Latent proteins are essential for the transformation of normal B lymphocytes into lymphoblastoid cell lines (LCLs), and they are involved not only in driving the overexpression of oncogenes, the silencing of tumor suppressors, the cell cycle, migration, but also in the regulation of adhesion.

Genetic instability is one of the major common features of cancer and can be observed at the chromosomal or genetic level in malignant cells (71). Integration of EBV into the host genome may be a common occurrence in lymphomas, but our understanding of this is limited. On the one hand, the large size of the EBV genome itself makes it difficult to determine the integration site with the host genome and to analyze it further, on the other hand, the highly methylated DNA hinders the mapping of the EBV genome, and not only that, multiple copies of the viral exons can generate interference noise at the integration site (72), which makes it more difficult to study it in depth. Previous studies have demonstrated the integration of EBV in the chromosomal genome of BL (73) and other B-cell lymphomas (74, 75). Takakuwa et al. (71) demonstrated in Raji that integration of EBV into 6q15 resulted in loss of expression of the human Bach2 gene (BACH2) at the mRNA and protein levels. BACH2 has been shown to have a significant inhibitory effect on cellular proliferation, and deletion of BACH2 expression may contribute to the development of B-cell lymphomas, including BL. Related studies have previously analyzed copy number alterations (CNAs) and gene expression profiles of EBV⁺ and EBV^-DLBCL samples confirming that EBV⁺ DLBCL has fewer genomic alterations (76). In a recent whole-exome sequencing of EBV⁺DLBCL, it was shown that a heterogeneous mutational landscape is associated with DNA double-strand break-homologous recombination repair failure, and genes found to have a high number and frequency of mutations include serine protease 3 (PRSS3), MUC3A and MUC16 (77). A recent study by Zhou et al. (78) demonstrated an elevated frequency of mutations in MYC and RHOA in patients with EBV⁺DLBCL. An updated mutational map of EBV⁺DLBCL has been comprehensively characterized, complementing previous studies with recurrent alterations in CCR6, CCR7, DAPK1, TNFRSF21, and YY1 (79), further elucidating the mechanism by which EBV leads to B-cell transformation.

EBV was the first virus to detect viral miRNAs (80). The EBV genome encodes 44 mature miRNAs belonging to two distinct classes, BamHI-A region rightward transcript (BART) and Bam HI fragment H rightward open reading frame 1 (BHRF1), which have different expression levels in different EBV latency types (81). Among them, BART transcripts encode 22 miRNA precursors and 40 mature miRNAs, while BHRF1 transcripts express three miRNA precursors to produce four mature miRNAs. Current published literature has demonstrated that EBV-encoded miRNAs play an important role in the development and progression of EBV-associated malignancies, including cell proliferation, apoptosis, invasion, and transformation (82, 83).Moreover, EBV miRNAs can even directly target immune-related genes, allowing infected cells to evade surveillance and destruction of the immune system (84), (85). However, EBV miRNAs have different expression profiles in different cancer types. In EBV-infected DLBCL, all EBV-miRNAs except BHRF1 cluster and EBV-miR-BART15 and -20 could be detected, as demonstrated in Imig et al. And in NK/T-cell lymphomas, the most highly expressed viral miRNAs were miR-BART1-5p, miR-BART5, miR-BART7, miR-BART11-5p, and miR-BART19-3p, accounting for 50% of viral miRNAs and approximately 1% of total miRNAs (86). Studies have described the presence and expression levels of EBV miRNAs and host miRNAs in different lymphomas, with some focusing on patient samples and others on different cell line models for in vitro experiments. EBV microRNA profiles and human microRNA profiles for EBV-associated lymphomas are detailed in a recent study by Soltani et al. (87) What’s more, published studies have confirmed that EBV-encoded miRNAs may interfere with host miRNAs, which actually leads to even more complications (87).

