Human immunodeficiency virus (HIV-1) is the causative agent of acquired immunodeficiency syndrome (AIDS). In 2015 36.7 million people were living with HIV/AIDS worldwide, 1.1 million of which died of AIDS related illnesses (WHO factsheet, 2016 ). In sub-Saharan Africa, the most affected region, 25.6 million people were living with HIV in 2015. Current treatment of HIV-1 infection consists of antiretroviral treatment (ART). This is a combination of HIV drugs that lowers the viral load number of HIV in the body, which in turn helps the immune system recover and reduces the chance of transmission (WHO factsheet, 2016). Currently 17 million people HIV-infected people with HIV are on ART therapy (UNAIDS). However, this treatment is not a cure and the HIV pandemic is still growing. In order to find a solution to the pandemic, the mechanisms behind HIV-1 transmission need to be unravelled.
HIV-1 is transmitted through bodily fluids such as blood, vaginal fluid, breast milk and semen (ref ). The most common way route of HIV-1 transmission is sexual transmission across the genital mucosa. Through which cells exactly the virus enters the body and what events occur after that remain under discussion (de Witte et al., 2008). The low transmission rate of HIV-1 suggests that the epithelium functions as a protective barrier and the cells within this layer might be protected from HIV-1 infection (de Witte et al., 2008). Viral loads, viral variants and genital co-infections are some of the factors that are also involved in whether HIV-1 is sexually transmitted to a person (de Jong et al., 2008 ).
HIV-1 transmission in humanshuman HIV infection requires the dissemination of the virus from the site of infection to secondary lymphoid organs and more specifically their T cell zones (Fauci, 1996). The initial HIV-1 infection of cells is mediated by binding of gp120, a surface subunit of the envelope protein (Env), to CD4 on the target cell (Kwon et al., 2002). The conformational change of gp120 that occurs next allows for its binding to one of the chemokine receptors, chemokine receptor type 5 (CCR5) and C-X-C chemokine receptor type 4 (CXCR4). Fusion of the viral and host membranes is then initiated by insertion of gp41, the other Env subunit(Ref freed 2001 bijv?).
After the virus has infected the CD4+ T cells it undergoes extensive viral replication (Fauci, 1996). The replication in the helper T cells leads to depletion of this cell type , which is needed in adaptive immune responses and renders the body susceptible to infection with opportunistic pathogens and cancers. A strong response in CD8+ T cells has been linked to slower disease progression, although it does not lead to elimination of the virus (Pantaleo et al., 1997).
Although CD4+ T cells seem to be the target of HIV-1 it is likely that innate immune cells within the mucosal tissues are the first to encounter or be infected HIV-1 since transmission mostly happens across these layers (ref).
Besides T helper cells, mucosal innate immune cells such as dendritic cells (DCs) and Langerhans cells (LCs) also express CD4 and CCR5, which means they could potentially be infected by HIV-1 (ref). These cells are antigen-presenting cells (APCs) that have the unique capacity to recognize antigens on pathogens in peripheral tissues and then mature and migrate to the draining lymph nodes to activate naïve naive T cells via antigen presentation on MHC molecules (Hamimi et al. 2016, Geijtenbeek et al., 2000b). An important factor in innate immunity is the expression of pattern recognition receptors (PRRs). Immature immune cells, such as DCs and LCs can recognize pathogen associated molecular patterns (PAMPs) on pathogens with their PRRs and initiate the internalization, degradation and antigen presentation of pathogens to T cells (ref). Furthermore, the recognition of pathogenic structures by PRRs triggers the activation and maturation of the cells (Mesman & Geijtenbeek, 2012). Initial contact with HIV-1 in the body might not be through co-receptors and CD4 but via PRRs on innate immune cells (Geijtenbeek et al., 2000a).
Examples of PRRs are the Toll-like receptors (TLRs) and the C-type lectins, or CLRs. The latter has the ability to bind carbohydrate structures on HIV-1 via their carbohydrate-recognition domains (CRDs) (ref). Recognition of the sugar motifs on pathogens can leads to internalization, degradation and loading of antigens onto MHC molecules , which can then be presented to cells of the adaptive immune system(ref). Polymorphisms in the CRD can alter CLR function (Ward et al., 2006). The function of innate immune cells in HIV-1 transmission is also influenced by expression of CLRs (Mesman & Geijtenbeek, 2012). The response of the immune cell is modulated by the type of pathogen that is recognized by the PRR (van Kooyk & Geijtenbeek, 2003). Therefore, Meaning that recognition of different pathogens leads to different responses.
