAs) (l-3) and a membrane antigen (LMP) (4). B lymphocytes productively infected with EBV express additional viral antigens, such as viral capsid antigens (VCA) and viral membrane antigens (5). EBNA- 1 is the nuclear antigen in latently infected cells that is recognized most frequently by IgM and IgG antibodies ’ This work was supported in part by the Danish Rheumatism Association, Wedell-Wedellborgs Foundation, Association pour la Recherche contre le Cancer, Philippe Foundation, and Grants AR21 175, AR33489, AR25443, and RR00833 from the National Institutes of Health. This is Publication 5476BCR from the Research Institute of Scripps Clinic. * Abbreviations used: EBNA, Epstein-Barr nuclear antigen; EBV, Epstein-Barr virus; IM, infectious mononucleosis; VCA; viral capsid antigen. 325 0008-8749/89 $3.00 Copyright 0 1989 by Academic Press, Inc. All rights of reproduction in any form reserved.
326
PETERSEN ET AL. EBNA-1 1
MSDEGPGTGP GNGLGEKGDT SGPEGSGGSG PQRRGGDNHG RGRGRGRGRG <-------p12---->
51
GGRPGAPGGS GSGPRHRDGV RRPQKRPSCIGCKGTHGGTG AGAGAGGAGA <--------p2,------->
101
GGAGAGGGAG AGGGAGGAGG AGGAGAGGGA GAGGGAGGAG GAGAGGGAGA <-GGGAGGAGAG GGAGGAGGAG AGGGAGAGGG AGGAGAGGGA GGAGGAGAGG -------p6*------->
151 201
GAGAGGAGGA GGAGAGGAGA GGGAGGAGGA GAGGAGAGGA GAGGAGAGGA
251
GGAGAGGAGG AGAGGAGGAG AGGGAGGAGA GGGAGGAGAG GAGGAGAGGA
301
GGAGAGGAGG AGAGGGAGAG GAGAGGGGRG RGGSGGRGRG GSGGRGRGGS
351
GGRRGRGRER ARGGSRERAR GRGRGRGEKR PRSPSSQSSS SGSPPRRPPP <--------=3----x--------=4---->
401
GRRPFFHPVG FADYFEYHQE GGPDGEPDVP PGAIEQGPADDPGEGPSTGP <-------E9----->
451
RGQGDGGRRK KGGWFGKHRG QGGSNPKFEN IAEGIJULIA RSHVERTTDE <-----Eli----->
501
GTWVAGVFVY GGSKTSLYNLRRGTALAIPQCRLTPLSRLPFGMAPGPGPQ
551
PGPLRESIVCYFMVFLQTHIFAEVLKDAIKDLVMTKPAPTCNIRVTVCSF
601
DDGVDLPPWF PPMVEGAAAE GDDGDDGDEG GDGDEGEEGQ E <-------E,-----> <------El$------->
FIG. 1. Peptide sequence of the EBNA- 1 molecule (one letter amino acid code). F12 begins at residue I. All peptides, except E7, contain an additional cysteine at the COOH-terminal end.
generated after IM. Both anti-VCA and anti-EBNA-1 IgG antibodies persist for life (6). IM is characterized by a high incidence of autoantibodies (reviewed in (6)), some of which are cross-reactive with EBNA- 1 (7). EBNA- 1 has a unique structure in that about one-third of the molecule consists of a glycine-alanine repeat (8). This repeat, which spans 248 amino acid residues, carries the major epitope(s) for IgM anti-EBNA- 1 antibodies that develops during IM (7,9). Synthetic peptides derived from the repeating sequence inhibit not only IgM anti-EBNA- 1 binding, but also the binding of IgM autoantibodies in IM sera to cellular proteins (7). The IgG anti-EBNA- 1 antibodies that occur post-EBV infection are also directed to the glycine-alanine repeat, but show no cross-reaction with host proteins. The mechanisms for the isotype switch from IgM to IgG anti-EBNA are poorly understood. The switch implies a role for T helper cells. EBV-specific cell-mediated immune responsescan be demonstrated in people with past EBV infection (10, 11); virus-specific cytotoxic T cells restricted to major histocompatibility antigens of class I are in part responsible for this response,although the
HUMAN
T CELL RESPONSES TO THE EPSTEIN-BARR
NUCLEAR ANTIGEN-l
321
TABLE 1 Proliferative Responsesto EBNA- 1 Peptides Post-IM
EBV normals
EBV+ normals
Peptide
1
2
3
1
1
2
3
4
F12 P27 P62 E3 E4 El1 E7
0.7 1.2 1.3 0.6 0.9 NT 0.7
0.7 0.7 0.9 0.8 1.0 0.8 NT
NT 0.6 0.7 1.0 0.8 0.4 NT
N AA 1.6 NT 1.2 NT 1.4
2.2 3.8 1.8 1.3 1.8 NT 0.9
1.6 1.9 4.2 2.3 3A 5.J NT
NT 0.9 14 NT 0.6 0.7 NT
NT 2.Q 0.5 NT 0.6 0.5 NT
Note. PLB were cultured for 7 days with peptides (0.2-40 wg/ml) and assayedfor [3H]thymidine incorporation (cpm/lO’). Maximal responses for each peptide are shown. Stimulation index was calculated by mean cpm with peptide/mean cpm of background. Stimulation indices 2 3 are regarded as significant. NT, not tested.
