Glycoproteins D of Equine Herpesvirus Type 1 ... - Journal of Virology

Glycoproteins D of Equine Herpesvirus Type 1 (EHV-1) and EHV-4 Determine Cellular Tropism Independently of Integrins Walid Azaba,b and Nikolaus Osterriedera Institut für Virologie, Freie Universität Berlin, Berlin, Germany,a and Department of Virology, Faculty of Veterinary Medicine, Zagazig University, Egyptb

Equine herpesvirus type 1 (EHV-1) and EHV-4 are genetically and antigenically very similar, but their pathogenic potentials are strikingly different. The differences in pathogenicity between both viruses seem to be reflected in cellular host range: EHV-1 can readily be propagated in many cell types of multiple species, while EHV-4 entry and replication appear to be restricted mainly to equine cells. The clear difference in cellular tropism may well be associated with differences in the gene products involved in virus entry and/or spread from cell to cell. Here we show that (i) most of the EHV-1 permissive cell lines became resistant to EHV-1 expressing EHV-4 glycoprotein D (gD4) and the opposite was observed for EHV-4 harboring EHV-1 gD (gD1). (ii) The absence of integrins did not inhibit entry into and replication of EHV-1 in CHO-K1 or peripheral blood mononuclear cells (PBMC). Furthermore, integrin-negative K562 cells did not acquire the ability to bind to gD1 when ␣V␤3 integrin was overexpressed. (iii) PBMC could be infected with similar efficiencies by both EHV-1 and EHV-4 in vitro. (iv) In contrast to results for equine fibroblasts and cells of endothelial or epithelial origin, we were unable to block entry of EHV-1 or EHV-4 into PBMC with antibodies directed against major histocompatibility complex class I (MHC-I), a result that indicates that these viruses utilize a different receptor(s) to infect PBMC. Cumulatively, we provide evidence that efficient EHV-1 and EHV-4 entry is dependent mainly on gD, which can bind to multiple cell surface receptors, and that gD has a defining role with respect to cellular host range of EHV-1 and EHV-4.

E

quine herpesvirus types 1 and 4 (EHV-1 and EHV-4) are members of the Alphaherpesvirinae subfamily, genus Varicellovirus (23, 58). Both viruses are endemic in horse populations throughout the world. EHV-1 and EHV-4 have significant genetic and antigenic similarity and have all of their 76 genes in common. Comparison of the complete DNA sequence of EHV-4 strain NS80567 to that of EHV-1 strain Ab4p showed a high degree of conservation that is reflected by the fact that the amino acid sequences of individual proteins are 55 to 96% identical. It is noteworthy that EHV-1 and EHV-4 gD homologues share an amino acid identity of approximately 77% (66). In horses, both viruses are spread from animal to animal by the respiratory route, with primary replication occurring in respiratory epithelia. Although both viruses cause respiratory disease, only infection with EHV-1 results in epidemic abortion, perinatal mortality, and neurological disorders that differ in severity but often result in complete paralysis (4, 21). The pathogenicity of EHV-1 is ascribed to the capacity of the virus to rapidly reach lymphoid tissues associated with the upper respiratory tract and to infect mononuclear cells that ultimately enter the bloodstream and lead to cell-associated viremia (41, 78). As a result, EHV-1 can spread throughout the body by infected peripheral blood mononuclear cells (PBMC). EHV-1 can reach within a short time frame the vasculature of the pregnant uterus or the central nervous system, where it can attach to, enter, and replicate in endothelial cells (EC) (63, 78). On the other hand, pathogenesis and cellular tropism of EHV-4 have been studied to a much lesser extent and most information has been accumulated from natural cases of infection or EHV-4 challenge experiments to evaluate the efficiency of vaccines. Intriguingly, and in contrast to EHV-1 infection, lytic infection with EHV-4 remains limited to the upper respiratory tract. Leukocyte-associated viremia is extremely rare and clearly is not a consistent feature of EHV-4 infections. Consequently, EHV-4 is only very rarely associated with abortion and neurological disor-