HIV-1 is one of the pathogens that has developed the ability to use CLRs for enhanced infection of target cells (van Kooyk & Geijtenbeek 2003). One of the CLRs that seems to be involved in this mechanism is DC-SIGN on DCs.
However, Langerhans cells are a subset of DCs that express a similar CLR called langerin langerin, a CLR expressed on LCs, has shown to have an oppositeopposite protective role againstin infection with HIV-1 by providing protection from infection (de Witte et al., 2007). These observations led to the question whether the expression of CLRs is beneficial to the host in preventing viral infections, and in particular HIV-1 infection. In order to answer this question, the function and regulation of different CLRs that are present on innate immune cells will be assessed. Additionally, the activation of virus specific adaptive immune responses via CLRs will be determined. Furthermore, there will be an evaluation on how CLRs might contribute to the bottleneck in founder viruses. Lastly, there will be an assessment on how conserved the receptors are.
The binding of HIV-1 by DCs is facilitated by DC-SIGN, which binds to the Env protein, gp120 on HIV-1 (Geijtenbeek et al., 2000a). DC-SIGN is heavily expressed in tissues that are considered to be target sites for transmission, such as the endo- and ectocervix (Hirbod et al., 2011; Hirbod et al., 2009). DC-SIGN expressing cells DCs in the rectum have also been demonstrated to bind and transport HIV very efficiently (Gurney et al., 2005). Additionally, foreskin tissue has also been shown to have abundant DC-SIGN expressing cells DCs in close proximity to potential target cells, the CD4+ T cells (Hirbod et al., 2010). The last observation could provide an explanation for the fact that circumcised men are less frequently infected than those who are not circumcised (Auvert et al., 2005).
Since its discovery, DC-SIGN has been known to have an enhancing role in the transmission of HIV-1. Virions that are bound by DC-SIGN can remain infectious for several days allowing for transmission in trans by DCs (Geijtenbeek et al., 2000a; Kwon et al., 2002). This mechanism allows the transfer of HIV-1 to target cells by DCs that capture virus via DC-SIGN, but the DCs does not appear to become infected themselves . Both immature and mature DCs facilitate trans-infection. When DC-SIGN is blocked with an antibody, the infection of T cells is abrogated showing the enhancing role of DC-SIGN (Geijtenbeek et al., 2000a). Interestingly, the in trans mechanism displayed by DCs does not result in the full degradation of the virus although DCs are rich in degradative compartments that are important in antigen processing (Turville et al., 2004; Kwon et al., 2002; Arrighi et al., 2004). Different models have been proposed addressing how the virus is protected from degradation and what compartments the virus is directed to upon internalization after recognition by DC-SIGN. It has been suggested that DCs infect CD4+ T cells in trans because HIV-1 virions remain surface-bound (Cavrois et al., 2007). However, most studies found that HIV-1 is rapidly internalized into compartments that do not belong to the endolysosomal pathway upon recognition by DC-SIGN (Turville et al., 2004; Kwon et al., 2002). The conditions in the compartment seem to be important in retaining viral infectivity, for example, in the compartment the virus is protected from neutralizing antibodies (Kwon et al., 2002). This in turn could lead to delivery of HIV-1 to uninfected T cells through an infectious synapse since HIV-1 remains competent for infection to a second target cell even after internalisation by DC-SIGN (Kwon et al., 2002; Turville et al., 2004).
However, there is controversy over whether DCs can get infected themselves when binding HIV-1. As mentioned, DCs efficiently transfer HIV-1 in trans via DC-SIGN without being infected but it has also been suggested that there is a second phase long-term transfer in which DCs are infected (Turville et al., 2004). Although virus in both immature and mature DCs is diverted from classical endolysosomal compartments, there still appears to be significant degradation of HIV-1 within 24 hours (Turville et al., 2004, Kwon et al., 2002). In the second phase of increased transmissibility immature DCs facilitate virus replication for transfer to T cells because the residual virus that has not been degraded is still infectious (Turville et al., 2004). Proviral DNA levels rise in immature DCs after 48 hours, indicating that there is de novo production of HIV-1. This suggests that it is a function of time that causes the DC to become infected or not. The infection of DCs could possibly lead to the formation of HIV-1 reservoirs leading to latency of the virus (Coleman & Wu, 2009). DC-SIGN interacting with gp120 has also been suggested to be important for the reactivation of latent HIV-1 in DCs through the NF-kB signalling pathway (Jin et al., 2016).
