Mechanisms of crosstalk between the oropharyngeal microbiome and human papillomavirus in oropharyngeal carcinogenesis: a mini review

Head and neck squamous cell carcinoma (HNSCC) is the sixth most common cancer globally. Notably, human papillomavirus (HPV)-positive oropharyngeal squamous cell carcinoma (OPSCC) is on the rise, accounting for 70% of all OPSCC cases. Persistent high-risk HPV infection is linked to various cancers, but HPV infection alone is not sufficient to cause cancer. Advances in next-generation sequencing have improved our understanding of changes in the human microbiome of cancerous environments. Yet, there remains a dearth of knowledge on the impact of HPV-microbiome crosstalk in HPV-positive OPSCC. In this review, we examine what is known about the oropharyngeal microbiome and the compositional shifts in this microbiome in HPV-positive OPSCC. We also review potential mechanisms of crosstalk between HPV and specific microorganisms. Additional research is needed to understand these interactions and their roles on cancer development and progression.


Introduction
While the overall incidence of head and neck squamous cell carcinoma (HNSCC) is declining globally, human papillomavirus (HPV)-positive oropharyngeal squamous cell carcinoma (OPSCC) has been increasing, with HPV accounting for 70% of all OPSCC (1)(2)(3).HPV is among the most prevalent viral infections worldwide, with nearly all sexually active people being infected with HPV in their lifetime, and half of these being high-risk HPV.Despite this, most people will clear the infection, and only a small minority will have persistent infection (4).It is well-established that persistent high-risk HPV infection is responsible for the development of cervical, anogenital, and oropharyngeal cancers.However, HPV alone is not sufficient to cause cancer (4).Factors such as chronic inflammation, host microbiome alterations, and immune deficiency may be related to pathogenesis, but the mechanisms remain unclear (5).
Recent advances in next generation sequencing have revolutionized our ability to characterize the human microbiome (6).Microbiome studies of different normal and cancer samples have led to improved understanding of associations between microbes and carcinogenesis with potential for risk stratification and prognostication in cancer (7).The oral microbiome harbors over 700 microorganisms (8)(9)(10), and disruptions to this microbiome are known to contribute to inflammatory environments with carcinogenic potential (11)(12)(13)(14).Firmicutes, Proteobacteria, Bacteroidetes, Fusobacteria, and Actinobacteria have been identified as the most abundant phyla in HNSCC, accounting for > 90% of the oral bacterial community (15,16).The association between oral dysbiosis and oral squamous cell carcinoma (OSCC) has been well established (15)(16)(17)(18)(19)(20).However, the oropharyngeal microbiome and its association with HPVpositive OPSCC are just being uncovered (5).Furthermore, as HPV infection alone is insufficient for cancer development, it is crucial to explore the possibility of HPV-microbial crosstalk and its role in carcinogenesis.
It is important to distinguish the microbiomes of oral cavity and oropharyngeal cancers.The oral cavity microbiome is heavily influenced by environmental factors and subsequent dysbiosis leads to HPV-negative OSCC (19,21,22).On the other hand, being primarily HPV driven, OPSCC is an entirely different disease which may be less influenced by environmental factors (23).This review will focus on the unique features of the oropharyngeal microbiome and its interplay with HPV infection in oropharyngeal carcinogenesis.Many reviews discuss the associations between the oral microbiome and OSCC, which will not be discussed.Given the lack of studies on HPV-microbial crosstalk in the context of OPSCC, we also aim to propose potential mechanisms of crosstalk between HPV and specific microbes and highlight areas requiring future study.

Literature search strategy
We conducted a review of the literature to identify articles addressing the oropharyngeal microbiome, the oropharyngeal microbiome in HPV-positive OPSCC, and HPV-oropharyngeal microbial crosstalk in OPSCC carcinogenesis.This work is not a systematic review; therefore, we did not use PRISMA guidelines.Each query of PubMed's online repository used the keywords "oropharyngeal microbiome" or "oropharynx [MeSH] microbiota [MeSH]."These keywords were paired with combinations of "human papillomavirus," "HPV," "human papillomavirus [MeSH]," and "oropharyngeal neoplasms [MeSH]."Articles were sorted based on relevance and evaluated for content.Exclusion criteria included non-English studies, studies with only pediatric participants, and studies that utilized non-specific oropharyngeal collection methods.

