Specific Interaction of Feline Immunodeficiency Virus Surface Glycoprotein with Human DC-SIGN

DC-SIGN, a specific C-type lectin expressed on dendritic cells,binds and transmits multiple strains of primate immunodeficiencyviruses to susceptible cells. Here, we report that human DC-SIGNalso captures feline immunodeficiency virus via high-affinity(1 nM), Ca²⁺-dependent, D-mannose-inhibited binding to the majorenvelope glycoprotein, gp95.

INTRODUCTION

Top

Introduction

DC-SIGN, a cell surface C-type lectin expressed on dendriticcells (DCs), is thought to play key roles in the interactionof DCs with T cells as well as in human immunodeficiency virus(HIV) pathogenesis (reviewed in reference 4). By binding tointracellular adhesion molecule 3 (ICAM-3), DC-SIGN facilitatesthe initial interaction between DCs and resting T cells, whichleads to antigen recognition and initiation of the immune response(6). DC-SIGN is also a unique type of attachment factor forHIV type 1 (HIV-1) in that it binds the envelope glycoproteingp120 of HIV-1 and promotes infection in trans of target CD4⁺T cells (5). It is believed that HIV may exploit DC-SIGN forits transport from mucosal sites of infection to permissiveT cells in secondary lymphoid organs. DC-SIGN also binds HIV-2and simian immunodeficiency virus surface glycoproteins (10)and can enhance infection in cis, especially when CD4 and coreceptorlevels are limiting (7). We were thus interested to study potentialDC-SIGN interaction with feline immunodeficiency virus (FIV),the causative agent of feline AIDS in the domestic cat (8).Feline DC-SIGN has not been cloned, and reagents and protocolsfor the preparation of feline DCs are in development. However,based on a previous demonstration of FIV binding to human CXCR4(13), we initiated studies utilizing human DC-SIGN.

DC-SIGN interactions with FIV.

Top

Dc-sign interactions with fiv.

CHO pgsA745 cells were transduced with a murine stem cell retrovirusvector (MIGR1-DC-SIGN-GFP) expressing human DC-SIGN in tandemwith green fluorescent protein (GFP) via an internal ribosomeentry site linker in which DC-SIGN is tagged with the AU-1 sequence(7) at the extracellular C terminus. The pgsA745 cell line isa CHO mutant cell line defective in the biosynthesis of glycosaminoglycansand thus does not express cell surface heparan sulfate proteoglycans(3), which also act as binding receptors for certain HIV isolates(11) and the 34TF10 (12) molecular clone of FIV (2). GFP-positivecells were sorted by cytofluorimetry, and the expression ofDC-SIGN on GFP-positive cells was confirmed with an anti-AU-1antibody (Fig. 1A). We first analyzed the binding of FIV surfaceglycoprotein gp95 in comparison with that of HIV gp120 and ICAM-3by using the immunoadhesins gp95-Fc (FIV), gp120-Fc (HIV-1),and ICAM-3-Fc. The gp95-Fc immunoadhesin has been previouslyreported (2), and ICAM-3-Fc was purchased from R&D Systems.For gp120-Fc, gp120 from the JR-CSF clone of HIV-1 was subclonedin frame with the Fc domain of immunoglobulin G1 (IgG1) andproduced in CHO cells by using the glutamate synthetase amplificationsystem as previously reported for FIV gp95 (2). DC-SIGN interactionwith gp95-Fc, gp120-Fc, and ICAM-3-Fc was assessed by fluorescence-activatedcell sorter (FACS) analysis with DC-SIGN-positive and -negativepgsA745 cells. Cells (10⁵) were incubated for 1 h at 4°Cwith 1 µg of immunoadhesin per ml, washed, and then incubatedfor another 45 min with an anti-Fc phycoerythrin-conjugatedantibody (Cappel, Durham, N.C.). As shown in Fig. 1B, bindingof FIV gp95-Fc to DC-SIGN-negative cells was not observed. However,we observed a strong binding of gp95-Fc to DC-SIGN-positivecells, but only in the presence of CaCl₂ (Fig. 1C). The bindingof gp95-Fc was specific in that no binding was observed withFc alone (data not shown). Furthermore, the presence of 10 mMEGTA abolished the binding of gp95-Fc, as did the preincubationof the cells with mannan at 100 µg/ml (Fig. 1C). Bindingof both gp120-Fc and ICAM-3-Fc also required calcium and wasinhibited by EGTA and mannan (Fig. 1C). Inhibition by mannanwas stereospecific in that D-mannose competed for gp95-Fc andgp120-Fc binding to human DC-SIGN much more efficiently thandid L-mannose (Fig. 2).

