The Role of Viral Proteins in Epstein-Barr Virus Induced Disease

From MicrobeWiki, the student-edited microbiology resource
Jump to: navigation, search
This is a curated page. Report corrections to Microbewiki.

By Jackson Cabo

Introduction

Figure 1. The original electron micrographs of cultured Burkitt’s Lymphoma tissue published by Epstein and Barr in 1964. V denotes the presence of EBV virions. [1].

The Epstein-Barr Virus (EBV)(Figure 1) is a common human herpes virus that can cause both infectious mononucleosis and lymphoproliferative disease. EBV is unique in that it infects about 95% of the adult population between 35-40 years old in the U.S. [1]. In most cases, EBV infectious cause no symptoms, particularly if the patient is infected as a child [1,8]. Infection during adolescence are associated with mononucleosis about half of the time [1]. EBV is associated with cancers such as Burkitt’s Lymphoma and nasopharyngeal carcinoma [1,2]. The virus is capable of infecting host B-cells and epithelia, and primarily proliferates via a non-lytic mechanism [2]. During this latent process, virus-derived nuclear proteins (EBNAs) and membrane proteins (LMPs) are expressed by infected host cells [2].

An advancing area of research is aimed at understanding how viral proteins may play a role in lymphoproliferative disease. LMP-1 is one of these viral membranous proteins that may induce indefinite, tumorigenic replication in infected B-cells [2,3]. In a sense, the LMP-1 acts to "transform" these B-cells into an immortalized, proliferating line. While EBV infections usually only cause mild symptoms, attempts to develop treatments and antivirals have generally been unsuccessful [1]. It is particularly difficult to control spread between hosts as chronic infections may reactivate, allowing symptom-free carriers to continue to transmit the virus years after initial infection [1]. While the latent life cycle appears to be the primary contributor to the incidence of lymphoproliferative disease in EBV infection, the lytic system makes a significant contribution. Lytic mechanisms are essential for horizontal spread of the virus, and several lytic cycle associated mechanisms contribute to both B-cell proliferation or success rates of infection [2,9,11]. In a sense, lytic mechanisms are employed for primary infection of humans, but once established in the body, latent mechanisms prevail in maintaining an infection in a host [2,22]. Also vital to EBV's success as a human pathogen is its ability to evade the host immune system. EBV proteins are involved in a series of signaling pathways that prevent detection by intracellular and extracellular components of the immune system. Due to EBV's putative role in carcinogenesis and the limited success in development of antivirals to combat infection, it is important to develop a deeper understanding of the mechanisms by which the virus alters the character of host cells, and evades human immune defenses. This review will focus on the activity of viral LMP-1 and EBNA1, and the roles they play in Epstein-Barr associated disease. Focus will also be placed on how the lytic and latent life cycles of EBV contribute to a successful infection. Some of the prevalent immune-avoidance techniques of EBV will be outlined. In context of these mechanisms, a brief outline of EBV pathology will be discussed.

Virus Classification and Genome

Figure 2. A diagram of a Epstein-Barr Virus dsDNA episome. Note the clustering of reading frames for related EBNA and LMP proteins. Note the location of the origin of replication, oriP. The green arrow denotes the direction of transcription during latency III. The short red arrow indicates the direction of transcription for EBNA1, which is activated during latency I and II. Source: [2]. .

The Epstein-Barr virus is a Baltimore Class I virus of the family Herpesviridae. It is a gammaretrovirus also designated as Human Herpes Virus 4 (HHV-4). M.A. Epstein and Yvonne Barr were the first identify EBV in tumor tissue associated with Burkitt's Lymphoma. The researchers used electron microscopy to determine that these viral particles were very similar in structure to Herpes simplex virions (Figure 1)[4]. The genome of EBV is composed of double-stranded DNA and is 172,282 base-pairs long [5]. This relatively long genome is characteristic of Herpes viruses; Herpes simplex has a 152-kilobase genome [6]. The open reading frames (ORFs) or EBV are generally broken up into separate lytic and latent sections [7]. While most of the viral genes encode proteins, some of the latent genes remain untranslated (EBERs: ENV encoded RNAs) and several micro-RNAs are coded for [7]. The relativity large dsDNA genome of the Epstein-Barr virus exhibits separation of ORF sections coding for genes associated with its two different replicative strategies.During latent infection, the EBV genome exists in a circularized form localized in the host cell nucleus [2]. The reading frames for the LMP and EBNA proteins are clustered separately within the episome (Figure 2).

EBV Virion Structure

Figure 3. A general diagram of a herpesvirus virion. Note that the dsDNA genome is wrapped around a central nucleo-protein. Spike glycoproteins (not labeled) on the surface play a role in host cell entry. [2].

Epstein and Barr observed that this lymphoma-associated virus was about 20% smaller than typical Herpes simplex virions [4]. Similar to other Herpesviruses, the innermost part of the EBV virion consists of DNA wrapped around a central nucleo-protein core (Figure 3) [8]. The core of the virion is surrounded by a nucleocapsid, a layer of protein tegument, and finally and outer envelope with spike glycoproteins [8]. Similar to many other viruses, many of these glycoproteins are vital for host-cell entry mechanisms. Infected host cells release EBV virions exclusively during the lytic cycle [7].


Lytic Life Cycle

While the EBV lytic life cycle is more rarely observed than the latent mode of replication, it is particularly important as it is the only way that the virus may make virions and be transferred horizontally between hosts (or cells) [8,9]. Immunosuppressive diseases like AIDS typically show increased free virion levels in the blood, a marker of increased lytic activity [9]. While virions are often found in the saliva of infected hosts, little to no lytic-infected cells are typically detected in the body [9]. Human cytotoxic T-cells are particularly adept at recognizing and destroying lytically infected cells expressing certain early stage lytic genes [10]. Hence, lytic activity appears to drive EBV spread in human populations. However, the latent replicative cycle is favored under normal conditions within a host, possibly as a means to evade host immune responses. Alpha and Beta herpesviruses have elaborate mechanisms for lytic gene concealment from the host immune system while EBV has few mechanisms to prevent immunomediated destruction of lytic-infected cells [10]. For example, Herpes simplex may be capable of inhibition of host Major Histocombatibility Complex, which reduces B-cell antigen presentation and recognition by cytotoxic T-cells [10]. In contrast to other herpesviruses, gammaherpesviruses such as EBV have a different mechanism for avoidance of host immune response. EBV primarily relies on a latent replication cycle in which its genome proliferates via clonal replication within dividing host B-cells [10]. Hence, unlike other herpesviruses that rely on lytic replication for spread within a host, EBV relies more on its latent mechanism. Virion production by EBV takes on a specific "niche role" vital for the initial infection of a new human host. Because latent mechanisms are responsible for sustained infections, EBV is not under significant selective pressure to "develop" elaborate lytic immune avoidance traits.

