Factors influencing the human cytotoxic T cell response to autologous lymphoblastoid cell lines in vitro.

Epstein-Barr virus (EBV)- and fetal calf serum (FCS)-specific cytotoxic human T cells can be generated in vitro, and have been shown to be HLA-antigen restricted. In the present work, peripheral blood mononuclear cells were stimulated with the gamma-irradiated autologous lymphoblastoid cell line (LCL) grown previously in FCS, human serum, or without serum. The induction and generation of cytotoxic T cells was carried out exclusively in culture medium containing autologous serum. With EBV-seronegative responders, FCS-grown stimulator LCL generated a FCS-specific cytotoxic T cell response. AB serum-grown LCL generated only a weak response, except at a high stimulatory dose, where the response tended to be essentially nonspecific. EBV-seropositive responders, in contrast, gave a typical secondary EBV-specific response regardless of the serum in which the LCL had been grown previously; no FCS response was detected. Dose-response and cold target inhibition studies confirmed these results. EBV immunity obviously plays a major role in the T cell response to the autologous LCL, which can no longer be viewed simply as a form of autologous mixed leucocyte reaction.

HLA-DR-antigen-associated restriction of EBV-specific cytotoxic T-cell colonies.

EBV-specific cytotoxic T cells can be generated in vitro in a secondary response. Several previous studies with bulk cultures provided evidence that cytotoxicity was restricted by HLA-A,B-related antigens. In the present family study, the EBV-specific cytotoxic T-cell response of a normal EBV-seropositive donor was analysed in detail by T-cell colony formation. Peripheral blood mononuclear cells were stimulated by the gamma-irradiated autologous lymphoblastoid cell line (LCL) and 3 days later seeded into agarose. Colonies were harvested, amplified by addition of interleukin-2 (IL-2), and analysed for T-cell markers and specificity in 51Cr-release assays. Twenty-two colonies were studied: all colonies were OKT3+, five were predominantly OKT4+, 9 were OKT8+ and 8 were mixtures. As expected from previous work, the OKT8+ colonies were cytotoxic for the autologous LCL target and cytotoxicity was blocked by monoclonal antibody (W6/32) to the nonpolymorphic determinants of HLA-A,B,C antigens. Significantly, the OKT4+ colonies tested also showed specific cytotoxicity, but lysis of the autologous LCL was blocked by the monoclonal antibody (OKlal) to the non-polymorphic determinants of HLA-DR antigens. Two interesting patterns of specificity were seen in cytotoxicity tests on sibling LCL targets. In one pattern, targets bearing the A11, B5, DR7 haplotype were lysed, while those bearing the A1, B8, DR3 were not, indicating haplotype preference. In the other pattern, there was lysis of the autologous cell line but not of the sibling targets. These results including HLA-DR-associated restriction, haplotype preference and strict self-preference, further illustrate the complexity of the EBV-cytotoxic T-cell response.

A comparison of Epstein-Barr virus-specific T-cell immunity in rheumatoid arthritis and osteoarthritis patients.

The level of Epstein-Barr virus (EBV)-specific T-cell-mediated immunity in 20 rheumatoid arthritis (RA) patients was compared with 16 age- and sex-matched osteoarthritis (OA) patients using the regression of EBV-transformation assay. The results show that the level of EBV-specific T-cell immunity in RA patients is significantly depressed compared with OA patients (P less than .001) or healthy laboratory controls (P less than .001). In contrast, lymphocytes from RA and OA patients showed a similar ability to act as a responder population in the mixed leucocyte reaction. It is unlikely that the difference in EBV-specific immunity is due to a general T-cell defect in RA patients since there was no correlation between EBV-specific T-cell immunity and mixed leucocyte reactivity. There was no correlation between EBV-specific T-cell immunity and any of the indicators of disease activity nor was there any difference in the anti-EBV antibody titre between both groups of patients. These results indicate that RA patients are deficient in the EBV-specific cytotoxic T-cell precursor population and may explain some of the reported observations of the involvement of EBV in this disease.

Epstein-Barr virus specific T-cell response in nasopharyngeal carcinoma patients.

