We have described a panel of MAbs that recognize and discriminate among DC-SIGN family molecules, providing an additional tool in characterizing DC-SIGN function and understanding its role in primate lentivirus pathogenesis. Staining analysis of DC-SIGN- and L-SIGN-expressing NIH 3T3, HEK293T, and THP-1 cells (data not shown for all cell types) demonstrated that among the seven MAbs, three were DC-SIGN specific, three were cross-reactive with DC-SIGN and L-SIGN, and one was L-SIGN specific.
In addition, myeloid-lineage DC also reacted with those MAbs that recognized DC-SIGN. In the course of creating DC through differentiation of CD14+
monocytes, we observed that treatment with either IL-4 or IL-13 was sufficient to induce DC-SIGN expression on these cells. IL-4 and IL-13 are the cytokines typically expressed by Th2 CD4+
T cells. Thus, it is conceivable that the interaction of Th2 CD4+
T cells with monocytes or macrophages might create a microenvironment in which HIV-1 is more readily transmitted from monocytes/macrophages to either Th2 CD4+
T cells or other monocytes/macrophages. Of interest, IL-4 and IL-13 signals activate the STAT-6 transcription factor, whereas cytokines that do not induce DC-SIGN do not activate STAT-6 (23
) Thus, it will be interesting to examine whether STAT-6 or transcription factors induced by STAT-6, such as GATA3 (28
), participate in the induction of DC-SIGN transcription.
Functional characterization of the panel of DC-SIGN family MAbs revealed that they may be useful in studying DC-SIGN-ligand interactions. The DC-SIGN-specific or -cross-reactive MAbs inhibited the adhesion between DC-SIGN and ICAM-3. As expected, the L-SIGN-specific MAb 604(L) was not able to block adhesion of DC to ICAM-3. However, 604(L) slightly decreased ICAM-3 adhesion to THP-1/DC-SIGN cells, suggesting cross-reactivity that was not detected by FACS staining. Notably, five times more antibody was used in the ICAM-3 blocking assays than in the FACS staining analysis. Background binding of ICAM-3 to THP-1 cells was uniformly less than 5% in repeat experiments.
Except for MAb 507(D), HIV-1 transmission via THP-1/DC-SIGN cells to the Hut/CCR5 target cells was completely inhibited to background levels with the MAbs recognizing DC-SIGN. Although MAb 507(D) reacted with DC-SIGN strongly in the FACS staining assays, its relatively weaker neutralization of virus transmission suggests different sites of DC-SIGN for MAb 507(D) binding and virus interactions. As expected, L-SIGN-specific MAb 604(L) could not block HIV-1 capture and transmission by THP-1/DC-SIGN cells.
The mechanism by which DC-SIGN MAbs block HIV Env or ICAM-3 interactions remains to be determined. Given that both ligands are blocked, the DC-SIGN MAbs might affect the conformation or exposure of the CRD. Further investigation of DC-SIGN mutant molecules will allow mapping of epitopes that also influence HIV-1 Env or ICAM-3 binding.
Neutralizing DC-SIGN or L-SIGN MAbs may also be used to elucidate the role of DC-SIGN and L-SIGN in vivo in animal models or in vitro in organ culture systems. Unexpectedly, we observed that virus transmission by DC could be only partially impaired with the cocktails of DC-SIGN-specific or -cross-reactive MAbs. In agreement with this result, virus transmission by DC was reduced only to 59 to 78% with each of six individual DC-SIGN-reactive MAbs (10 μg/ml) relative to mouse IgG control antibody (data not shown). Consistent with earlier data, the L-SIGN-specific MAb 604(L) was not able to block the virus transmission mediated by DC. We also found that incubation with 20 μg of mannan per ml could only block HIV-1 transmission by DC to 39%, despite the fact that virus transmission via THP-1/DC-SIGN cells could be completely inhibited by mannan incubation (data not shown). These data suggest that although DC-SIGN is important for DC-mediated HIV-1 transmission to target cells, there may be additional factors that the virus uses in this process. In fact, Rhesus
macaque DC efficiently transmit primate lentiviruses in the absence of DC-SIGN expression (33
Besides the role of DC-SIGN in HIV-1 transmission, direct contact of HIV-1-exposed DC with T cells through adhesion molecules or other unknown factors and mechanisms may also contribute to the general property of virus transmission by DC. In the initial description of DC-SIGN function in HIV-1 transmission, we noted that CD4+
T cells were consistently better targets than 293T-CD4-CCR5 cells (14