0022-538X/12/$12.00

Journal of Virology

p. 2031–2044

ders (46, 53, 55, 69). The differences in pathogenicity between EHV-1 and EHV-4 seem to be reflected in the host range of both viruses for various cultured cells. EHV-1 can readily be propagated in many cell lines, including primary cells and cell lines derived from horse, bovine, rabbit, hamster, mouse, monkey, pig, and cat (75). In contrast, EHV-4 appears to be restricted mainly to cells derived from horses and replicates only poorly in very few other cell lines, e.g., African green monkey kidney (Vero) cells. The clear difference in cellular tropism may well be associated with differences in the gene products involved in virus entry and/or spread from an infected to a neighboring uninfected cell. Such functions are regularly executed by herpesviral envelope glycoproteins. As is the case with other alphaherpesviruses, EHV-1 can enter cells through direct fusion of its envelope with the plasma membrane at neutral pH, a process that is mediated by glycoprotein B (gB), gC, gD, and presumably the gH/gL complex (22, 49, 52). gD was shown to be the essential receptor-binding protein of many alphaherpesviruses (13, 64), and the gD receptors identified so far include members of the tumor necrosis factor (TNF) receptor family (HveA), the poliovirus receptor family (HveB and HveC, members of the immunoglobulin superfamily), and a modified form of heparan sulfate called 3-O-sulfated heparan sulfate. All of these receptors have been identified to mediate entry of HSV-1 and HSV-2 (30, 39, 47, 62, 72). It was shown that EHV-1 is also able to infect cell lines which express these receptors, as well as other cells that are resistant to infection with HSV-1,

Received 13 October 2011 Accepted 5 December 2011 Published ahead of print 14 December 2011 Address correspondence to Nikolaus Osterrieder, [email protected]. Copyright © 2012, American Society for Microbiology. All Rights Reserved. doi:10.1128/JVI.06555-11

jvi.asm.org

2031

Azab and Osterrieder

including Chinese hamster ovary (CHO-K1) and J1.1-2 cells. The latter represent a subpopulation of thymidine kinase-negative baby hamster kidney (BHK) cells (27). Recently, two independent studies have identified equine major histocompatibility complex class I (MHC-I) molecules as a gD receptor for EHV-1 entry into equine cells, including EC of the central nervous system (CNS) (40, 61). Yet, other nonequine cells, e.g., CHO-K1, can still be infected with EHV-1 independently of equine MHC-I (61), and experimental data suggest that EHV-1 utilizes a unique entry receptor(s) that differs from that used by other alphaherpesviruses. It also remains to be elucidated which glycoprotein(s) are used for the entry process in various cell types. Furthermore, EHV-1 can use different cellular entry pathways to infect important target cell populations that include direct fusion at the plasma membrane as outlined above and a nonclassical, endocytic/phagocytic pathway (27, 34, 68). Contrary to the EHV-1 entry receptors and receptor-binding proteins, no data have been accumulated regarding the receptors or glycoproteins that may trigger entry of EHV-4. Taken together, it was suggested that the broad tissue tropism of EHV-1, but not EHV-4, may be due to its ability to use multiple cellular receptors and different pathways to initiate virus entry, with gD suggested as the main player in all entry processes. Integrins are cell surface proteins composed of ␣ and ␤ transmembrane subunits. The ␣V integrin subunit can pair with ␤1, ␤3, ␤5, ␤6, and ␤8 subunits (36). Integrins trigger endocytic pathways and mediate cell-cell and cell-matrix adhesion (17). Several viruses are now known to utilize integrins for their entry into cells, and examples include adenoviruses (77), Epstein-Barr virus (EBV) (16), human cytomegalovirus (HCMV) (71), Kaposi’s sarcoma-associated herpesvirus (KSHV) (2), rotaviruses (32), and echoviruses (11). The RGD motif is the minimal peptide region of many proteins known to interact with cell surface integrins, such as ␣V␤3, ␣V␤5, and ␣3␤1 (65), and RGD motifs are essential for integrin receptor binding of many viruses, such as foot-andmouth disease virus (26) and coxsackievirus (57). For EHV-1, it was postulated that the RSD integrin motif, present in EHV-1 gD but not EHV-4 gD, can bind to ␣V␤3 and ␣V␤5 integrins found on PBMC and CHO-K1 cells, respectively (68). We hypothesized that EHV-1 is able to bind to different cell receptors that are unavailable to EHV-4. To address our hypothesis, the gD genes were swapped between both viruses. The newly generated viruses included an intertypic EHV-1 mutant having EHV-4 gD (gD4) in the place of EHV-1 gD (gD1) and a corresponding EHV-4 carrying gD1. These mutants, together with appropriate revertant viruses in which the original sequences were restored, were used to examine the viruses’ ability to infect different cell lines and as such to define the host range of each virus in vitro. Furthermore, a possible role of integrins to mediate entry of EHV-1 and EHV-4 into PBMC and other cell types through interaction with the RSD motif found in gD was further evaluated. MATERIALS AND METHODS Viruses. EHV-1 strain L11⌬gp2 (59) was reconstituted after transfection of 2 ␮g of bacterial artificial chromosome (BAC) DNA into rabbit kidney (RK13) cells, using Lipofectamine 2000 (Invitrogen). Recombinant WA79 was derived from an EHV-4 infectious BAC clone that was generated by the insertion of a loxP-flanked BAC vector into the intergenic region between genes 58 and 59 (7).

