The most common mode of HIV-1 spread is through heterosexual intercourse, and a vast majority of new infections globally occur in females. Immunobiological investigations of HIV infection involving the cervical mucosa are of paramount importance, especially to better understand the events associated with initial infection as well as viral dissemination in the host. To attain successful establishment of infection in the host, HIV virions need to cross the genital mucosal barrier and/or encounter certain immune cells that can transfer the viruses to target cells in the epithelium (1, 2). The target tissue for initial viral invasion appears to be the simple columnar and the stratified squamous epithelia of the endocervix and ectocervix, respectively (2). Evidence suggests that 24h following exposure, dendritic cells (DCs) with captured virions reach the draining lymph nodes where they transfer infectious HIV to bystander cells (2, 3). The primary target cells of HIV and productive viral infection appear to be tissue-resident memory (TRM) CD4+ T cells (3).

The mucosal barrier has several defense mechanisms to deploy against diverse types of invading pathogens: the mucus as physical barrier and a multitude of antimicrobial factors that are produced both by epithelial cells as well as leukocytes to protect the host. Given that mucosal epithelial cells represent the front-line barrier for HIV to surpass before establishing an infection, the cells do play a paramount role in viral acquisition and dissemination (4). Besides, epithelial cells also produce several antimicrobial factors as well as regulate mucosal integrity toward molecules and microbes that translocate across the cellular layer. In addition, they also express pathogen recognition receptors (PRRs) that respond to pathogenic invasion by triggering the initiation of type I and type III interferon (IFN) responses. Such epithelial responses have been shown to restrict HIV replication in macrophages (4). Furthermore, HIV exposure also increases the production of inflammatory mediators such as TNF by epithelial cells (5). Secreted antimicrobial factors include α- and β-defensins and lysozymes are part of the mucosal defense (6). In addition to secreted factors, local microbiota and underlying infections also appear to impact the mucosal defense and prevent the initial binding of HIV. Inflammatory conditions of the vagina, such as bacterial vaginosis, are associated with increased risk of HIV acquisition (7, 8). Pre-existing sexually transmitted infections, such as herpes simplex virus type 2 (HSV-2), enhance the risk of HIV acquisition due to recruitment of more target cells into the genital mucosa (9). Furthermore, HSV-2 modulates the mucosal immune cells to be more receptive to HIV infection (10, 11). The presence of pro-inflammatory cytokines produced during infection and inflammation, especially TNF, is associated with enhanced HIV infection (12, 13). Interestingly, the levels of type I IFN can have pleotropic effects depending on the circumstances, and have been shown to increase both uptake and dissemination of HIV as well as limiting viral replication and protecting against HIV acquisition (14). Other factors that are likely to influence the initial mucosal responses to HIV and other pathogens include semen, which has been shown to induce inflammation and increase HIV acquisition (15). The viral transmission fluid also contains factors that inhibit viral infection, such as antibodies, and several types of lectins (16).

The female genital compartment contains an array of soluble factors that participate in local innate immune defenses; notable among them are complement proteins such as C1q and C3. During the initial phases of infection, the transferred HIV virions are opsonized by complement proteins (17) such as inactivated C3b (iC3b) and C1q that are attached to the viral surface (18). The complement protein C3 when activated forms C3b that can be part of the complement cascade that destroys pathogens or infected cells by cytolysis and virolysis. Then again in the case of HIV, C3b is converted to iC3b by complement regulatory factors found in the viral envelope and thereby protecting the virus from virolysis (19). In addition, C1q adheres to gp120 and gp41 and this process appears to be particularly efficient in the presence of anti-gp41 antibodies (20). The opsonization pattern of the virions will be affected by the surrounding microenvironment and the type of tissue or body fluid (18) due to different levels of antibodies and complement factors as well as tissue/fluid specific factors. Virus particles that are transferred via mucosal transmission should be opsonized by complement and other factors present in cervical secretions or seminal fluids, including HIV specific antibodies if the donor is HIV-1 seropositive (18). We have previously shown that complement opsonization influences HIV infection by enhancing viral uptake by emigrating cervical mucosal DCs (21). Furthermore, complement opsonization of HIV dampens the inflammatory and antiviral responses of in vitro cultured immature DCs (22) and affects the pattern of chemotactic factors produced by the DCs, which in turn alters the migration of NK cells (23). In this light, it is critical to explore the effects of opsonization on the initial phases of HIV infection to advance our understanding of what factors that are important for the establishment infection.

