HIV-1 Infection: The Role of the Gastrointestinal Tract


  • Mariangela Cavarelli,

    1. Viral Evolution and Transmission Unit, Division of Immunology, Transplantation and Infectious Diseases, San Raffaele Scientific Institute, Milan, Italy
    Search for more papers by this author
  • Gabriella Scarlatti

    Corresponding author
    1. Viral Evolution and Transmission Unit, Division of Immunology, Transplantation and Infectious Diseases, San Raffaele Scientific Institute, Milan, Italy
    • Correspondence

      Gabriella Scarlatti, Viral Evolution and Transmission Unit, Division of Immunology, Transplantation and Infectious Diseases, San Raffaele Scientific Institute, Via Olgettina 58, 20132 Milan, Italy.


    Search for more papers by this author


The intestinal mucosa has an important role as portal of entry during mother-to-child transmission of HIV-1 and during sexual transmission. Tissue morphology and integrity, as well as distribution of relevant cell types within the mucosa, spanning from the oropharynx to the rectum, can greatly influence viral infection, replication, presentation, and persistence. The relative contribution to transmission by cell-associated or cell-free virus is still not defined for the different routes of transmission. Although the main target cells for HIV-1 replication are the CD4+ T lymphocytes, which are rapidly depleted both in the periphery and in the mucosal tissues, dendritic cells, Langerhans' cells, and macrophages are players in each of these processes. The predominant cells involved may differ according to the tract of the gut and the route of transmission. The microenvironment of the intestinal mucosa, including mucus, antibodies, or chemo-cytokines, can as well influence infection and replication of the virus: their role is still under investigation. The understanding of these processes may help in developing efficient prevention strategies.


The majority of HIV transmissions occurs across a mucosal surface. The interaction of HIV-1 with the mucosal epithelia of the genital, oropharyngeal, and gastrointestinal tract is the first step in the establishment of systemic HIV-1 infection. In specific, the intestinal mucosa is the portal of entry during mother-to-child transmission (MTCT) of HIV-1, which occurs across the upper intestinal tract after ingestion of HIV-1 from maternal blood and cervical mucus during delivery or from breast milk, and during sexual transmission, which instead occurs in the lower intestinal tract through infected body fluids and semen.[1, 2] The colo-rectal mucosa is the primary route of HIV infection in men who have sex with men, and is also an underappreciated contributor to heterosexual infections, as clearly presented by Jacqueline Campbell during the “HIV Workshop: Hormone Regulation of the Mucosal Environment in the Reproductive Tract and the Prevention of HIV Infection” held in Boston in 2013. The risk of transmission during unprotected anal intercourse has been estimated to be at least 10 times greater than during unprotected vaginal intercourse.[3-5] Indeed, anal sex has a considerably higher risk with 1 in 20–300 infections for each sexual act as compared to vaginal sex with 1 in 1000–10,000 infections.[6, 7] As expected, the risk of infection is strongly dependent on the phase of the disease and is almost 10-fold higher during acute infection.[8] Other factors, such as coexisting sexually transmitted infections (STIs), strongly increase the risk of infection.[6, 8] For example, women with cervical human papilloma virus (HPV) infection have a considerable risk for coexisting anal HPV infection,[9] but its role in favoring HIV infection is not known. The potential contribution of anal intercourse to HIV transmission has significant implications for the usage, safety, and delivery mechanisms of microbicides and other HIV prevention modalities.

HIV target cells present in the gastrointestinal mucosa

Tissue morphology and integrity, as well as distribution of relevant cell types within the mucosa, spanning from the oropharynx to the rectum, can greatly influence viral transmission. After crossing the mucosal barrier, the main target cells for HIV-1 replication are the CD4+ T lymphocytes, which are rapidly depleted both in the periphery and in the mucosal tissues. However, many cell type variants of both lymphocytic and monocytic lineage are residing in the mucosal tissues and are potential targets for the incoming virus.