2032

jvi.asm.org

Cells. Fetal horse kidney (FHK), kindly provided by V. Svansson, University of Iceland, human embryonic kidney (cell line 293), RK13, HeLa, Vero, feline kidney (CrFK), and Madin-Darby bovine kidney (MDBK) cells were propagated in Dulbecco’s modified Eagle’s medium (DMEM) (Biochrom) supplemented with 10% fetal bovine serum (FBS) (Biochrom), 100 U/ml penicillin, and 100 ␮g/ml streptomycin (1% penicillinstreptomycin). Equine dermal (NBL-6) and CHO-K1 cells were grown in Iscove’s modified Dulbecco’s medium (IMDM) (Invitrogen) supplemented with 10% FBS. CHO-A, CHO-B, and CHO-C were a kind gift from P. Spear, Northwestern University, Chicago, IL, and express the HveA, HveB, and HveC receptors, respectively. They were grown in IMDM supplemented with 10% FBS and 500 ␮g/ml G418 (Invitrogen). The human erythroleukemia cell lines K562 and K562␣V␤3, a kind gift from Scott D. Blystone, SUNY Upstate Medical University, NY, were grown in IMDM and IMDM supplemented with 750 ␮g/ml G418. PBMC were isolated from heparinized blood collected from healthy horses by density gradient centrifugation over Histopaque 1077 (Sigma), following the manufacturer’s instructions. After two washing steps, cells were resuspended in RPMI 1640 supplemented with 10% FBS, 0.3 mg/ml glutamine, 100 ␮g/ml kanamycin, nonessential amino acids (Biochrom), and 1% penicillin-streptomycin. To isolate primary equine vascular endothelial cells (EC) from carotid arteries of healthy horses, collagenase (Sigma) treatment was used as described previously (43). To separate EC from contaminating cells, cultures were labeled with 10 ␮g/ml of the low-density lipoprotein 1,1=-dioctadecyl-3,3,3=,3=-tetramethylindocarbocyanide perchlorate (DiI-Ac-LDL) (Biomedical Technologies Inc.) for 4 h at 37°C. After trypsinization, EC were washed, resuspended in medium, and sorted by fluorescence-activated cell sorting (FACS) using a FACS Aria high-speed flow cytometer (Becton Dickinson). The DiI-AcLDL fluorescent signal was detected with a 585/45 BP filter, and the brightest 22% of cells were gated and sorted at 65 lb/in2 into medium. Antibodies. The anti-human CD51/61 monoclonal antibody (MAb), an ␣V␤3 integrin antagonist, and MAb P1F6, an ␣V␤5 integrin antagonist, were obtained from Biolegend and Millipore, respectively. An unrelated mouse immunoglobulin G (IgG) isotype control was obtained from Cell Signaling Technologies. EHV-4 polyclonal anti-gD antibodies were kindly provided by Ken Maeda, Yamaguchi University, Japan. AntiEHV-1 gD 19-mer polyclonal antibodies were kindly provided by Dennis O’Callaghan, Louisiana State University Health Sciences Center, Shreveport, LA. EHV-1 gB MAb 3F6 was described before (5, 48). Anti-equine MHC class I monoclonal antibodies PT85A (isotype IgG2a) and H58A (isotype IgG2a) were obtained from VMRD. Plasmids. Transfer plasmids encoding either EHV-1 or EHV-4 gD with a kanamycin resistance (Kanr) gene were constructed. All of the primers are listed in Table 1. The EHV-1 and EHV-4 gD genes were amplified by PCR using primers P1 and P2 or P3 and P4, respectively (Table 1). The PCR products were digested with the appropriate restriction enzymes and inserted into vector pcDNA3 (Invitrogen), resulting in recombinant plasmids pcDNAgD1 and pcDNAgD4. To construct pcDNAgD1Kan and pcDNAgD1-4-Kan, the Kanr gene was amplified from plasmid pEPkan-S by PCR using primers P5, P6, P7, and P8 (Table 1). The PCR products were digested with the appropriate restriction enzymes and inserted into pcDNAgD1 or pcDNAgD4. Correct amplification and insertion were confirmed by nucleotide sequencing (Starseq). BAC mutagenesis. (i) pL11⌬gD1 and pYO⌬gD4. EHV-1 strain RacL11 cloned as a BAC (pL11) contains the enhanced green fluorescent protein (EGFP) gene instead of the nonessential gene 71 (59). The EHV-4 BAC clone pYO03 was generated by the insertion of mini F plasmid sequences flanked by loxP sites into the intergenic region between genes 58 and 59 (7). Both pL11 and pYO03 BACs were maintained in Escherichia coli GS1783 cells (a kind gift from Greg Smith, Northwestern University, Chicago, IL). Viruses reconstituted from pL11 and pYO03 were used in this study to make use of EGFP expression for rapid identification of infected cells. Deletion of gD1 and gD4 was done by two-step Red recombination as described before (67). Briefly, PCR primers, P9, P10, P11, and

