Acute HIV infection exerts dramatic and perhaps irreversible MALT damage (27, 28). The fact that HIV replication has been associated with a loss of anti-inflammatory bacteria (20, 29) has spurred research into the hypothesis that HIV infection affects the microbiota and that this altered microbiota may contribute to persistent inflammation, increasing the risk of comorbidities. Mechanistically, HIV infection could affect the microbiota by inducing depletion of Th17 cells in MALT, enteropathy, mucosal inflammation, aberrant cytokine production, and intestinal epithelial cell damage (25, 30–34). In the context of HIV infection, microbial-induced immune activation occurs and correlates with markers of intestinal damage, suggesting that the microbiota is a relevant driver of systemic inflammation (35–41). Even the changes appreciated in the oral microbiota of people living with HIV (PLWH), who exhibit an increased prevalence of dental caries and periodontal inflammation, seem to be connected to shifts in systemic immune responses (reviewed in (42)). Specific Lactobacillus species-rich vaginal microbiota have been associated to protection from HIV infection (last reviewed in (43)).

It is now widely accepted that impairment of intestinal integrity and dysbiosis lead to translocation of bacterial derivatives from the gut to the bloodstream, resulting in chronic inflammation. This may occur by immunosuppressive or immunostimulatory mechanisms and via various non-mutually exclusive processes, including augmented antigenicity, adjuvanticity, or bystander T-cell activation (44, 45). This fact has been studied before for HIV-associated inflammation, which has been associated with an increase of active microorganisms leading to different pathways related to immune modification (46), These pathways include (i) decreased amino acid catabolism, leading to nutritional deficits (47). (ii) induction of indolamine-2,3-dioxygenase-1 (IDO1) leading to an increased transformation of tryptophan into the immunosuppressive kynurenine derivatives, bacterial translocation, and systemic inflammation, which has been linked with excess mortality risk ,45). (iii) increased butyrate synthesis, which, among other functions, tempers intestinal inflammation (48). and (iv) accumulation of inflammatory molecules, such as arachidonic acid and leukotriene-B4 (49).

Inflammatory biomarkers levels remain increased in PLWH even when ART is started early (50). Chronic inflammation has consistently been associated with an excess risk of comorbidities during treated HIV infection and is suggested as a contributing risk factor (51, 52). Thus, the HIV field has pursued whether the microbiota affects inflammation during treated infection. For example, microbiota metabolic profiles affect HIV inflammation by promoting changes in glutathione metabolism and zeatin biosynthesis, butyrate production, or tryptophan catabolism (46, 49, 50, 53). Furthermore, a well-defined deleterious consequence of HIV infection is bacterial translocation triggering immune activation (54–56). A few sequence-based and ultramicroscopic studies have uncovered a blood bacterial DNA profile in HIV. Following acute SIV infection in macaques, analysis of bacterial DNA isolated from the colon, liver, and mesenteric lymph nodes demonstrated a preference for the phylum Proteobacteria to translocate to these compartments and an increased metabolic activity of Proteobacteria within the colonic lumen (57). In a study in PLWH diagnosed with advanced disease and starting ART, we also found that Proteobacteria was the predominant phylum in the blood, indicating commonalities in the mechanisms by which bacterial translocate from the gut into the bloodstream between SIV and HIV infections. The same study showed that ART initiation in late-presenters attenuated the bacterial signature of untreated HIV infection, characterized by the presence of DNA from commensal bacteria with pathogenic potential (58). Relationships between the translocated microbiome, systemic inflammation, and clinical outcomes were described in a different study showing increased CD4 T cell counts following one year of ART that were associated with high Serratia abundance, innate proinflammatory cytokines and metabolites driving Th17 gene expression signatures, and restoration of mucosal immunity (59).