EBV miRNAs are essential for regulating the viral life cycle. It was demonstrated as early as lizasa et al. (88) that EBV-miRNA-BART6-5p targets four sites within the 3’-UTR of human Dicer mRNA and comprehensively affects the maturation of the miRNAs, resulting in the total repression of these molecules, which helps to maintain latent infection. Of particular note, in addition to EBV miRNAs, EBV-associated products also contribute to the downregulation of Dicer, such as the EBNA1 protein, which has been described in detail in Mansouri et al. (89) EBV miRNA biogenesis and action are also affected by adenosine to inosine (A-to-I) RNA editing. A-to-I editing of pri-miR-BART6-5p was found in EBV-infected BL to activate Zta and Rta viral proteins encoding EBNA2 viral oncogenes and essential for lysis and replication, leading to the transition of the viral cycle to type III latency (88). Of interest, EBV-encoded miRNAs are also involved in host cell growth, cell cycle, and apoptosis. PRDM1/Blimp1 is a major regulator of terminal B-cell differentiation and is well known as an oncogene in aggressive lymphomas. Nie et al. (90) have demonstrated that the cellular target of the EBV-miRNA-BHRF1-2 is PRDM1, and that by inhibiting the PRDM1-mediated function and conferred a growth advantage to EBV-infected B cells, promoting lymphoma development. Another study confirmed that the EBV-miRNAs-BART9 were involved in the proliferation of Nasal NK/T cell lymphomas (NKTCL) by regulating the level of LMP-1 (91).

The success and persistence of any viral infection depends on a complex balance with the host immune system, and EBV miRNAs are also involved in the regulation of the host immune system. ( Figure 1 ) EBV-miRNAs-BART6-3p was found to mediate down-regulation of the interleukin-6 receptor (IL-6R) in BL (92), which is involved in regulating key cellular processes, including cell proliferation, survival, and response to host pathogens after dimerization receptor binding to interferon-α, IL-12, or IL-27 (93). In addition, the EBV-miRNAs-BART20-5p were shown to inhibit T-bet translation through secondary inhibition of p53 (94). The role of EBV-encoded miRNAs in immunomodulation has been well and exhaustively described (84, 95–97). The latest research has confirmed that in DLBCL, EBV-miRNA-BHRF1-2-5p targets LMP1 to drive the expression of PD-L1 and PD-L2, exerting context-dependent immune counter-regulation, leading to immune escape and contributing to persistent viral infection (98). In another study, Murer et al. (99) used NOD-SCID _γc null (NSG) and HLA-A2 transgenic NSG mice to construct a mice model infected with an EBV variant infection lacking viral miRNAs and a mice model infected with wild-type EBV, which found that the viral load in mice infected with EBV variants lacking viral miRNAs was significantly reduced, and the proliferation frequency of EBV-infected B cells was also decreased. What’s more, the depletion of T CD8⁺ cells led to the formation of lymphomas n the mouse model infected with the viral miRNA-deficient variant, which supports the notion that EBV miRNAs play a major role in immune evasion in vivo and support tumor development. The role of EBV virus-encoded microRNAs in human lymphomas can be found in the review by Navari et al. (82).

The hypothesis that chronic antigenic stimulation causes lymphomas remains controversial. Chronic local antigen-stimulated immune responses caused by HBV infection might be associated with the development of lymphoma (116). A large 14-year follow-up cohort study in Korea (106) consistently associated HBsAg⁺ with elevated risk of NHL, suggesting that chronic infection promotes the development of lymphoma. Risk of B-NHL is not increased in individuals previously infected with HBV or vaccinated against HBV (117, 118). Nucleic acid sequences specific to HBV have been detected in peripheral blood nuclei and hematopoietic tumor cells of patients with HBsAg⁺ (3, 119, 120), which might result in chronically stimulated B cells that transform into B-cell NHL. Peripheral blood mononuclear cells (PBMCs) derived from patients with chronic HBV infection have immortalization potential when cultured in vitro (121). New cells identified in the peripheral blood of some patients with non-lymphoid chronical HBV infection were later confirmed as being of B-cell origin. Moreover, the immunophenotype of these cells was similar to that of most HBsAg⁺ B-cell NHL. This supported the relevance of HBV-induced B-cell NHL, although none of the patients developed lymphoma during > 1 year of follow-up. Furthermore, a 42.1% and 65.5% bias towards Immunoglobulin Heavy Variable 4-34 (IGHV4-34) heavy, and Immunoglobulin Kappa Variable 4-1 (IGKV4-1), light-chain genes respectively in HBsAg⁺ DLBCL, exceeded that in normal peripheral blood B cells and B-cell NHLs (107). However, these results were contradicted by a study that found no evidence of biased IGVH gene usage or the stereotyped third complementarity determining region (CDR3) (122). Unlike classical antigen-driven hepatitis C virus-associated lymphoma, the chronic antigen stimulation model seems less applicable to HBV-associated DLBCL.