precise target(s) expressedby the infected B cells remains elusive. Recent data suggest that a hydrophilic part of the latent membrane protein (LMP) (4) contributes at least some of the target for the cytotoxic T cells ( 12). Szigeti et al. ( 13), using leucocyte migration inhibition (LMI) as their assay,showed that purified EBNA- 1 preparations elicited positive cellular responses in patients with past infection, and Dillner et al. (14) using the same assay,showed positive responsesto a synthetic peptide from the glycine-alanine repeat and to one of two peptides derived from sequencesoutside the glycine-alanine repeat. We have been interested in obtaining further information on the cellular immune response to EBNA-1 peptides. To do this we have examined proliferative responses by T cells and used limiting dilution techniques to assessspecific T cell precursor numbers. We demonstrate here that in humans with past EBV infection, T cell proliferative responsesmay occur to a peptide derived from the glycine-alanine repeat of EBNA- 1, but more frequently for the non-glycine-alanine sequences.In EBV-negative persons no such T cell responsescan be found. MATERIALS AND METHODS Donors. Healthy normal persons with or without serologic evidence of post-EBV infection, as indicated by the presence or absence of antibodies to the VCA, were volunteer employees of Scripps Clinic and Research Foundation donating to the normal blood pool of the General Clinical Research Center. Patients with IM were students at San Diego State University and were studied at the onset of the diseaseand 4-6 weeks after diagnosis. All IM caseswere heterophile-positive and all had elevated IgM and low or no IgG anti-peptide P62 antibody levels (9). Synthetic peptides. A panel of synthetic peptides (Fig. l), derived from EBNA-1, was synthesized by the solid phase method of Merrifield (15), modified slightly as described previously ( 16). The purity was >90% as determined by high-pressure liquid chromatography. The peptides (see Fig. 1) were all neutral or hydrophilic. All peptides, except E7, have an additional COOH-terminal cysteine that is not present
328
PETERSEN ET AL.
; --------
p*7
-- ____-
--:
Primary Cultures W Control m +Peptide Secondary Cultures COMean *2 SD of Control .
. P62 E4
El1
F12
50
50
50
0 50
25
10
5
50
Responder Cells Added(xl,OOO) FIG. 2. (A) Limiting dilution experiment with synthetic peptides. The limiting dilution experiment is shown in detail for peptide P27; the result of the primary culture is shown with vertical bars. The P27stimulated cells from the primary cultures were distributed to microtiter plates under limiting dilution condition. For each dilution 18 wells were cultured with peptide; 6 wells were cultured without peptide. Positive wells were scored as having cpm > mean + 2 SD of control cultures (bars). For the remaining peptides, P62, E4, El 1, and F12, only responseswith 50,000 stimulated cells are shown. (B) Plot of the percentage of negative wells (from the experiment shown in A) versus the numbers of peptide-stimulated cells added to each well.
in EBNA- 1, but was added to facilitate coupling to carrier for experiments not shown herein. Stock solutions of each peptide were stored at 10 mg/ml in Hz0 at -20°C. Cell isolation. Blood mononuclear cells (MNC) were isolated by isopycnic centrifugation through Ficoll-Hypaque. MNC were washed three times prior to resuspension in culture medium consisting of RPM1 1640 supplemented with 100 U/ml penicillin, 10 pg/ml streptomycin, and 2 rniW L-glutamine. Primary cultures. In initial experiments, MNC were cultured in triplicate at 0.5 X lO’/ml, 200 &well, in RPM1 1640 supplemented with 10% pooled human AB serum, in flat-bottomed 96-well microtiter plates. Peptides were added at 0.2 to 40 &ml. In some experiments purified protein derivative of tuberculin was used as control antigen (State Serum Institute, Copenhagen, Denmark, 1 &ml). Cultures were maintained for 7 days in a humidified atmosphere of 95% air and 5% COz. Eighteen hours before harvest, 1 &i of [5-3H]thymidine was added to each well. Thymidine incorporation was quantified by liquid scintillation counting. In subsequent experiments, bulk cultures were performed in 25-ml flasks containing 10 ml cell suspension, with or without peptides (1 pg/ml). Bulk cultures were terminated after 8 days; 18 hr before harvest, 1O5cells were transferred in triplicate to microtiter plates together with 1 &i of [5-3H]thymidine. Enumeration of peptide-reactive T lymphocytes in secondary cultures. The T cell precursor assay was carried out as described (17). MNC stimulated in the primary bulk cultures were collected at 7-8 days, washed, and resuspended in RPM1 1640
HUMAN
T CELL
B
RESPONSES
TO THE
EPSTEIN-BARR
NUCLEAR
ANTIGEN-l
329
Cells Added( x 1,000)
FIG.2-Continued
with 10% human AB serum. Cells were seeded in 96-well flat-bottomed microtiter plates under limiting dilution conditions with 50,000,25,000, 10,000, or 5000 viable cells per well in medium containing 5% human AB serum. Each well also received irradiated autologous MNC to achieve a total of lo5 cells per well. For each density of responder MNC, peptide was added to 18 wells, whereas 6 wells were cultured without peptide. The secondary cultures were maintained for 4-5 days. Eighteen hours before harvest, 1 &i was added to each well. For each responder cell dilution, positive wells were defined as having cpm higher than the mean f 2 standard deviations of the cpm obtained for the six control wells. Precursor T cell frequency was evaluated by plotting the percentage of negative wells for each dilution against the number of responder cells per well, as previously described ( 17).
330
PETERSEN
ET AL.