Although DC-SIGN seems to be involved in the infection of DCs by HIV-1, DCs are probably not directly infected through HIV-1 binding to DC-SIGN since the fusion of HIV-1 with target cells is not mediated by gp120 binding to DC-SIGN., in contrast to CD4 and chemokine receptors that do mediate entry into the cell (Kwon et al., 2002).
Besides capturing and internalizing HIV-1, DC-SIGN has also been suggested to mediate DC-T cell interaction (Arrighi et al., 2004). When DC and T cells cluster, the virions that have been internalized by DCs accumulate to the sites of contact with CD4+ T cells at the infectious synapse (Turville et al., 2004). The formation of this synapse is impaired when DC-SIGN is knocked-out. However, absence of DC-SIGN does not prevent the formation of DC-T cell conjugates (Arrighi et al., 2004). The internalization of HIV-1 X4 virions also still occurs in DCs lacking DC-SIGN although binding is reduced and transfer to target cells is severely impaired(ref). Indicating that DC-SIGN is an important component in the formation of the infectious synapse that can facilitate the transfer of HIV-1 to T cells.
DC-SIGN also has also been found to bind ICAM-3 on resting T cells with high affinity (Geijtenbeek et al., 2000b). This interaction is very important in the initial contact between DCs and T cells to bring them in close proximity but does not seem to be required for HIV-1 transmission to T cells (Geijtenbeek et al., 2000a). Experiments using CD4+ T cells that were lacking ICAM-3 were still infected by dendritic cells expressing DC-SIGN but lacking CD4 or CCR5(ref? ik snap dit exp niet helemaal). Thus, in trans infection it seems that it is the interaction between gp120 and DC-SIGN that is important rather than ICAM-3 binding to gp120.
Through binding to DC-SIGN HIV-1 has evolved to take advantage of the cellular machinery and the physiological functions of dendritic cells (Kwon et al., 2002).
LCs specifically express langerin, a CLR that recognises carbohydrates on pathogens including gp120 on HIV-1. It has similar carbohydrate specificity to DC-SIGN suggesting similar roles (van der Vlist & Geijtenbeek, 2010). However, langerin expression induces the formation of Birbeck granules. These granules are cytoplasmic organelles that are thought to be involved in the processing and degradation of antigens and are uniquely expressed in LCs (Valladeau et al., 2000). In addition to that, HIV-1 co-localises with langerin at the cell surface but also in the intracellular vesicles (de Witte et al., 2007). Providing evidence that when the virus is captured, it is internalized into Birbeck granules. This internalization pathway, that is specific to LCs, is likely central to the ability of langerin to prevent the transmission of HIV-1 and is distinct from the internalization pathway of DC-SIGN, which protects HIV-1, in part, from degradation and promotes transmission. Indeed, blocking langerin with an antibody resulted in enhanced transmission of HIV-1 (de Witte et al., 2007). Additionally, transmission to CD4+ T cells by LCs was also observed when high viral loads were used. This could indicate that when langerin is saturated it is unable to protect against infection of LCs when using high concentrations of HIV-1 and this subsequently leads to HIV-1 transmission to T cells.
Another observation made was that mature LCs down-regulate the expression langerin (de Witte et al., 2007), which could explain why mature LCs are more efficiently infected (Kawamura et al., 2001). Emphasizing that capturing of HIV-1 by langerin serves as a natural barrier against HIV-1 by protecting LCs from infection and transmission to T cells by LCs.
However, not all LCs express langerin, Ballweber et al. (2011) showed that a subset of vaginal LCs do not express langerin. Although these cells lack langerin, they did not become infected. These vaginal LCs did however, migrate to the to CD4+ T cell zones and passed on infectious HIV-1 virus, a mechanism that resembles the first phase transfer via DC-SIGN seen in DCs. This indicates that there might be a pathway in LCs that aids HIV-1 to avoid being directed to the langerin-mediated degradation in Birbeck granules (Ballweber et al., 2011). Apart from vaginal LCs, infection with R5 virus in skin LCs has also been shown to result in a first phase transfer mediated by langerin (Nasr et al., 2014). Furthermore, a second-phase transfer was observed where de novo replication was observed. This in cis transfer was mediated through CCR5 and CD4 and facilitated by HIV-1 binding to langerin, resembling the second-phase transfer in DCs. Administration of a monoclonal antibody to langerin could inhibit both phases, showing that both phases are mediated by langerin. Infection with X4 virus resulted only in first phase transfer in LCs, this in accordance with reports stating that immature LCs only transmit CCR5 (Sarrami- Forooshani et al., 2014). The first phase viral transfer in LCs did take longer than the one in DCs. This difference in kinetics may be due to differences in degradative pathways between the two cell types, resulting in slower degradation of infectious viruses in LCs and longer preservation of HIV-1 (Nasr et al., 2014).