There is a substantial body of evidence suggesting an association between Epstein-Barr virus (EBV) and undifferentiated nasopharyngeal carcinoma (NPC). The present study has compared a group of NPC patients (newly diagnosed and long-term survivors) and controls for EBV-specific T-cell immunity using the regression of transformation assay. Newly diagnosed patients (17 tested) when compared with either long-term survivors (20 tested) or controls (30 tested) showed a significant impairment in virus-specific T-cell immunity (p = 0.036, p = 0.043 respectively). Furthermore, donors with IgA antibody to EBV showed a significant depression in virus-specific T-cell immunity compared with donors without IgA antibody (19 IgA-positive, 48 IgA-negative; p = 0.0025). These results may be important in explaining the postulated role of EBV in the aetiology of NPC.

A comparison of Epstein-Barr virus-specific T-cell immunity in malaria-endemic and -nonendemic regions of Papua New Guinea.

Epstein-Barr virus genome-positive Burkitt's lymphoma is endemic in Africa and Papua New Guinea and in both countries the tumour is restricted to regions with holoendemic malaria. The present work has compared groups of healthy indigenous individuals living in malarious and non-malarious regions of Papua New Guinea for Epstein-Barr virus-specific T-cell-mediated immunity using the in vitro regression assay. Residents of the malarious region (55 tested), when compared with either residents of the non-malarious area (35 tested) or Caucasian controls (27 tested) showed a significant (p less than 0.0001) impairment of virus-specific T-cell immunity but no obvious disturbance (p greater than 0.05) of anti-viral antibody titres. These results may be important in explaining the postulated role of malarial infection as a co-factor in the pathogenesis of Burkitt's lymphoma.

Evidence for the involvement of HLA-DR antigens in restricted cytotoxicity by fetal calf serum-specific human T cells.

Fetal calf serum (FCS) generated at least two distinct populations of human cytotoxic cells in vitro. One population expressed natural killer (NK) cell-like activity and lysed K562 and HSB-2 targets more effectively than autologous or allogeneic lymphoblastoid cell lines (LCLs). The other population contained FCS-specific cytotoxic T cells which preferentially lysed the autologous LCLs and showed minimal lysis of K562. E-rosette separation and cold target competition experiments clearly established that NK cells were not involved in the self-reactive lysis. Moreover, the lytic activity of the E-rosetted T cells was reduced by up to 95% when autologous target cells were grown in human AB serum rather than FCS, showing that FCS-associated determinants on targets were essential in the cytolytic phase. Autologous LCLs grown in FCS were also considerably stronger competitors than human serum-grown LCLs. The consistent self-preferred lysis suggested that HLA antigen-related restriction was involved, but the patterns of lysis did not implicate HLA-A or B antigens, and monoclonal antibody (W6/32) to an A, B, and C monomorphic determinant failed to block FCS-specific lysis. In contrast, monoclonal antibody (DA.2) to a monomorphic determinant of DR effectively blocked FCS-specific lysis. Cytotoxicity tests with a small panel of DR-typed donors indicated that strong cross-reactions were invariably associated with sharing of DR antigens, particularly DR2, and to a lesser but significant extent DR7. Although DR antigen sharing did not always result in lysis of allogeneic targets, the overall evidence strongly suggests that FCS-specific T-cell cytotoxicity in humans is restricted by products encoded by or associated with the DR genes.

Generation in vitro of EBV-induced specific cytotoxic T cells in autologous serum avoids complications due to self-preferred foetal calf serum-specific T-cell cytotoxicity.

A previous report has established that in cultures of human mononuclear leukocytes, foetal calf serum (FCS) is capable of generating high levels of T cells preferentially cytotoxic for the autologous lymphoblastoid cell line (LCL). The present study compared the capacity of Epstein-Barr virus (EBV) to generate cytotoxic T cells in cultures of mononuclear cells grown in FCS in this system. Five EBV-seropositive and three seronegative donors were used and cultures were harvested at 14 days. With cultures from seropositive donors, whether grown in FCS or in autologous serum, EBV infection generated T cells cytotoxic for the autologous LCL; the response in uninfected control cultures was markedly lower. With seronegative donor cultures grown in FCS, there was virtually no difference in the capacity of T cells generated in infected or uninfected cultures to lyse the autologous LCL. Moreover, cells from seronegative donors cultured in human serum gave no detectable lysis of autologous LCL in either infected or uninfected cultures, clearly showing the absence of a response to EBV. This evidence shows that it is possible to distinguish the generation of specific cytotoxic T cells by FCS from generation by EBV, and with certain donors the apparently EBV-induced response may actually include a significant component induced by FCS in the medium. The cytotoxicity patterns of EBV-induced and FCS-induced T cells for autologous and allogeneic LCL targets showed a degree of parallelism, stressing the need for caution in interpretation of data obtained from cultures using FCS.