Hepatitis B viral DNA is integrated into the chromosomal DNA of lymph node cells (123). A genome-wide investigation of HBV integration in HCC found that HBV integration alters chromosomal stability and gene expression, and shortens the overall survival of infected individuals (124). Approximately 50% of woodchuck hepatitis virus (WHV) is integrated into the myelocytomatosis oncogene (MYC) family of genes and it affects the proto-oncogene in woodchuck models of HCC with chronic WHV infection (125). In fact, HBV integration is common, occurring in 80%-90% of HBV-associated HCC (126, 127).

Hepatitis B viral DNA can be integrated into the genome of NHL cells, and like HCC, it has preferential targets in NHL, since exons of the protein-coding genes FAT Atypical Cadherin 1 (FAT2), Senataxin (SETX), Integrin Subunit Alpha 10 (ITGA10) and Granulophysin (CD63) are disrupted by HBV DNA and the expression of seven HBV preferential target genes, Ankyrin Repeat and Sterile Alpha Motif Domain Containing 1B (ANKS1B), Histone Deacetylase 4 (HDAC4), EGF Like, Fibronectin Type III And Laminin G Domains (EGFLAM), Mannosidase Alpha Class 1C Member 1 (MAN1C1), XK-Related 6 (XKR6), Zinc Finger And BTB Domain Containing 38 (ZBTB38), and Coiled-Coil Domain Containing 91 (CCDC91) is significantly altered in NHL (128). The expression of six of these genes is increased in NHL whereas that of HDAC4 is not, suggesting that HBV integration leads to the cis-activation of primary oncogenes rather than the inactivation of tumor suppressor genes. However, no evidence of HBV DNA integration into the tumor genome has been found in either HBV-associated FL (129, 130) or DLBCL (122). A trend towards an increased genome-wide mutational load has been identified by whole genome, or whole exon sequencing in the coding regions of HBsAg⁺ follicular lymphoma (FL) with significantly more non-silent mutations per tumor (129). The most significantly mutated genes were Histone-Lysine N-methyltransferase 2D (KMT2D), Immunoglobulin Lambda‐Like Polypeptide 5 (IGLL5), CREB-binding protein (CREBBP), B Cell Lymphoma 2 (BCL2), Tumor Necrosis Factor Receptor Superfamily 14 (TNFRSF14), Interferon Regulatory Factor 8 (IRF8), Enhancer Of Zeste 2 Polycomb Repressive Complex 2 (EZH2), Signal Transducer And Activator Of Transcription 6 (STAT6), AT-Rich Interaction Domain 1A (ARID1A), and Guanine Nucleotide-Binding Protein Subunit Alpha-13 (GNA13). Furthermore, the most obvious mutational pathways were HBV infection-associated, followed by the Forkhead Box O (FoxO), Wingless/Integrated (Wnt), Janus Kinase/STAT (JAK-STAT), B-Cell Receptor (BCR), Phosphatidylinositol-3 Kinase (PI3K), and Nuclear Factor Kappa B (NF-κB) signaling pathways.

