Epstein–Barr virus (EBV) could be considered as a component of the human microbiome as a consequence of its roughly sustaining 95% of the adult populations worldwide, and majority of it persists lifelong as an asymptomatic carrier (Dreyfus, 2013; Young et al., 2016; Connolly et al., 2021). In addition, EBV is the first identified human oncogenic virus (De Martel et al., 2020) which was discovered 50 years ago (Young et al., 2016). The importance of EBV reactivation is emphasized in the progression of EBV-driven carcinogenesis since the antibodies for capsid antigen and EBV-DNase of EBV were observed to be increased prior to the tumorigenesis in nasopharyngeal carcinoma (NPC) (Chien et al., 2001). Therefore, factors which might induce the EBV reactivation will be emphasized in this review for the pathogenesis of EBV-driven cancers.

The commensal microbiome refers to the diverse microorganisms, which consist mainly of bacteria and virus, as well as other components such as archaea, fungi, and protozoa that colonize barrier surfaces of different niches of mammals, such as the skin, vaginal, upper respiratory, and gastrointestinal tracts (Lloyd-Price et al., 2016; Barko et al., 2018). For the past few years, microbiome studies focus mainly on the composition and function of bacteria and archaea (Carding et al., 2017; Nikolich-Zugich et al., 2017). However, recent virome studies emphasized that viruses, which are abundant in divergent tissues (such as oral cavity, skin, gut, and blood) as well as in the feces of individuals in sickness and in health (Norman et al., 2015; Neil and Cadwell, 2018; Schmidt, 2018; Clooney et al., 2019), are the largest proportion of the human microbiome instead of bacteria (Wylie et al., 2012). The virome of human, which consists of diverse viruses that could infect not only eukaryotic cells but also prokaryotic cells (Handley, 2016), is an important factor in host health and diseases (Cadwell, 2015).

As EBV is sustained in almost all adult humans, it is puzzling why only a few of them evolved to induce malignant transformation. Some cooperative triggers must occur in the tumorigenesis of EBV-driven cancers. In the 1960s, the role of the commensal microbiome in modulating virus infections was first suggested (Robinson and Pfeiffer, 2014). The susceptibility of host to virus infection was enhanced in germ-free (GF) mice. Over the past few years, host–microbiota interactions have been reported to be fundamental for the regulation and maintenance of the mammalian immune system (Belkaid and Hand, 2014; Belkaid and Harrison, 2017; Rowan-Nash et al., 2019), which could be the implication for how we understand the persistence and reactivation of EBV, as well as the progression of EBV-associated cancers. In the present review, we aimed to discuss the impact of the microbiome on EBV infection. In addition, we focus on how the microbiome reactivates latently infected EBV and pertain to the etiology of EBV-driven cancers in patients who could be with asymptomatic life-long infection.

EBV is a globally spread virus that infects and persists lifelong in about 95% of the world’s population (Torniainen-Holm et al., 2018). The life cycle of EBV includes a latent infection phase, during which the virus persists by attaching to the host chromosomes, and a lytic replication phase, predominantly occurring in oropharyngeal epithelial cells (Hochberg et al., 2004) (shown as Figure 1 ). During the latency, a few EBV viral promotors are dynamically regulated, and the latency of EBV is classified from latency 0 to 3 according to the differential expressed sets of viral genes in a cell-dependent manner (Woisetschlaeger et al., 1990; Ohga et al., 2002). The latent EBV genomes are spontaneously reactivated by various stimuli, including pathogenic infections and other commensal microbiomes. The virion episome is spliced to linear and released from infected cells during the transformation of EBV latency into a lytic/replicative phase.

EBV harbors high tumorigenic potential preferentially infecting B-cells, T-cells, natural killer (NK) lymphocytes, and epithelial cells, steadily promoting the uncontrolled proliferation of infected cells (Miller and Lipman, 1973a; Miller and Lipman, 1973b), leading to a wide spectrum of EBV-positive cancers (Cai et al., 2015). According to the EBV-infected cell type involved in the tumorigenesis, the divergent EBV-driven cancer could be grouped into three mainly groups, including lymphoproliferative disorders (LPD) (Heslop, 2020), epithelial malignancies (Yamaguchi et al., 2018), and lymphoepithelioma-like carcinoma (LELC) (Ose et al., 2021) ( Table 1 ).