Current evidence supports investigating therapeutic strategies for immune modulation in HIV. However, so far, no intervention targeting the microbiota of PLWH either by using prebiotics (16, 16, 60), probiotics (61–63), synbiotics (64), rifaximin (65, 66), or even fecal microbiota transplants (67, 68) have convincingly proved to effectively temper inflammation or enhance boost immune recovery following ART initiation. In general, there was lack of standardization in the outcomes assessed (ranging from studies designed to assess T cell changes (60, 64) to exploratory studies evaluating multiple markers of T cell activation (69) or soluble markers of inflammation or bacterial translocation), the duration of the intervention (from weeks (16, 61, 62, 69) up to one year), the disease status (from ART naive patients followed without ART (60) or patients presenting at advanced stages of the disease (64) starting ART to patients under ART-mediated HIV RNA suppression (61–63), and even the dosage and components of the prebiotic or probiotic mixtures (16, 62–64, 69). Therefore, this still represents a field of active research, and has been extensively reviewed elsewhere (70).

We know less about the impact of HPV infection on the microbiota epithelial surface integrity, mucosal state, and immune regulation, all factors related to HPV persistence and progression to cancer (46, 71–74). For example, metabolites associated with the vaginal microbiome, including biogenic amines, glutathione, and lipids, have been implicated in HPV persistence (75). It has been described that the microbiota composition can affect all these factors in the context of HPV infection (76–80). A meta-analysis found that Lactobacillus iners and non-Lactobacilli species dominance in the vaginal microbiota is associated with a higher risk of persistent HPV infection and dysplasia (22) compared to the dominance of L. iners and L. crispatus (81, 82).

A Lactobacillus-depleted microbiome has been associated with a proinflammatory environment that may increase malignant cell proliferation and HPV E6 and E7 oncogene expression (22, 83, 84) and promote coinfections by other pathogens such as Chlamydia trachomatis (80). Specifically, it has been shown that HPV down-regulates some innate molecules, such as SLPI, S100A7, elafin, HβD1, and TNFα/LPS that are used by some Lactobacillus species as an amino acid source sustaining their growth, in keeping with their decreased abundance in microbiome analyses of HPV infected individuals (80). Even virome alterations are associated with features of the vaginal microbiota and genital inflammation changes related to HPV infection (85).

Some authors have connected the expression of proinflammatory and chemotactic cytokines related to HPV-induced carcinogenesis with an increased presence of Sneathia or Gardnerella in the vaginal microenvironment (86–90). Furthermore, even though a certain level of inflammation has been described as potentially beneficial to decrease HPV dissemination, several studies have shown specific inflammation markers as related to the progression to a carcinogenic status that could be used as clinical markers to prevent high-grade squamous intraepithelial lesions (46, 86, 90–95) and specific metabolic profiles (96–98). However, most studies in the field are cross-sectional, so it is hard to assess whether the microbiota influences HPV infection or vice versa.

Viruses infecting epithelial cells can profoundly affect the mucosal immune system—the central habitat of the mucosal microbiota—altering the immunological signals required to orchestrate commensal colonization and possibly affecting systemic immune responses and other processes. While from an applied perspective, the gut microbiota functionalities are more relevant for health, most studies have focused on the compositional level, and only fewer studies have focused on the functional consequences. Some microbiota-associated mechanisms possibly influencing the clinical course have been characterized for HIV and HPV ( Table 1 ).

Technologies such as Next Generation Sequencing (NGS), RNA sequencing (RNA-seq), and mass spectrometry (MS/MS) and all their different variations have already been used to describe the global landscape of viral-host interactions. Typically, gain and loss-of-function studies are performed to study the differential expression of DNA or RNA, either of human or bacterial origin, after viral infections. However, since this cannot capture the whole picture of the complex interactions between the virus and the host, mass spectrometry started being used to study the complete proteome, secretome, and metabolome, and even for bacterial identification in clinical microbiology (133). In addition, some studies have used these technologies, even performing integration of some of them (134), to study the role of the microbiome in the inflammation state produced by HIV infection (20) and reviewed in (24, 46, 135) and in pathogenesis and progression to cancer after HPV infection (74, 93, 136).