Journal of Virology

EHV-1 and -4 gD Have a Crucial Role in Cellular Tropism

TABLE 1 Olignucleotide primers used in this study Product and primer

Sequencea

EHV-1gD P1 P2

ataggatccatgtctaccttcaagcttat atagcggccgcttacggaagctgggtatatt

EHV-4gD P3 P4

aatggatccatgtctaccttcaagcctat tatgcggccgcttacggaagctgagtatatt

Kan-1 P5 P6

ggtgaattcaacttcccacaaggagagtagggataacagggtaatcgattt gttgaattcaccaagaaaccgacgtggccagtgttacaaccaattaacc

Kan-4 P7 P8

agtgaattcaactacagacaaggtgtagggataacagggtaatcgattt gttgaattcactgagaaaacgacttggccagtgttacaaccaattaacc

gD1 deletion P9 P10 gD4 deletion P11 P12 gD4-Kan P13 P14 gD1-Kan P15 P16 gD152D P17 P18 gD152N P19 P20 Primers for sequencing P21 P22 P23 P24 P25 P26 P27

cgtgccaccgccctggtacgtgtttttcaataaacgaagcacagtgttgcgt aacctgctaggatgacgacgataagtaggg taaggccgtggacacctcccagcaggttacgcaacactgtgcttcgtttatt gaaaaacacaaccaattaaccaattctgattag cgcgccgctactttagtgggttttttttaataaacgcggtacagtgttgcgta acatgctaggatgacgacgataagtaggg taaggccgtgggtacctcccagcatgttacgcaacactgtaccgcgtttatt aaaaaaaacaaccaattaaccaattctgattag tatgcagaagcgtgccaccgccctggtacgtgtttttcaataaacgaagca tgtctaccttcaagcctat cgcgaagctttaaggccgtggacacctcccagcaggttacgcaacactgt ttacggaagctgagtatatt tgtgtagaagcgcgccgctactttagtgggttttttttaataaacgcggtatg tctaccttcaagcttat agcgtagctttaaggccgtgggtacctcccagcatgttacgcaacactgtt tacggaagctgggtatatt aatccaaagttattatttggaatgtgtgatgagcgatcaGtcaatagt aggatgacgacgataagtaggg cgcagcagttgtaatcaaactattgagccataatatatCtgatcgctcat cacacattcccaaccaattaaccaattctgattag aatccaaagttattatttggaatgtgtgatgagcgatcaAatatattatgg ctcaatagtaggatgacgacgataagtaggg ccgcagcagttgtaatcaaactattgagccataatatatTtgatcgctcatc acacattccaaccaattaaccaattctgattag