The HBx protein encoded by the X gene was once named “viral oncoprotein.” This protein is involved in hepatocyte transformation through regulation of the cell cycle and the pleiotropic activity of DNA repair and signaling pathways (131–133). The expression of HBV antigens, especially HBx protein, is abundant in HBV⁺ DLBCL sera (103). These findings were consistent with the significantly elevated HBx levels in HCC due to stable HBV integration (124, 134). The HBx protein inhibits p53 in hepatocytes, which leads to abnormal hepatocyte division and HCC (135, 136). A similar B cell mechanism might contribute to the malignant transformation and development of B cell NHL (3). Among the various activities of HBx, its transactivation might play a crucial role in carcinogenesis. Interaction between HBx and the acetyltransferase CREBBP/p300 facilitates the recruitment of these cofactors to the CREB-responsive promoter, which leads to the activation of gene expression (112). A Chinese study of HBV-associated FL found significantly upregulated CREBBP-binding genes in HBsAg⁺, compared with HBsAg^- FL (129). This could explain the low dependence of HBsAg⁺ FL on CREBBP mutations in that study, as interaction between HBx and CREBBP/p300 might mimic the role of mutant CREBBP during the early stages of lymphoma. The contribution of HBx to the pathogenesis of lymphoma remains obscure, and further investigation is needed to verify its mechanism of action.

The tumor microenvironment is a complex system of cellular and subcellular components with reciprocal signaling pathways that play key roles in carcinogenesis (137). Tumorigenesis is dependent on the TME, and stroma is uniformly and inappropriately activated in cancer, thus contributing to the malignant features of tumors (138). Chronic and persistent HBV infection induces immune cell dysfunction, T-cell failure, as well as the extensive activation and production of numerous cytokines, chemokines and growth factors that constitute a sophisticated TME that might affect cancer development (139, 140). Hepatitis B surface antigen-positive FLs might have an altered TME with increased infiltration of cluster of differentiation (CD)8⁺ memory T cells, CD4⁺ Th1 cells, M1-macrophages and increased T cell failure (129). This was consistent with similar findings in HCC associated with HBV.

The unique biological characteristics of HBV complicates exploring curative mechanisms, and animal models have various strengths and weaknesses. This might explain to some degree, the limited progress of investigations into HBV-related lymphoma.

The defined pathogenic link between chronic Helicobacter pylori infection and the development of mucosa-associated lymphoid tissue (MALT) gastric lymphoma suggests that chronic antigenic stimulation can determine the likelihood of NHL (167). Notably, the regression of MALT lymphoma after HP eradication makes this possibility more plausible (168). Splenic lymphoma regression after antiviral therapy similarly eradicates HCV (169). About 10% of patients with type II mixed cryoglobulinemia (MC) develop overt B-NHL after 5-7 years of follow-up (170), and HCV is a major etiological factor in MC and might also be the cause of its evolution to overt NHL (171–173). HCV-associated type II MC expresses immunoglobulins encoded mainly by germline V_H1-69 and VκA27 genes. A preference for the V_H1-69/VκA27 combination in HCV-associated lymphomas is consistent with the possible role of antigen selection in the expansion of B cell clones (174). In addition, B-cell receptors expressed by lymphomas in patients infected with HCV rarely react with viral proteins (175). Notably, the highly biased stereotyped BCR sequence of HCV⁺ B-NHL has also been found in other HCV-B-cell malignancies (176). This confirmed that HCV-associated lymphoma cells originate from precursors with autoimmune properties rather than from B cells that express antiviral BCR.