The oropharyngeal cancer microbiome
Few studies have compared the microbiomes of normal oropharynx and oropharyngeal cancers (Table 2).Differences in sample collection ranging from saliva samples or oral washes, which is non-specific to the oropharynx, to oropharyngeal swabs and tissues, which is more specific to the oropharynx, has led to discordance in the literature (38,(44)(45)(46)(47)(48)(49)(50).As the normal flora of the oropharynx is dominated by a small number of phyla and genera of bacteria, identifying significant differences between normal and cancer requires careful evaluation using precise methods.Some studies report significant differences in bacteria that account for less than 0.1% of the overall abundance of the whole oropharyngeal microbiome, putting in question the real biological relevance of these differences (38,44,49).In this review, we focus on the shifts in relative abundance in the more abundant bacteria of the oropharynx as well as those with evidence of pathogenic roles in association with HPV (Figure 1).
o No significant differences in the microbial community composition of tumor-affected vs contralateral tonsils of same patient.

Metagenomics
• Shift towards more gram-negative bacteria in specimens with cancer present • Burkholderia pseudomallei was unique to all specimens that contained cancer.
• Specimens from patients with nodal metastasis contained unique signatures for the gram negative, facultative anaerobe Shigella dysenteriae; Orientia tsutsugamushi and Neisseria gonorrhoeae.
• Three species signatures were unique to negative-node primary: o Leptospira interrogans, Sphingomonas wittichii, and Coxiella burnetiï decreased abundance of Fusobacterium and a relatively increased abundance of Streptococcus (48).One study found enriched presence of Chlamydia trachomatis in HPV-positive tonsil SCC and patients with more advanced disease were colonized with Neisseria gonorrhoeae (47).
After our review of the literature, no study has independently validated the interplay between specific microbes and HPV in OPSCC.Some have reported the oral bacterial signatures of HPV-positive OPSCC, but these are primarily descriptive (9, 17, 63-65).In the subsequent sections, we propose potential mechanisms of crosstalk between HPV and specific microbes identified from our literature search in promoting HPVrelated carcinogenesis.
HPV-Neisseria/Chlamydia: overview HPV-positive OPSCC patients exhibit tumor microbiome enriched in Neisseria, particularly in regionally metastatic disease (Table 2) (45-47).Furthermore, one study identified Neisseria gonorrhoeae (NG) as the only signature present in patients with nodal disease and Chlamydia trachomatis (CT) as a shared signature amongst all HPV-positive OPSCC patient cohorts (Table 2) (47).These findings contribute to the growing evidence linking Neisseria and Chlamydia to persistent HPV infection and HPV-positive cancers of the cervix and oropharynx (66)(67)(68).
The carcinogenic potential of the crosstalk between HPV-NG and HPV-CT in the oropharynx may stem from their abilities to enable persistent HPV infection within tissues.Cervicitis is associated with prolonged duration of HPV infection as well as an increased likelihood of high-grade cervical dysplasia (69, 70).Furthermore, high-risk HPV (HR-HPV) and NG co-infection showed an increased risk of atypical squamous cells of undetermined significance (≥ ASC-US) and high-grade squamous intraepithelial lesions (HSIL) (66).HPV-CT co-infection resulted in 5 times higher risk of both ≥ ASC-US and HSIL.A meta-analysis of 22 studies confirmed that HPV-CT co-infection conferred a higher risk of cervical cancer than either alone (71).As NG and CT are Summary findings of enriched genera/species identified in normal, HPV-infected premalignant, and HPV-positive OPSCC tonsils based on our focused review of the literature.The enriched genera/species outlined in this figure result from studies using sampling methods specific to the tonsils.
primarily GU pathogens, they are thought to be transplanted to the oropharynx during oral sex (59) and subsequently cause chronic subclinical infections, with a predilection to tonsillar tissues where they are difficult to eradicate (72,73).

