Epstein-Barr virus (EBV) is a cancer-causing human herpesvirus that persists in B cells for the life of the host in greater than 95% of the world’s population. EBV is associated with endemic Burkitt lymphoma in equatorial Africa, nasopharyngeal cell carcinoma in South-east Asia, Hodgkin’s disease, AIDS lymphomas, and B cell lymphoproliferative diseases (LPD)/lymphomas during T cell immunocompromise worldwide. EBV is also the causative agent of infectious mononucleosis, a form of EBV-driven lymphoproliferation during primary infection that is eventually controlled by the immune system. EBV-LPD/lymphomas during immunocompromise are aggressive tumors that can affect as many as 20% of solid-organ transplant recipients. Treatment of EBV-mediated diseases is challenging and no specific therapy or vaccine directed against the virus exists.
We find EBV exciting not just for the multitude of diseases it can cause, but also for the unique and long history of peaceful coexistence of humans with this virus. This evolutionary coexistence has likely shaped the human immune system while at the same time providing mechanisms for the virus to escape immune surveillance. Our goal is to better understand EBV-host relationships and the pathogenesis of EBV-mediated disease by adopting a three-pronged approach. The first focuses on dissecting the early cellular events that are critical for susceptibility of B cells to EBV-driven proliferation. The second is to reveal the determinants of resistance or susceptibility of latently-infected B cells to stimuli that trigger a switch from EBV latency to lytic cycle. The third focuses on delineating EBV-directed immune responses during health and disease.
EBV-driven B cell proliferation
Following infection of B cells with EBV, cells must undergo proliferation for viral latency and persistence to be established. Ex vivo, EBV-driven proliferation gives rise to EBV-infected lymphoblastoid cell lines that can persist indefinitely in culture. Such cell lines as well as viral oncoproteins that mediate cell proliferation continue to be subjects of widespread investigations. However, examination of cellular processes involved in the early stages following infection with EBV has been lacking. Our studies on early cellular events that follow exposure of primary B cells to EBV reveal that B cells are differentially susceptible to EBV-mediated proliferation despite expression of viral oncogenes and CD23, a molecule necessary for subsequent immortalization of cells. By day 3 after exposure to EBV, two distinct sub-populations of cells emerge. One sub-population is marked as CD23hiCD58+IL6- and the other is marked as CD23loCD58+. While both sub-populations of cells express CD23, only CD23hiCD58+IL6- cells undergo proliferation. CD23loCD58+ cells express IL6, a B cell growth factor, but fail to proliferate. Neither differences in levels of expression of LMP1, a critical viral oncoprotein, nor differences in differentiation stages of B cells appear to govern this dichotomy in outcome. Since IL6 signals via STAT3, a molecule recognized for its pro-proliferative and anti-apoptotic functions, the roles on IL6, STAT3, and IL6-expressing CD23loCD58+ cells in EBV-driven B cell proliferation are the focus on ongoing research.
Susceptibility to EBV lytic cycle induction
While on the one hand, viral latency ensures persistence of EBV in the host, on the other hand, latently infected B cells must periodically undergo lytic replication to produce infectious virus that can infect new cells in the same host and spread to another host. However, a balance between maintenance of latency and exit from latency to support lytic replication is necessary to ensure that all cells do not support lytic replication simultaneously. Indeed when B cell lines latently infected with EBV are exposed to chemicals that induce the viral lytic cycle, greater than half of the population is resistant to lytic induction. Our goal is to identify host molecules and pathways that govern susceptibility or resistance to lytic induction by comparing total cellular gene expression patterns in cells susceptible to those in cells resistant to lytic induction. To perform these experiments, we first developed a method to identify and physically separate lytic cells from those refractory to lytic cycle induction following exposure of EBV-infected cell lines to lytic cycle inducing agents. This flow cytometry-based separation technique exploits the ability of EBV lytic protein-directed serum antibodies to detect EBV-infected cells supporting lytic replication. Such antibodies are routinely present in healthy individuals persistently infected with EBV. By probing human genome microarrays with RNA isolated from separated lytic and refractory cells derived from a Burkitt lymphoma cell line latently infected with EBV, we identified Stat3 as a candidate gene. Stat3 is of interest because of the association between constitutively activated (phosphorylated) STAT3 and several types of human cancers. We find that cells that are refractory to lytic cycle induction are marked by increased expression of Stat3 message and unphosphorylated STAT3 protein. Transcriptional targets of unphosphorylated STAT3 such as IL8, CCL5, and ICAM1 are also expressed at higher levels in refractory cells. We are investigating whether STAT3 is causal to maintenance of latency and whether other cellular proteins are linked to lytic cycle susceptibility or resistance. This line of investigation is relevant to development of EBV-LPD during immunocompromise as lytic replication is thought to precede lymphoproliferation in such patients. Additionally, cancer-therapeutic approaches that use antiviral agents to kill EBV-infected cancer cells following induction into the viral lytic cycle are not likely to be successful if a substantial fraction of cancer cells latently infected with EBV resist induction into the EBV lytic cycle.
EBV-directed immune response
Primary infection with EBV is a mucosal infection, at least in the initial stages, and is characterized by lytic EBV replication resulting in salivary shedding of infectious EBV particles. We asked whether IgA antibodies directed against EBV replication proteins are a marker for primary infection. Using a sensitive flow cytometry-based assay that we developed, we found that IgA antibodies directed against EBV lytic antigens are present in the sera of individuals undergoing primary EBV infection but not in healthy individuals persistently infected with EBV. These IgA antibodies, which react with both early and total lytic EBV antigens, are relatively transient compared to IgM and IgG antibodies to lytic antigens. The decline in early lytic antigen-directed IgA antibodies coincides with clinical improvement and highlights the importance of mucosal immune responses to components of the viral DNA replication machinery during the initial stages of primary EBV infection.
Cell mediated immune responses to EBV are critical for prevention of outgrowth of EBV-infected B cells into EBV-lymphomas. For modalities such as adoptive immunotherapy to be successful it is essential to determine the relative contributions made by different types of immune cells towards the total EBV-directed response, determine the frequencies of reactivated EBV-directed CD8+ and CD4+ T cells, and variation of such responses in healthy individuals. Using a short-term functional assay that we developed, we were able to examine immediate cellular immune responses to antigens presented by autologous EBV-LCL allowing direct comparison of variation of responses by different types of cells in the immune system. Our results demonstrate that CD8+ and CD4+ T cells (effector memory cells of both types) make the bulk of the response with each type contributing nearly equally towards the total EBV-directed immune response. In addition, γ-δ T cells also make a significant response. In contrast to existing dogma that EBV-directed CD8+ T cells are ten-fold more frequent than EBV-directed CD4+ T cells, we find that CD8+ T cells are less than two-fold more frequent than similarly reactive CD4+ T cells. These findings shed light on the identity and function of immune cells that keep EBV-infected B cells in check without allowing outgrowth into cancer in vivo. Of translational significance, our findings indicate that CD4+ T cells and γ-δ T cells are likely to be important components of reactivated T cell lines used in adoptive immunotherapy against EBV-lymphomas.
Using this experimental system, we are also able to clearly distinguish healthy individuals with EBV-directed cell mediated immune responses from individuals who are EBV-naïve and lack EBV-directed cell mediated immune responses, thus forming the basis for a potentially clinically useful diagnostic tool especially in the setting of immunosuppression.