The HCV envelope protein E2 can bind to CD81 expressed on B cells (164). This receptor is upregulated in HCV infection and MC and positively correlates with viral load (177). Moreover, CD81 forms a conjugate complex with CD19 and CD21 in human B cells (178, 179), and the attachment of the B cell antigen receptor (BCR) to any component of this complex decreases the threshold required for BCR-mediated B-cell proliferation (180). Bound E2-CD81 is also involved in activating the transcription factor NF-κB, which subsequently increases the expression of Bcl-2 protein, thus enhancing B cell survival and protecting human B lymphocytes from Fas-mediated apoptosis (181). In addition, HCV E2 binds to CD81 antibodies on neonatal human B cells, which leads to the activation and sequential proliferation of the C-JUN N-terminal kinase pathway (166). Furthermore, HCV-E2 binding to CD81 directly prevents the functional activation of NK cells, providing an effective immune escape strategy for the virus (182). Overall, the interaction between HCV and CD81 promotes chronic infection and facilitates the development of HCV-associated B-cell lymphoma.

Some evidence indicates that intracellular viral replication is not required for tumor transformation (183). The hit-and-run theory suggests that viruses play a predisposing role in cancer formation and that the viral genome can be completely lost after the host cell has accumulated numerous mutations (184). This mechanism was suggested for HCV (185). Infection with HCV results in a 5-10-fold increase in the frequency of mutations in the Ig heavy chain, B cell Lymphoma 6 (BCL-6), Protein 53 (p53) and Catenin genes in HCV-infected B-cell lines and HCV-associated peripheral blood mononuclear cells, lymphomas, and HCC in vitro. The authors concluded that HCV induces a mutator phenotype by causing changes in proto-oncogenes and oncogenes that successively lead to oncogenic B cell transformation, even when the virus might have already left the cells. The same group also conducted RNA interference experiments and found that HCV induced error-prone DNA polymerases ζ, ι, and activation-induced cytidine deaminase. All these together contribute to increased mutation frequency, complementing the oncogenic mechanism of HCV causing lymphoma. Some controversy remains regarding the clinical applicability of these findings, as they have not been confirmed in vivo (186, 187).

Infection with HCV stimulates nitric oxide (NO) production by activating the inducible NOS (iNOS) gene through the viral core and NS3 protein (188). Nitric oxide causes DNA breaks and enhances DNA mutations. The HCV core protein binds to NBS1 and inhibits formation of the Mre11/NBS1/Rad50 complex, thus affecting Ataxia Telangiectasia-Mutated (ATM) activation and inhibiting DNA binding by repair enzymes (189). Infection with HCV inhibits multiple DNA repair processes and leads to chromosomal instability, which explains its oncogenicity from a different perspective.

Hepatitis C viral RNA and protein were detected in an established HCV-infected B-NHL cell line in vitro using RNase protection assays and immunoblotting (190). That study confirmed that HCV can infect primary human hepatocytes, PBMCs and established Raji B cell lines in vitro, indicating that HCV can replicate in B cells. Ample evidence supports the notion that intracellular viral proteins contribute to oncogenic transformation. Interferon regulatory factor-1-null (irf-1(-/-)) mice with inducible and persistent expression of HCV structural proteins (irf-1/CN2 mice) have been established (191). These mice have a high incidence of lymphoma and lymphoproliferative disorders. The HCV core and E2 proteins are responsible for the expression of interleukin (IL)-2, -10, and -12, as well as the induction of Bcl-2 in the presence of nucleocapsid proteins in the context of complex signaling networks in these mice (191). Another transgenic mouse model expressing HCV core protein frequently developed follicular center cell-type lymphoma, and HCV core mRNA was detected in lymphoma tissues (192). Transgenic RzCD19Cre mice express the complete HCV genome in B cells (193). However, the incidence of DLBCL in RzCD19Cre mice was only 25%. The incidence of B-cell lymphoma correlated significantly with serum levels of soluble interleukin-2 receptor α subunit (sIL-2Rα) only in the RzCD19Cre mice.

Small non-coding MicroRNAs (miRNAs) sequence-specifically regulate gene expression at the post-transcriptional level (194). They play roles in controlling various biological functions such as developmental patterns, cell differentiation, proliferation, genomic rearrangement and transcriptional regulation (195). MicroRNA-26b is significantly downregulated (P = 0.0016) in HCV⁺ splenic marginal zone lymphoma (SMZL) and this might cause miR-26b to stop inhibiting never in mitosis gene A (NIMA)-related Kinase 6 (NEK6) and have oncogenic potential in HCV-associated SMZL (196). MicroRNA-26b functions not only in the specific area of HCV-associated SMZL, but also in HCV-associated NHL, including MZL and DLBCL (197). Overall, these findings suggest that miRNA network dysregulation is involved in the development of HCV-associated lymphomas.