Figure 4. A diagram of EBV proteins involved in target cell binding and membrane fusion. Note that host cell binding and dusion appear to be distinct mechanisms, as fusion can occur without active binding ligands. The cell represents a typical EBV host, such as human B-cells. [3].

Latent activity is the primary contributor to the development of lymphoproliferative diesease (LPD) in EBV infected individuals. However, lytic-incompetent EBV strains are less effective generators of LPD associated lesioning than the wild-type [2,11]. In addition, lytic infection results in the secretion of paracrine factors that may stimulate the growth of latently infected B-cell lines [9, 11]. While it may have a more diffuse effect than the latent mechanism, EBV lytic infection mediated signaling may contribute to the development of lymphoproliferative disease (partially through enhancement of latent infections).

For the purposes of this review, focus will be placed on viral infection of B-lymphocytes. The lytic cycle begins with mature EBV virions reaching target host cells, such as B-cells. Contact between the virion and B-cell is initiated by the binding of EBV glycoprotein 350 to B-cell CD-21 [12]. For the most part, EBV infection is specific to cells expressing membranous CD-21 (namely B-cells and some epithelia), though some CD-21 independent attachment mechanisms are possible albeit with low efficacy [8]. EBV 350 is an example of a lytically expressed gene that is specifically targeted by the human immune system [13]. This gp350 binding is complemented by the binding of EBV gp42 to B-cell MHC-II [14]. In order for the viral envelope and B-cell membrane to fuse, the EBV virion must have functional gH, gL, and gp42 spike glycoproteins (Figure 4)[15]. There is also evidence that a third glycoprotein, gB, contributes to membrane fusion [21].

Figure 5. A diagram of EBV proteins (yellow), DNA (red), RNA (purple) transduced by virions to B-cell hosts. Human host factors are shown in blue. EBV RNA packaged within virions may contribute to successful infection, immune response evasion, and transition to a latent replicative mechanism. Hence this lytic mechanism may reinforce lymphoproliferative disease through supporting the latent mechanism. Green arrows indicate upregulation or stimulation, and red arrows denote inhibitory processes .





Once the gH/gL/gp42 complex penetrates the B-cell membrane, the viral capsid is dissolved, allowing the viral dsDNA genome to enter the cytoplasm. Interesting, viral RNA is often packaged in the virions in addition to the dsDNA genome. These transduced elements are referred to as tvRNA and include noncoding RNAs and mRNAs [16].

EBV tvRNA contribute to successful infection, cytokine signaling, and initiate transition towards a latent life cycle [16] (Figure 5). One of these mRNAs, BZLF1 (a transcription factor) helps to activate viral promoters that trigger a transition to a pre-latent phase [16]. Another virion carried RNA, BNLF2a transcript, contributes to the protection of the infected B-cells from cytotoxic T-cells [16]. While cytotoxic T-cells are less adept at detecting BNLF2a(+) EBV infected cells [16], the exact mechanism by which this tvRNA reduces B-cell recognition is not known.

In addition to tvRNA mediated B-cell alteration, viral tegument proteins also contribute to the progression of infection. EBV tegument released into the cytoplasm of host cells upregulates the transcription of IE (intermediate-early) genes which typically code for other transcription factors [9]. These factors subsequently upregulate genes involved in viral DNA replication [9]. Hence, virion transported viral RNA and tegument protein play immediate roles in altering B-cell character to ensure successful initial infection and immune system evasion. EBV tvRNA appear to directly contribute to transition from a lytic life cycle to a latent mode of proliferation.

In the lytic life cycle, viral dsDNA generally localizes to the nucleus for replication [17]. Viral dsDNA is replicated using host machinery, which stimulates the production of viral structural proteins. Viral particles are then assembled in the nucleus [9,18]. After full particles are assembled, they bud out of the nuclear membrane, then the cell membrane. When exiting the host cell, these virions may acquire their primary envelope from the nuclear membrane, and the outer envelope from the cell membrane [19]. Unlike many other herpesviruses, this exit mechanism does not initiate mandatory cell death [19,20].

In conclusion, the lytic life cycle of the Epstein-Barr virus plays a less central role in the maintenance of viral infection than what is seen in closely related herpesviruses. However, the lytic life cycle is indispensable to host-host spread and horizontal intercellular transmission. In Epstein-Barr, lytic immune response avoidance mechanisms are less robust than those seen in other herpesviruses. This reflects the fact that EBV relies more on clonal replication of infected B-cells (latent mediated) for maintenance of an infection within a host. Virion carried proteins and RNA change the character of infected B-cells so as to maximize chances of successful infection, and in some cases to initiate latent mechanisms. Interestingly, lytic virion budding by EBV is not very lethal to host cells, a finding that has potential applications to putative anti-EBV agents.

Latent Life Cycle

Figure 6. The spectrum of viral latent gene expression in EBV infected B-cells. Below the schematic for each latency pattern are the lymphoproliferative disorders associated with each type of expression. Source:[22]. [4].

While the lytic life cycle of EBV is vital to host-host transmission and had mechanisms that support successful infections within human hosts, the latent life cycle makes a more direct contribution to lymphoproliferative disease. In most cases, once EBV virions achieve primary infection of B-lymphocytes, the virus primarily replicates by a latent mechanism [2,16]. This results in the transformation of B-cells into ever-proliferating lymphoblastoid cell lines (LCL's) [2]. The latent phase in the EBV viral life cycle is defined by: 1)No production of virions, and 2) the production of a select few viral proteins and transcripts [22]. Many of these select latent viral proteins modulate the character of host B-cells and contribute to lymphoproliferation [2,22]. In contrast to the lytic cycle where the viral genome is brought to the nucleus and copied, under the lytic cycle infected B-cells contain several copies of extrachromosomal EBV episomes [2]. Hence, as the B-cells proliferate, so does the virus.

Latency can be divided into 3 distinct stages defined by which latent genes are being expressed (Figure 6). In latency I, only EBNA1 is expressed [22], while in latency II EBNA1 is expressed along with intermediate levels of other EBNAs and LMP proteins [2,22]. Latency II can be subdivided into IIa, where EBNA2, but not LMP1 is expressed, and IIb, where the expression of these two proteins is reversed [23].Latency III is characterized by unregulated expression of the complete latent viral proteome which includes EBNAs 1-6, as well as 3 LMP proteins (1,2A, and 2B)[22]. The transcription of the circularized viral genome begins at either the Wp or Cp promoter (Figure 2)[2]. Differential splicing of the same long transcript (in green, figure 2) yields the different EBNA mRNA. In vivo LCLs typically display latency III-like expression profiles [22]. B-cells must have latency III expression profiles for successful generation of LCLs in vitro [23]. In addition to these viral proteins, several non-coding RNAs (EBERs) and micro-RNAs are also produced during all three stages of latency [2,7,22,24].