HPV-NG crosstalk: proposed mechanisms
Although both NG and CT contribute to persistence of HPV infection in the oropharynx, their underlying mechanisms may differ.NG disrupts the mucosal barrier by inhibiting epithelial cell renewal and exfoliation.More specifically, NG can engage human carcinoembryonic antigen-related cell adhesion molecules (CEACAMs) present within pharyngeal and urogenital mucosal surfaces by using outer membrane adhesins (74)(75)(76).Once engaged, NG blocks detachment of infected epithelial cells from the extracellular matrix (ECM) by triggering the full-length expression of CD105.CD105 is sufficient to inhibit infectioninduced detachment, suggesting a mechanism by which NG can disrupt the innate defense mechanism of epithelial exfoliation (77).
In vitro studies have demonstrated that NG can also induce DNA damage.Normal vaginal mucosa and cervical adenocarcinoma cell lines infected with NG displayed significantly more DNA strand breaks compared to uninfected controls (78).NG was also shown to induce the cell cycle inhibitors, p21 and p27 (79).Interestingly, NG infected cells can evade DNA damage-induced apoptosis via downregulation of tumor suppressor p53 (78).Through expression of Tfp/pilT, NG can activate the extracellular signal-regulated kinase (ERK) prosurvival pathway resulting in downregulation of proapoptotic proteins Bim and Bad (80).By inhibiting apoptosis in host cells, NG can delay the immune response, triggering a "carrier state" and/or intracellular survival.In HPV-infected epithelium, such a survival mechanism may confer greater immortality in combination with well described oncoproteins HPV E6/E7.

HPV-CT crosstalk: proposed mechanisms
At a molecular level, CT can also traumatize the epithelium of susceptible mucosal tissues, creating opportunities for microabrasions that promote HPV entry (81).Persistent CT infection in epithelial cells increases the secretion of proinflammatory cytokines IL-8, GROɑ, GM-CSF, and IL-6, which can cause oxidative stress (82).Moreover, the epithelial cells continue to release IL-1ɑ after CT infection, triggering neighboring noninfected epithelial cells to produce additional cytokines, sustaining the inflammatory response and ultimately leading to epithelial tissue damage (82).Of note, HPV-CT co-infected cervical lesions demonstrate significantly upregulated expression of NF-kB compared to non-CT infected cervical lesions.NF-kB is a family of transcription factors that play critical roles in intracellular regulation of immune response and inflammation (83,84).The upregulated NF-kB pathway caused by HPV-CT co-infection can exacerbate IL-8 mediated epithelial tissue damage, facilitating HPV virion penetrance (85).Furthermore, CT infection has been shown to disrupt Ncadherin-dependent cell-cell junctions in human cervical epithelial cells, resulting in an increase in epithelial paracellular permeability (86).This may explain how CT can increase the risk of HPV acquisition and persistence and how HPV-CT co-infection can increase risk of carcinogenesis (87,88).
Beyond epithelial disruption, CT can also disturb the immune response necessary to clear HPV, leading to persistent infection.CT can cause a shift in the immune landscape from a cell-mediated response to a humoral response due to its chronic infection in mucosal tissues (83,89,90).As cell-mediated immune responses are critical for clearance of HPV infections, a shift to a humoral response creates an environment where HPV infection can thrive (91).Indeed, a cohort study of Swedish women showed that a selfreported history of CT infection was the most significant risk factor for presence of HPV DNA in the blood (92).Furthermore, the CT outer membrane pore protein OmpA can inhibit apoptosis in infected host cells by targeting the pro-apoptotic proteins Bax and Bak.Similar to HPV-NG crosstalk, inhibition of apoptosis may compound the anti-apoptotic effect of HPV E6/E7 and is a hallmark of carcinogenesis (93).
Chronic CT infection can also promote carcinogenesis via DNA damage caused by reactive oxygen species (94).One study used an HPV16 infected ectocervix organoid and CT co-culture system to model HPV-CT crosstalk (61).Using this model, they found that CT impedes HPV-induced mechanisms that maintain genome integrity, including mismatch repair, creating a genotoxic effect in host cells.Furthermore, HPV16 E6/E7 integration in the host genome slows the CT life cycle and induces persistent infection.This study demonstrates the potential complexity of this bidirectional relationship between CT and HPV in cancer development and progression (61).

Summary of HPV-NG/CT crosstalk
The crosstalk between HPV-NG and HPV-CT may contribute to carcinogenesis in the oropharynx by disrupting the normal mucosal barrier and facilitating entry of HPV virions, delaying immune response, and promoting genotoxicity.Additional research is required to evaluate the impact of HPV-NG and HPV-CT coinfection in oropharyngeal models.