The cervical mucosa consists of multiple layers of different tissue types, with the endocervix being lined with a single layer of columnar epithelium, a transformation zone where the columnar cells change into squamous cells, and the ectocervix lined with stratified squamous epithelial cells. The different regions of the female reproductive tract contain diverse subsets of DCs and T cells (24–27). DCs play a major role in the dissemination of HIV infection in the host (25). DC subsets are found both at the basal membrane as well as in the outer layers of the epithelium, and represent one of the first cell types to encounter HIV in the cervical mucosa (28). DCs and myeloid DCs in cervical mucosa play an important early role in the capture the HIV and harboring of infectious virions and in the transmission of HIV to CD4+ T cells (29, 30). For instant, the epithelium also contains CD1a+ DCs that reportedly preserve HIV virions and support their replication (31).

DCs are efficient in capturing, effectively preserving, and transporting the virus via the mucosa to infect T cells in the lymphoid organs (32). The CD4+ T cells in the cervix are located close to the basal membrane but can also be found within the stratified epithelia of the ectocervix, as well as within the columnar epithelia of the endocervix (33). It is also worth noting that the T helper (TH) 17 T cell subsets are more abundant in the ectocervix and endocervix than in the endometrium. Further, it has also been reported that endocervical CD4+ T cells are highly susceptible to HIV infection in vitro, whereas the same cell type present in the endometrium are relatively resistant (27). In addition to DCs and T cells, the cervical mucosa also contains macrophages and CD14+ myeloid cells; the latter being the most abundant immune cell type in the tissue (33).

It has been reported that the HIV transmitter/founder virions are foremost CCR5-tropic, and that a vast majority of the new infections is established by one single transmitted founder virus (34, 35). The ability to enter target cells is attributed to the HIV envelope (Env) protein that encompasses the gp120 and gp41 subunits, and notably, on the founder viruses the gp120 appears to have unique phenotypes (16, 36). During the initial phases of HIV infection, the viral population is markedly homogenous. After the initial phase, a rapid virus evolution occurs, likely attributed to the high replicative potential of HIV, leading to increased rates of genetic diversity, and hence a highly heterogenous HIV population in an infected individual (16, 34, 35).

The ex vivo HIV infection of cervical mucosa has been studied by several groups including ours (1, 2, 28, 31, 37, 38), but the initial immune responses of the human cervical mucosa to HIV exposure have not been well characterized. So, we endeavored to investigate the effects of either free or complement-opsonized HIV on the initial events during viral transmission, using a cervical explant model to get insights of factors that facilitate or protects against viral transmission that could be utilized for the development of effective protective vaccines/microbicides. We used RNA sequencing and proteomics to delineate the molecular events underlying the differences in gene and protein expression after exposure to the free and the complement-opsonized HIV. We found that the infection levels of the cervical mucosal tissue and emigrating immune cells were enhanced after exposure to complement-opsonized HIV and complement and antibody-opsonized HIV compared to free HIV. This increase in infection was mirrored in the biological function and disease pathway analyses of the transcriptome and proteome of the tissues, with multiple pathways supporting viral infection and replication significantly enriched/activated. Signaling pathways involved in immune activation were induced 6h post exposure, but this activation was downregulated at 24h post exposure, both at the transcript and protein levels indicating an active immune suppression induced by HIV. These early events occurring in the cervical tissue are most likely crucial for the establishment of the viral infection and for persistence.