RESULTS
ProIifrative responsesto EBNA-1 peptidesin primary cultures.In our initial experiments, a panel of synthetic peptides was used to screen for in vitro proliferative responsesin microtiter plates, using a range of concentrations. As displayed in Table 1, no stimulation was achieved with MNC from VCA- individuals. Low, but significant proliferation was observed in VCA+ individuals. The peptides that gave responsesin VCA+ donors were F12, P27, P62, E4, and El 1. Generally, maximal responseswere obtained with peptides used at l-10 pg/ml (data not shown). Experiments with MNC from two patients during the acute phase of IM showed no stimulation (data not shown) with any of the peptides listed in Table 1. Another IM patient was tested 4 weeks after diagnosis. MNC from this individual were stimulated by peptide P27 (Table 1). T cell precursor numbers in secondarycultures. The low stimulation indices seen after primary in vitro stimulation indicate that the frequencies of precursor T cells specific for the peptides are low. Secondary cultures were carried out in microtiter plates after preliminary culture in flasks. In the secondary cultures with 18 replicates, it was possible to estimate the precursor frequency for T cells specific for peptides. Figure 2A shows a limiting dilution experiment with secondary responsesto peptide P27. When the number of responding cells was plotted semilogarithmically against the number of cells assayed, the precursor frequency could be calculated to be 1: 20,000 (Fig. 2B). In this individual, no secondary responseswere seenwith four other EBNA-1 peptides. In the experiment shown in Fig. 3A primary responses in one donor to two of the EBNA-1 peptides, P62 and P27, were seen, and precursor frequencies corresponding to 1:30,000 and 155,000, respectively, could be calculated in secondary cultures. Lower responseswere seenwith E4 and E 11 in this individual. Table 2 summarizes all of our experience in enumerating peptide-specific T cells in secondary cultures of EBV- and EBV+ donors. T cells specific for one or more of peptides P27, P62, E3, E4, and E 11 were detected in six of seven VCA+ individuals. The frequencies were in the range of 1:20,000 to < 1:100,000. Generally, T cells from a given donor recognized not more than four peptides. The remaining peptides (F12, E9, E7, and E 13) did not yield detectable responses. Peptide-specific precursor T cells were not detected in two VCA- donors tested. As a positive control, one of these donors had PPD-specific T cell precursor cells at a high frequency (> 1:5000). We studied three patients during the early convalescent phase of IM; as in the case of VCA- individuals, no precursors were found (not shown). DISCUSSION These experiments demonstrate that selected EBNA-1 peptides can stimulate T cells in a proliferation assay. The peptides selected had been synthesized previously for studies of antibody reactivity with glycine-rich sequences in the molecule (I@, and thus they did not represent peptides preconceived to serve asgood T cell antigens. We show that five of these peptides, one of which lies inside the glycine-alanine repeat of EBNA-1, one of which spans the N-terminal end of the glycine-alanine repeat, and three of which were outside the repeat, were stimulatory. Four other EBNA-lderived peptides failed to induce detectable T cell proliferation. Control experiments established that only T cells from VCA+ responded; none of the peptides stimulated
HUMAN
T CELL RESPONSES TO THE EPSTEIN-BARR
NUCLEAR ANTIGEN-l
331
A 60 50 r --__..-____ pfj*--- _____.._: Ei g
40-
2 x
r .-___...__p2,--
---__I I Primary Cultures EaControl FZI + Peptide
. j-
30Secondary Cultures
P
IXMean k2 SD of Control
Li ci
CellsAdded(x 1.0001
6
Cells Addedix 1,000)
60
I
0 P62 P27 0 E4
0
l
\
FIG. 3. (A) Limiting dilution experiments showing responsesto two peptides, P62 and P27. (B) Plot of the percentage of negative wells (from the experiment shown in A) versusthe number of peptide-stimulated cells added to each well.
332
PETERSEN ET AL. TABLE 2
T Cell Precursor Frequencies Calculated from Secondary Cultures of MNC of VCA+ Individuals Peptides
1
2
3
F12 P27 P62 E3 E4 E7 E9 El1 El3
None I :20,000 None NT None NT NT None NT
NT” 1:55,000 1:30,000 NT 1:75,ooo NT NT <1:1oo,ooo NT
NT None None None 1:4o,ooo None None None
4 NT None None None :1:1oo,ooo None None 1:25,000 None
5
6
I
NT None None 1:30,000 None None None None None
NT None None None None None None None None
NT 1:25,000 None 1:50,ooo 1:45,ooo None None None None
’ NT, not tested.
T cells from VCA- donors, nor from subjects in early convalescence from EBV infection. The latter results are in accord with the findings of Szigeti et al. (13) who also reported that IM patients do not display a cell-mediated immune response to EBNA1, as measured by the production of leukocyte migration inhibitory factor. Dillner et al. (14) also studied EBNA- 1 peptides in cell-mediated immune responses,using LMI assaythat may not be specific for T cells (19). They showed that LMI could be elicited by two peptides derived from EBNA-1, one of which was a glycine-alanine peptide identical to P62 and the other of which encompassed residues 368-383 and was almost identical to peptide E4. In our own studies, peptide P62 revealed measurable T cell precursors in only one-seventh of the VCAi donors tested, and E4 in four-sevenths of the VCA+ donors. Two other peptides studied by Dillner et al. ( 14) were nonstimulatory in their system and differed from the peptides we report here. As expected from the low stimulation indices in primary responsesin VCA+ donors (Table l), the frequencies of T cell precursors specific for the peptides were low, ranging from 1:20,000 to < 1:100,000. Considering that these frequencies were calculated after secondary stimulation, it can be estimated that the frequencies in the original blood specimens were at least one order of magnitude less. It has been observed that peptides antigenic for T cell responsesoften have a-helical and amphipathic structure (20-22). It is worth noting, therefore, that neither of two of the stimulatoty peptides we have studied, P27 or E 11, have structures that predict cr-helicity or amphipathicity (22). Contrariwise, peptides E7 and E13, which were nonstimulatory in our system, have strong prediction for amphipathicity. Thus amphipathicity may not be necessaryfor initiating T cell responsesto EBV peptides. We do not yet know whether the T cells responding to EBNA- 1 peptides recognize the peptides complexed to classI or to classII molecules of the major histocompatibility complex. If the former alternative is correct, then EBNA- 1, despite its not being a protein of the outer cell membrane, could contribute to the LYDMA target seen by cytotoxic lymphocytes. Similar roles for EBNA-2 in EBV-infected cells (23) and for influenza-encoded proteins in influenza-infected cells have been suggested (24). If EBNA-1 peptides are recognized principally in association with class II molecules, then EBNA-1 could be expected to stimulate predominantly CD4 T helper cells.