Thus, although there have been reports of a protective function of langerin, there have also been reports showing both in trans and in cis infection transmission toof T cells.
The role of Birbeck granules in the protective function of LCs
As mentioned before, HIV-1 can be captured by langerin and can then is subsequently directed to the Birbeck granules. The origin and function of these rod-shaped granules remains unresolved (Valladeau et al., 2000; McDermott et al., 2004). However, it is clear that langerin induces their formation and antibodies bound by langerin are internalized into the Birbeck granules(ref). Birbecks granules partly overlap with the endosomal recycling pathway and are thus thought to be part of the clathrin-mediated pathway (McDermott et al., 2002). Besides clathrin-mediated endocytosis, lipid raft internalization is a major internalization route (van den Berg et al., 2014). There is evidence that Birbeck granules are part of this internalization route rather than the clathrin-mediated pathway. It has been reported that langerin co-localizes with caveolin-1 (van den Berg et al., 2014). This is an integral membrane molecule that is part of the caveolar endocytosis pathway that arises via lipid rafts. Additionally, they showed that caveolin-1 is needed for uptake of HIV-1 by langerin into Birbeck granules. Inhibition of caveolar uptake and silencing caveolar-1 resulted in increased HIV-1 integration and infection. When inhibiting clathrin this effect was not seen, which means that clathrinthis structure is probably not involved in HIV-1 restriction in LCs. Nevertheless, when inhibiting clathrin there was a decrease in HIV-1 uptake, however there was no effect observed in HIV-1 integration. The decrease in HIV-1 uptake when inhibiting clathrin suggests that HIV-1 uptake is dependent on both caveolin and clathrin mediated endocytosis but HIV-1 is only restricted by caveolin-mediated endocytosis. Indicating that particularly caveolar internalization contributes to the antiviral function of Birbeck granules and langerin(van den Berg et al., 2014). .
How do genital co-infections influence LC function?
There are various reasons that could account for differences in HIV-1 susceptibility of the host (de Witte et al., 2007). Variations in langerin function due to genetic or cellular factors might influence the susceptibility but also the presence of another infectious agent. Studies have shown that sexually transmitted diseases cause a higher susceptibility to HIV-1 infection (De Jong et al., 2008, 2010; Sheffield et al., 2007). There could be several mechanisms behind this; disruption of the epithelial layer by lesions or ulcerations caused by genital diseases might provide better access for HIV-1, as well as the recruitment of immune cells to the site of infection. Langerin on immature LCs in the mucosa have shown to provide a protective barrier to HIV-1 infection. De Jong et al. (2008) investigated whether cytokines and PAMPs present during bacterial genital co-infection breach this protective barrier of LCs to provide access to HIV-1 by inducing LC activation. 24 hours after incubation with TLR agonists or heat killed Neisseria gonorrhea or Candida albicans, two common pathogens that cause infection in HIV-1 positive individuals, skin and vaginal biopsies showed increased TNF-alpha levels compared to non-stimulated biopsies. TNF-alpha is a cytokine that has been shown to enhance HIV-1 replication in T cells and macrophages suggesting a role as therapeutic target (De Jong et al., 2008). Not surprisingly, immature LCs were very inefficient in mediating transmission of HIV-1 in an ex vivo model using human epidermal sheets. However, upon stimulation with inflammatory agents TNF-alpha and Pam3CSK4, a ligand for the TLR1/TLR2 heterodimer, HIV-1 transmission by LCs was increased.
The enhanced transmission was due to different mechanisms, TNF-alpha enhanced viral replication whereas Pam3CSK4 increased viral capture by LCs. Even at low viral titers Pam3CSK4 induced HIV-1 transmission to T cells, which suggests that it overcomes the previously described protective langerin function. The enhanced trans-infection is probably due to increased capture. The fact that TNF-alpha does not enhance transmission at low viral concentrations means that when transmission is enhanced by TNF-alpha it is a prerequisite that LCs already have a level of infection. The enhanced transmission of HIV-1 via TNF-alpha is therefor dependent on langerin.