The latency of EBV has been grouped according to the immune status of the patient and expression of EBV proteins (Carbone et al., 2008; Taylor et al., 2015), and the differences among these latency groups determine the treatment responses of EBV. Latency 0 only expresses the EBV-encoded messenger RNA (EBER) and BamH1-a rightward reading frame transcript (BART) in cells of healthy previously infected individuals. In Burkitt lymphoma, gastric cancer, HIV-associated diffuse large B-cell lymphoma (HIV-DLBCL), and approximately two-thirds of nasopharyngeal carcinoma, EBV exists in a latency 1 pattern and only expresses EBV-nuclear antigen 1 (EBNA1), EBER, and BARTs (Tempera and Lieberman, 2014), thereby evading immune responses to EBV (Burkitt, 1958; Arvey et al., 2015). Latency 3, which persists only in severely immunocompromised hosts such as EBV-positive post-transplant lymphoproliferative disorder (PTLD), expresses all virus-specific latent nuclear antigens and membrane proteins (Gottschalk et al., 2005; Lacasce, 2006). For latency 2, late membrane protein-1 (LMP1) is expressed in addition on the basis of latency 1, and this phase has been observed in Hodgkin, some nasopharyngeal carcinoma, and so on (Sekihara et al., 2014). As a consequence of the restricted antigen expression, EBV-CTLs (Bollard et al., 2014) which act well in latency 2 fail in latency 1 tumors.

During the viral infections, many viruses would enter the latency state with their genetic materials integrated into the genome of the infected cells (Nikolich-Zugich et al., 2017) and regulate the host immune system (Virgin, 2014). Viral infections are reported to modulate the host immune system by stimulating the production and release of a variety of cytokines, including interferon and other cytokines such as interleukin (IL)-10 (Nikolich-Zugich et al., 2017). As for the role of EBV in infected hosts, EBV and related herpesvirus are assumed to encode a copy of human cytokines that regulate the host immune system to satisfy their own colonization and expansion needs (Irons and Le, 2008; Slobedman et al., 2009). A human IL-10-like protein that acts in the TH2 family was reported to be encoded by EBV (Hsu et al., 1990; Macneil et al., 1990; Moore et al., 1990). The LMP1 expression of EBV could act as the tumor necrosis factor (TNF) receptors to transmit the growth signals through TNF-receptor-associated factors (TRAFs) (Liebowitz, 1998) and thus is vital for the EBV-induced tumorigenesis (Iwakiri et al., 2013). EBER was reported to act as a substitute of interferon and induce the expression of IL-9/10 and insulin-like growth factor, resulting in the promotion of EBV-infected cells growth (Kitagawa et al., 2000; Iwakiri et al., 2003; Yang et al., 2004; Iwakiri et al., 2005).

It is vital to elucidate the mechanisms involved in the EBV reactivation since the chronic EBV reactivation is considered as a key mechanism in the pathogenesis of EBV-driven or -associated cancers.

Despite that mechanisms for EBV latency have been clarified, those required to reactivate latent EBV have not yet been elucidated well. Since reactivation of latent EBV is associated with EBV-driven cancers (Chien et al., 2001), the further understanding of the mechanisms that promote EBV reactivation is of great significance for revealing the pathobiology of EBV-driven cancers and developing novel therapies against it. The epigenetic regulation such as deacetylation by histone deacetylases (HDACs) is reported to play central roles in viral latency and reactivation (Li et al., 2007; Berger, 2007). EBV tends to establish a latency state in infected cells, and once transformed to the lytic replication cycle, cells are regulated by the “open” and “closed” conformations of chromatin (Tsurumi et al., 2005; Murata et al., 2021). For the maintenance and disruption of EBV latency, HAT and HDAC take part in the post-translational modification (hypoacetylation) of DNA-associated histone in the BZLF1 promoter (Liu et al., 1997a; Liu et al., 1997b; Gruffat et al., 2002). Besides, methylation also plays an important role in EBV latency maintenance (Murata et al., 2012; Imai et al., 2014).

During lytic infection, the viral genome of EBV is amplified up to 1,000-fold and expresses a variety of EBV genes to maintain the cell cycle progression in the S-phase which is necessary for viral replication (Tsurumi et al., 2005). The expression of the BamHI Z/R fragment leftward open reading frame 1 (BZLF1/BRLF1) genes can induce the lytic cycle by cascade transactivating both early and late EBV genes (Speck et al., 1997; Tsurumi et al., 2005; Murata et al., 2021). For the final step of the EBV lytic cycle, the virion genome is replicated, cleaved, packaged, and released to infect other susceptible cells (Tsurumi et al., 2005). During the EBV infection, the expressions of LMP1 and EBNA2 (Sekihara et al., 2014) during the lytic/replicative phase are considered as oncogenic; thus, regulation of LMP1 and EBNA could suppress the tumorigenesis.