The cervical mucosa represents the most common route for HIV transmission and most of all new infections occur in women (1, 2). The effects of initial HIV exposure, i.e., the first 24h, on the cervical tissue has not been well elucidated and was therefore the scope of the current investigations. By using RNA sequencing and LS-MS proteomics to get the transcriptome and proteomic profiles, respectively, the overall effects HIV exposure exerted on the cervical mucosal tissue were investigated. Considering the stringent statistical analyses adopted, the findings assume paramount importance for advancing our understanding of the events during the initial phases of exposure to HIV and the establishment of infection. To start with, we verified that the complement-opsonized HIV presented a higher level of infection of the cervical tissue compared to free HIV and in the emigrating cells had a preference to infect DCs over CD4 T cells, which is in agreement with previous finding from our laboratory (21). The higher HIV-1 infection of DCs is in line with findings from our and several groups for monocyte derived DCs (53–56). The observed increased HIV infection of the emigrating cells with the complement-opsonized virus, was associated with a significant upregulation of the pathways and factors involved in the diverse aspects of viral infection, viral entry, and viral replication, both at the transcriptome as well as the proteome levels in the cervical mucosa and should reflect the effects in vivo as the transferred HIV virions should be complement-opsonized (17). Importantly, the effects seen in the tissues on the transcriptome and protein levels are not restricted to a direct infection of immune cells, but could also be attributed to the activation of PRRs and other innate pathways in the epithelial cells and immune cells that are exposed to but not productively infected by HIV but produce pro-inflammatory chemokines and anti-viral factors as shown by other groups (4, 57, 58). Thus, the effects seen in the tissues on the transcriptome and protein levels in this study are likely to be attributed to the activation of mucosal cells and not restricted to a direct infection of immune cells. Besides the activation of PRRs the virions interactions with target cell in the mucosa involve an array of receptors including a4b7 for CD4 T cell infection (59) and Siglec-1 for DCs capture and transfer of cell associated virus to new cells (60). This important early role of DCs and CD14+ myeloid cells in the capture the HIV and harboring of infectious virions and in the transmission of HIV to CD4+ T cells (29) might also play an important role in our system. Besides the role of DCs in capturing the virions in the upper epithelium, Rhodes et al. showed that HIV-1 penetrated to the sub-epithelial layer where the myeloid cells, i.e., CD14+ CD1c+ monocyte derived DCs and cDC2s, became HIV-1 infected and transmitted the virus to CD4+ T cells (30). Integrated HIV DNA was found early in T cells but not myeloid cells indicating the initial replication occurred in the T cells not myeloid cells in the cervical mucosa (29). We found that both the mucosal T cells and DCs to productively infected after three days and this is accordance to findings by several other groups (1, 2, 28, 31, 37, 38).

There are differences in tissue composition throughout the female genital tract, i.e. between the endocervix and ectocervix, and it is clear that both these sites are targets for HIV infection (61). We found infection in the cervical endo and ecto tissue and in emigrating DCs and T cells, with higher infection induced by complement-opsonized HIV in the tissue and DCs compared to infections with free virions, which confirms our previous findings (21). The immune response of cervical mucosa is affected by the menstrual cycle (62). A “window of vulnerability” to HIV-infection during the secretory phase has been proposed (63). The majority of tissue donors in this study were older than 45 years, i.e., pre-menopausal or menopausal. Thus, we consider the menstrual cycle to have little or no impact on HIV infectivity in our model.