Genital co-infections can activate LCs through TLR triggering, as shown before with the TLR1/TLR2 ligand Pam3CSK4, indicating that CLRs are not the only PRRs involved in the enhanced transmission when there is co-infection. Furthermore, not only Pam3CSK4 causes an increase in transmission, agonists for other TLRs such as TLR2, TLR4 and TLR5, also cause an increase in transmission in the tissues of some of the donors (De Jong et al., 2008). TLR2 agonists have been shown to increase the HIV-1 susceptibility in LCs, but not in DCs through APOBEC3G down-regulation, a cellular restriction factor of HIV (Ogawa et al., 2009). Strikingly, LPS, a TLR4 agonist, also induced HIV-1 transmission by LCs in part of the donors (de Jong et al., 2008), even though previous research showed that there is no expression of TLR4 on LCs (Flacher et al., 2006). The increase observed in the case of LPS might due to indirect activation of LCs, by for example inflammatory cytokines produced by surrounding cell types (de Jong et al., 2008). When skin biopsies were stimulated with LPS, this led to the production of TNF-alpha, which in turn caused an increase in transmission. There are often variations observed in transmission between donors which could possibly be explained by differences in TLR expression and activity, which results in differences in susceptibility. These observations show that the protective barrier of langerin on LCs can be overcome by inflammatory conditions caused by bacterial or fungal pathogens or high viral loads (de Jong et al., 2008).
In addition to that, ulceration disease such herpes, caused by the herpes simplex virus-2 (HSV-2) has also been shown to enhance HIV-1 infection of LCs in multiple ways (De Jong et al., 2010; Ogawa et al., 2013). LC maturation induced by HSV-2 infection caused a decrease in langerin expression levels (de Jong et al., 2010). Furthermore, HSV-2 competed with HIV-1 for binding to langerin. By decreasing langerin expression and function, the capture of HIV-1 by langerin is reduced and allows for LC infection, which in turn led to increased transmission to T cells. However, this increased transmission of HIV-1 was not due to trans-infection to T cells but it seemed to be due to increased infection of LCs.
As well as HSV-2, viruses that contain TLR ligands might induce the same effects. Indeed, Polyinosinic:polycytidylic acid (poly I:C), a TLR3 ligand, increased transmission by LCs similarly to HSV-2 by inducing maturation and decreasing langerin expression.
In contrast, the study by Ogawa et al. (2013) revealed that other epithelial cells than LCs were primarily infected upon stimulation of HSV-2. Through production of antimicrobial peptide (AMP) LL-37 epithelial cells caused enhanced HIV infection of LCs, thus there is indirect infection of LCs.
The AMP LL-37 produced by epithelial cells under the stimulation of HSV-2 strongly upregulated the expression of CD4 and CRR5 on adjacent LCs and thereby enhanced HIV infection of LCs (Ogawa et al., 2013). Contrary to LCs, MoDCs HIV infectivity was inhibited by LL-37 probably due to down-regulation of DC-SIGN surface expression. Emphasizing the differences in effect of AMPs on HIV-1 infectivity in LCs and DCs and also the effects of langerin and DC-SIGN.
All in all, there seems to be strong evidence that viral, fungal and bacterial co-infections result in increased susceptibility to HIV-1 and there have been observations that langerin on LCs might be involved.
Dendritic cell immunoreceptor (DCIR) is another CLR that is highly expressed on DCs. Like DC-SIGN it acts as an attachment factor of HIV-1 and it can actively participate in both in trans and in cis infection of CD4+ T cells (Lambert et al., 2008). DCIR has also been found to be expressed on CD4+ T cells and the expression is driven by HIV-1 (Lambert et al., 2010). Soluble factors produced by infected cells are responsible for the up-regulation of DCIR in uninfected cells and might promote viral dissemination since the higher surface expression is accompanied by enhancement of attachment, replication and transfer of virus (to bystander cells). Although the receptor plays a role in attachment and transfer of the virus it is less capable of capturing and transferring HIV-1 than DC-SIGN (Jin 2014).
The mannose receptor (MR) can be found on vaginal epithelial cells as well as dermal DCs. Recognition of gp120 by the MR leads to trafficking to lysosomes (Turville et al., 2004). Emigration and maturation of DCs does not only lead rapid down-regulation of DC-SIGN expression but also MR expression (Turville et al., 2002). Correlating with lower binding of gp120 and an increase in CD4-dependent binding of gp120.
Vaginal epithelial cells have also been shown to get infected via MR rather than CD4 (Jadhav).
Higher expression of MR on cells of HIV-1 controllers (Hamimi et al., 2016).