gctgcttgtactgtatgtta acatgctcatatgttctccg atgtgaaatgatagcgctga attgcctttgagcaaaactt gaggtcatagagcccgtaac ctttcttttgagcagaactt aatgccaaagaaatcaccac

a Restriction enzyme sites are given in lowercase bold letters; sequences in italics indicate additional bases which are not present in the EHV-1 or EHV-4 sequence. Underlined sequences indicate the template binding region of the primers for PCR amplification with pEPkan-S. Uppercase bold letters indicate the nucleotides that were mutated.

February 2012 Volume 86 Number 4

P12 (Table 1) were selected such that the recombination arms of 50 nucleotides (nt) enabled the substitution of nt 1 to 1209 of the EHV-1 or EHV-4 gD gene by the Kanr gene. PCR products were digested with DpnI in order to remove residual template DNA. The transfer fragments were then electroporated into GS1783 containing the BACs. Kanamycinresistant colonies were purified and screened by PCR and restriction fragment length polymorphism (RFLP) to detect E. coli harboring mutant clones. Positive clones were subjected to a second round of Red recombination to obtain the final constructs, pL11⌬gD1 and pYO⌬gD4, after excision of the Kanr gene (Fig. 1). (ii) pL11gD4 and pYOgD1. The 2,209-bp DNA transfer fragments, gD4Kan and gD1Kan, were amplified by PCR using pcDNAgD4Kan or pcDNAgD1Kan as templates and primers P13, P14, P15, and P16 (Table 1). The resulting PCR products were digested with DpnI and electroporated into GS1783 harboring the BACs. Electroporated cells were selected on LB agar plates containing 25 ␮g/ml chloramphenicol and kanamycin. Kanamycin-resistant colonies were purified and screened by PCR and RFLP to detect E. coli harboring recombinant pL11gD4Kan and pYOgD1Kan. Positive clones were subjected to a second round of Red recombination to obtain the final constructs, pL11gD4 and pYOgD1, after excision of the Kanr gene (Fig. 1). (iii) pL11gD1R and pYOgD4R. To reintroduce the authentic gD sequences in the respective BACs, we first deleted the gDs from both pL11gD4 and pYOgD1 and inserted back gD1 and gD4 into the corresponding BACs (Fig. 1). (iv) pYOgD4152D. A point mutation targeting the RSN motif present in EHV-4 gD was engineered by converting nucleotide 454 of gD from an adenine to a guanine, changing the asparagine into aspartic acid (gD152D), by employing two-step Red-mediated recombination. Primers P17, P18, P19, and P20 (Table 1) used for mutant (gD152D) and revertant (gD152N) were described earlier. The respective genotypes of all the mutants and revertants were confirmed by PCR, RFLP, and nucleotide sequencing using primers P21 to P27 (Table 1). Generation of recombinant viruses. The EHV-1 mutants and revertants, EHV-1gD4 and EHV-1R, were reconstituted after transfection of 2 ␮g of purified BAC DNA into RK13 cells using Lipofectamine 2000 (Invitrogen) as described previously (59). For EHV-4 mutants and revertants, EHV-4gD1, EHV-4152D, and EHV-4R, the viruses were reconstituted by transfection of purified DNA into 293 cells as described earlier (7, 8). Three days later, the supernatant and cells were collected and used to infect confluent NBL-6 cells. Western blotting. For Western blot analyses, pellets of infected FHK cells were resuspended in radioimmunoprecipitation assay buffer (50 mM Tris [pH 7.4], 1% Triton X-100, 0.25% Na-deoxycholate, 150 mM sodium chloride, 1 mM EDTA) with a protease inhibitor cocktail (Roche). Sample buffer (1 M Tris-HCl [pH 6.8], 0.8% sodium dodecyl sulfate [SDS], 0.4% glycerol, 0.15% ␤-mercaptoethanol, 0.004% bromophenol blue) was added, the mixture was heated at 95°C for 5 min, and proteins were separated by 10% SDS-polyacrylamide gel electrophoresis (PAGE) exactly as described before (70). Expression of gD was detected with EHV-4 polyclonal anti-gD antibodies (1/500 dilution) and anti-EHV-1 gD 19-mer polyclonal antibodies (1/1,000 dilution). Goat anti-mouse or anti-rabbit IgG coupled to peroxidase (Southern Biotech, Birmingham, AL) at 1/10,000 dilutions were used as secondary antibodies. Reactive bands were visualized by enhanced chemiluminescence (ECL Plus; Amersham). Virus growth assays. To determine virus replication, single-step growth kinetics and plaque areas were determined as described before (9, 70). Briefly, confluent NBL-6 cells were infected at a multiplicity of infection (MOI) of 1 or 0.01. After 1 h of adsorption, cells were washed and overlaid with DMEM containing 10% FBS. Infected cultures were harvested at the indicated times postinfection (p.i.) and stored at ⫺80°C. Viral titers were determined by plating onto NBL-6 cells. Three days p.i., cells were fixed with absolute ethanol and stained with 0.3% crystal violet,