Cytokines are small glycoproteins and peptides that usually have relatively short half-lives and act via autocrine and paracrine signaling. Cytokines mediate interactions between immune and non-immune cells in tumors and can promote or inhibit cancer cell growth (198). B-cell activating factor (BAFF) is a key survival factor for B cells that is upregulated during HCV infection (199). An excess of BAFF in the absence of protective tumor necrosis factor (TNF) leads to a high incidence of lymphoma in BAFF transgenic mice, suggesting that BAFF functions in promoting B-cell malignancy (200). Notably, other cytokines and growth factors, including IL-6, -17, -10 and TGF-ß, also contribute to B-cell proliferation in HCV infection (201–203).

However, the molecular mechanisms underlying the development of HCV-associated lymphomas remain poorly understood. The prevailing views are not mutually exclusive and might involve parallel pathways leading to HCV-associated lymphoma, as it is likely that a combination of translational conditions is required to eventually lead to the development of lymphoma. Additional bridging studies combining in vivo and ex vivo investigations are required to further explore this topic.

Although HIV infection is characterized by a reduction in the function or number of CD4⁺ T cells (213), the obviously increased B cell activation in HIV infection is primarily driven by the abnormal production of B cell-stimulating cytokines such as IL-6 and chronic antigenic stimulation. Elevated levels of circulating free immunoglobulin light chains in patients at increased risk of HIV-associated lymphoma might represent a marker for polyclonal B-cell activation (214). In addition, evidence indicates a skewed IGHV repertoire in specific HIV-NHL categories. Heterogeneous expression of IGHV genes in HIV-NHL might be related to specific pathways of antigenic stimulation (215).

Serum levels of IL6, IL10, C-reactive protein (CRP), soluble (s)CD23, sCD27, and sCD30 are significantly higher in patients with HIV-NHL compared with HIV⁺ or AIDS controls after adjusting for numbers of CD4⁺ T-cells (216). The CD40 ligand (CD40L) can insert itself into the surface of HIV-1 particles when budding from activated CD4⁺ T cells (217), and HIV containing CD40 ligand (CD40L) activates B cells, which leads to secretion of the cytokines, IL-6, IL-10, IL-12 and TNF-α (218), in a way that mimics physiological stimulation. The role of CD40L in cancer has been detailed in a review (219). The HIV-1 trans-activator of transcription (Tat) induces the expression of IL-6 and IL-10 at the cellular level. Findings were similar at the individual level by in transgenic mice (220), and numerous spleens from Tat-transgenic mice had malignant lymphomas of B-cell origin. The HIV Tat also enhances the intrinsic antibody diversification mechanism by increasing activation-induced deaminase (AID)-induced somatic mutations in the variable heavy chain (VH) region of human B cells (221), which might lead to genome-wide mutations in malignant B cells among patients with HIV.

Mice transgenic for a defective HIV-1 provirus lacking part of the gag-pol region overexpress the HIV proteins p17, gp120, and negative regulatory factor (nef), then develop B-cell lymphoma (222). This supports the pathogenic role of aberrant HIV protein and B-cell-stimulating cytokine expression during lymphoma formation. Indeed, the HIV-1 matrix protein p17 persists in the germinal center after HIV-1 drug inhibition, and its variants (vp17s) activate Akt signaling and promote the growth of transformed B cells. This protein might also upregulate LMP-1 in B lymphocytes infected with EBV, leading to lymphoma development (223). Infection HIV can directly induce lymphoma formation. The oncogenic effects of HIV-1 proteins have been reviewed in detail elsewhere (224) and are not discussed herein.