Improvements in meta-omic techniques have mainly been used to study the totality of the aimed compounds (genes, proteins, and metabolites) in a set of commensal organisms (metagenomics, metaproteomics, and metametabolomics). Currently, sophisticated versions of these methodologies are becoming more commonly used. For example, shallow metagenomics sequencing is being used to obtain strain-level resolution (137). This, together with the development of advanced computational methods (138), is increasing our resolution allowing the identification of novel strains with probiotic potential, an unmet need by previous studies with prebiotics or probiotics in PLWH (39, 64). Other thriving methods include single-cell technologies, which allow isolating, culturing, and characterizing the genomes and transcriptomes of individual microbes in complex communities (139), or tridimensional mapping of the host microbiota interactions within the mucosa, which is advancing our understanding of the microbiota-immune response interactions to the next level (140).

Inside and outside the HIV and HPV fields, the lack of methodological standardization is one of the main limitations in the study of the microbiome and challenges reproducibility (141). For example, a comparison of the clinical impacts of the use of probiotic showed very different results (142) (see Table 2 ). Although, as discussed before, technologies are improving, and now is possible to perform whole genome shotgun sequencing to enhance the detection of diversity, prediction of genes, and accuracy of bacterial species detection (166). However, it is also important to complement the studies by using omics other than genomics to obtain information at the functional level, although there are also challenges regarding the standardization of these methodologies (136, 167, 168). One of these challenges is the integration of datasets (169), which has led to the proposal of the use of machine learning and artificial intelligence for this task, which also have intrinsic limitations (170).

Omic technologies result in compositional profiles and large taxonomic lists for which we lack culture methods in most cases. ‘Culturomics’—a high-throughput culture method— and MALDI-TOF mass spectrometry allow the growth of fastidious bacteria together with the identification of several bacterial species and longer incubation periods. However, these techniques have only allowed us to partially overcome the previously mentioned limitations (171). Furthermore, validation of results obtained from the omic techniques is challenging since, in most of the cases, if validation is performed, only a few of the most statistically significant hits are selected for validation. Even when results are validated, any assumption made or reductionist approach used in the experimental design need to be revisited in order to ensure that the results are physiologically relevant and translation to their clinical use can be performed.

In the case of microbiota studies, omic techniques may often result in compositional profiles and large taxonomic lists for which we lack culture methods in most cases. ‘Culturomics’—a high-throughput culture method—allows the growth of fastidious bacteria and more extended incubation periods, and MALDI-TOF mass spectrometry allows the identification of several bacterial species. However, these techniques have only allowed us to partially overcome the previously mentioned limitations (171).

Thus, the previously described shortcomings pose an enormous challenge to unleashing the clinical potential of microbiota role in medicine. If we want to assess the causal-effect relationship better and move towards applied microbiome research, it will be necessary to start with a clinical question and use the most consistent methodology to perform hypothesis-driven research that identifies convincing interactions and confounders. For this, we will need to define first the best hypothesis inspired by a clinical question. Then, from the hypothesis, we should carefully design the experimental approaches (e.g. different omics) and analysis (e.g. network models) and further perform experimental validation, including controls and complementary data sets (e.g., qPCR to confirm sequencing, immunoblot to confirm proteomic, fluorescence resonance energy transfer to confirm AP-MS data, infection kinetics, results validation in external cohorts, etc).

For example, we recently sought to solve a clinical need using applied microbiome research. We asked whether the microbiome could be harnessed to improve the prevention of anal precancer—a leading neoplasia in PWLH— for which we need better screening tools. After investigating a discovery and a validation cohort of at-risk patients, we discovered twelve proteins, previously reported to be associated with cancer progression, that were overexpressed in the anal bacteria from subjects with precancerous lesions. Since these proteins contribute to succinyl-CoA and cobalamin production, we measured the intracellular bacterial concentrations of these metabolites. We discovered that cobalamin and succinyl-CoA were increased in the anal microbiome of patients with anal precancer and overperformed the reference test—anal cytology—. Furthermore, we validated the findings in an external validation cohort, and we demonstrated greater in vitro production of succinyl-CoA and cobalamin in bacteria associated with HSIL or cancer vs. those presumably protective (172). Therefore, starting from a clinical question and integrating data from different omic levels we were able to define a new microbiome-based tool that could help in the prevention of a common cancer in PLWH by discovering two powerful biomarkers of anal precancer that could improve anal cancer prevention.