A broad and continually evolving area of research is centered on determining how these latent viral factors transform their B-cell hosts. Do latent factor contribute to evasion of immune response? Do different latent expression profiles result in different malignancies? What are the signaling cascades behind viral latent protein modulation of B-cell character? Why is the EBV latent mode of replication so successful in sustaining long-lasting infections in human hosts? Questions such as these have been well-studied, revealing an expansive and intricate mode of B-cell manipulation by EBV.

Latent Mechanisms Contributing to Immune Response Evasion

Figure 7. (A) Immunostaining of human Nasopharyngeal Carcinoma cells transfected with a plasmid containing either EBNA1 or a CK-2 binding mutant EBNA1 (Δ387-394). Red staining is for EBNA1, green for PML and blue (control) is a DAPI stain. Note the significant reduction in PML levels between wild-type EBNA1 infected cells and CK-2 binding incompetent EBNA1 infected cells. (B) The number of PML-NBs detected per cell is significantly reduced in EBNA1 infected cells. This effect is abrogated when EBNA1 is not able to bind CK-2. OriP is an empty vector control. (C) A Western blot showing that PML protein levels are significantly lower in wild-type EBNA-1 infected cells as compared to CK-2 binding-incompetent strains (Δ387-394). The second EBNA1 mutant (Δ395-450) cannot bind human USP7 (Ubiquitin specific protease), did not appear to reduce PML levels. This suggests that EBNA1-CK-2 interaction is required for EBV mediated degradation of PML, while association with USP7 is not required. However, prior work indicates that association with USP7 is required for inhibition of PML-NB complexes. Source:[41]. [5].
Figure 8. Immunostaining of human Nasopharyngeal Carcinoma cells (CNE2E) that stably express EBV EBNA1 (and thus mimic Latency 1). Cell were treated with two siRNA to knockdown protein levels. siGFP was used as a control. siCK2α was used to inhibit the activity of the human kinase CK-2. CK2 is stained in red, and PML in green (with DAPI as a control). Inhibition of CK2 in EBNA1 positive cells shows abrogation of PML knockdown. This implicates CK2 as a driving force in EBNA1 mediated PML degradation. (F) Comparison of phosphorylation of PML in EBNA1 positive (CNE2E) and negative (CNE2) Nasopharyngeal Carcinoma cell lines. PML exists primarily in its phosphorylated form (p517 stained, red)in EBNA1 positive cells. Source:[41]. [6].

As noted above, once primary EBV infection establishes a small infected B-cell population in a host, further viral proliferation occurs primarily by way of a latent mechanism in which EBV is replicated in association with lymphoproliferation. Though the latent mechanism may afford a lower risk of detection by the host immune system than virion production, all EBV viral proteins are immunogenic [22]. As such, lymphoproliferative diseases associated with the high-expression type III latency occur primarily in immunocompromised hosts [22]. Type I and II latent infections may be successful in immunocompetent hosts due to lessened antigen presentation. Interestingly, EBNA1 further reduces antigen presentation by inhibiting host HLA-1 activity (human MHC) [25,26]. EBNA1 has long alanine-glycine repeats that prevent host proteases from breaking up the protein into antigens that can be presented by HLA-1[25,27]. EBNA1 is also capable of inhibiting its own synthesis [27]. This fine tuning prevents EBNA1 levels from becoming too high in host cells, which lowers the chance that an immune response will be triggered. Due to this mechanism, it appears that cytotoxic T-cells are generally incapable of EBNA1 recognition, which enhances the survival of type I EBV latent infections in humans [26].

The Epstein-Barr Virus also evades the host immune system through inhibition of some intracellular signaling pathways. An example of this is EBNA1 modulated inhibition of ProMyelocytic Leukaemia (PML) protein activity in Nasopharyngeal Carcinoma. PML plays a key role in triggering an intracellular anti-viral response and has been identified as a tumor-suppressor (possibly through an apoptotic mechanism)[27,42]. EBNA1 activity triggers the phosphorylation and degradation of PML complexes called PML Nuclear Bodies (PML-NBs)[43]. This phosphorylation triggers inhibition of PML via a complex ubiquitination pathway outlined by Sivachandran et al.[43]. Sivachandran et al. demonstrate that EBNA1 mediated inhibition of PML-NB activity is driven my the viral proteins interaction with a host kinase called CK-2. The researchers showed that EBV strains expressing EBNA1 with a mutant CK-2 binding domain were incapable of PML-NB degradation (Figure 7a,b). This indicates that this EBNA1-CK2 interaction is indispensable in EBNA1 driven anti-PML activity. What role does CK-2 play? Sivachandran also showed that siRNA knockdown (by siCK2α) of CK-2 abolishes the anti-PML activity of EBNA1 (Figure 8). Clearly, CK-2 is a key player in this pathway. Sivachandran et al. went on to find that PML, EBNA1, and USP7 colocalize in the nucleus of infected cells. Thus, they proposed a mechanism my which EBNA1 sequesters CK-2 to PML-NBs in the nucleus, which results in phosphorylation of these proteins. In fact, the researchers found that PML exists primarily in its phosphorlyated form in EBNA1 positive cells (Figure 8F). CK-2 phosphorylation of PML signals for the protein to be polyubiquinated, which targets it for digestion by proteases [43,44]. Interestingly, preliminary unpublished work by the same authors suggests that EBNA1 may alter USP7 activity in an independent anti-PML mechanism [43]. EBNA1 mediated inhibition of human PML-NBs provides a strong example of EBV's capability to hijack vital host intracellular signaling pathways. In this case, EBNA1 drives the degradation of a putative pro-apoptotic protein involved in immune response. EBV manipulates other signaling pathways as a means to curtail host immune responses. Latent infection may induce the transcription of c-myc oncogene, resulting in a reduction of host immune-stimulating signaling [28].

