Previous Article | Next Article ![]()
Journal of Virology, July 2002, p. 7203-7208, Vol. 76, No. 14
0022-538X/02/$04.00+0 DOI: 10.1128/JVI.76.14.7203-7208.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Department of Microbiology-Immunology, The Feinberg School of Medicine, Northwestern University, Chicago, Illinois 60611
Received 7 February 2002/ Accepted 24 April 2002
|
|
|---|
|
|
|---|
Human and animal representatives of the alphaherpesviruses have a broad host range and similar requirements for entry into cells (30). The first step is usually binding of virion envelope glycoprotein C (gC) and/or gB to cell surface glycosaminoglycans, preferentially heparan sulfate. The subsequent interaction of virion gD with one of its receptors triggers the penetration of virus, which occurs through fusion of the viral envelope with the cell membrane and requires four virion glycoproteins, gB, gD, gH, and gL.
Three classes of cell surface molecules can serve as gD receptors for the entry of HSV-1 or HSV-2 (30). These include a member of the tumor necrosis factor receptor family (22), two members of the immunoglobulin (Ig) superfamily (6, 12, 18, 33) and specific sites in heparan sulfate generated by the action of certain isoforms of 3-O-sulfotransferase (28). The Ig superfamily members were recently named nectin-1 and nectin-2 (26, 32) and were previously called Prr1 and Prr2 (9, 19), respectively, or HveC and HveB (12, 33), respectively. Human nectin-1 and nectin-2 can serve as entry receptors for PRV as well as HSV. Many human cell types, such as epithelial cells, express multiple HSV entry receptors, including herpesvirus entry mediator (HVEM), nectin-1, and nectin-2 (12, 33).
Nectin-1 and nectin-2 are related to nectin-3 and nectin-4 as well as to the poliovirus receptor (9, 19, 24, 26). All five proteins, each encoded by a different gene, have three extracellular Ig-like domains, an N-terminal V-like domain, and two C2-like domains and are homologous with respect to amino acid sequence. The nectins are Ca2+-independent cell adhesion molecules that can engage in homophilic or heterophilic interactions. In some instances, HSV-1 gD can block these interactions (17, 25). Differentially spliced transcripts from the nectin genes can be translated to yield multiple isoforms with identical ectodomains. Most of the membrane-bound isoforms can bind through their cytoplasmic tails to the PDZ domain in l-afadin. The nectins and l-afadin colocalize with E-cadherin and catenins in adherens junctions of epithelial cells (31, 32). Depletion of extracellular Ca2+ can cause rapid disruption of adherens junctions with redistribution of nectin and turnover of E-cadherin (1, 15).
If nectin-1 or nectin-2 is localized to adherens junctions in epithelial cells, are these gD receptors accessible for binding to virus? In this study, we used Madin-Darby canine kidney (MDCK) cells to examine the effects of junctional disruption on cell surface localization of nectin-1, binding to the cells of HSV-1 gD and PRV gD, and susceptibility of the cells to viral entry. MDCK cells expressing transfected human nectin-1 or an endogenous entry receptor were used because, in either case, the only receptors available for HSV and PRV entry were human nectin-1 and/or a molecule antigenically related to human nectin-1, probably dog nectin-1. Thus, use of the MDCK cells facilitated investigation of the relationship between cell surface distribution of a single type of entry receptor and susceptibility of the cells to HSV and PRV entry.
|
|
|---|
(32), designated here MDCK-nectin-1 cells, were provided by Y. Takai (Osaka University Medical School, Osaka, Japan) and grown in DME with 10% FCS and G418 at 300 µg/ml (Gibco-BRL). The epitope tag in Flag-nectin-1
is on the natural N terminus resulting from signal peptidase cleavage. Control experiments done, as described previously (12), to compare the entry activities of nectin-1
and Flag-nectin-1
revealed that the epitope tag did not prevent HSV-1 or PRV entry, although entry was not as efficient as that observed with the untagged form of receptor (data not shown). The ß-galactosidase reporter viruses have been described previously (2, 12, 33). HSV-1(KOS)tk12 was propagated and titers were determined on Vero cells, and PRV(Kaplan)gH- (a gift from T. Mettenleiter, Federal Research Center for Virus Diseases of Animals, Insel Riems, Germany) was propagated and titers were determined on gH-expressing Vero-SW78 cells.
Ca2+ switch protocol.