jvi.asm.org 2033


FIG 1 Schematic diagram of the procedures used to construct mutant genomes. (a and b) Schematic representation of the genomic organization and the BamHI restriction map of the pL11 EHV-1 BAC and the pYO03 EHV-4 BAC. The two unique regions (UL and US), the terminal and internal repeat sequences (TRS and IRS), and the inserted mini-F cassettes are shown. The PCR cassette containing the Kanr gene was first inserted into the gD locus of pL11 or pYO03 using Red recombination. The transfer cassettes (gD4Kan and gD1Kan) were then electroporated into E. coli harboring pL11⌬gD1 or pYO⌬gD4, respectively. Recombinant pL11gD4 and pYOgD1 were ultimately transfected into eukaryotic cells to reconstitute infectious viruses (EHV-1gD4 and EHV-4gD1). (c) The Kanr gene was inserted at nucleotide position 456 of the gD4 gene in pYO03. A point mutation targeting the RSN motif present in gD4 was engineered by converting nucleotide 454 of gD from an adenine to a guanine, changing the asparagine into aspartic acid (gD152D).

and plaques were counted. Growth kinetics were determined in three independent experiments. For plaque size measurements, NBL-6 cells were seeded in six-well plates, infected with viruses at an MOI of 0.001, and overlaid at 1 h p.i. with DMEM containing 0.5% methylcellulose (Sigma). At 3 days p.i., 50 fluorescent plaques were photographed for each virus and average plaque areas were measured by using ImageJ software vl.32j (http://rsb.info.nih.gov/ij/). Values were calculated and compared to those for plaque areas induced by parental viruses, which were set to 100%. Average percentages of plaque areas and standard deviations (SD) were determined from three independent experiments. Virus infection assay. A cluster assay, similar to a plaque assay, for measuring the infectivity (efficiency of plating) of recombinant viruses was developed. Cell monolayers were inoculated with different viruses at an MOI of 0.1. After 1 h of adsorption, cells were washed and overlaid with DMEM containing 10% FBS and the infection was allowed to proceed for another 48 h. Infected cells were inspected by using an inverted fluorescence microscope (Zeiss Axiovert 100) and photographed with an Axiocam charge-coupled-device (CCD) camera (Zeiss) for each virus. Flow cytometry. To evaluate integrin expression, CHO-K1, Vero, NBL-6, FHK, K562, K562␣V␤3, and PBMC cells were incubated with 2 ␮g/ml of MAbs targeting ␣V␤3 (CD51/61) or ␣V␤5 (P1F6) for 1 h at room temperature (RT). An isotype control mouse IgG (2 ␮g/ml) was also included. After two washes with phosphate-buffered saline (PBS), cells were incubated with Alexa Fluor 488-labeled goat anti-mouse IgG (1/500