Suppression of apoptosis not only allows for a sustained infection, it also allows for avoidance of host immune response. Apoptosis not only deprives EBV of the host it needs to survive, it also triggers the release of "danger signals" including cytokines and viral particles that may be more easily detected by the human immune system [27,37,38]. One of the mechanisms by which EBV prevents apoptosis is through upregulation of existing apoptosis-suppressing pathways in host cells [27]. EBV upregulates the expression of anti-apoptosis proteins such as Bcl-2, A20, and Surivin [27]. Studies have shown that transfection of Burkitt Lymphoma cell lines with LMP-1 expressing vectors yields a marked increase in Bcl-2 expression[39]. Hence, it appears that LMP-1 may act as a stimulant of Bcl-2 activity that acts to prevent apoptosis in Latency II/III infected host cells. The Epstein-Barr Virus also produces its own Bcl-2 homolog, BHRF1, that contributes to apoptosis prevention during the lytic [40] and latent [41] life cycles. The mechanism by which Bcl-2 (or BHRF1) blocks apoptosis remains a subject of debate. However, molecular studies have determined that Bcl-2 (or relevant viral homologs) prevent cytochrome-c efflux from mitochondria, which in turn prevents the release of cell-destroying caspases [42]. Thus, EBV mediates Bcl-2 signaling in infected cells in order to prevent apoptosis, which in turn reduces the chance of detection by the immune system. Beyond modulation of host cell proteins or signaling pathways, EBV also can also moderate the expression of its own genes to dampen the ability of the immune system to detect the infection.

EBV may sustain successful infection even in the face of an extreme host immune response through adopting latency 0. During latency 0, the EBV genome is sequestered to the nucleus of infected cells and all expression is turned off, which essentially renders the virus undetectable [27]. This allows for the establishment of (non-dividing) EBV reservoirs in the host B-cells that can survive an immune response firestorm.

Figure 9. Comet assays showing an increases in DNA damage associated with EBNA1 activity. The BJAB-E1 (B-Cell line) is EBNA1 positive, while BJAB is EBNA1 (-). Comet assays quantify DNA fragmentation by using a fluorescent tag to mark DNA molecules. If a genome is broken up, pieces will break off during the electrophoresis process. The longer the "comet" the more fragmentation in the DNA. DNAse is used as a positive control. Source: [31]. [7].

Collectively, these mechanisms allow EBV infections to survive and evade immune defenses. The Epstein-Barr Virus evades immune reponse through modulation of existing cellular pathways genes as well as modulation of its own gene expression. A key area of focus appears to be inhibition of apoptosis and inhibition of antigen presentation. At its core, apoptosis is a anti-viral immune response as it prevents the virus from accessing the machinery it needs to sustain itself and proliferate. EBNA1 mediated inhibition of PML-NBs and LMP1 mediated upregulation of Bcl-2 are just two of many ways by which EBV attempts to prevent apoptosis. EBV also carefully regulates the expression of its own genes. Genes like EBNA1 can regulates their own expression. In addition, EBV can completely shut off its expression in Latency 0, and by doing so, "keeps a low profile." There a certainly tradeoffs to this strategy because while downregulation of expression reduces the chance of a successful immune response, it also reduces the ability of the virus to proliferate. While these mechanisms give us insight into how EBV fight human defenses, it is important to develop an understanding of how its latent viral proteins such as EBNA1 [31] and LMP1 [2,29,30] contribute to carcinogenesis.

EBNA1 and its Role in Lymphoproliferation

Figure 8. A ribbon model of the EBNA1 dimer associating with an EBV episome. Note the presence of the primary N-terminal DNA binding domain (in gold). The viral DNA is is green. Source: [33]. [8].
Figure 10. (A)DCDFA fluorescence assays showing an increases in ROS activity associated with EBNA1 activity. DCDFA fluoresces when oxidized, and is thus a good measure of ROS activity or oxidative stress. The BJAB-E1 (B-Cell line) is EBNA1 positive. The BJAB-tTAE1 strain expresses EBNA1 only in the absence of tetracycline. BJAB is EBNA1 (-). Note the abrogation of ROS induction by EBNA1 upon treatment with anti-NOX2 drugs Apocynin and DPI. (B) Similarly, comet assays indicate a significant reduction in DNA fragmentation in EBNA1 positive lines upon treatment with Apo or DPI. Source: [31]. [9].
Figure 11. A summary of the EBNA1-mediated oncogenesis cascade outlined by Gruhne et al.. [31] Note that it remains a subject of debate whether EBNA1 activity directly contributes to oncogenesis.

As noted above, EBNA1 is the only EBV viral protein expressed in all major types of latent infection, and is the only viral protein expressed in latency I infections [22,31]. EBNA1 is a viral nuclear antigen that binds specifically to the EBV episome in host cells. Interestingly, EBNA1 primarily exists in a dimeric form in vivo . EBNA1 has two primary domains: a DNA-specific domain that binds to the viral origin of replication (oriP), and a host-DNA tethering domain that associates with the human B-cell genome [32]. The crystal structure of EBNA1 has been resolved [33] and shows the presence of two unique binding domains (Figure 8). Interestingly, the N-terminal region of EBNA1 appears to wrap around a portion of the EBV DNA helix.

This binding between EBNA1 and the viral episome reflects a strong structure-function relationship. EBNA1 has been shown to act as an important "manager" of the EBV episome in latently infected cells. For example, EBNA1 is vital for the replication of the EBV episome in replicating B-cell lines [2,32,34]. Hence, EBNA1 appears to indirectly assist with carcinogenesis in that it ensure that the EBV genome is retained in proliferating B-cell lines. However, whether EBNA1 play a more direct role in oncogenesis remains a subject of debate.

Recently, a pair of mechanisms have been proposed by which EBNA1 may contribute to oncogenesis. Gruhne and her colleagues at the Karolinska Institute recently presented a study in which they provide evidence that EBNA1 may play a role in reactive oxygen species (ROS) stress in infected B-cells [31]. The authors suggest that this generation of ROS induces oxidative damage in the host cell genome, which has been shown in the past to support tumor formation and immortalization of B-cell lines [35]. First, Gruhne et al. used comet assays to demonstrate that genomic damage in B-cells was significantly higher in EBNA1 infected cells as compared to a control (Figure 9). Using a florescent marker for DNA, comet assays allow for quantification of DNA fragmentation on a cellular scale. These comets form upon separation of cell lysates using gel electrophoresis. The greater the degree of genomic fragmentation, the larger the "tail" of the comet will be. While prior studies had established a link between EBNA1 and host cell genomic damage, none had established a comprehensive mechanism by which the protein initiates this damage.