Cells were grown in DME containing 2 mM Ca2+ with 10% FCS (NC medium). To deplete the cells of Ca2+ as described previously (4), the cells were washed with phosphate-buffered saline and transferred for 1 or 2 h to Ca2+-free DME with 1 mM EDTA and 10% FCS that had been depleted of Ca2+ by passage over Chelex-100 (Bio-Rad) (LC medium,
2 µM Ca2+).
gD:Fc binding assay. The production and quantitation of herpesvirus gD:Fc hybrid proteins and assay for binding of the gD:Fcs to cells were described previously (10, 11). HSV-1 gD:Fc has the first 345 amino acids (first 320 amino acids after signal peptide cleavage) of HSV-1(KOS) gD fused through a linker of 6 amino acids (YRARIH) to 231 amino acids at the C terminus of rabbit IgG heavy chain. The PRV gD:Fc is similar except that the first 408 amino acids (first 391 amino acids after signal peptide cleavage) of PRV(Kaplan) gD is fused through a 3-amino-acid linker (RIH) to the rabbit sequence. MDCK cells grown in 96-well plates were incubated with serial dilutions of HSV-1 gD:Fc or PRV gD:Fc for 1 h at 37°C. Cells were washed, fixed with 2% formaldehyde and 0.2% glutaraldehyde, and sequentially incubated with biotinylated anti-rabbit IgG (Sigma), Amdex streptavidin-conjugated horseradish peroxidase (HRP; Amersham), and HRP substrate (BioFx Lab). Binding was monitored at 370 nm in a Spectra Max 250 enzyme-linked immunosorbent assay reader. Alternatively, cells grown on coverslips were incubated with gD:Fc, and binding was visualized with Alexa 488-conjugated goat anti-rabbit IgG (Molecular Probes). To assess the ability of antibodies to block gD:Fc binding, serial dilutions of monoclonal antibodies recognizing nectin-1 (R1.302, Immunotech) (6) or nectin-2 (R2.477, Immunotech) (18) were added to cells 1 h prior to the addition of a constant amount of gD:Fc.
Viral entry assay. The assay for viral entry was described previously (22). Serial dilutions of ß-galactosidase-expressing viruses were added to cells grown in 96-well plates and incubated for 6 h. Cells were washed, permeabilized, and incubated with ß-galactosidase substrate, O-nitrophenyl-ß-D-galactopyranoside (ONPG; Sigma). The reaction was monitored at 410 nm to quantitate viral entry. Alternatively, infected cells were permeabilized and incubated with the ß-galactosidase substrate, X-Gal (5-bromo-4-chloro-3-indolyl-ß-D-galactopyranoside; Gibco-BRL), which yields an insoluble blue reaction product. The stained cells were photographed with an Olympus microscope (model CK2), equipped with a 35-mm camera, controlled by the Olympus automatic exposure photographic system (PM-10AK3).
Immunofluorescence and microscopy. Indirect immunofluorescence was carried out as described previously (32). The primary antibodies used were mouse monoclonal anti-nectin-1 antibodies, R1.302 (6) and CK6 and CK8 (16), mouse anti-Flag (M1, Sigma), and rat anti-E-cadherin (ECCD2; Zymed) antibodies. The secondary antibodies were Alexa 488-conjugated goat anti-mouse IgG and Alexa 568-conjugated goat anti-rat IgG. Immunofluorescence observations were made with a Zeiss LSM510 confocal microscope equipped with a 100x, 1.4 NA oil immersion objective. Orthogonal sections were made in a z stack of 30 images (1-µm interval) according to the LSM510 operating manual.
|
|
|---|
(referred to here as MDCK-nectin-1 cells) were grown to confluence in medium containing standard levels of Ca2+ (2 mM; NC medium) and then either maintained in this medium or switched to medium with greatly reduced levels of Ca2+ (
2 µM; LC medium). As shown previously (32), Flag-tagged nectin-1 and E-cadherin colocalized at cell junctions in MDCK-nectin-1 cells maintained in NC medium (Fig. 1A to C). Following depletion of Ca2+ from the culture medium, dissociation of cell-cell junctions was in initial stages by 1 h (Fig. 1D to F) and was nearly complete by 2 h (Fig. 1G to I). E-cadherin was no longer detectable after 2 h (Fig. 1G and I), as previously shown (15). Under the same conditions, nectin-1 remained detectable on the cell surfaces (Fig. 1H) as previously described (1). Orthogonal sectional views demonstrated that nectin-1 was localized only at cell-cell junctions in NC medium and redistributed over large areas of the cell surface after Ca2+depletion (Fig. 1J and K).