The depletion of CD8+/CD4+ T cells and NK cells is reported to promote the occurrences of EBV-driven cancers (Murer et al., 2019; Münz, 2021), and extra evidence has also been claimed that the coinfection of EBV and some pathogen could have induced the EBV reactivation and impaired the EBV-targeted immune cell response. The commensal microbiota is considered as fundamental for the development of secondary lymphoid structures, as well as the differentiation, maturation, and function of T and B cells including virus-specific effector CD4⁺ and CD8⁺ T cells, FoxP3⁺ CD4⁺ T regulatory cells (Tregs) and Th17 cells, CD4⁺ helper T cells, and B cells (Hill and Artis, 2010; Furusawa et al., 2013; Khosravi et al., 2014; Josefsdottir et al., 2017; Luu et al., 2018; Zhao and Elson, 2018; Lee et al., 2021). It is reasonable to believe that other commensal microbiota coinfection with EBV could also be a trigger for EBV-driven cancer.

The disturbance of commensal microbiota is reported to result in impaired lymphoid tissue development and alter susceptibility to infectious diseases (Abt and Artis, 2009; Atarashi et al., 2011; Pickard et al., 2017). For example, the colonization of segmented filamentous bacteria (SFB) in the intestine is associated with increased CD4⁺ T helper 17 cells in the intestine (Wang et al., 2019b). The oral colonization of Lactobacillus paracasei has also been reported to induce an increased number of tissue resident and circulatory myeloid cells in mice lungs (Belkacem et al., 2017).

As for the treatment of EBV-driven cancers, studies have tried to develop some EBV-targeted therapies in addition to traditional chemotherapy and radiotherapy. However, limit progressions have been made by targeting the deregulated signal pathway in EBV infection or focusing on the drugs targeting EBV antigens, such as EBNA1 (Thompson et al., 2010; Messick et al., 2019). Up to now, no vaccine against EBV has been made successfully (Dugan et al., 2019; Cai et al., 2021). The application of nucleoside analogs which could act during the lytic phase of EBV, such as ganciclovir and zidovudine, in addition to IL-2 and CAR-T treatment, worked well in EBV-positive PCNSL (Aboulafia et al., 2006; Dugan et al., 2019). For latent EBV-driven cancer, the induction of the EBV lytic phase could be a rationale chosen, and BA was reported to take effect in refractory EBV-driven lymphoma (Ghosh et al., 2007; Perrine et al., 2007). Therefore, whether there would be another microbial product that could be efficient in the treatment of EBV-driven cancer needs further exploration. In recent years, increasing evidence suggested that the microbiome could regulate the host responses toward anticancer therapy, including chemotherapy, radiation, and targeted therapy (Cogdill et al., 2018). A recent study of advanced colorectal cancer reported that the diversity of blood microbiota influences the host response to chemotherapy and adoptive T cell immunotherapy (Yang et al., 2021). Consequently, whether the regulation of microbiome constitution would favor the therapeutic efficiency in EBV-driven cancer also needs to be explored.

Over the past few years, the roles of the commensal microbiome in modulating host immunity have been studied (Brosschot and Reynolds, 2018; Neil and Cadwell, 2018). The function of the commensal microbiome ranges from aiding in metabolism to competing with invasive pathogens (Abt et al., 2012; Belkaid and Hand, 2014). EBV, as a member of the commensal microbiome, has also been reported to regulate the host immune system and interact with other components of the microbiome. Some of these interactions are considered to induce the reactivation of EBV. However, a more thorough understanding of the molecular mechanisms by which specific constituents of the commensal microbiota would promote the reactivation of EBV, as well as driven EBV-associated cancers, is in great need.

YW wrote the paper. HX, JH, and HC supported and supervised the manuscript revision. All of the authors discussed and commented on the manuscript. All authors contributed to the article and approved the submitted version.

This research was supported by funds from the National Natural Science Foundation of China (31701207, 82000175) and the Chen Xiao-ping Foundation for the Development of Science and Technology of Hubei Province.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.