For immunity, although multiple mechanisms may contribute to the development of lymphoma in HIV-infected individuals, two mechanisms appear to be involved: (1) loss of immunoregulatory control of EBV and/or KSHV; (2) chronic B-cell activation due to immune dysfunction caused by HIV infection. The cooperation of HIV, EBV, and KSHV in the pathogenesis of lymphoma and resulting microenvironmental abnormalities have been reviewed in detail elsewhere (225, 226). Table 3 shows associations between HIV-associated lymphoma and EBV and KSHV infections. It has long been shown that B-cell activation and immature phenotypic changes in vivo are accompanied by polyclonal Ig production in HIV-infected individuals (228). Notably, recent studies suggest that HIV may contribute to lymphomagenesis by acting directly on B lymphocytes as a key microenvironmental factor. It is worth noting that recent studies have shown that HIV may lead to lymphomagenesis by acting directly on B lymphocytes as a key microenvironmental factor. Various HIV-encoded proteins, including gp120, may trigger and maintain abnormal activation of B cells, abnormal secretion of cytokines IL6 and IL10, and so on, which have been stated in other subsections of HIV-associated lymphomas in this paper. Perhaps it is time to revisit the second immune-related mechanism.

Retroviruses damage DNA via various mechanisms such as genome integration, replication, inflammation, and direct interaction of viral proteins with DNA and HIV might be randomly integrated into the human genome. However, a pattern of integrated duplicated Alu elements and introns of Breast Cancer Gene 1 (BRCA1) has been identified (229) that supports the tendency of HIV-1 to integrate near the Alu class of human repetitive elements (230).

A genome-wide analysis of 57 HIV lymphomas found that genes associated with fragile sites such as Fragile Histidine Triad (FHIT; FRA3B), WW domain-containing oxidoreductase (WWOX; FRA16D), Deleted in Colon Cancer (DCC; FRA18B), and Parkinson Protein 2 (PARK2; FRA6E), are frequently inactivated by mesenchymal deletions in HIV-NHL, and that the prevalence of FHIT alterations is significantly higher in HIV-DLBCL (231). Among these, FHIT, WWOX and DCC are tumor suppressor genes that are frequently inactivated in various human malignancies (232–234). Thus, HIV might act directly at the genomic level to promote the pathogenesis of HIV-NHL, and this translational effect is partially independent of the expression of viral oncogenes. Human immunodeficiency virus induces c-myc dysregulation in B cells, and levels of viral RNA and myc expression correlate (235). Expression of the highly oncogenic transcription factor c-myc is enhanced at the transcriptional and translational levels in the presence of HIV-1 Tat protein (236).

Viruses and their components manipulate the expression of host miRNAs and play important roles in cancer pathogenesis. Hsa-miR-200c-3p is significantly downregulated in HIV-associated BL, and the zinc finger E-box binding homeobox epithelial-mesenchymal transition (EMT) transcription factors ZEB1 and ZEB2 are upregulated and actively help to promote tumor metastasis and invasion (237). Moreover, miRNA-21 is significantly elevated in peripheral B cells of patients infected with HIV, suggesting that it might contribute to the maintenance of B cell hyperactivation (238). A proteomic analysis of plasma proteins from AIDS-NHL recently identified 20 host proteins and a set of protein combinations that might serve as biomarkers for the pathogenesis of AIDS-NHL (239). This indicates a new direction towards a better understanding of the pathogenesis of HIV lymphoma.

KSHV miRNAs are generated from 12 pre-miRNA transcripts in the latency region, ultimately producing at least 17 mature miRNAs. (263) The biogenesis of KSHV miRNAs and their role in the development of KSHV-associated malignant tumors has recently been described in detail. (242, 264) Among the large number of miRNAs encoded by KSHV, KSHV-miRNA-K11 compares particularly because it shows significant homology to cellular miRNA-155. (265) MiRNA-155/bic overexpression can be observed in many human B-cell lymphomas, (266) and B-cell lymphomas can be induced in mice. (267).