HUMAN
T CELL RESPONSES TO THE EPSTEIN-BARR
NUCLEAR ANTIGEN-l
333
However, it is equally possible that some EBNA-1 peptides will be bound to class I MHC molecules and others to class II MHC molecules in EBV-infected B cells and other antigen presenting cells. In this respect, it must also be recognized that we have not yet examined peptides from the entire EBNA- 1 sequence.Therefore, the peptides that react with immune T cells must represent a minimal estimate of those that can be recognized. ACKNOWLEDGMENTS Dr. Richard Hot&ten and Johnson & Johnson are acknowledged for preparation of the panel of synthetic peptides and Dr. Dennis Carson for reviewing the manuscript. Ms. Jane Uhle is thanked for typing the manuscript.
REFERENCES 1. Ricksten, A., Kallin, B., Alexander, H., Dillner, J., Fahreus, R., Klein, G., Lemer, R., and Rymo, L., Proc. Nat]. Acad. Sci. USA 85,995, 1988. 2. Dillner, J., Kallin, B., Alexander, H., Emberg, I., Uno, M., Uno, Y., Klein, G., and Lemer, R. A., Proc. Nati. Acad. Sci. USA 83,6641, 1986. 3. Sawada, K., Yamamoto, M., Tabata, T., Smith, M., Tanaka, A., and Nonoyama, M., I’irologv 168, 22, 1989. 4. Hennesy, K., Fennewald, S., Hummel, M., Cole, T., and Kieff, E., Proc. Nat/. Acad. Sci. USA 81, 7207, 1984. 5. Henle, G., and Henle, W., J. Bacterial. 91, 1248, 1966. 6. Tosato, G., Adv. Cancer Res. 49,75, 1987. 7. Rhodes, G., Rumpold, H., Kurki, P., Patrick, K., Carson, D. A., and Vaughan, J., J. Exp. Med. 165, 1026,1987. 8. Hennesy, K., and Kieff, E., Proc. Natl. Acad. Sci. USA 80,5665, 1983. 9. Rumpold, H., Rhodes, G. H., Bloch, P. L., Carson, D. A., and Vaughan, J. H., J. Immunol. 138,593, 1987. 10. Thorley-Lawson, D. A., Chess, L., and Strominger, J. L., J. Exp. Med. 146,495, 1977. I 1. Slovin, S. F., Scooley, R. T., and Thorley-Lawson, D. A., J. Immunoi. 130,2127, 1983. 12. Thorley-Lawson, D. A., and Israelsohn, E. S., Proc. Nutl. Acud. Sci. USA 84,5384, 1987. 13. Szigeti, R., Masucci, M. G., Henle, G., Henle, W., Purtillo, D. T., and Klein, G., Clin. Immunol. Immunopathol. 22, 128, 1982. 14. Dillner, J., Szigeti, R., Henle, W., Henle, G., Lemer, R. A., and Klein, G., Int. J. Cancer40,495, 1987. 15. Menifield, R. B., J. Amer. Chem. Sot. 85,2 149, 1963. 16. Hot&ten, R. A., Chang W. C., and Li, C. H., Int. J. Pept. Protein Res. 16,311, 1980. 17. Ford, D., and Burger, D., Cell. Immunol. 79,334, 1983. 18. Petersen, J., Rhodes, G., Roudier, J., Carson, D. A., and Vaughan, J. H., In “Epstein-Barr Virus and Human Disease II” (V. Ablashi, Ed.), in press. 19. Masucci, G., Szigeti, R., Stevens, D., Masucci, M. G., Klein, E., Petersen, J., Bendtzen, K., and Klein, G. Cell. Immunol. 85,5 11, 1984. 20. Livingstone, A. M., and Garrison Fathman, C., Annu. Rev. Immunol. 5,477, 1987. 21. Chou, P. W., and Fasman, G. D., Adv. Enzymol. 47,45,1978. 22. Berkower, I., Buckenmeyer, G. K., and Berzofsky, J. A., J. Immunol. 136,2498, 1986. 23. Moss, D. J., Misko, 1. S., Burrows, S. R., Burman, K., McCarthy, R., and Sculley, T. B., Nature (London)331,719,1988. 24. Townsend, A. R. M., Rothbard, J., Gotch, F. M., Bahadur, G., Wraith, D., and McMichael, A. J., Cell 44,959, 1986.