). Some of the known interactions that occur between DC and T cells are contacts between DC-SIGN and ICAM-3, LFA-1 and ICAM-1, LFA-3 and CD2, and the antigen-presenting complexes with their cognate T-cell receptors. Because the role of adhesion interactions mediated by DC-SIGN had not been formally examined within the context of HIV-1 transmission, we investigated whether DC-SIGN/ICAM-3 interactions create a microenvironment that favors transmission to T cells. In addition, others have demonstrated that direct contact of HIV-1-infected DC with T cells is required for efficient virus transmission and subsequent virus production (32
In examining the possible role of ICAM-3, we found that exposure of DC-SIGN to soluble ICAM-3 was not sufficient to fully neutralize HIV-1 transmission mediated by THP-1/DC-SIGN cells, even if the concentration of ICAM-3 was increased to 30 μg/ml. Similar results were obtained when DC were used as donor cells in the HIV-1 capture and transmission assay (data not shown). Nonetheless, a more careful examination of DC-SIGN interactions with HIV-1 Env should provide the basis for the development of potential antiviral drugs that can be used prophylactically or therapeutically in disrupting DC-SIGN capture and transmission of HIV-1. Indeed, recent crystallographic studies of the DC-SIGN and DC-SIGNR CRD complexed with high-mannose oligosaccharide highlight contact points within the CRD which might serve as targets in developing novel antiviral agents against HIV-1 (10
Initial experiments had indicated that HIV-1 capture and transmission could not be inhibited by using soluble ICAM-3 or ICAM-3 MAbs in the coculture stage of DC or THP-1/DC-SIGN donor cells and target T cells (data not shown). However, it was possible that the soluble ICAM-3 or the antibodies did not fully block DC-SIGN/ICAM-3 interactions between the donor and target cells. We thus chose to employ a system in which DC-SIGN/ICAM-3 interactions could be strictly controlled. Because ICAM-3-negative GHOST/R5 cells are readily infected by HIV-1, it was tested whether ICAM-3 expression might increase their susceptibility to DC-SIGN-mediated HIV-1 transmission. Our data indicate that the presence or absence of ICAM-3 on target GHOST/R5 cells did not affect their susceptibility to direct infection by HIV-1 or transmission mediated by DC-SIGN.
Similarly, attempts to block DC-SIGN/ICAM-3 interactions during the coculture of THP-1/DC-SIGN donor cells and GHOST/R5/ICAM-3 target cells did not impair HIV-1 transmission. In fact, we found that the transmission of HIV-1 mediated by DC-SIGN to the ICAM-3-positive GHOST/R5/ICAM-3 cells was slightly lower than that to ICAM-3-negative GHOST/R5 cells (Fig. ). In summary, no direct role for ICAM-3 in the DC-SIGN-mediated transmission of HIV-1 is apparent. The fact that different nonhematopoietic target cells can be used in HIV-1 transmission from DC-SIGN-expressing donor cells suggests that the effect of DC-SIGN is virus specific or restricted to the HIV-1 receptor molecules.
A recent study (2
) mapped the determinants recognized by a panel of 16 MAbs raised against recombinant DC-SIGN to the repeat region, the lectin-binding domain, and the extreme C terminus of DC-SIGN. Although all of the MAbs bound to DC-SIGN on the surface of cells, none of the MAbs potently inhibited binding of soluble ICAM-3 to DC-SIGN, nor did any of the MAbs effectively block virus transmission. The DC-SIGN MAbs used in this study were developed against cell surface-expressed DC-SIGN and were competent in blocking ICAM-3 binding and neutralizing HIV-1 capture and transmission, suggesting that the HIV-1 Env and ICAM-3 binding sites are conformational. Three of these MAbs that recognized Rhesus
macaque DC-SIGN also blocked HIV-1 and SIV transmission mediated by Rhesus
macaque DC-SIGN (33
). Thus, these MAbs may be useful to test the role of DC-SIGN in primate lentivirus transmission in in vivo models.
In summary, we have characterized a panel of seven MAbs reactive to DC-SIGN family molecules, some of which recognize primary human DC. Examination of cytokine-treated monocytes indicates that these MAbs will be useful in characterizing DC development. These studies also suggest that DC-SIGN may contribute to viral pathogenesis in other cell types, given the ability to induce DC-SIGN expression in human monocytes. The DC-SIGN-reactive MAbs compete with ICAM-3 for specific adhesion to DC-SIGN and inhibit R5- and X4-tropic HIV-1 transmission to T cells, suggesting that the DC-SIGN-MAb interaction sites may be useful as targets in antiviral therapy. Consistent with our studies using Rhesus macaque DC, less-efficient MAb neutralization of human DC-mediated HIV-1 transmission indicated that these cells are also capable of transmitting HIV-1 to target cells in a DC-SIGN-independent manner. Finally, although interactions between DC-SIGN and ICAM-3 may be important in initiating DC and CD4+ T-cell contact, they appear to be neither essential nor contributory in the cell-to-cell transmission of HIV-1 via DC-SIGN.