2034

jvi.asm.org

dilution) for 1 h at RT. After a final wash, 10,000 cells were analyzed with a FACSCalibur flow cytometer (BD Biosciences), and the intensity of fluorescence was analyzed using FlowJo software (Treestar). For infection experiments, a total of 2 ⫻ 105 CHO-K1, Vero, NBL-6, and FHK cells as well as K562, K562␣V␤3, and PBMC suspensions were seeded in 24-well plates. Before the addition of antibodies, maintenance media were removed and all cells were washed with PBS containing 2% FBS. Cells were then incubated with integrin and/or anti-equine MHC-I antibody H58A (20 ␮g/ml) at 4°C for 1 h. After being washed, cells were infected with the parental, mutant, or revertant virus at an MOI of 1 or 5. CHO-K1, Vero, NBL-6, and FHK cells were trypsinized at 24 h p.i. and washed twice, while K562, K562␣V␤3, and PBMC were washed twice at 48 h p.i. After centrifugation, cells were resuspended in PBS supplemented with propidium iodide (PI) (Molecular Probes) at a final concentration of 10 ␮g/ml, and the intensity of fluorescence of 10,000 cells was analyzed to determine the percentage of infected cells. Analysis of virus binding to cells by flow cytometry. K562, K562␣V␤3, and NBL-6 cells were washed with PBS containing 2% FBS. The cells were incubated with either rL11⌬gp2 or rWA79 at an MOI of 5 for 2 h at 4°C. In other experiments, CHO-K1 cells were incubated with parental and mutant viruses for 2 h at 4°C. Cells were then washed twice and incubated with either EHV-1 gB MAb 3F6 or EHV-4 polyclonal anti-gD antibodies. The binding levels of both viruses were measured by using Alexa Fluor 488-labeled goat anti-mouse IgG (1/500 dilution) and quantified using

Journal of Virology


FIG 2 Identification of recombinant viruses by PCR, RFLP, and Western blotting. (a) PCR analysis of EHV-1 (left panel) and EHV-4 (middle and right panels) mutants to detect genes 72 encoding gD. PCR products from parental and mutant viruses were electrophoresed in a 1% agarose gel. A molecular weight marker (lane M) was included. (b) Purified DNAs from EHV-4, EHV-4⌬gD, EHV-4gD1, and EHV-4gD152D (left panel) as well as EHV-1, EHV-1⌬gD, and EHV-1gD4 (right panel) were digested with ScaI, ClaI, and KpnI. Fragments in the mutants that appeared as a consequence of the deletion or insertion of gD sequence are marked by arrows. (c and d) Parental and mutant viruses express gD at similar levels. FHK cells were infected with parental, mutant, or revertant viruses at an MOI of 0.1. For Western blot analysis, cell lysates were prepared and proteins separated by SDS-10% PAGE before transfer to a polyvinylidene difluoride (PVDF) membrane. The blots were incubated with anti-EHV-1 gD 19-mer (1/1,000 dilution) (c) or EHV-4 polyclonal anti-gD antibodies (1/500 dilution) (d) followed by anti-rabbit and anti-mouse IgG peroxidase antibodies (1/7,500 dilution), respectively. Cell lysates from noninfected FHK cells were included as a control.

flow cytometry. The effect of anti-integrin ␣V␤3 antibodies on virus binding was also studied by incubating cells with the anti-integrin ␣V␤3 MAb CD51/61 for 1 h before virus infection. Binding of both viruses was detected and quantified as described above. As a control, cells were also stained with the primary and secondary antibodies without prior virus binding. Statistical analysis. Using Microsoft Excel, Student’s t test for paired data was used to test for statistically significant differences. Data given are means, and bars show standard deviations.

RESULTS

Generation and genotypic characterization of mutant viruses. Two-step Red-mediated recombination was used to exchange the gD-encoding sequences between EHV-1 and EHV-4 (Fig. 1). The newly generated viruses included an EHV-1 mutant harboring gD4 in place of authentic gD1 and a corresponding EHV-4 carrying gD1 (Fig. 1a and b). Furthermore, EHV-4 with a point mutation in the RSN motif present in gD was engineered by changing the asparagine into aspartic acid (gD152D) (Fig. 1c). The appropriate revertant viruses, in which the original sequences were restored, were also generated using the same methodology. PCR analysis showed that all constructs were of the correct size and position (Fig. 2a), and further analysis of the mutant clones was done using RFLP (Fig. 2b) and nucleotide sequencing (data not shown). To determine whether gD1 and gD4 were expressed properly by the mutant viruses, FHK cells were infected with the mutant or revertant viruses and the cell lysates were analyzed by Western