As mentioned before, DCs are very potent APCs and have the unique capacity to activate naïve CD4+ T cells after the DCs have matured and migrated to the secondary lymphoid organs (Hamimi et al. 2016, Geijtenbeek et al., 2000b). The ability of DCs to present exogenous antigen to CD4+ T helper cells relies on the processing of the antigen within the MHC class II-rich endosome/lysosome compartments (Valladeau et al., 2000). In addition to that, DCs have the remarkable ability for cross-priming. This is a mechanism by which exogenous antigen is presented to CD8+ cytotoxic T cells through alternative routing via the MHC class I pathway. The interaction between DCs and T cells thus play an important role in the induction of virus-specific CD8+ T cell responses (Hamimi et al., 2016). Capture of exogenous antigens by MMR and DC-SIGN promote this cross-presentation process. This could possibly mean that higher expression of the surface molecules on MoDCs from HICs aid in the uptake of HIV-1 and they could favour cross-presentation of HIV-1 antigens to CD8+ T cells. (Hamimi et al., 2016) A study by Hamimi et al. (2016) showed that monocyte derived DCs (MoDCs) from HIV-1 controllers (HICs) have a higher capacity of binding HIV probably due to the fact that they have higher levels of surface molecules such as MMR, DC-SIGN and syndecan-3. HICs are rare HIV-1 infected individuals that have an efficient HIV-specific CD8+ T cell response that can control viral replication without the need for therapeutic intervention such as antiretroviral therapy.
In contrast, LCs seem unable to cross-present antigens to CD8+ T cells (Allan et al., 2006). Research in mice with HSV skin infection showed that the cytotoxic T lymphocyte (CTL) activation depended on lymph node-resident DCs rather than on migrating LCs, suggesting that there is antigen transfer between DC subsets. Furthermore, human LCs have also been shown to be unable of cross-presenting antigens derived from measles virus (van der Vlist et al., 2011).
The DC-LC clustering that appears to be required for antigen transfer between these two cell types and subsequent anti-HIV-1 CTL activation is mediated by binding of LCs langerin to hyaluronic acid (HA) on DCs (van den Berg et al. 2015). After treatment with hyaluronidase (hyal)-2, which removes HA, soluble langerin binding to HA was abrogated. In addition, clustering of LCs and DCs was also decreased when hyal treatment was administered. These results indicate that the binding of langerin on LCs to HA on DCs is needed for LC-DC clustering and that it is regulated by the hyal-2 enzyme. Clustering of immature primary LCs and DCs in turn seemed to be required for the transfer of HIV-1 antigens from LCs to DCs since infected LCs alone could not activate CTLs but in the presence of DCs there was CTL activation measured. This suggests that there is antigen transfer between LCs and DCs and which then leads to cross-presentation of the antigens to CD8+ T cells to induce a virus specific response. The transfer of antigen between the two DC subtypes might be an evolutionary conserved mechanism to enhance the speed of the antiviral response of the skin and mucosa, since migration of DCs is much faster than LCs (van den Berg et al., 2015).
and might be useful for vaccine development.
The inability of LCs to cross-present antigen might be affected by the maturation state of the cells. LC from umbilical chord stimulated with interferon (IFN)-gamma led to cross-presentation of exogenous protein antigen by potentiating maturation of the LC (Matsuo et al., 2004).
More infection and proliferation of T cells was seen when LCs formed clusters with CD4+ T cells (Sugaya et al., 2004). This suggests that infection occurs through clustering rather than free virions produced by infected LCs.
Infections by HIV are generally established through a single HIV-1 variant termed the transmitted or founder (T/F) virus (Joseph et al., 2015). As disease progresses the homogenous HIV-1 population that is observed in individuals shortly after transmission, changes into a genetically diverse viral population. This suggests that there is a selection for the transmission of a single founder virus. If the factors involved in the selection for a certain viral phenotype can be elucidated this might inform new approaches for prevention of HIV-1 infection and transmission.
One of the suggested viral phenotypes that could possibly correlate with transmission seems to be the choice of co-receptor HIV-1 uses. As mentioned, the main receptor for infection is CD4, however HIV-1 also requires chemokine receptors for membrane fusion (Sarrami- Forooshani et al., 2014). The most important co-receptors for the two main HIV-1 variants, R5 and X4 viruses, are CCR5 and CXCR4 respectively. Out of these two, CCR5 using strains are the viruses that predominantly cause infection. Interestingly, primary infection rarely occurs with X4 strains although these strains are present in chronically infected patients. In about 50% of chronically infected individuals the viral tropism is switched from R5 to R5X4 or X4 viruses. The switch in co-receptor usage has been associated with accelerated loss of CD4+ T cells resulting in rapid progression to AIDS and death.