HPV-Fusobacterium: overview and proposed mechanisms
Fusobacterium spp are well established periodontal pathogens (95).Of these, Fusobacterium nucleatum (FN) is well known for its role in OSCC development (96-98).FN colonization increases risk of OSCC by augmenting IL-6-STAT3 signaling pathways, promoting tumor growth and invasiveness (15,99).However, studies on the oropharyngeal microbiome suggest the contrary: HPV-positive OPSCC exhibits a reduction of Fusobacterium spp compared to the normal oropharynx (48).As with NG and CT, such differences can be partially explained by differences in experimental methods: lack of stratification of HPV-positive to HPV-negative, sampling methods, and sequencing methodology (100).
Although less prominent in HPV-positive OPSCC compared to otherwise healthy adults, progressive over-representation of FN has been characterized with stage III HPV-positive OPSCC patients, showing a significantly higher relative abundance of FN compared to earlier stages.This correlation between FN representation and stage is consistent with that of OSCC and gastrointestinal squamous cell carcinoma studies (46,98,(101)(102)(103).Furthermore, some have proposed that patients with stage III HPV-positive OPSCC are at a higher risk of recurrence despite definitive concurrent chemoradiation due to differences in FN enrichment (104).Ultimately, the etiology explaining these associations remain unclear.HPV-FN co-infection in the tonsillar tissues of susceptible patients can be a contributing factor.
The potential mechanisms by which FN contributes to cancerous growth have been studied in the colon and oral cavity.FadA, a virulence factor expressed by FN, is a cell surface adhesion protein involved in the attachment and invasion of epithelial cells.It can interact with E-cadherin of epithelial cells to stimulate the Wnt/ b-catenin pathway (105).As a result, b-catenin can activate the expression of cyclin D1, c-Myc, and Wnt signaling genes, which promote cell proliferation and tumor growth.HR-HPV oncoproteins also activate the canonical Wnt/b-catenin pathway, which further contribute to the onset, progression, and maintenance of transformed cells (106).Studies in cervical cancer have highlighted the role of Wnt signaling as the "second hit," responsible for transforming HPV-immortalized cells (107).Therefore, HPV-FN crosstalk may result in a multiplicative effect on the growth of HPV-positive cancers (Figure 2).
HPV-FN crosstalk may also alter mucosal integrity, predisposing to HPV infection and OPSCC progression.FN produces matrix metalloproteinases, which are capable of degrading components of the extracellular matrix (ECM) (109).FN can also promote the epithelial-mesenchymal transition (EMT), predisposing cells to malignant transformation.One study in oral epithelial cells demonstrated that FN infection can trigger EMT in both normal and cancerous cells via the lncRNA MIR4435-2HG/ miR-296-5p/Akt2/SNAIl signaling pathway, a process that also downregulates E-cadherin (110), thereby breaking down the epithelial barrier and allowing HPV infection and malignant transformation (111).Additional research must explore the mechanisms by which HPV-FN crosstalk can foster tumor growth, progression, and clinical outcomes.Proposed multiplicative effect HPV-Fusobacterium nucleatum (FN) crosstalk in contributing to HPV-positive OPSCC based on our focused review of the literature.FadA binding enhances when Annexin A1 level increases (105).Annexin A1 is significantly elevated in the margins of HPV-positive OPSCC compared to HPV-negative OPSCC (108).FadA can interact with E-cadherin of epithelial cells to stimulate the Wnt/b-catenin pathway (105).As a result, b-catenin can activate the expression of cyclin D1, c-Myc, and Wnt signaling genes, which promote cell proliferation and tumor growth.HR-HPV oncoproteins E6/E7 can also activate the canonical Wnt/b-catenin pathway, which further contribute to the onset, progression, and maintenance of transformed cells (106).E7 binds to PP2A in the structural and catalytic domain to contribute to the inhibition of GSK3b activation (106).The E6/E6AP complex can stabilize b-catenin, avoiding its proteasomal degradation and promoting its nuclear translocation.The binding of E6 to Dishevelled can disrupt the b-catenin degradation complex.