![]() View larger version (120K): [in a new window] |
FIG. 1. Localization of Flag-tagged nectin-1 and E-cadherin in MDCK-nectin-1 cells before and after a switch to medium depleted of Ca2+. Cells were maintained in NC medium (A to C) or switched to LC medium for 1 h (D to F) or 2 h (G to I) and then double stained with mouse anti-Flag (B, E, H, J, and K) and rat anti-E-cadherin (C, F, and I) antibodies. (A, D, and G) Phase-contrast images. (J and K) Orthogonal sections of anti-Flag-stained MDCK-nectin-1 cells maintained in NC medium (J) or in LC medium for 2 h (K). Bar, 10 µm.
|
Enhanced binding of gD to epithelial cells depleted of Ca2+. To assess the presence and distribution of gD receptors, MDCK cells and MDCK-nectin-1 cells were maintained in NC medium or switched to LC medium for 2 h and then were incubated with soluble forms of HSV-1 gD and PRV gD (gD:Fc hybrid molecules), followed by fixation and addition of labeled anti-Fc antibodies. Figure 2 shows that both the HSV-1 and PRV gD:Fcs bound to the MDCK-nectin-1 cells. Also, PRV, but not HSV-1, gD:Fc bound to the untransfected MDCK cells. PRV gD is known to bind to human nectin-1 with higher affinity than HSV-1 gD (8). One explanation for the results in Fig. 2A, C, E, and G is that PRV gD also binds with higher affinity to the endogenous dog receptor than does HSV-1 gD and that the latter interaction is too weak to be detected by the use of soluble gD:Fc. MDCK cells must express an HSV-1 receptor, because MDCK cells can be infected by HSV-1.
![]() View larger version (71K): [in a new window] |
FIG. 2. Binding of HSV-1 gD:Fc and PRV-gD:Fc to MDCK cells (A, C, E, and G) and MDCK-nectin-1 cells (B, D, F, and H). The cells were maintained in NC medium (A to D) or switched to LC medium for 2 h (E to H) and then incubated with either HSV-1 gD:Fc (A, B, E, F, I, and J) or PRV gD:Fc (C, D, G, and H), followed by fixation and incubation with Alexa 488-conjugated goat anti-rabbit IgG. (I and J) Orthogonal sections of HSV-1 gD:Fc-stained MDCK-nectin-1 cells maintained in NC medium (I) or LC medium for 2 h (J). Bar, 10 µm.
|
To investigate whether depletion of Ca2+ and the redistribution of gD receptors were accompanied by altered amounts of gD:Fc bound to cells, a quantitative binding assay was performed on live MDCK cells or MDCK-nectin-1 cells maintained in NC or LC medium. There was no stable binding of HSV-1 gD:Fc to MDCK cells, regardless of the conditions (Fig. 3), consistent with the immunofluorescence observations in Fig. 2. Greater amounts of HSV-1 gD:Fc bound to MDCK-nectin-1 cells incubated for 2 h in LC medium than to cells maintained in NC medium. There was also enhanced binding of PRV gD:Fc to both MDCK cells and MDCK-nectin-1 cells in response to Ca2+ depletion (Fig. 3).
![]() View larger version (29K): [in a new window] |
FIG. 3. Enhanced binding of HSV-1 gD:Fc and PRV-gD:Fc to MDCK cells and MDCK-nectin-1 cells after a switch to LC medium. Cells grown in 96-well plates were maintained in NC medium or switched to LC medium for 2 h and incubated with either HSV-1 gD:Fc or PRV gD:Fc for 1 h. Cells were then washed, fixed, and sequentially incubated with biotinylated goat anti-rabbit IgG, streptavidin-conjugated HRP, and HRP substrate. Values are optical densities at 370 nm; means of triplicate determinations with standard deviations for one representative experiment of five replicates are shown.
|
![]() View larger version (29K): [in a new window] |
FIG. 4. Anti-nectin-1 antibody blocks the enhanced binding of HSV-1 gD:Fc and PRV gD:Fc to MDCK cells and MDCK-nectin-1 cells. Cells in 96-well plates were maintained in NC medium or switched to LC medium for 2 h and incubated with serial dilutions of monoclonal antibodies against nectin-1 (R1.302.12) or nectin-2 (R2.477.1) for 1 h prior to addition of a single dose of HSV-1 gD:Fc (24 µg/ml) (A) or PRV gD:Fc (4 µg/ml) (B). After 1 h of incubation, cells were washed, fixed, and incubated with biotinylated secondary antibody, streptavidin-conjugated HRP, and HRP substrate. Values are optical densities at 370 nm; means of triplicate determinations with standard deviations for one representative experiment of five replicates are shown.