Since it seems that most new HIV-1 infections are established via the mucosal surfaces, it is thus important to know how the HIV-1 co-receptors CCR5 and CXCR4 are distributed on the target cells at these surfaces and underlying tissues (Zaitseva et al., 1997). Additionally it might be relevant to learn whether the CLRs on these cell types could be part of causing the bottleneck in selecting for R5 viruses.
R5 and X4 HIV-1 viruses have both been shown to be able to infect immature LCs (Sarrami- Forooshani et al., 2014; Tchou et al., 2001). However, in ex vivo experiments with immature LCs only R5 viruses were transmitted to T cells (Sarrami- Forooshani et al., 2014). The restriction in transmission of X4 virus might be due to the activation state of the LC. Indeed when cells were matured transmission of both virus types could be observed. It might be possible that the immune activation alters the viral internalization pathway or vesicle transport in LCs and this could subsequently allow for efficient HIV-1 transmission to T cells (Sarrami- Forooshani et al., 2014). The infection of LCs by X4 needed to occur at a mature state in order to observe transmission of the virus. LCs that had been infected at an immature state did not transmit virus even after maturation.
These results suggest that selection for R5 occurs at the transmission phase rather than the capability to infect LCs. This is further proven by the fact that both R5 and X4 viruses can be bound by langerin, indicating that if langerin on LCs is involved in the selection for R5 viruses it is probably not due to differences in binding (Sarrami- Forooshani). Furthermore, infection levels of LCs also did not seem to affect transmission since LCs were more efficiently infected by one of the X4 strains compared to the R5 strain, although it did not result in transmission. Immune cells isolated from women with genital ulceration diseases such as syphilis and herpes had increased CCR5 expression on their CD14+ immune cells, such as dermal DCs, obtained from within the tissue of the genital ulcers (Sheffield et al., 2007). If taken into consideration that the viral variants that are usually transmitted use the CCR5 receptor on the host immune cells, this could also explain the higher susceptibility to HIV-1 when there is a co-infection.
A factor that seems to be of importance in creating a bottleneck as well is glycosylation of the HIV-1 glycoprotein Env. This glycoprotein is heavily glycosylated, with the N-linked glycans accounting for approximately 50% of the protein its mass (Leonard et al., 1990). It has been suggested that HIV-1 Env glycosylation leads to shielding of epitopes on Env, which prevents (neutralizing) antibodies from binding to it (Joseph et al., 2015). The shielding function of glycans on Env suggests that their primary role is immune evasion (aangehaald in ping).
Founder viruses on average have under-glycosylation of the Env protein compared to the viruses found in chronically infected individuals (Chohan et al., 2005; Ping et al., 2013). The under-glycosylation of T/F HIV Env suggests that it must somehow be advantageous to the virus in transmission. It might be possible that CLRs select for lower glycosylation of Env because it is easier to bind. Under-glycosylated HIV is bound by DC-SIGN (). However, the course the virus takes after binding the receptor is not sure. Binding of DC-SIGN has been shown to result in transmission in trans (Shen et al., 2014) however, this is contradicted by the results of Montfort et al., (2011) who showed that recognition by DC-SIGN led to degradation of the virus in the endocytic pathways.
Secretions from uterine epithelial cells surrounding innate immune cells have also been shown to decrease the expression of DC-SIGN on immature DCs (Ochiel et al., 2010). Consequently, they inhibited the trans infection of T/F virus by immature DCs. When DC-SIGN was blocked with specific antibodies the inhibition of trans infection of T/F viruses was lifted, indicating that other receptors might be involved in the selection of founder viruses. This suggests a protective role of DC-SIGN induced by the secretions of uterine epithelial cells.
Conservation of CLRs/Polymorphisms
In the study done on the protective function of langerin by de Witte et al. (2007), they found differences in langerin function between skin donors. One donor displayed efficient transmission by LCs, which contradicted the findings that LCs have a barrier function. Polymorphisms might contribute to the differences in efficiency (de Witte et al., 2007). There are several single nucleotide polymorphisms (SNPs) that result in changes in amino acid residues in the carbohydrate-recognition domain (CRD) of human langerin (Ward et al., 2006). The CRD of langerin binds mannose and other related sugars but only have low affinity for monosaccharides. Changes in amino acid residues of the CRD have shown to alter binding affinity of langerin to sugars. This change could potentially mean that certain haplotypes may have increased susceptibility to infection since they bind pathogens less effectively, including HIV-1. Furthermore, there have been recordings of one individual with a very rare heterozygous point mutation in the CRD of langerin that resulted in the loss of Birbeck granules (Verdijk et al., 2005).