IMMUNOLOGY
123,325-333 (1989)
Human T Cell Responses to the Epstein-Barr Nuclear Antigen-l (EBNA-1) as Evaluated by Synthetic Peptides’ J~RGEN PETERSEN, GARY RHODES, KEVIN PATRICK,* JEAN ROUDIER, AND JOHN H. VAUGHAN Research Institute ofScripps Clinic, Department ofBasic and Clinical Research, 10666 North Torrey Pines Road, La Jolla, California 92037, and *San Diego State University, Student Health Services, 5300 Campanile Street, San Diego, California 92182 Received April 24, 1989; accepted July 6, 1989
A panelof synthetic peptides derived from Epstein-Barr virus (EBV) nuclear antigen 1 (EBNA-1) was used to examine human T cell responses to this antigen. In six of seven normal persons with past EBV infection, T cell precursors specific for five peptides (P27, amino acid residues 83-101; P62, 148-166; E31, 353-367; E41, 368-381; and El 1,461-474) were detectable. The precursor frequencies were in the range of 1:20,000 to < 1:100,000 peripheral blood mononuclear cells as determined by limiting dilution analyses. Only two of these peptides were predicted as a-helices; all peptides were glycine-rich. Four other peptides were not reactive in the seven individuals tested. T cell responseswere not detectable in donors without prior EBV infection. Infectious mononucleosis patients investigated 4-6 weeks after diagnosis had likewise no detectable peptide-specific T cell precursors. Thus, it appears that T cells recognizing peptides from EBNA-I arise and persist in people with past EBV infection. o 1989 Academic press, IX. INTRODUCTION
Epstein-Barr virus (EBV)2 is a ubiquitous human herpes virus that infects the majority of the population. The infection persists for life, B cells serving as a reservoir. EBV infection is the cause of infectious mononucleosis (IM) and has been associated with other clinical disorders, such as Burkitt lymphoma, nasopharyngeal carcinoma, and immunodeficiency syndromes. Human B lymphocytes latently infected with EBV express at least five nuclear antigens (EBNAs) (l-3) and a membrane antigen (LMP) (4). B lymphocytes productively infected with EBV express additional viral antigens, such as viral capsid antigens (VCA) and viral membrane antigens (5). EBNA- 1 is the nuclear antigen in latently infected cells that is recognized most frequently by IgM and IgG antibodies ’ This work was supported in part by the Danish Rheumatism Association, Wedell-Wedellborgs Foundation, Association pour la Recherche contre le Cancer, Philippe Foundation, and Grants AR21 175, AR33489, AR25443, and RR00833 from the National Institutes of Health. This is Publication 5476BCR from the Research Institute of Scripps Clinic. * Abbreviations used: EBNA, Epstein-Barr nuclear antigen; EBV, Epstein-Barr virus; IM, infectious mononucleosis; VCA; viral capsid antigen. 325 0008-8749/89 $3.00 Copyright 0 1989 by Academic Press, Inc. All rights of reproduction in any form reserved.
326
PETERSEN ET AL. EBNA-1 1
MSDEGPGTGP GNGLGEKGDT SGPEGSGGSG PQRRGGDNHG RGRGRGRGRG <-------p12---->
51
GGRPGAPGGS GSGPRHRDGV RRPQKRPSCIGCKGTHGGTG AGAGAGGAGA <--------p2,------->
101
GGAGAGGGAG AGGGAGGAGG AGGAGAGGGA GAGGGAGGAG GAGAGGGAGA <-GGGAGGAGAG GGAGGAGGAG AGGGAGAGGG AGGAGAGGGA GGAGGAGAGG -------p6*------->
151 201
GAGAGGAGGA GGAGAGGAGA GGGAGGAGGA GAGGAGAGGA GAGGAGAGGA
251
GGAGAGGAGG AGAGGAGGAG AGGGAGGAGA GGGAGGAGAG GAGGAGAGGA
301
GGAGAGGAGG AGAGGGAGAG GAGAGGGGRG RGGSGGRGRG GSGGRGRGGS
351
GGRRGRGRER ARGGSRERAR GRGRGRGEKR PRSPSSQSSS SGSPPRRPPP <--------=3----x--------=4---->
401
GRRPFFHPVG FADYFEYHQE GGPDGEPDVP PGAIEQGPADDPGEGPSTGP <-------E9----->
451
RGQGDGGRRK KGGWFGKHRG QGGSNPKFEN IAEGIJULIA RSHVERTTDE <-----Eli----->
501
GTWVAGVFVY GGSKTSLYNLRRGTALAIPQCRLTPLSRLPFGMAPGPGPQ
551
PGPLRESIVCYFMVFLQTHIFAEVLKDAIKDLVMTKPAPTCNIRVTVCSF
601
DDGVDLPPWF PPMVEGAAAE GDDGDDGDEG GDGDEGEEGQ E <-------E,-----> <------El$------->
FIG. 1. Peptide sequence of the EBNA- 1 molecule (one letter amino acid code). F12 begins at residue I. All peptides, except E7, contain an additional cysteine at the COOH-terminal end.
generated after IM. Both anti-VCA and anti-EBNA-1 IgG antibodies persist for life (6). IM is characterized by a high incidence of autoantibodies (reviewed in (6)), some of which are cross-reactive with EBNA- 1 (7). EBNA- 1 has a unique structure in that about one-third of the molecule consists of a glycine-alanine repeat (8). This repeat, which spans 248 amino acid residues, carries the major epitope(s) for IgM anti-EBNA- 1 antibodies that develops during IM (7,9). Synthetic peptides derived from the repeating sequence inhibit not only IgM anti-EBNA- 1 binding, but also the binding of IgM autoantibodies in IM sera to cellular proteins (7). The IgG anti-EBNA- 1 antibodies that occur post-EBV infection are also directed to the glycine-alanine repeat, but show no cross-reaction with host proteins. The mechanisms for the isotype switch from IgM to IgG anti-EBNA are poorly understood. The switch implies a role for T helper cells. EBV-specific cell-mediated immune responsescan be demonstrated in people with past EBV infection (10, 11); virus-specific cytotoxic T cells restricted to major histocompatibility antigens of class I are in part responsible for this response,although the
HUMAN
T CELL RESPONSES TO THE EPSTEIN-BARR
NUCLEAR ANTIGEN-l
321
TABLE 1 Proliferative Responsesto EBNA- 1 Peptides Post-IM
EBV normals
EBV+ normals
Peptide
1
2
3
1
1
2
3
4
F12 P27 P62 E3 E4 El1 E7
0.7 1.2 1.3 0.6 0.9 NT 0.7
0.7 0.7 0.9 0.8 1.0 0.8 NT
NT 0.6 0.7 1.0 0.8 0.4 NT
N AA 1.6 NT 1.2 NT 1.4
2.2 3.8 1.8 1.3 1.8 NT 0.9
1.6 1.9 4.2 2.3 3A 5.J NT
NT 0.9 14 NT 0.6 0.7 NT
NT 2.Q 0.5 NT 0.6 0.5 NT
Note. PLB were cultured for 7 days with peptides (0.2-40 wg/ml) and assayedfor [3H]thymidine incorporation (cpm/lO’). Maximal responses for each peptide are shown. Stimulation index was calculated by mean cpm with peptide/mean cpm of background. Stimulation indices 2 3 are regarded as significant. NT, not tested.