Gruhne makes a strong case that EBNA1 manipulates existing cellular components to drive genomic destabilization. The authors suggest that this modulation begins with EBNA1 mediated activation of seemingly innocuous, small protein. Specifically, Gruhne demonstrated that EBNA1 upregulates the levels of GTP bound Rac1 in host cells. Rac1 is a small G-protein found in human cells that must be activated by GTP to contribute to the formation of an activated NADPH oxidase called NOX2. Gruhne et al. demonstrated that EBNA1-positive B-cells showed elevated levels of active NOX2. NOX2 is membranous protein the generates the reactive oxygen species superoxide. Gruhne and her colleagues went on to show that NOX2 activity induced genomic damage in EBNA1-positive EBV infected B-cells [31]. The researchers used DCFDA to detect ROS activity because this molecule fluoresces when oxidized. DCFDA fluorescence intensity (and thus ROS activity) was higher in EBNA1-positive cell lines in comparison to a control (Figure 10A). Interestingly, when EBNA1-positive cells were treated with NOX2 inhibitors such as diphenylene iodonium (DPI) and apocynin (Apo), there was a significant reduction in ROS activity to near-control levels (Figure 10A). Hence, EBNA1 mediated upregulation of NOX2 generates ROS stress. Interestingly, this decrease in ROS activity via NOX2 inhibition is also accompanied by a reduction in genomic fragmentation. Gruhne and her colleagues used comet assays (Figure 10B) to confirm that DPI and Apo inhbition of NOX2 causes a reduction in genomic fragmentation.

Hence, the researchers diagram a cascade in which EBNA-1 activates a ROS generator NOX2, which in turn induces DNA damage to host B-cells (Figure 11). Overall, the researchers establish a strong link between EBNA1 mediated ROS production and genomic instability. However, it remains to be determined whether this instability is a causative agent of lymphoproliferation [34]. While Gruhne notes that DNA damage is ubiquitous in EBV positive tumor cells, it is unknown whether this evolved as a mechanism to ensure virus proliferation or whether this damage has some other role. Gruhne even goes so far as to suggest that this ROS mediated damage may be an "accident" in the evolutionary pathway of EBV [31]. Studies have shown that host NF-kB (involved in carcinogenesis, see next section) may be upregulated by genomic damage [48]. It appears that EBNA1 mediated DNA damage supports B-cell proliferation, which allows EBV to successfully and permanently colonize a host B-cell pool. Excessive replication of B-cells by this EBNA1 driven mechanism is generally kept in check in immune-competent individuals [31].

However, EBNA1 likely is not the primary contributor to oncogenesis and B-cell immortalization in most EBV infected hosts. While the evidence presented by Gruhne and her coworkers is certainly convincing, several studies have found that EBNA1 is not sufficient for oncogenesis. For example, Kang et al. found that high expression of EBNA1 in B-cells introduced to several mouse lineages failed to induce lymphoma [45]. In addition, it was been shown that some "immortal" B-cell lines are capable of proliferation without EBNA1, albeit at a slower rate [46]. Hence, it remains unclear how important EBNA1 is to B-cell transformation. However, since EBNA1 is the only viral protein expressed in Latency I malignancies such as Burkitt's Lymphoma, it is possible EBNA1 plays a central role in B-cell transformation in a select few cancers. While EBNA1 has been shown to induce DNA damage to B-cells [31,36], this effect is greatly magnified in Latency III expressing cells [36]. Thus, it is important to consider the role of the latency II/III protein LMP-1 in carcinogenesis.

Latent Membrane Protein 1 (LMP1) and Its Role in Lymphoproliferation

Figure 12. A summary of EBV LMP-1 interactions with human host proteins. Notice that many cascades terminate with stimulation of Nf-kB complex. EBV modulation of Nf-kB is one of the primary modes of carcinogensis induced by the virus. [10].
Figure 13. A nude mouse injected with LMP-1 expressing RAT-1 cells. Note the formation of tumors on the right leg and near the head. This provides evidence that LMP1 is a key factor in EBV mediated carcinogenesis. Source:[3]. [11].

Latent Membrane Protein 1 is perhaps one of the most-well studied latent genes expressed by the Epstein-Barr Virus. This protein is a potent inducer of lymphoproliferative disease through a variety of cascades involving interactions with host cell proteins. LMP1 is expressed on the membrane of infected B-cells during latency type II and III [22]. Interestingly, LMP1's activity mimics that of a constitutively active tumor necrosis factor receptor, CD40 [22,2,29]. The lymphoproliferative capabilities of LMP1 are primarily centered around its ability to mimic the signaling profile of a constitutively active CD40. Hence, LMP-1 modulates several complex signaling cascades within host cells, many of which contribute to cell proliferation, growth, or suppression of apoptosis [2](Figure 12).

How important is LMP1 to generation of lymphoblastoid cell lines? LMP1's carcinogneic activity was first described by Wang, Liebowitz, and Kieff [3]. The researchers found that Rat-1 cell lines constructed to express LMP-1 showed notable increases in proliferation rates and resistance to growth inhibition [3]. Further, injection of this cell line into nude mice induced tumor formation (Figure 13). In another early study, Kaye et al. found that B-cells inoculated with an EBV episome containing a mutant LMP1 protein were incapable of growth transformation indicative of EBV associated cancers [46]. B-Cell transformation was restored upon introduction with wild-type LMP1 via EBV P3HR-1 strain co-infection.

It is widely accepted that LMP-1 directly interacts with human Tumor Necrosis Factor Receptor associated proteins (TRAFs) [2]. These TRAFs bind the cytoplasmic carboxy terminus of LMP-1, which are termed C-terminal activation regions [2,30].The LMP1 protein may be subdivided into 3 subdomains: membrane spanning, cytoplasmic N-Terminal, and cytoplasmic C-terminal, all of which are required for induction of lymphoproliferation in vivo [30](Figure 12). LMP-1 manages complex signaling cascades by recruiting TRAFs to its CTAR-1 (C-terminal activation region) and TRADD (death-domain containing protein)[2]. TRADD contributes to indirect LMP1-TRAF signaling as it binds TRAF2 [2]. Hence, while the majority to TRAF signaling by LMP1 is localized to CTAR1, indirect signaling can occur indirectly at CTAR-2. LMP1 recruitment of TRAFs triggers a host of pathways that ultimately contribute to blocking of apoptosis and initiation of cell growth and division.

One of the primary pathways by which LMP1 mediates lymphoproliferation is through upregulation of the human transcription factor NF-κB. Chemical inhibition of NF-κB in some proliferating lymphocyte lines has been shown to abolish rapid cell growth and division [47]. NF-κB appears to drive the establishment of inflammation-associated tumor growth and proliferation [48]. NF-κB appears to have a diverse set of functions, primarily involving inflammation, growth induction, and immune response to stimuli such as cytokines, free radicals, and viral infection [49]. NF-κB has also been linked to inflammatory response in the cardiovascular system [49]. It is possible that EBV "hijacking" of NF-κB expression may magnify its normal growth induction function role, resulting in cancer. How does LMP1 manipulate signaling pathways to upregulate Nf-kB expression?