Indeed, a substantial proportion of the total lymphocyte pool of the body is within the gut-associated lymphoid tissue (GALT), which thus represents an important site for HIV infection and replication. The GALT is organized to have a dual function: on one side, the lymphoid structures and, in the small intestine, the Peyer's patches are specialized local inductive sites, considered as zones of T-cell education; and on the other side, the diffuse yet dense populations of lymphocytes and antigen-presenting cells distributed throughout the mucosa, and the intra-epithelial lymphocytes are widespread effector sites, where foreign antigens are neutralized. Mucosal surface antigens need to cross the epithelial barrier to reach the immune inductive site and the effector sites.[10]

Inductive sites are placed in the lingual and palatine tonsils and adenoids of the oral mucosa, in the very last portion of the small intestine and throughout the whole tract of the large intestine with an increasing frequency in the colon and rectum.[11, 12] These primarily consist of resting T and B cells; however, the continuous stimulation by luminal antigens results in cell activation, proliferation, homing, and turnover, which in the case of CD4+ T-cells conceivably serves as a mechanism for viral persistence. In contrast, the proximal small intestine, the jejunum, is practically devoid of organized lymphoid tissue and primarily harbors immunologically ‘activated’ memory CD4+ T-cells with high levels of CCR5, which again can be target of HIV replication.

CD4+ T-cells having a memory CCR5+ phenotype are the predominant cell type also of the imm-une effector sites. Indeed, due to their proximity to the external environment and constant exposure to myriad food and microbial antigens, the mucosal CD4+ T-cells in the gastrointestinal tract are predominantly activated and well differentiated to express a clear memory phenotype.[13] Recently, memory CD4+ CCR5+ T-cells have been identified also in the gut of fetuses and newborn infants.[14] These cells are prone to infection with HIV, and once depleted do not repopulate the lamina propria as shown in untreated SIV-infected macaques with a progressive disease.[11, 15-17] Analogously, HIV-infected patients administered highly active antiretroviral therapy (HAART) have a slow and incomplete restoration of CD4+ T-cells in the gut as compared to the peripheral blood, which was attributed to residual viral replication and persistent immune activation in the gut.[18]

More recently, the population of α4β7 + CD4+ T-cells, which are homing to the gut, was described as a major target of HIV/SIV infection, especially during the acute stage of the disease.[19-21] Indeed, at few days post-infection, SIV gag DNA levels are five times more abundant in α4β7high than in α4β7−CD4+ T-cells.[20] In addition, their frequency in blood correlates with the loss of CD4+ T-cells in the gut.[21] Moreover, animals with a high frequency of a4b7high CD4+ T-cells in the circulation and in rectal tissue seem more susceptible to SIV via rectal transmission.[22]

The intestine and associated lymphoid tissues are home to an extensive network of innate immune cells with antigen-presenting function, such as DCs and macrophages.[23, 24] CD11c+ DCs expressing the HIV-1 receptor CD4 and co-receptors CCR5 and CXCR4, as well as DC-SIGN, are present in the jejunum and can capture HIV-1 when inoculated ex vivo on an explant culture.[25] Immature DCs, which express CD1a and CD207 (Langerin), are detected surrounding the colonic/rectal glands and are sparse in the lamina propria just beneath the epithelium.[26] DC-SIGN+ DCs, which express CCR5 and CD4, present in colon and rectum, are subepithelial and across the entire thickness of the mucosa.[27] Indeed, we showed that colonic CD11c+DC-SIGN+ DC, but not CD68+ macrophages (DC-SIGN+ and DC-SIGN−), extend transepithelial dendrites, and are capable of sampling luminal virus and transfer infection to CD4+ T-cells.[28] We may envisage that these DCs may not only become infected within the rectal mucosa and ferry HIV to the draining lymph nodes but also form a local viral factory in DC-T-cell conjugates.

The intestinal lamina propria contains also a large amount of macrophages; however, intestinal macrophages apparently did not express CCR5 or CXCR4 and consequently were not permissive to HIV-1 infection.[29, 30] However, more recently, a population of macrophages, that express CD14, CD68, CD163, and DC-SIGN (CD209), and that are thus a potential target for HIV, has been described in normal human sigm-oid colon and rectum.[26] Moreover, a comprehensive comparative analysis of potential HIV target cells in healthy individuals revealed a tissue specificity of macrophages. Interestingly, in the rectum, there are more than three times as many CD68+ macrophages expressing the HIV co-receptor CCR5 than in the colon, and as such, rectal macrophages seem biologically closer to the HIV-susceptible CCR5 phenotype in the vagina than the mostly HIV-resistant CCR5 phenotype in the colon.[31] The higher expression of CCR5 on rectal macrophages suggests that the most distal sections of the gut may be especially vulnerable to HIV infection. Thus, the different distribution of potential target cells for the virus along the gastrointestinal tract may suggest that different pathways are used accordingly by the virus to infect and persist.