precise target(s) expressedby the infected B cells remains elusive. Recent data suggest that a hydrophilic part of the latent membrane protein (LMP) (4) contributes at least some of the target for the cytotoxic T cells ( 12). Szigeti et al. ( 13), using leucocyte migration inhibition (LMI) as their assay,showed that purified EBNA- 1 preparations elicited positive cellular responses in patients with past infection, and Dillner et al. (14) using the same assay,showed positive responsesto a synthetic peptide from the glycine-alanine repeat and to one of two peptides derived from sequencesoutside the glycine-alanine repeat. We have been interested in obtaining further information on the cellular immune response to EBNA-1 peptides. To do this we have examined proliferative responses by T cells and used limiting dilution techniques to assessspecific T cell precursor numbers. We demonstrate here that in humans with past EBV infection, T cell proliferative responsesmay occur to a peptide derived from the glycine-alanine repeat of EBNA- 1, but more frequently for the non-glycine-alanine sequences.In EBV-negative persons no such T cell responsescan be found. MATERIALS AND METHODS Donors. Healthy normal persons with or without serologic evidence of post-EBV infection, as indicated by the presence or absence of antibodies to the VCA, were volunteer employees of Scripps Clinic and Research Foundation donating to the normal blood pool of the General Clinical Research Center. Patients with IM were students at San Diego State University and were studied at the onset of the diseaseand 4-6 weeks after diagnosis. All IM caseswere heterophile-positive and all had elevated IgM and low or no IgG anti-peptide P62 antibody levels (9). Synthetic peptides. A panel of synthetic peptides (Fig. l), derived from EBNA-1, was synthesized by the solid phase method of Merrifield (15), modified slightly as described previously ( 16). The purity was >90% as determined by high-pressure liquid chromatography. The peptides (see Fig. 1) were all neutral or hydrophilic. All peptides, except E7, have an additional COOH-terminal cysteine that is not present
328
PETERSEN ET AL.
; --------
p*7
-- ____-
--:
Primary Cultures W Control m +Peptide Secondary Cultures COMean *2 SD of Control .
. P62 E4
El1
F12
50
50
50
0 50
25
10
5
50
Responder Cells Added(xl,OOO) FIG. 2. (A) Limiting dilution experiment with synthetic peptides. The limiting dilution experiment is shown in detail for peptide P27; the result of the primary culture is shown with vertical bars. The P27stimulated cells from the primary cultures were distributed to microtiter plates under limiting dilution condition. For each dilution 18 wells were cultured with peptide; 6 wells were cultured without peptide. Positive wells were scored as having cpm > mean + 2 SD of control cultures (bars). For the remaining peptides, P62, E4, El 1, and F12, only responseswith 50,000 stimulated cells are shown. (B) Plot of the percentage of negative wells (from the experiment shown in A) versus the numbers of peptide-stimulated cells added to each well.
in EBNA- 1, but was added to facilitate coupling to carrier for experiments not shown herein. Stock solutions of each peptide were stored at 10 mg/ml in Hz0 at -20°C. Cell isolation. Blood mononuclear cells (MNC) were isolated by isopycnic centrifugation through Ficoll-Hypaque. MNC were washed three times prior to resuspension in culture medium consisting of RPM1 1640 supplemented with 100 U/ml penicillin, 10 pg/ml streptomycin, and 2 rniW L-glutamine. Primary cultures. In initial experiments, MNC were cultured in triplicate at 0.5 X lO’/ml, 200 &well, in RPM1 1640 supplemented with 10% pooled human AB serum, in flat-bottomed 96-well microtiter plates. Peptides were added at 0.2 to 40 &ml. In some experiments purified protein derivative of tuberculin was used as control antigen (State Serum Institute, Copenhagen, Denmark, 1 &ml). Cultures were maintained for 7 days in a humidified atmosphere of 95% air and 5% COz. Eighteen hours before harvest, 1 &i of [5-3H]thymidine was added to each well. Thymidine incorporation was quantified by liquid scintillation counting. In subsequent experiments, bulk cultures were performed in 25-ml flasks containing 10 ml cell suspension, with or without peptides (1 pg/ml). Bulk cultures were terminated after 8 days; 18 hr before harvest, 1O5cells were transferred in triplicate to microtiter plates together with 1 &i of [5-3H]thymidine. Enumeration of peptide-reactive T lymphocytes in secondary cultures. The T cell precursor assay was carried out as described (17). MNC stimulated in the primary bulk cultures were collected at 7-8 days, washed, and resuspended in RPM1 1640
HUMAN
T CELL
B
RESPONSES
TO THE
EPSTEIN-BARR
NUCLEAR
ANTIGEN-l
329
Cells Added( x 1,000)
FIG.2-Continued
with 10% human AB serum. Cells were seeded in 96-well flat-bottomed microtiter plates under limiting dilution conditions with 50,000,25,000, 10,000, or 5000 viable cells per well in medium containing 5% human AB serum. Each well also received irradiated autologous MNC to achieve a total of lo5 cells per well. For each density of responder MNC, peptide was added to 18 wells, whereas 6 wells were cultured without peptide. The secondary cultures were maintained for 4-5 days. Eighteen hours before harvest, 1 &i was added to each well. For each responder cell dilution, positive wells were defined as having cpm higher than the mean f 2 standard deviations of the cpm obtained for the six control wells. Precursor T cell frequency was evaluated by plotting the percentage of negative wells for each dilution against the number of responder cells per well, as previously described ( 17).