View larger version (34K):
[in this window]
[in a new window]
FIG. 1. Binding of FIV, HIV-1, and ICAM-3 immunoadhesins to CHO cells as a function of DC-SIGN expression. (A) FACS analysis of pgsA745 cells transduced with the MIGR1 vector either expressing GFP (left panel) or coexpressing GFP and human DC-SIGN (right panel). (B) Control binding of immunoadhesins to pgs745 cells transduced with GFP alone, gp95_PPR-Fc (envelope from FIV-PPR), gp95₃₄-Fc (envelope from FIV-34TF10), gp120_JRCSF-Fc (envelope from HIV-1 JRCSF strain), and ICAM-3-Fc. (C) Calcium-dependent binding of immunoadhesins to pgs745 cells transduced with human DC-SIGN. The top four panels show a lack of binding in the absence of CaCl₂, the second four panels show binding in the presence of 1 mM CaCl₂, the third four panels show a lack of binding in the presence of 1 mM CaCl₂ and 10 mM EGTA, and the bottom four panels show a lack of binding in the presence of 1 mM CaCl₂ and 100 µg of mannans per ml.

View larger version (21K):
[in this window]
[in a new window]
FIG. 2. Stereospecificity of mannose competition for gp95-Fc and gp120-Fc binding to soluble DC-SIGN. Data were obtained from an assay based on an enzyme-linked immunosorbent assay in which a high-binding-capacity 96-well plate was coated with 200 nM recombinant His₆-tagged extracellular domain of DC-SIGN overnight at 4°C. Either gp95-Fc (1 nM) or gp120-Fc (2 nM) was added with an increasing amount of D-mannose or L-mannose. Bound SU-Fc adhesin was detected with an anti-human Fc IgG1-specific antibody conjugated with horseradish peroxidase.

Binding affinity measurements.

Top

Binding affinity measurements.

Comparisons were made of the relative binding affinities ofFIV gp95-Fc and HIV-1 gp120-Fc (Fig. 3). Interestingly, FIVgp95-Fc had a higher binding affinity for DC-SIGN than did HIV-1gp120-Fc, with K_d values of 1.03 and 2.99 nM, respectively (Fig.3). In toto, these binding data suggest that the phenomenonof DC-SIGN binding to the HIV envelope protein is recapitulatedin the feline lentivirus model.

View larger version (11K):
[in this window]
[in a new window]
FIG. 3. Relative binding affinities of FIV gp95-Fc and HIV gp120-Fc for DC-SIGN. Inhibition curves were derived from an assay based on an enzyme-linked immunosorbent assay as described in the legend to Fig. 2, except that 125 nM His-tagged extracellular domain was used. Increasing concentrations of ligand were added, and the amount of bound ligand was detected with anti-human Fc IgG1-specific antibody conjugated with horseradish peroxidase. Experimental values were plotted, and the K_d was computed with GraphPad Prism software.

DC-SIGN facilitates FIV infection in trans.

Top

Dc-sign facilitates fiv...

To determine whether DC-SIGN could present FIV to target cellsas observed for HIV-1, DC-SIGN-positive pgsA745 cells were seededin 12-well dishes and incubated with either FIV-PPR (molecularlycloned primary isolate) (9) or FIV-34orf2rep (molecularly clonedlaboratory-adapted isolate) (12) at a multiplicity of infection(MOI) of 0.01 for 3 h at room temperature. After extensive washing,cells were cocultured overnight with 2.5 x 10⁴ 104-C1 cells.On the following day, the T cells were transferred to freshwells, and virus production was monitored at 7 days post-cocultureby a reverse transcriptase activity assay as previously described(2). As shown in Fig. 4, DC-SIGN-positive pgsA745 cells effectivelytransmit FIV to the feline 104-C1 T-cell population. FIV capturewas DC-SIGN dependent, since preincubation of DC-SIGN-positivecells with mannan at 50 µg/ml blocked binding of FIV andsubsequent transfer to target cells. Additionally, DC-SIGN-negativeparental pgs745 cells failed to capture and facilitate transmissionof FIV (Fig. 4).

View larger version (28K):
[in this window]
[in a new window]
FIG. 4. DC-SIGN transmits FIV to susceptible target cells. Control pgsA745 cells (DC-SIGN^-) and pgsA745 cells transduced with human DC-SIGN (DC-SIGN⁺) were exposed to tissue culture supernatants containing either FIV-PPR or FIV-34orf2rep. The cells were thoroughly washed and then cocultured with susceptible feline 104-C1 T cells. The 104-C1 cells were recovered and cultured separately, and virus production was detected at 7 days postinfection by a reverse transcriptase (RT) activity assay. No infectivity was noted in cocultivations with DC-SIGN^- cells, but both strains of FIV were efficiently transmitted to 104-C1 cells in cocultures with DC-SIGN⁺ cells. DC-SIGN⁺ cells that were pretreated with 50 µg of mannan per ml failed to transmit FIV infection, consistent with dependence on DC-SIGN for virus transmission.