As noted above, LMP1-Nf-kB signaling pathways begin with TRAF interactions with the CTAR-1 region of LMP1. Multiple LMP1 proteins may aggregate together on the membrane to present multiple cytoplasmic domains in one area, which allows for more efficient LMP1-TRAF interactions [50]. Mutation of the CTAR-1 region inhibits LMP1-TRAF binding and is associated with a significant reduction in the ability of LMP1 to induce lymphoproliferation (13.9% of wild-type B-cell proliferation induction) [51]. In EBV infected B-cells, most TRAF1 and 3 exists in direct association with LMP1, while TRAF 3 less frequently binds to LMP1 [52]. Hence, LMP1-TRAF1 association appears top be the first association in the primary cascade used by EBV to transform human B-cells into immortal lymphoblastoid lines.

Beyond this first step, the LMP1-NF-kB cascade becomes a little less clear. There are multiple proposed mechanisms by which LMP1 activated TRAFs ultimately signal for NF-kB activation [53]. A widely accepted pathway is diagrammed in figure 12. By this pathway, LMP1 activated TRAFs recruit a host protein called NIK (NF-κB-inducing kinase). It has been suggested that aggregation of multiple TRAF2 to TRAF1 or TRADD bound to LMP1 drives activation of NIK [54](Figure 13). NIK then phosphorylates IKKα, another human kinase. IKKα subsequently phosphorylates IκBα [54]. IκBα normally inhibits NF-kB by sequestering it to the cytoplasm [54]. After it is phosphorylated by IKKα, IκBα is ubiquitinated and degraded, which allows NF-kB (as a p65-p50 heterodimer) to enter the nucleus and begin modulating transcription [2,54]. Another well-accepted mechanism is the processing of host p100 (a precursor) to NK-kB via IKKα/NIK activity (Figure 12). While these mechanisms are very complex and new intricacies continue to be discovered, the main "take away" point is that the primary mode of EBV-initiated lymphoproliferation is a LMP-1 initiated cascade that terminates with the upregulation of NF-kB. It is important to note that LMP1 also upregulates other putative tumorigenic signaling pathways that do not involve NF-kB, such as the JNK kinase pathway [55](Figure 12).

However, EBV may induce oncogenesis even without LMP1, albeit with low efficacy. Complete knockout of LMP1 reduces the ability of EBV to transform B-cells to 1.2% (significantly non-0) of control levels [51]. This provides direct evidence that LMP-1 is not the only carcinogenic factor encoded by EBV, and "backup" mechanisms exist.

In conclusion, EBV associated lymphoproliferative disease appears to be largely driven by LMP1 hijacking of host signaling pathways. The direct oncogenic effect of EBNA1 remains of subject of debate, but its role as a manager of the viral episome in sustained latent infection is well-established. Numerous lytic and latent EBV factors contribute to immune system evasion. An important mechanism is the prevention of apoptosis by viral proteins like LMP1. While an understanding of the molecular mechanisms behind EBV-associated cancer is important, it is essential to address current questions facing today's medical professionals. How do EBV infections manifest? Who is at risk?

EBV Pathology

Figure 14. A child suffering from Burkitt's Lymphoma. [12].

Epstein-Barr infections are particularly hard to understand as the vast majority of them are largely asymptomatic [23]. However, terrible diseases such as nasopharyngeal carcinoma, Burkitt's Lymphoma, and gastric carcinoma may be also be linked to EBV activity. Mononucleosis is a sort of middle ground, that is typically can be left untreated but occasionally causes life-threatening complications. In fact over 90% of the world's adult population harbors some kind of EBV infection [1]. If so many people are infected, why are cancer rates not sky high?

The answer may lie in EBV's survival strategy. Without its host, EBV will die, thus it has an evolutionary incentive to keep its host alive. For the most part, EBV carriers have infections that fall into lower-stange latencies, meaning that few viral proteins are expressed [23]. This means that while the viral episome will continue to survive in host cells, it is not expressing a lot of proteins to avoid detection by the immune system. Low-activity infections such as these typically are asymptomatic [23]. Cancers tend to show up primarily in immunosuppressed individuals: those suffering from AIDS, malnutrition, or other infections [23]. Most of these infections typically adopt a latency III expression profile and are much more common in the developing world. Recent transplant patients are also at high risk for developing EBV-associated cancers [2,31]. Interestingly, this effect may be due to a synergy between EBNA1 activity and activity of immunosuppressive drugs used during the transplant [31]. Some immunosuppressive drugs used in these surgeries such as cyclosporin A generate ROS stress [31]. This could create a situation in which the drug compounds EBNA1 induced DNA damage, theoretically contributing to B-cell transformation.

Most EBV-associated cancers are identified by biopsies of tumor tissue using probes for viral proteins. Monoclonal antibodies used in conjunction with fluorescence microscopy have allowed medical professionals to more easily identify different types of EBV-associated cancers in recent years [56].

Burkitt's Lymphoma is perhaps the most interesting B-cell EBV-associated cancer as it has notable prevalence in both the "developing" and "developed" world. "Endemic" Burkitt's lymphoma is highly prevalent in sub-Saharan Africa particularly in regions that exhibit very high rates of malaria [2]. EBV is found in close to 100% of these cases. The cancer can manifest as large tumors typically located near the mouth (Figure 14)[2,57]. It is possible that high rates of malaria contribute to high rates of immunosuppression in these populations. Interestingly, Burkitt's lymphoma also is highly prevalent in AIDS patients in the first world, with EBV found in about 40% of cases [2].

References

[1]"Epstein-Barr Virus and Infectious Mononucleosis." Centers for Disease Control. http://www.cdc.gov/ncidod/diseases/ebv.htm. Accessed: 11/3/12.

[2]Young LS,Rickinson AB. 2004. Epstein-Barr Virus: 40 Years On. Nature Reviews. 4:757-768.

[3]Wang D, Liebowitz D, Kieff E. 1985. An EBV Membrane Protein Expressed in Immortalized Lymphocytes Transforms Established Rodent Cells. Cell. 43: 831-840.

[4]Epstein MA, Achong BG, Barr YM. 1964. Virus Particles In Cultured Lymphoblasts from Burkitt's Lymphoma. The Lancet. 283:702-703.

[5] Baer R, Bankier AT, Biggin MD, Deininger DL, Farrell PJ, Gibson TJ, Hatfull G, Hudson GS, Satchwell SC, Seguin C, Tuffnell PS, Barrell BG. 1984. DNA sequence and expression of the B95-8 Epstein-Barr Virus Genome. Nature. 310: 207-211.