330
PETERSEN
ET AL.
RESULTS
ProIifrative responsesto EBNA-1 peptidesin primary cultures.In our initial experiments, a panel of synthetic peptides was used to screen for in vitro proliferative responsesin microtiter plates, using a range of concentrations. As displayed in Table 1, no stimulation was achieved with MNC from VCA- individuals. Low, but significant proliferation was observed in VCA+ individuals. The peptides that gave responsesin VCA+ donors were F12, P27, P62, E4, and El 1. Generally, maximal responseswere obtained with peptides used at l-10 pg/ml (data not shown). Experiments with MNC from two patients during the acute phase of IM showed no stimulation (data not shown) with any of the peptides listed in Table 1. Another IM patient was tested 4 weeks after diagnosis. MNC from this individual were stimulated by peptide P27 (Table 1). T cell precursor numbers in secondarycultures. The low stimulation indices seen after primary in vitro stimulation indicate that the frequencies of precursor T cells specific for the peptides are low. Secondary cultures were carried out in microtiter plates after preliminary culture in flasks. In the secondary cultures with 18 replicates, it was possible to estimate the precursor frequency for T cells specific for peptides. Figure 2A shows a limiting dilution experiment with secondary responsesto peptide P27. When the number of responding cells was plotted semilogarithmically against the number of cells assayed, the precursor frequency could be calculated to be 1: 20,000 (Fig. 2B). In this individual, no secondary responseswere seenwith four other EBNA-1 peptides. In the experiment shown in Fig. 3A primary responses in one donor to two of the EBNA-1 peptides, P62 and P27, were seen, and precursor frequencies corresponding to 1:30,000 and 155,000, respectively, could be calculated in secondary cultures. Lower responseswere seenwith E4 and E 11 in this individual. Table 2 summarizes all of our experience in enumerating peptide-specific T cells in secondary cultures of EBV- and EBV+ donors. T cells specific for one or more of peptides P27, P62, E3, E4, and E 11 were detected in six of seven VCA+ individuals. The frequencies were in the range of 1:20,000 to < 1:100,000. Generally, T cells from a given donor recognized not more than four peptides. The remaining peptides (F12, E9, E7, and E 13) did not yield detectable responses. Peptide-specific precursor T cells were not detected in two VCA- donors tested. As a positive control, one of these donors had PPD-specific T cell precursor cells at a high frequency (> 1:5000). We studied three patients during the early convalescent phase of IM; as in the case of VCA- individuals, no precursors were found (not shown). DISCUSSION These experiments demonstrate that selected EBNA-1 peptides can stimulate T cells in a proliferation assay. The peptides selected had been synthesized previously for studies of antibody reactivity with glycine-rich sequences in the molecule (I@, and thus they did not represent peptides preconceived to serve asgood T cell antigens. We show that five of these peptides, one of which lies inside the glycine-alanine repeat of EBNA-1, one of which spans the N-terminal end of the glycine-alanine repeat, and three of which were outside the repeat, were stimulatory. Four other EBNA-lderived peptides failed to induce detectable T cell proliferation. Control experiments established that only T cells from VCA+ responded; none of the peptides stimulated
HUMAN
T CELL RESPONSES TO THE EPSTEIN-BARR
NUCLEAR ANTIGEN-l
331
A 60 50 r --__..-____ pfj*--- _____.._: Ei g
40-
2 x
r .-___...__p2,--
---__I I Primary Cultures EaControl FZI + Peptide
. j-
30Secondary Cultures
P
IXMean k2 SD of Control
Li ci
CellsAdded(x 1.0001
6
Cells Addedix 1,000)
60
I
0 P62 P27 0 E4
0
l
\
FIG. 3. (A) Limiting dilution experiments showing responsesto two peptides, P62 and P27. (B) Plot of the percentage of negative wells (from the experiment shown in A) versusthe number of peptide-stimulated cells added to each well.
332
PETERSEN ET AL. TABLE 2
T Cell Precursor Frequencies Calculated from Secondary Cultures of MNC of VCA+ Individuals Peptides
1
2
3
F12 P27 P62 E3 E4 E7 E9 El1 El3
None I :20,000 None NT None NT NT None NT
NT” 1:55,000 1:30,000 NT 1:75,ooo NT NT <1:1oo,ooo NT
NT None None None 1:4o,ooo None None None
4 NT None None None :1:1oo,ooo None None 1:25,000 None
5
6
I
NT None None 1:30,000 None None None None None
NT None None None None None None None None
NT 1:25,000 None 1:50,ooo 1:45,ooo None None None None
’ NT, not tested.