Proinflammatory cytokines and chemokines are constitutively expressed in the female genital tract (64, 65), which has been considered to be immunologically active (65). This may explain why only restricted changes of many inflammatory factors in response to HIV were observed at the protein level, despite the triggering of inflammatory responses at the on gene/mRNA level. Another important aspect of the initial events on both mRNA and protein levels, during the early phases of HIV exposure, is the responses exerted by the epithelial cells lining the cervical mucosa. SIV transmission via vaginal mucosal challenge has been shown to induce MIP-3α/CCL20 signaling in endocervix in combination with innate immune and inflammatory responses to infection in both cervix and vagina (66). In the initial response after SIV exposure, the cervical mucosal epithelium played an important role in the production of inflammatory chemokines including CCL3, CCL20 and CXCL8 (58). These chemokines are crucial for the recruitment of macrophages, plasmacytoid DCs and CD4+ T cells to the site of infection, which preferentially are clustered below the cervical epithelium in a chemokine dependent manner (58). In addition, SIV rapidly induced a broad spectrum of proinflammatory responses in the epithelium of the cervical mucosa in rhesus macaques, characterized by an upregulation of the NF-kB signaling pathway and a reduction in the glucocorticoid receptor signaling pathway mainly in the epithelial cells lining the mucosa (67). In our data we also found an upregulation of CCL20, CCL2 and CXCL11 after 6h of HIV exposure. In addition, we could measure an increase in CXCL8 on mRNA level induced by F-HIV, and on protein levels for all HIV groups. At the 24h time point, transcription of these chemokines was downregulated. Thus, it can be hypothesized that the expression of cytokines seen in the SIV model might be driven by immune cells migrating into the tissue. In accordance with Shang et al. (67) NF-kB signaling was upregulated during the early phase of HIV exposure and establishment of infection. Combined, these results suggest that female genital tract epithelial cells are of major importance in our system since their modulation of the local mucosal environment will affect the recruitment of immune cells and the establishment of HIV infection.

HIV exposure has previously been shown to induce a transient activation of immune responses in different models (22, 46, 68), which is in line with our present findings. The initial events and activation of e.g., inflammatory responses in the mucosal tissue at the 6h post HIV exposure may facilitate the establishment of infection. It has been reported that inflammation, even at low asymptomatic levels, could increase the risk of HIV and SIV acquisition (66, 69, 70). Transcriptomics showed an activation of inflammatory pathways in the tissues exposed to complement-opsonized HIV at 6h. Enrichment of factors related to inflammatory pathways were not reflected at the protein levels at 24h. Active suppression of mRNA translation into protein caused by, for instance micro RNAs may contribute to lack of congruence (71), as well as proteasomal protein degradation via ubiquitination, a pathway know to be exploited by HIV (72), or a threshold value of mRNA levels for protein synthesis that is not reached (73). HIV has previously been shown to modulate mRNA processing and translation at multiple levels and to redirect protein synthesis toward a more beneficial state for viral replication and dissemination (68).

In the combined endo and ecto cervical tissues, pathway analysis showed the activation of the IFN signaling pathway with IFNA and IFNL1 as top enriched upstream regulators after 6h exposure. The enrichment of factors related to IFNA and IFNL1 was more pronounced in tissues exposed to F-HIV than C-/CI-HIV at this time point. This is in accordance with previous findings, wherein we have shown a CR3-mediated suppression of antiviral pathways in DCs exposed to complement-opsonized HIV (22).

Metabolic changes have been studied in many aspects of inflammation and infection, but not in cervical tissues in the context of HIV infection. After HIV exposure, there was enrichment of factors of the PPAR and LXR/RXR pathways at the 24h time-point. Both transcriptomics and proteomics indicated activation of the LXR/RXR signaling pathways. Resting and memory immune cells are mainly metabolically inactive but can after stimulation rearrange their metabolism towards a faster production of ATP and important building blocks for the induction of a strong immune response (74). LXRs forms heterodimers with RXRs and the complex acts as a modulator of lipid and cholesterol homeostasis as well as inflammatory signaling (75). LXR signaling is associated with repression of inflammation through inhibiting NF-κB (50). Activation of these pathways are therefore indicative of active suppression of inflammatory signaling in tissues exposed to HIV. The transcriptomic data also suggest inhibition of inflammation after 24h of HIV exposure, since there were negative Z-scores, i.e., inhibition/suppression for “Toll-like Receptor signaling” and “Role of Pathogen Recognition Receptors in Recognition of Bacteria and Viruses” at this timepoint. The anti-inflammatory signaling at 24h was most pronounced after exposure to complement-opsonized HIV. The protein kinase A (PKA) signaling pathway (47, 76, 77), was activated for C-HIV and CI-HIV. The PKA together with cAMP signaling is an important modulator of many different pathways involved in immune activation and inhibit transcription of several inflammatory cytokines induced by the NF-κB pathway (77). Furthermore, cAMP have previously been shown to be increased in HIV positive individuals (76, 77). Together these data suggest suppression of the cervical tissue inflammatory response, and possibly a shift to an immunosuppressive milieu in the tissue, preferentially induced by opsonized HIV at 24h post exposure.