Mechanisms of HIV transmission across the intestinal epithelium

Most of the vast mucosal surfaces of the gastrointestinal tract is lined by a single layer of epithelial cells sealed by tight junctions (TJ) that exclude pepti-des and macromolecules with antigenic potential.[32] Moreover, many antigens and microorganisms are prevented from contact with epithelial surfaces by local secretions containing mucins and antibodies.[10] The role of the mucus in the establishment of infection has been subject of recent studies focused on the cervicovaginal mucosa.[33, 34] Interestingly, HIV can interact with and penetrate the squamous epithelium of vagina and ectocervix as well as the columnar epithelium of the endocervix, whose structure by the way resembles the columnar rectal epithelium. However, in vivo macaque experiments indicate that the mucus impedes the efficiency of virus penetration in this tissue. In line with this observation, in the intestine, the epithelial cells themselves are well equipped to face a pathogen-rich foreign environment. The apical surfaces of enterocytes are modified by rigid, closely packed microvilli, the brush border, coated with a thick layer of large membrane-anchored, negatively charged mucin-like molecules called the glycocalyx,[35, 36] which constitutes a diffusion barrier and a highly degradative microenvironment. However, the intestinal epithelium is not an impenetrable barrier, and although sexual transmission of HIV through unprotected anal intercourse appears to be facilitated by epithelial damage such as micro-trauma,[37, 38] studies of SIV transmission in monkeys have established that infection may also occur without damage of epithelial linings of the rectal mucosa as it does through the vaginal one.[39, 40]

HIV itself may also cause an alteration of the intestinal epithelial barrier and an increased epithelial permeability. This intestinal barrier dysfunction is observed already during primary HIV infection, coincides with the CD4+ T-cell loss in GALT, and has been implicated as the cause of systemic immune activation during chronic phase of HIV infection (for a review, see[41]). In vitro studies using primary genital cells and intestinal cell lines have shown that dir-ect exposure of the epithelium to HIV determines destruction of the tight junctions and enhances permeability.[42] The mechanism appears to be mediated by the interaction between the gp120 and TLR2 and TLR4 present on the female upper genital tract (endometrial and endocervical) epithelium, which leads to the activation of intracellular NF-kB pathway, and in turn increased production of inflammatory cytokines, specifically TNF-α, by the genital epithelial cells.[43] Similarly, the barrier function of intestinal epithelial cell lines is altered by TNF-α and IFN-γ produced by the epithelium itself after incubation with HIV.[44, 45] In the intestine, also HIV-1 Nef protein has been proposed to induce alterations of epithelial integrity and induction of proinflammatory mediators.[46]

Although epithelial cells express low to negligible levels of CD4, the primary receptor for HIV, virus attachment is mediate by galactosyl ceramide (GalCer)[47] and the co-receptors CXCR4 or CCR5 allow for virus entry in polarized epithelial cell lines.[48, 49] However, there is no clear evidence that intestinal enterocytes become infected in vivo.[50] When SIV is applied to the tonsil's surface of macaques, infectious virus appears rapidly within the T-cells underlying tonsillar crypts, but no infection is detected in the epithelium itself.[51]

HIV spread across oropharyngeal epithelia may be an important route of MTCT, but apparently it is less frequent in adult oral transmission. Virions can traverse adult and fetal oral epithelial cells by tra-nscytosis in vitro.[52] Cell-free HIV-1, which crossed the fetal oral epithelia, is able to further transfer infection to CD4+ T-cells, DCs, and macrophages, whereas infectivity of the virions is greatly diminished during the passage through the adult cells. This is ascribed to the high level expression of anti-HIV innate proteins, the β-defensins, and the secretory leukocyte protease inhibitor (SLPI) by the adult oral epithelial cells. Antibodies against GalCer and heparan sulfate proteoglycan but not against CCR5 and CXCR4 reduce infection, suggesting that chemokine receptors do not play a primary role in transmission through the fetal/infant oral mucosa.[52] Interestingly, virus transmission may be mediated also by infected cells, indeed macrophages but not CD4+ T-cells can transmigrate across the oral mucosa to disseminate infection.