[6] Mahiet C, Ergani A, Huot N, Alende N, Azough A, Salvaire F, Bensimon A, Conseiller E, Wain-Hobson S, Labetoulle M, Berradeau S. 2012. Structural Variability of the Herpes Simplex Virus 1 Genome in Vitro and In Vivo. Journal of Virology. 86(16): 8592-8601.

[7] "Epstein-Barr Virus." International Agency for Research on Cancer. IARC Monographs. 1997. 49-92. http://monographs.iarc.fr/ENG/Monographs/vol100B/mono100B-6.pdf. Accessed: 11/3/12.

[8]Liebowitz D, Kieff E. 1993. Epstein-Barr virus. In: The Human Herpesvirus. Roizman B, Whitley RJ, Lopez C, editors. New York. 107–172.

[9] Swaminathan S, Kenney S. "The Epstein-Barr Virus Lytic Life Cycle." 2009. In: DNA Tumor Viruses. 285-315.

[10] Steven, N. M., Annels, N. E., Kumar, A., Leese, A. M., Kurilla, M. G. and Rickinson, A. B. 1997. Immediate early and early lytic cycle proteins are frequent targets of the EpsteinBarr virus-induced cytotoxic T cell response. J Exp Med. 185: 1605–1617.

[11] Hong, G. K., Gulley, M. L., Feng, W. H., Delecluse, H. J., Holley-Guthrie, E. and Kenney, S.C. 2005. Epstein-Barr virus lytic infection contributes to lymphoproliferative disease in a SCID mouse model. Journal of Virology. 79: 13993–14003.

[12] Nemerow, G. R., Mold, C., Schwend, V. K., Tollefson, V. Cooper, N. R. 1987. Identification of gp350 as the viral glycoprotein mediating attachment of Epstein–Barr virus(EBV) to the EBV/C3d receptor of B cells: sequence homology of gp350 and C3 complment fragment C3d. J. Virol. 61: 1416–1420.

[13] Khyatti M, Patel PC, Stefanescu I, Menezes J. 1991. Epstein-Barr virus (EBV) glycoprotein gp350 expressed on transfected cells resistant to natural killer cell activity serves as a target antigen for EBV-specific antibody-dependent cellular cytotoxicity. Journal of Virology. 65(2): 996-1001.

[14] Borza, C. M. Hutt-Fletcher, L. M. 2002. Alternate replication in B-cells and epithelial cells switches tropism of Epstein–Barr virus. Nature Med. 8: 594–599.

[15] Kirschner AN, Omerovic J, Popov B, Longnecker R, Jardetzky TS. 2006. Soluble Epstein-Barr Virus Glycoproteins gH, gL, and gp42 Form a 1:1:1 Stable Complex That Acts Like Soluble gp42 in B-Cell Fusion but Not in Epithelial Cell Fusion. Journal of Virology. 80(19): 9444-9454.

[16]Jochum S, Ruiss R, Moosman A, Hammershmidt W, Zeidler R. 2012. RNAs in Epstein-Barr virions control early steps of infection. PNAS. 109(21): 7970-7971.

[17] Daikoku T, Kudoh A, Fujita M, et al. 2005. Architecture of replication compartments formed during Epstein-Barr virus lytic replication. Journal of Virology.79(6):3409–3418.

[18] Henson BW, Perkins EM, Cothran JE, Desai P. 2009. Self-assembly of Epstein-Barr virus capsids. J Virol. 83:3877–90.

[19]Gong M, Kieff E. 1990. Intracellular trafficking of two major Epstein-Barr virus glycoproteins, gp350/220 and gp110. Journal of Virology. 64(4):1507-1516.

[20] "Epstein-Barr Virus." http://virology-online.com/viruses/EBV.htm Accessed: 11/10/12.

[21] Spear PG, Longnecker R. 2003. Herpesvirus Entry: An Update. Journal of Virology. 77(19): 10179.

[22] Cesarman E. 2011. Gammaherpesvirus and lymphoproliferative disorders in immunocompromised patients.Cancer Letters. 305: 163-174.

[23] E Klein, LL Kis, G Klein. 2007. Epstein-Barr virus infection in humans: from harmless to life endangering virus-lymphocyte interactions.Oncogene. 26:1297–1305.

[24] Fok V, Friend K, Seitz JA. 2006. Epstein-Barr virus noncoding RNAs are confined to the nucleus, whereas their partner, the human La protein, undergoes nucleocytoplasmic shuttling. Journal of Cell Biology. 173(3): 319-325.

[25] Apcher A, Daskalogianni C, Manoury B, Fahraeus R. 2010. Epstein Barr Virus-Encoded EBNA1 Interference with MHC Class I Antigen Presentation Reveals a Close Correlation between mRNA Translation Initiation and Antigen Presentation. PLoS Pathogens. 6(10): e1001151.

[26] Hochberg D, Middeldorp JM, Catalina M, Sullivan JL, Luzuriaga K, Thorley-Lawson DA. 2004. Demonstration of the Burkitt's lymphoma Epstein-Barr virus phenotype in dividing latently infected memory cells in vivo. PNAS. 101(1): 239-244.

[27] Ning S. 2001. Innate immune modulation in EBV infection. Herpesviridae. 2:1.

[28]God JM, Haque A. 2010. Burkitt lymphoma: pathogenesis and immune evasion, J. Oncol.

[29] Gires O, Zimber-Strobl U, Gonnella R, Ueffing M, Marschall G, Zeidler R, Pich D, Hammerschmidt W. 1997.Latent membrane protein 1 of Epstein–Barr virus mimics a constitutively active receptor molecule. The EMBO Journal. 16(20): 6131–6140.

[30] Brodeur SR, Cheng G, Baltimore D, Thorley-Lawson DA. 1997. Localization of the Major NF-kB-activating Site and the Sole TRAF3 Binding Site of LMP-1 Defines Two Distinct Signaling Motifs. Journal of Biological Chemistry. 272(32): 19777–19784.

[31]Gruhne B, Sompalle R, Marescotti D, Kamranvar SA, Gastaldello S, Masucci MG. 2009. The Epstein-Barr virus nuclear antigen-1 promotes genomic instability via induction of reactive oxygen species. PNAS. 106(7): 2313-2318.

[32] Saha A, Robertson E. 2011. Epstein-Barr Virus–Associated B-cell Lymphomas: Pathogenesis and Clinical Outcomes. Clinical Cancer Research. 17(10): 3056-3063.

[33] Bochkarev A, Barwell JA, Pueftzner RA, Furey Jr. W, Edwards AM, Frappier L. 1995. Crystal Structure of the DNA-Binding Domain of the Epstein-Barr Virus Origin-Binding Protein EBNA1. Cell. 83: 39-46.