HIV opsonized by serum and serum-containing HIV-specific antibodies increase the internalization of virions via receptor-mediated endocytosis (56, 78) and the receptors involved in the viral interaction and uptake. The receptors involved in endocytosis include complement receptors and Fc-receptors. Increased levels of proteins involved in endocytosis after C-HIV exposure, in the cervical tissue explants, may correlate with increased viral uptake and increased viral dissemination. We have previously shown that C-HIV binding and signaling via complement receptor 3, induce suppression of the antiviral and the inflammatory response in immature DCs via the complement receptor 3 engagement (22). This was mainly due to fixation of iC3b molecules on the viral surface during HIV opsonization (19). The effects complement HIV-1 exert on the DCs has been shown to depend on the balance of CR3 and CR4 on the DCs, with high level of CR3 give rise to suppression of inflammation and antiviral activities, whereas high CR4 induce and activation (79). The presence of pre-existing HIV specific antibodies such as anti-gp120 and anti-gp41 during complement opsonization will increase C1q deposition on the viral surface (20). Activation of Fc-receptors by the complement-opsonized virions will also influence the ensuing response (80). For example, an activation of FcγRIIB induced by the formation of SIV specific immune complexes following SIV-vaccination of rhesus macaques have been shown to induce immune suppression in the mucosal epithelium (67). Modulation of Fc-receptor signaling might also have affected the present findings, since these receptors can transduce a wide variety of different immunological and HIV specific effector functions (80). The differences in pathways activated by C-HIV and CI-HIV could be attributed to differences in Fc-receptor engagement. In addition, differences in micro-RNA expression between C-HIV and CI-HIV exposed tissues may contribute to differences in inflammatory signaling by modulation of Fc-receptor induced signaling.

In addition to receptor-mediated immune suppression, there may also be a regulation of translation of mRNA into protein via miRNAs. miRNA dysregulation has been shown during HIV infection and is also involved in breaches in the gastrointestinal integrity found in HIV-infected individuals (81). Nonetheless, we did not find these affected in our system, and for the miRNAs with significant fold change in our system, the functions remain unknown. Potentially, they might be involved in limited inflammation and antiviral responses as well as increased HIV infection, seeing their important role in regulating responses, seen in our complement-opsonized HIV- exposed tissues and hence needs further investigation.

One limitation of the study is the use of spinoculation to infect the mucosal tissue (37) seeing that this method allows the virions access to the tissue from more than the apical/epithelial side. The rational for using this method was that it achieves a high level of HIV-1 infection compared to other infection methods.

In conclusion, we observed enhanced HIV infection in the cervical tissue when the virions were opsonized with complement and/or immunoglobulins. Pathway analysis of cervical tissue gene expression showed enrichment of factors related to the upregulation of different immune modulatory and inflammatory pathways that were time dependent. Proteomics showed that most of the immune pathway genes activated on transcriptional level did not translate into proteins. We found evidence of active suppression of immune responses, which offers an environment advantageous for HIV infection. Complement opsonization is critical in the early events of sexual transmission of HIV infection. The effects exerted by the complement-opsonized virions should be considered in the design of drugs and strategies to prevent HIV transmission.

The data presented in the study are deposited in the NCBI GEO repository, accession number GSE160774.

The study was approved by the Linköping University Ethical Review Board (Ethical permit EPN M206-06). Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

CS, RE, MK, EC, MS, SN, ML, ES, KB, and AB performed experiment and/or analyzed data., NB provided human cervical tissue samples. CS, RE, SN, ES, and ML wrote the manuscript. NB, KB, EC, MK, and AB edited the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported in part with Federal funds from the National Cancer Institute, National Institutes of Health, under Contract No. HHSN261200800001E.

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.