|
Viral entry assays were performed using cells maintained in NC medium or switched for 2 h to LC medium prior to incubation with serial dilutions of HSV-1(KOS)tk12 or PRV-gH-. These viruses express ß-galactosidase from a lacZ cassette inserted within their genome. At 6 h after addition of virus, cells were lysed to quantitate ß-galactosidase activity using either ONPG or X-Gal as the substrate. Expression of ß-galactosidase signals the entry of virus into cells and expression of viral and reporter genes.
Depletion of Ca2+ from the culture medium significantly enhanced the entry of both HSV-1 and PRV into MDCK cells and MDCK-nectin-1 cells. Quantitative results presented in Fig. 5A show that viral entry was more efficient for the cells depleted of Ca2+ than for cells maintained in NC medium at each of the input doses of virus tested. The levels of ß-galactosidase activity were related to the numbers of cells infected at each virus dose, as shown by the X-Gal-stained cells in Fig. 5B to I. Thus, the disruption of cell junctions and release of junctional nectin-1 to larger areas of the cell surface correlated with enhanced susceptibility of the cells to viral entry. We estimate that this enhancement was at least 10-fold, based on the amounts of virus required to achieve equivalent numbers of infected cells under the two conditions.
![]() ![]() View larger version (176K): [in a new window] |
FIG. 5. Enhanced entry of HSV-1 and PRV after a switch to LC medium. Cells in 96-well plates (2 x 104 to 4 x 104 cells per well) were maintained in NC medium or switched to LC medium for 2 h and incubated with serial dilutions of ß-galactosidase-expressing HSV-1 or PRV. After 6 h, cells were processed for incubation with the ß-galactosidase substrate ONPG (A) or X-Gal (B to I). (A) ß-Galactosidase activity was quantitated as a measure of viral entry. Values are optical densities at 410 nm; means of triplicate determinations with standard deviations for one representative experiment of five replicates are shown. (B to I) X-Gal staining at a single virus dose of 107 PFU per well in MDCK cells (B, C, F, and G) and MDCK-nectin-1 cells (D, E, H, and I), cultured in either NC medium (B, D, F, and H) or LC medium (C, E, G, and I) prior to infection. Bar, 100 µm.
|
Insights into viral entry of epithelial cells. Recently, two cell surface glycoproteins, the coxsackievirus and adenovirus receptor (CAR) and the junction adhesion molecule, were identified as transmembrane components of the tight junction in epithelial cells as well as entry receptors for coxsackie virus and adenovirus and for reovirus, respectively (3, 7). Disruption of tight junctions by Ca2+ depletion enhanced CAR-mediated coxsackievirus and adenovirus infection in polarized T-84 epithelial cells (7). Sequestration of CAR in tight junctions apparently limits viral infection, similar to our findings that the localization of nectin-1 to adherens junctions actually impairs its ability to mediate HSV-1 and PRV entry. Although soluble forms of gD could bind to nectin-1 even when it was localized to cell junctions and presumably engaged in homophilic trans interactions, the junctions probably do not allow access of virions to the nectin-1 so engaged. Thus, if nectin-1 is the principal receptor for viral entry into epithelial cells at the portal of entry into the natural host, then some damage to the epithelium may be required for efficient viral entry. Alternatively, other gD receptors, including forms of nectin-1 that do not bind to l-afadin (nectin-1ß), could mediate entry of virus into the first cells of an intact epithelium to be infected.
This work was supported by National Institutes of Health grant R37 AI36293.
|
|
|---|
(PRR2
or HveB) and nectin2
are low-efficiency mediators for entry of herpes simplex virus mutants carrying the Leu25Pro substitution in glycoprotein D. J. Virol. 74:1267-1274.
serves as a species nonspecific mediator for entry of human and animal
herpesviruses in a pathway independent of a detectable binding to gD. Proc. Nat. Acad. Sci. USA 97:4867-4872.
to human nectin-1
(HveC) in sequence and activity as a gD receptor for alphaherpesvirus entry. J. Virol. 74:11773-11781.
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»