The tonsil's mucosa contains M cells lying above regions where DCs are juxtaposed with CD4+ lymphocytes, whereas LCs cluster in the subepithelial papillae of the oral mucosa. Accordingly, in the primate MTCT SIV model, infection was shown to occur through the surface mucosa of tonsils, where specialized M cells and DCs may transport HIV to the interior of the tissue. DCs of mucosal surfaces can serve as antigen-presenting cells after migrating from mucosal tissues to draining lymph nodes, or in the case of tonsils, they can present antigens locally to cells of mucosal lymphoid follicles.

Transmission of HIV-1 through the lower gastrointestinal tract has been object of intensive studies. In cultures of polarized epithelium, cell-associated, but not cell-free virus is rapidly delivered to the basal surface by a process of transcytosis, and thus, transported to submucosal lymphocytes for infection.[53] Primary intestinal epithelial cells express both GalCer and CCR5, but unlike some cell lines, they do not express CXCR4; as consequence, they apparently allow transcytosis only of HIV isolates capable of interaction with CCR5.[54] CCR5-dependent transcytosis and presentation to subepithelial T-cells could act as a filter, accounting for the limited transmission of CXCR4-dependent isolates from chronically infected patients. However, we have shown that HIV-1 of either phenotype, R5 and X4, can enter colonic epithelial cell lines in vitro.[28] Interestingly, primary jejunal epithelial cells incubated with HIV-1 carry over R5 viruses to receptive target cells,[54] whereas M cells in the Peyer's patches of the digestive epithelium, which deliver samples of foreign material directly to the close intra-epithelial lymphoid cells, selectively transport X4 viral variants through a chemokine-receptor-mediated mechanism.

DCs in jejunum explant cultures are the predominant target cell of R5 HIV-1 early after infection, and leave thereafter the tissue to transmit in trans the virus to lymphocytes.[25] More recently, our group described that cell-free R5 but not X4 HIV-1 attracts colonic lamina propria resident CD11c+ DCs to extend their cellular processes through the intact epithelium. Migrated DCs sample luminal virions and are able to transfer infection to CD4+ T-cells.[28] The migratory process was dependent upon the interaction of the HIV envelope protein with CCR5, and therefore, providing CCR5-using viruses with an additional mechanism to gain access to the intestinal mucosa. In sampling the mucosal surface environment, DCs run the risk of infection. Their antigen uptake and migratory activities, coupled with their ability to form close associations with T-cells, may lead to replication and dissemination of T-cell-tropic viruses that contact mucosal surfaces. It will be important to understand whether these DCs will reside in the mucosa to be involved in antigen presentation and/or be able to reach the lymph nodes to spread further the infection.

LCs of the intestinal and vaginal mucosa were repeatedly described to be the first target for HIV-1 and the vehicle for the virus to reach replication competent cells. Since LC expresses CD4, CCR5, and CXCR4 within the genital tract and also in the intestinal lamina propria,[26] it has been suggested that they can capture HIV and migrate to the draining lymph nodes to transmit the virus to CD4+ T-cells.[55, 56] Whether or not LCs can be infected after exposure to virus is still controversial. Some studies of vaginal mucosal administration of SIV in rhesus macaques demonstrated that submucosal LCs rapidly become infected after virus inoculation while other studies suggest that the CD4+ T-cells are the cells infected and the major source of infectious virus during the acute stages of infection.[57, 58] Ex vivo studies have demonstrated that LCs are not required for, but may aid in viral transmission.[59]

Conclusions and perspectives

The gut serves as vehicle of HIV-1 infection as well as site of replication, persistence, and antigen presentation. The predominant cells involved in each of these processes may differ according to the portion of the gut, where different subsets of cells are present and distributed. The microenvironment of the different portions of intestinal mucosa, including mucus, antibodies, chemokines, or cytokines, can as well influence infection and replication of the virus. Analogously, the role of biological fluids, such as breast milk or seminal plasma, in driving infection is still an open question. Last, we have still to clearly understand the relative proportion of transmission, which is occurring through cell-free or cell-associated virus for the different routes of transmission. Unraveling these processes may help to identify the relevant cells involved during the early phases of infection and develop effective prevention strategies.