The pathogenesis of Epstein-Barr virus persistent infection

Abstract

Epstein-Barr virus (EBV) maintains a lifelong infection. According to the germinal center model (GCM), latently infected B cells transit the germinal center (GC) to become resting memory cells. Here, the virus resides quiescently, occasionally reactivating to infect new B cells, completing the cycle of infection. The GCM remains the only model that explains EBV biology and the pathogenesis of lymphoma. Recent work suggests modifications to the model notably that the virus contributes only modestly to the GC process and predictions from mathematical models that quiescence within memory B cells shapes the overall structure of viral infection but is not essential for persistence. Rather, it is the cycle of infection which allows viral persistence at the very low levels observed.

Figure 1

The Germinal Center Model (GCM) of EBV persistence. Infectious virus enters the lymphoid tissue of Waldeyer's ring and then crosses the epithelial barrier where it directly infects naive B cells, activating them into proliferating latently infected Blasts expressing all nine known latent proteins (the growth transcription program). These cells then move into the germinal center (GC) to participate in the GC reaction. Here they express a more restricted pattern of latent proteins, the default program. Eventually these cells leave as latently infected memory B cells that either only express the viral genome tethering protein EBNA1 (the EBNA1 only program) or no viral proteins at all (the latency program). The memory compartment has been considered the site of long-term persistence because the virus is quiescent [42] and therefore invisible to the immune response. At any time a small subset of latently infected memory B cells initiates lytic reactivation in association with terminal differentiation signals [5,43]. Reactivation of the virus is subdivided into three phases; Immediate early when the transcription factors initiating viral replication are expressed, Early when the proteins involved in viral DNA replication are produced, and Late when viral DNA and structural proteins are assembled into virions [44]. This process results in the release of infectious virus that may be shed into saliva for infectious spread or infect new naive B cells, thus completing the cycle.

Figure 2

Diagrammatical Representation of the Cyclic Pathogen Model (CPM). CPM is a mathematical description of the GCM. The CPM consists of a cycle of infected stages (blue circles based on the biological GCM illustrated in Figure 1.) For CPM there are 6 stages: Blast, GC, Memory, Immediate Early, Early and Late infected B cells, each of which is potentially controlled by a distinct CTL response (red circles). Note that the single lytic stage in the GCM is broken down into three discrete stages which are known to be recognized independently by the immune response. Note also that under biological conditions the late [45] and GC [46] stages are not always recognized by CTL and there is never a CTL response against the memory stage however the model allows analysis of theoretical conditions for example where the memory compartment is regulated by CTL. Each stage progresses to the next stage (blue arrows). Late lytic B cells release free virions which produce new infected Blasts. The latter has an amplification factor since loss of a single Late lytic cell produces many infected Blasts (small blue circle). Each infected population may have an inherent rate of proliferation or death (double green arrows). Each stage promotes proliferation of its cognate CTL population (if present) (curved blue arrows) and is in turn controlled by those CTLs (red inhibitory arrows). These CTLs (if present) have an inherent rate of loss in the absence of stimulation (not shown in diagram). B. Equations of the CPM. For each stage there are two equations governing the rate of change of the infected population S_i and the CTL population T_i for a total of 12 equations. Solution of these equations at steady state reveals one and only one solution that is stable and biologically credible. The parameters in these equations give the rates of the processes described above. Except for the rate of loss of CTLs (b), these parameters are stage-specific. C. Correspondence between the parameters of the equations in B. and processes shown in diagram A