Although SNPs are found in the langerin gene, structurally and functionally there is not much difference between species, which is in contrast with DC-SIGN that shows considerable divergence between species (Feinberg 2013).
Polymorphisms in the DC-SIGN gene have been observed and can alter the progression of diseases such as dengue fever (Lozach et al., 2005).
DC-SIGN also recognizes Ebola and cytomegalovirus (CMV) (Alvarez et al., 2002; Halary et al., 2002). Recognition of both these viruses by DC-SIGN leads to infection of target cells in trans and recognition of Ebola also leads to cis infection (Alvarez et al., 2002). Thus, DC-SIGN also acts as a receptor for Ebola and CMV and facilitates their capture and viral infection in a similar manner as for HIV-1. Indicating that it has a conserved function in humans?.
In this review the role of different CLRs on HIV-1 transmission was assessed in order to determine whether expression is beneficial to the host. CLRs have a very important function as PRR and in the case of DC-SIGN also as adhesion receptor (Geijtenbeek et al., 2000). Although, the CLRs discussed here can bind the same protein on HIV-1, the gp120 glycoprotein, the effects of binding are not the same.
DC-SIGN on DCs have been known to promote viral transmission in trans to CD4+ T cells, whereas LCs show protection from transmission via langerin. There is no consensus on what compartment HIV-1 is directed to upon internalization by DC-SIGN, that is, if it is internalized at all. Langerin on LCs, however, appears to exert a protective function by directing cells to a unique degradative compartment, the Birbeck granules.
Also the ability of DCs to evade infection by HIV-1 is also heavily disputed. Some say that in cis transfer of HIV-1 by DC-SIGN is possible, whereas others are only able to show transfer in trans. Either way DC-SIGN appears to promote viral transmission, an effect that is also observed in the transmission of Ebola and CMV.
The barrier function of langerin is further illustrated by a study where blocking of the receptor resulted in efficient transmission. More evidence supporting a barrier function of langerin can be observed when there is co-infection with HSV-2. This resulted in down-regulation of langerin as well as occupation of the langerin receptor by HSV-2, which in turn led to infection of the LC and transmission to T cells. Furthermore, more proof for a protective role is shown by the fact that there is a lack of langerin expression on a subset of vaginal LCs and that showed transmission.
However, the degradtive compartment after recognition by langerin can be evaded by maturation of the cells and high viral loads. The latter causes transmission through a mechanism similar to HSV-2, namely saturation of the receptors. Nasr and colleagues (2014) were able to infect immature LCs however this might have been due to higher viral loads used. Maturation, on the other hand, seems to down-regulate langerin levels.
MR and DCIR are two other CLRs, besides DC-SIGN and langerin that have been associated with transmission of HIV-1.
In the selection for R5 viruses it has been suggested that although recognition and internalization is through CLRs but the CCR5 and CD4 receptors are responsible for the bottleneck that is observed in T/F viruses (Turville et al., 200). Otherwise there cannot be selection for R5 viruses.
Although innate immune cells are important, DCs and LCs are not the only cells in the epithelium so even if they are among the first to encounter HIV-1, they are not the only type of cells that encounter HIV-1. Other epithelial cells, like keratinocytes, might come across the virus and transmit it to T cells in the secondary lymphoid organs. Or they can infect LCs indirectly as described in the herpes study by Ogawa et al. (2013).
The role of CLRs is extremely difficult to study because there are no ideal models. In vitro culturing of human monocytes in the presence of IL-4 and GM-CSF causes these cells to acquire the typical DC morphology and express different DC markers such as DC-SIGN, CD80 and CD86 (de Witte et al., 2008). TGF-beta is added to monocytes in addition to IL-4 and GM-CSF for creating MDLCs. However these are not the same as primary cells, in vitro generated LCs from CD34+ stem cells mediate first-phase transmission after maturation with LPS or TNF-alpha. However, the mechanisms through which CD34+ stem cell-derived LCs govern transmission might be different from primary LCs since these do not express TLR4. Determining the role of the different HIV-1 receptors on primary DCs and primary LCs rather than on cultured moDCs or moLCs, respectively, is therefor essential (de Witte et al., 2008).
...(download the rest of the essay above)