Influence of DC-SIGN on FIV infection in cis.

Top

Influence of dc-sign on...

Identification of GFP- and DC-SIGN-positive or GFP-positive104-C1 cells was established by retroviral transduction, andcells were then sorted by FACS analysis, with DC-SIGN expressionconfirmed by staining with an anti-AU-1 antibody (Fig. 5A).DC-SIGN-negative and -positive 104-C1 cells were infected atdifferent MOIs with FIV-PPR and FIV-34orf2rep strains, and virusproduction was monitored at 7 days postinfection by a reversetranscriptase assay. At MOIs higher than 0.1, substantial levelsof cytopathicity for both DC-SIGN-negative and -positive cellswere observed, and this effect was more pronounced in the DC-SIGN-positivecell population; i.e., at a high MOI, the enhancing effect ofDC-SIGN expression in cis was masked (data not shown). However,at MOIs of 0.1 to 0.001, an enhancing effect of DC-SIGN expressionwas evident, and viral infectivity was two- to threefold morepronounced, on average, in the DC-SIGN-positive cell population(Fig. 5B). The enhancing effect of DC-SIGN expression in cisis more evident when CD4 and coreceptor levels are limiting(7). Here, the feline 104-C1 T cells that we used expressedrelatively high levels of FIV primary receptor and CXCR4 (2),which might explain the moderate enhancing effect in cis ofDC-SIGN that we observed.

View larger version (26K):
[in this window]
[in a new window]
FIG. 5. Influence of DC-SIGN expressed in cis on the infection of feline T cells by FIV. (A) FACS analysis of 104-C1 cells transduced with the MIGR1 vector either expressing GFP (left panel) or coexpressing GFP and human DC-SIGN (right panel). (B) Cells were infected with the two virus strains, FIV-PPR and FIV-34orf2rep, at a range of MOIs. Virus expression was monitored by measuring reverse transcriptase (RT) levels in the culture supernatant at 7 days postinfection. A two- to threefold enhancement of FIV infection was noted as a function of DC-SIGN expression.

The cloning and characterization of feline DC-SIGN and consistentsuccessful culture of feline DCs have yet to be achieved. Thus,in vivo relevance of FIV-DC-SIGN interactions must still bedemonstrated. However, the findings reported here suggest theexistence of an additional parallel between FIV and HIV-1 infectionwhich may lead to the use of the cat model as a tool for thestudy of the significance of DC-SIGN association to virus spreadin a natural infection. The findings from such investigationsmay be relevant not only to the study of retrovirus infectionsbut also to the treatment of unrelated agents such as Ebolavirus, where it has been shown that the surface glycoproteinof that virus also interacts with DC-SIGN (1). As with DC-SIGN-HIV-1gp120 interaction, the binding of FIV gp95 to DC-SIGN is dependenton Ca²⁺, is completely inhibited by EGTA, and is blocked bymannan (Fig. 1). The exquisite specificity of DC-SIGN for theparticular glycan structures found on the FIV envelope is underscoredby the stereospecificity shown in competition for binding byD-mannose, both for FIV gp95 and for HIV gp120 (Fig. 2). Additionally,the binding affinity of FIV gp95 for human DC-SIGN is high,with a K_d of approximately 1 nM (Fig. 3), The findings alsodemonstrate that human DC-SIGN can, as with HIV, act to presentFIV for infection of permissive cells (Fig. 4) and, further,that the expression of human DC-SIGN in cis results in the enhancedinfection of permissive feline cells (Fig. 5). Whether DC-SIGNalone can serve to facilitate FIV infection remains to be determined.However, all other observations suggest that ultimate uptakeand productive infection by FIV will require entry via CXCR4binding (2). Rather, the findings suggest that DC-SIGN may serveas an additional binding receptor to increase the effectiveconcentration of viruses at the cell surface and thus increaseviral infection via the entry receptor, CXCR4.

ACKNOWLEDGMENTS

We thank Stacie Ngo for excellent technical assistance.

This work was supported in part by grants R01 AI25825 (J.H.E.)and R01 AI52021 (B.L.) from the National Institutes of Health.B.L. is also a recipient of the Culpepper Biomedical Scholaraward and a Burroughs Wellcome Fund Career Development award.

FOOTNOTES

* Corresponding author. Mailing address: The Scripps Research Institute, Department of Molecular Biology, La Jolla, CA 92037. Phone: (858) 784-9707. Fax: (858) 784-2750. E-mail: jelder{at}scripps.edu.

REFERENCES

Top