HPV-Streptococcus: overview and proposed mechanism
Oropharyngeal microbiome studies have also revealed that HPV-positive OPSCC patients demonstrate enriched signatures of the genus Streptococcus (38,46,48).Streptococcus has been heavily associated with both periodontal inflammation as well as OSCC, and is one of the most common microbes in the oral cavity and oropharynx (11,15,16,112).Furthermore, one study of salivary samples from patients with OPSCC revealed that the relative abundance of Streptococcus could be used to differentiate tumors from control samples (17).As Streptococcus spp are commensal to the oral cavity and oropharynx, environmental insults that trigger dysbiosis may allow for opportunistic organisms to colonize (113).Streptococcus mutans (SM), a viridans group streptococcus (VGS), represents one of these species.SM has been shown to induce IL-6 expression in human and mouse oral cancer cell lines, representing a specific mechanism for its potential carcinogenicity (114).As with the other microbes mentioned, it is uncertain whether the changes observed in streptococci are the cause or consequence of cancer.
One study demonstrated that Streptococcus may promote HPV16 entry into basal keratinocytes by supplying a furin-like peptidase (115).HPV16 infection is initiated by binding to receptor heparan sulfate proteoglycans (HSPGs) on the exposed basement membrane (81,115,116).Following binding to HSPGs, the L2 cleavage site is exposed and subsequently cleaved by furin, facilitating HPV16 entry into tissues (116).Unlike HSPGs, the expression of furin varies based on tissues (117).In the human oropharyngeal epithelium, furin expression is low in the basal layers where HPV16 typically initiates infections (118).One study revealed that PepO, the furin-like peptidase of S. gordonii, promoted HPV16 infection (119).Altogether, this suggests that HPV-Streptococcus crosstalk may allow for better penetration of HPV into the basal keratinocytes, where it can proliferate.

Conclusions and future directions
The intricate composition of the oropharyngeal microbiome is only beginning to be understood.Previous studies have focused on saliva due to ease of access (16,65,120).Although these findings have been used as a mirror of the oropharyngeal microbiome, targeted collection methods have shown unique signatures of bacteria and other microbes in this region (37,47,48).T r a d i t i o n a l p e r i o d o n t a l p a t h o g e n s f o u n d i n O S C C (Porphyromonas gingivalis and Treponema denticolum) did not present as key players (19,121).Instead, FN and SM provide proinflammatory environments, in conjunction with chronic HPV infection, that may increase propensity for the development of cancer (46,48).High quality mechanistic studies have yet to provide a comprehensive landscape of these interactions critical for the development of novel therapy.
We propose that HPV-microbial crosstalk impacts patient outcomes at multiple levels.First, it facilitates HPV virion entry to the basal keratinocytes and fosters an immune environment that also allows for HPV persistence.Furthermore, as demonstrated in other solid tumors, direct and secreted microbial features may confer treatment resistance in advanced stages of HPV-positive OPSCC (101,103).Key microbial signatures of HPV-positive OPSCC may provide useful biomarkers for deintensification of patients.
Novel reductionist approaches utilizing tissue-microbe coculture may be one way to model microbial effects on HPVassociated carcinogenesis (61).These studies would also benefit from including fungi, proteus, and bacteriophages, which occupy their own niche roles in the microbiome.In cervical cancer, specific fungal communities (Candida, Malassezia, and Sporidiobolaceae) were identified as significantly associated with HR-HPV and premalignant cervical lesions (122).We believe similar associations occur in the oropharynx, triggering carcinogenesis in the HPV-infected tonsil.As the incidence of HPV-positive OPSCC begins to exceed that of cervical cancer, it is necessary to understand the functional relationships between microbes and HPV to characterize oropharyngeal premalignant lesions and enhance our understanding of HPV-related carcinogenesis (123).

TABLE 1
Summary of the key taxonomic findings in the normal oropharyngeal microbiome based on sampling methods specific to the oropharynx.Prevotella, Neisseria, Veillonella, Fusobacterium, Streptococcus.Prevotella, Neisseria, Veillonella, Fusobacterium, and Streptococcus made up more than 70% of the bacteria.

TABLE 2
Summary of the key taxonomic findings in the HPV-positive OPSCC microbiome based on sampling methods specific to the oropharynx.Comparing relative abundance in patients with OPSCC to those without: