Characteristics of EBV-infected B cell lines and B-CLL clones which determine their interaction with lymphocytes

Abstract: CHARACTERISTICS OF EBV-INFECTED B CELL LINES AND B-CLL CLONES WHICH DETERMINE THEIR INTERACTION WITH T LYMPHOCYTES Javier Federico Avila-Cariho Doctoral dissertation from the Microbiology and Tumor Biology Center, Karolinska Institutet, Stockholm, Sweden The results concerning T Iymphocyte mediated recognition of Epstein-Barr virus (EBV) carrying Burkitt Iymphoma (BL) cell lines and EBV-infected chronic Iymphocytic leukemia (CLL) B cells contribute to the understanding of the generally harmless host-EBV interaction and may be extrapolated to the fate of normal B cells after EBV infection. The phenotype of EBV negative BL lines resembles resting B cells. We have observed that the virus-positive sublines stimulated allogeneic T cells more efficiently compared to the original virus-negative BL cells. We compared also the allostimulatory capacity of sublines developed from one EBV carrying BL, which were selected for differences in phenotype and expression of EBV encoded proteins. The sublines that have the latency I program express only EBNA-I, are similar to resting B cells and do not stimulate allogeneic T Iymphocytes, whilst the sublines'that exhibit the latency Ill program express the viral proteins EBNA-2 to -6 and LMP-1, have the phenotype of activated B cells and stimulate the allogeneic T Iymphocytes. Surface marker analysis of the populations suggested that the T cell stimulatory capacity correlated with the expression of the integrins LFA-I and LFA-3. The T cell response to the EBV-carrying BL lines and to the LCLs was not EBV specific since EBV-seronegative donors responded as strongly as the EBV-seropositive ones. The significance of this results is the demonstration that the EBV genome in the BL cells does not influence their interaction with allogeneic T Iymphocytes as long as they keep the latency I program (resting B cell phenotype). B-CLL is clonal and each patient represents thus a different B-cell clone. Clones differ in their sensitivity to experimental EBV infection but even if they are infected they yield rarely immortalized lines. Their degree of sensitivity to EBV infection suggests differences in the maturation state of the CLL cells. It may therefore reflect the variation in the response to EBV infection within a normal B cell population. The majority of the CLL clones has low susceptibility to EBV infection and do not stimulate proliferation of T Iymphocytes. Exposure of the EBV-infected CLL cells to a mixture of B cell mitogens (SAC-lL2-tioredoxin) stimulated T Iymphocytes. Usually EBV-infected CLL cells express EBNA-1 to -6 but not LMP-l, suggesting a new pattern of viral latency. Analysis of surface markers, adhesion and costimulatory molecules on the cells of the clones did not correlate with their T cell stimulatory capacity. Comparison of cytokine production by the differently activated CLL cells showed that the EBV-infected CLL clones with high susceptibility to be infected have high frequency of cells producing all the searched cytokines (IL-1a, IL-1,b, IL-2, IL-4, IL-6, IL-10, IL-13, IFN-y, TNF-a, TNF-b and GM-CSF), whilst the clones with low susceptibility to EBV, the diversity and frequency of the cytokines was lower. In general, the frequency of cytokine-expressing B- CLL cells correlated with their capacity to stimulate T cells. The in vitro EBV-infected CLL clones were Iysed with the same efficiency as the established LCLs by the autologous EBV specific T Iymphocytes. It can be assumed therefore that if such cells are infected in vivo, they eventually become activated, express viral encoded proteins and are eliminated by the T cell response. Key words: EBV, Burkitt Iymphoma, CLL, MLC, antiviral response phenotype, cytokines.