T cells from VCA- donors, nor from subjects in early convalescence from EBV infection. The latter results are in accord with the findings of Szigeti et al. (13) who also reported that IM patients do not display a cell-mediated immune response to EBNA1, as measured by the production of leukocyte migration inhibitory factor. Dillner et al. (14) also studied EBNA- 1 peptides in cell-mediated immune responses,using LMI assaythat may not be specific for T cells (19). They showed that LMI could be elicited by two peptides derived from EBNA-1, one of which was a glycine-alanine peptide identical to P62 and the other of which encompassed residues 368-383 and was almost identical to peptide E4. In our own studies, peptide P62 revealed measurable T cell precursors in only one-seventh of the VCAi donors tested, and E4 in four-sevenths of the VCA+ donors. Two other peptides studied by Dillner et al. ( 14) were nonstimulatory in their system and differed from the peptides we report here. As expected from the low stimulation indices in primary responsesin VCA+ donors (Table l), the frequencies of T cell precursors specific for the peptides were low, ranging from 1:20,000 to < 1:100,000. Considering that these frequencies were calculated after secondary stimulation, it can be estimated that the frequencies in the original blood specimens were at least one order of magnitude less. It has been observed that peptides antigenic for T cell responsesoften have a-helical and amphipathic structure (20-22). It is worth noting, therefore, that neither of two of the stimulatoty peptides we have studied, P27 or E 11, have structures that predict cr-helicity or amphipathicity (22). Contrariwise, peptides E7 and E13, which were nonstimulatory in our system, have strong prediction for amphipathicity. Thus amphipathicity may not be necessaryfor initiating T cell responsesto EBV peptides. We do not yet know whether the T cells responding to EBNA- 1 peptides recognize the peptides complexed to classI or to classII molecules of the major histocompatibility complex. If the former alternative is correct, then EBNA- 1, despite its not being a protein of the outer cell membrane, could contribute to the LYDMA target seen by cytotoxic lymphocytes. Similar roles for EBNA-2 in EBV-infected cells (23) and for influenza-encoded proteins in influenza-infected cells have been suggested (24). If EBNA-1 peptides are recognized principally in association with class II molecules, then EBNA-1 could be expected to stimulate predominantly CD4 T helper cells.
HUMAN
T CELL RESPONSES TO THE EPSTEIN-BARR
NUCLEAR ANTIGEN-l
333
However, it is equally possible that some EBNA-1 peptides will be bound to class I MHC molecules and others to class II MHC molecules in EBV-infected B cells and other antigen presenting cells. In this respect, it must also be recognized that we have not yet examined peptides from the entire EBNA- 1 sequence.Therefore, the peptides that react with immune T cells must represent a minimal estimate of those that can be recognized. ACKNOWLEDGMENTS Dr. Richard Hot&ten and Johnson & Johnson are acknowledged for preparation of the panel of synthetic peptides and Dr. Dennis Carson for reviewing the manuscript. Ms. Jane Uhle is thanked for typing the manuscript.
REFERENCES 1. Ricksten, A., Kallin, B., Alexander, H., Dillner, J., Fahreus, R., Klein, G., Lemer, R., and Rymo, L., Proc. Nat]. Acad. Sci. USA 85,995, 1988. 2. Dillner, J., Kallin, B., Alexander, H., Emberg, I., Uno, M., Uno, Y., Klein, G., and Lemer, R. A., Proc. Nati. Acad. Sci. USA 83,6641, 1986. 3. Sawada, K., Yamamoto, M., Tabata, T., Smith, M., Tanaka, A., and Nonoyama, M., I’irologv 168, 22, 1989. 4. Hennesy, K., Fennewald, S., Hummel, M., Cole, T., and Kieff, E., Proc. Nat/. Acad. Sci. USA 81, 7207, 1984. 5. Henle, G., and Henle, W., J. Bacterial. 91, 1248, 1966. 6. Tosato, G., Adv. Cancer Res. 49,75, 1987. 7. Rhodes, G., Rumpold, H., Kurki, P., Patrick, K., Carson, D. A., and Vaughan, J., J. Exp. Med. 165, 1026,1987. 8. Hennesy, K., and Kieff, E., Proc. Natl. Acad. Sci. USA 80,5665, 1983. 9. Rumpold, H., Rhodes, G. H., Bloch, P. L., Carson, D. A., and Vaughan, J. H., J. Immunol. 138,593, 1987. 10. Thorley-Lawson, D. A., Chess, L., and Strominger, J. L., J. Exp. Med. 146,495, 1977. I 1. Slovin, S. F., Scooley, R. T., and Thorley-Lawson, D. A., J. Immunoi. 130,2127, 1983. 12. Thorley-Lawson, D. A., and Israelsohn, E. S., Proc. Nutl. Acud. Sci. USA 84,5384, 1987. 13. Szigeti, R., Masucci, M. G., Henle, G., Henle, W., Purtillo, D. T., and Klein, G., Clin. Immunol. Immunopathol. 22, 128, 1982. 14. Dillner, J., Szigeti, R., Henle, W., Henle, G., Lemer, R. A., and Klein, G., Int. J. Cancer40,495, 1987. 15. Menifield, R. B., J. Amer. Chem. Sot. 85,2 149, 1963. 16. Hot&ten, R. A., Chang W. C., and Li, C. H., Int. J. Pept. Protein Res. 16,311, 1980. 17. Ford, D., and Burger, D., Cell. Immunol. 79,334, 1983. 18. Petersen, J., Rhodes, G., Roudier, J., Carson, D. A., and Vaughan, J. H., In “Epstein-Barr Virus and Human Disease II” (V. Ablashi, Ed.), in press. 19. Masucci, G., Szigeti, R., Stevens, D., Masucci, M. G., Klein, E., Petersen, J., Bendtzen, K., and Klein, G. Cell. Immunol. 85,5 11, 1984. 20. Livingstone, A. M., and Garrison Fathman, C., Annu. Rev. Immunol. 5,477, 1987. 21. Chou, P. W., and Fasman, G. D., Adv. Enzymol. 47,45,1978. 22. Berkower, I., Buckenmeyer, G. K., and Berzofsky, J. A., J. Immunol. 136,2498, 1986. 23. Moss, D. J., Misko, 1. S., Burrows, S. R., Burman, K., McCarthy, R., and Sculley, T. B., Nature (London)331,719,1988. 24. Townsend, A. R. M., Rothbard, J., Gotch, F. M., Bahadur, G., Wraith, D., and McMichael, A. J., Cell 44,959, 1986.