[34] Schulz TF, Cordes S. 2009. Is the Epstein-Barr virus EBNA-1 protein an oncogen? PNAS. 106(7): 2091-2092.

[35] Ranjan D, Siquijor A, Johnston TD, Wu G, Nagabhuskahn M. 1998. The effect of curcumin on human B-cell immortalization by Epstein–Barr virus. Am Surg. 64:47–51.

[36] Kamranvar SA, Gruhne B, Szeles A, Masucci MG. 2007. Epstein–Barr virus promotes genomic instability in Burkitt's lymphoma. Oncogene. 26:5115–5123.

[37]Everett H, McFadden G. 2001. Viruses and Apoptosis: Meddling with Mitochondria. Virology. 288: 1-7.

[38]Hockenbery E, Nunez G, Milliman C, Schreiber RD, Korsmeyer SJ. 1990. Bcl-2 is an inner mitochondrial membrane protein that blocks programmed cell death. Nature. 348: 334-336.

[39]Henderson S, Rowe M, Gregory C, Croom-Carter D, Wang F, Longneckr R, Kieff E, and Rickinson A. 1991. Induction of bcl-2 Expression by Epstein-Barr Virus Latent Membrane Protein 1 Protects Infected B Cells from Programmed Cell Death. Cell. 65: 1107-1115.

[40] Henderson S, Huen D, Rowe M, Dawson C, Johnson G, Rickinson A. 1993. Epstein-Barr virus-coded BHRF1 protein, a viral homologue of Bcl-2, protects human B cells from programmed cell death. PNAS. 90: . 8479-8483.

[41] Oudejans JJ, van den Brule AJ, Jiwa MN, de Bruin PC, Ossenkoppele GJ, van der Valk P, Walboomers JM, Meijer CJ. 1995. BHRF1, the Epstein-Barr virus (EBV) homologue of the BCL-2 protooncogene, is transcribed in EBV-associated B-cell lymphomas and in reactive lymphocytes. Blood. 86(5): 1893-1902.

[42] Yang J, Liu X, Bhalla K, Kim CN, Ibrado AM, Cai J, Peng TI, Jones DP, Wang X. 1997. Prevention of Apoptosis by Bcl-2: Release of Cytochrome c from Mitochondria Blocked. Science. 275: 1129-1132.

[43] Sivachandran N, Cao JY, Frappier L. 2010. Epstein-Barr Virus Nuclear Antigen 1 Hijacks the Host Kinase CK2 to Disrupt PML Nuclear Bodies. Journal of Virology. 84(21): 11113.

[44] Scaglioni, P. P., T. M. Yung, L. F. Cai, H. Erdjument-Bromage, A. J. Kaufman, B. Singh, J. Teruya-Feldstein, P. Tempst, and P. P. Pandolfi. 2006. A CK2-dependent mechanism for degradation of the PML tumor suppressor. Cell 126:269-283.

[45]Kang MS, Lu H, Yasui T, Sharpe A, Warren H, Cahir-McFarland E, Bronson R, Hung SC, Kieff E.2005.Epstein–Barr virus nuclear antigen 1 does not induce lymphoma in transgenic FVB mice. PNAS. 102:820–825.

[46] Kaye KM, Izumi KM, Kieff E. 1993. Epstein-Barr Virus latent membrane protein 1 is essential for B-lymphocyte growth transformation. PNAS. 90: 91.50-9154.

[47] Cavallini L, Francesconi MA, Zoccarato F, Alexandre A. 2001. Involvement of nuclear factor-kappa B (NF-κB) activation in mitogen-induced lymphocyte proliferation: inhibitory effects of lymphoproliferation by salicylates acting as NF-κB inhibitors. Biochemical Pharmacology. 62: 141-147.

[48] Karin M, Greten FR. 2005. NF-kB: linking inflammation and immunity to cancer development and progression. Nature Reviews Immunology. 5: 749-759.

[49] Brasier AR. 2006. The NF-kB Regulatory Network. Cardiovascular Toxicology. 6: 111-130.

[50] Mosialos G, Birkenbach M, Yalamanchili R, Van ArsDale T, Ware C, Kieff E. 1995. The Epstein-Barr Virus Transforming Protein LMP1 Engages Signaling Proteins for the Tumor Necrosis Factor Receptor Family. Cell. 80: 389-399.

[51] Dirmeier U, Neuhierl, Kilger E, Reisbach G, Sandberg ML, Hammerschmidt W. 2003. Latent Membrane Protein 1 Is Critical for Efficient Growth Transformation of Human B Cells by Epstein-Barr Virus. Cancer Research. 63: 2982-2989.

[52] Devergne O, Hatzivassilou E, Izumi KM, Kaye KM, Kleijnen MF, Kieff E, Mosialos G. 1996. Activation of TRAF1, TRAF2, and TRAF3 with an Epstein-Barr Virus LMP1 Domain Important for B-Lymphocyte Transformation: Role in NF-kB Activation. Molecular and Cellular Biology. 7098-7108.

[53] Luftig M, Yasui T, Kang MS, Jacobson N, Cahir-McFarland E, Seed B, Kieff E. 2003. Epstein-Barr virus latent infection membrane protein 1 TRAF-binding site induces NIK/IKKα-dependent noncanonical Nf-kB activation. PNAS. 101(1): 141-146.

[54] Sylla BS, Hung SC, Davidson DM, Hatzivassiliou E, Malinin NL, Wallach D, Gilmore TD, Kieff E, Mosialos G. 1998. Epstein–Barr virus-transforming protein latent infection membrane protein 1 activates transcription factor NF-kB through a pathway that includes the NF-kB-inducing kinase and the IkB kinases IKKa and IKKb. PNAS. 95: 10106-10111.

[55] Liljeholm S, Hughes K, Grundstrom T, Brodin P. 1998. NF-kB only partially mediates Epstein-Barr virus latent membrane protein 1 activation of B-cells. Journal of general Virology. 79: 2117-2125.

[56] Lawrence Young, Ph.D., Caroline Alfieri, Ph.D., Kevin Hennessy, Ph.D., Helen Evans, Carl O'Hara, M.D., Kenneth C. Anderson, M.D., Jerome Ritz, M.D., Ralph S. Shapiro, M.D., Alan Rickinson, Ph.D., Elliott Kieff, M.D., Ph.D., and Jeffrey I. Cohen, M.D. 1989. Expression of Epstein–Barr Virus Transformation–Associated Genes in Tissues of Patients with EBV Lymphoproliferative Disease. New England Journal of Medicine. 321:1080-1085.

[57] http://en.wikipedia.org/wiki/Burkitt's_lymphoma. Accessed: 12/10/12.