In this study, we show that the direct interaction between Siglec-7 and HIV-1 Env gp120 contributes to HIV-1 entry and infection of CD4pos cell targets, including T cells and macrophages. These findings are consistent with recent studies reporting that the interactions between sialic acids present on the viral envelope and Siglec receptors facilitate HIV-1 infection in several cellular targets [28–30].
Siglec molecules have been recognized as ligands of different sialic-acid bearing pathogens [13–15, 40–44], although it is still debated whether resting leukocytes with masked Siglecs are able to directly engage sialic acids on pathogens or if cell activation is required for unmasking these lectin-type receptors. Originally, the interactions between Siglec receptors and pathogens were reported to dampen host immune responses and to set appropriate activation thresholds for regulating cell proliferation and secretion of inflammatory mediators. Subsequently, it became evident that this biological phenomenon represents an advantage for pathogens that evolved to express sialic acids on their surface to evade host innate immune responses [45, 46]. However, recent evidence showed that sialic acid-expressing pathogens could also promote infection and mediate immune responses following their binding to Siglecs [13, 47, 48].
In line with a previous report , we demonstrate that Siglec-7 is able to directly interact with HIV-1 Env gp120 although we found a certain degree of variability among different HIV-1 strains. This phenomenon may have several explanations, including deletions or mutations of critical residues involved in Siglec-7 binding or, most likely, heterogeneity in HIV-1 glycosylation patterns [1–4, 13]. Evidence in favor of the latter hypothesis comes from our results showing that Siglec-7 interaction with HIV-1 is indeed sialic acid-dependent. In this regard, it has also been reported that sialic acid depletion from human CD4pos target cells enhances HIV-1 infection . This suggest that removing cis-interactions between sialic acids and Siglecs on the same cell leads to exposure of Siglecs that, in turn, become accessible to sialic acids exposed on gp120. Moreover, oligomeric modeling of the gp120 trimer shows that complex glycans are exposed on the outer rim, potentially accessible to Siglec recognition . Furthermore, it has recently been shown that different Siglecs, and Siglec-1 in particular, are able to recognize both R5 and X4 HIV-1 gp120 through interaction with sialic acids, thus contributing to HIV-1 infection of monocytes, macrophages and DCs [28–30].
The present study also provides evidence that Siglec-7 binds HIV-1 and enhances viral entry into CD4pos T cells when the soluble form of this lectin-type molecule is added to the culture. Indeed, our data clearly showed that Siglec-7-Fc fusion protein is able to enhance IIIB and BaL infections both in a CD4pos T cell line (PM1) and in primary CD4pos T cells, even though the levels of HIV-RNA were higher for IIIB compared to BaL. This latter phenomenon is likely due to the well-known preferential tropism of the IIIB HIV-1 strain for T cells . Further investigations focusing on the structures, biochemical properties and affinity of these interactions are needed to confirm this working hypothesis. Moreover, we also demonstrate that constitutive expression of Siglec-7 on MDMs markedly contributes to their HIV-1 infection, since specific Siglec-7 blockade induced a reduction of HIV-RNA in MDMs in a 4-hour entry assay. The degree of surface levels of Siglec-7 on monocytes and macrophages is still being debated, as there is one study showing that expression of this molecule on monocytes and MDMs is very low , while several other reports demonstrated that monocytes express high levels of Siglec-7 [9, 11, 12, 46, 52]. In our experiments, the constitutive expression of Siglec-7 on monocytes paralleled the ones observed on NK cells , and remained very high on MDMs even after 7 days of culture with GM-CSF although at lower levels if compared to freshly isolated monocytes.
Given the ability of Siglec-7 fusion protein to enhance in vitro HIV-1 infection of CD4pos T cells, we then determined whether different levels of sSiglec-7 released during the course of HIV-1 infection are associated with different clinical outcomes. We did not find any significant differences in the serum levels of sSiglec-7 between healthy controls and several different cohorts of HIV-1 infected patients. However, AIDS patients showed a trend toward increase of Siglec-7 in their sera compared to healthy donors and, in particular, we found a direct and statistically significant correlation between HIV-RNA and serum levels of Siglec-7 in naïve HIV-1 infected patients, suggesting that a higher viral burden could induce the shedding of Siglec-7 in serum. In this regard, it is conceivable to hypothesize that a possible source of sSiglec-7 are NK cells, since we demonstrated that the exposure of these cells to high chronic HIV-1 viremia induce the expansion of pathological NK cell subsets showing an aberrant receptor repertoire and an unusual negative expression of Siglec-7 [53, 54]. The fact that Siglec-7 fusion protein enhances the in vitro susceptibility of CD4pos T cells to be infected by HIV-1 is also consistent with the persistent high levels of HIV-1 viremia, with the high amount of sSiglec-7 and with low CD4pos T cell counts found in AIDS patients. Interestingly, a CD56neg/Siglec-7neg population associated with high levels of HIV-1 viremia has been described as a dysfunctional NK cell subset that is significantly expanded in the late stages of HIV-1 infection . In agreement with this, sSiglec-7 was found to be significantly increased in the sera of patients with viral load over 10,000 copies/ml, suggesting a possible role for this lectin-type receptor as a biological marker involved in the disease progression. We are currently testing this hypothesis in a larger patient cohort. Finally, an inverse correlation between the serum levels of Siglec-7 and CD4pos T cell counts was observed in AIDS patients, which were also the cohort of HIV-1 infected subjects with the highest serum levels of sSiglec-7. These findings lead us to hypothesize that the presence of remarkable amounts of sSiglec-7, together with a low CD4pos T cell count, high levels of HIV-1 viremia and expansion of Siglec-7neg NK cells, characterize the advanced stages of HIV infection.
For our experimental approaches testing the ability sSiglec-7 to enhance the infection of CD4pos PM1 cell line and primary T cells, we used a dose of fusion protein (10 μg/ml) much higher compared to amount of sSiglec-7 found in the sera of both healthy donors and HIV-1 infected patients (floating within a ng/ml range). We have chosen to use a higher sSiglec-7 concentration in our in vitro approach to maximize ligand-receptor binding. However, it must be emphasized that identical concentrations of an irrelevant protein such as soluble NKp44 was unable to promote infection using the same experimental conditions. Moreover, even though the serum concentrations of sSiglec-7 in HIV-1 infected patients or healthy donors are remarkably lower compared to those used in our in vitro approach, it remains to be determined the amounts of this soluble protein during acute phases of infection when HIV-1 viral load is in the ranges of several hundred-thousand copies . Finally, we need to assess whether higher levels of sSiglec-7 are also observed at gut or cervix mucosal levels, two of the main gates of viral entrance [55, 56]. Indeed, macrophages as well as NK and T cells play an active role within these tissue sites in context of innate and adaptive immune responses and establishment of viral reservoirs in the CD4pos targets [57–59].