blotting. All viruses expressed gD, approximately 55 kDa in size, at levels comparable to those of the parental viruses (Fig. 2c and d). The experiments showed that all mutant viruses expressed either gD1 or gD4 according to the genetic manipulation and as predicted. Virus growth analyses. Three independent growth kinetic experiments and plaque size measurements were performed using NBL-6 cells. The results showed that parental, mutant, and revertant viruses grew to similar titers in NBL-6 cells, regardless of whether they harbored gD1 or gD4 (Fig. 3a and b). In addition, there was no significant difference observed between plaque areas of the mutants and their corresponding parental viruses (Fig. 3c and d). We concluded, therefore, that the growth rates of the mutant viruses were not significantly affected by the exchange of gD genes. Type-specific gD determines the host range of EHV-1 and EHV-4 in vitro. EHV-1 can readily be propagated in many cell lines, including primary equine cells and cell lines derived from other species (75). In contrast, EHV-4 appears to be restricted mainly to cells derived from horses, such as FHK and NBL-6 cells, or replicates only poorly in very few other cell lines, such as Vero cells (44). To test the hypothesis that gD is a major determinant of EHV-1 and EHV-4 tropism, gD mutant viruses were used to explore the viruses’ ability to infect different cell lines, using the same infectious doses (MOI, 0.1), which allowed us to define the host range of each virus in vitro. Indeed, most of the cell lines permissive for EHV-1 became resistant to EHV-1 harboring gD4

jvi.asm.org 2035


FIG 3 In vitro growth properties of parental and mutant viruses. (a and b) For growth kinetics, NBL-6 cells were infected at an MOI of 1 or 0.01. Infected cells and supernatants were collected and virus titers were determined at the indicated times p.i. The data presented are means ⫾ SD of triplicate measurements. The asterisk indicates a P value of ⬎ 0.05 for means when compared to the parental viruses. (c and d) Means ⫾ SD of diameters of 50 plaques measured for each virus are shown. The plaque diameter of parental viruses was set to 100%. The asterisk indicates a P value of ⬎ 0.05 for means when compared to the parental viruses.

instead of gD1, and the opposite was observed for EHV-4 expressing gD1 (Fig. 4). We were able to show that RK13, CHO-K1, CHO-A, CHO-B, and CHO-C cells, all of which are highly resistant to infection with EHV-4, can be infected with EHV-4_gD1 at levels comparable to those of EHV-1. In contrast, all of these cell lines became highly resistant to EHV-1_gD4 (Fig. 4). To examine whether EHV-4 gD has a similarly defining role for cellular tropism, we infected Vero and CrFK cells with all mutant viruses (Fig. 5). Vero cells were shown previously to be virtually resistant to EHV-1 but not EHV-4 infection (27, 80); however, no data were available on EHV-1 or EHV-4 infection of CrFK cells. We were able to show that EHV-4 can infect both cell types efficiently, while EHV-1 only poorly infected these cells. Interestingly, EHV-1_gD4 was able to infect Vero and CrFK cells at levels comparable to those of EHV-4; however, the plaque morphology was strikingly different from that of EHV-4. On the other hand, EHV-4_gD1 became virtually unable to infect Vero or CrFK cells (Fig. 5). In addition, we discovered that HeLa cells, MDBK cells, PBMC, EC, and 293 cells can be infected with both EHV-1 and EHV-4 at similar efficiencies (Fig. 6). We concluded from the latter finding that gD1 and gD4 have the same role and facilitate entry of either virus into these cells. However, additional factors such as other glycoproteins, including gB and/or gH/gL or host factors, may contribute to the cellular tropism of the two viruses (Fig. 6). Taken

2036

jvi.asm.org

together, the data clearly indicated that gD1 can use host entry receptors, which are not accessible to gD4, and that gD1 confers an extended cellular host range to EHV-4. EHV-1 gD binds to CHO-K1 cells independently of integrin ␣V␤5. Recently, it was described that EHV-1 enters CHO-K1 cells predominantly via the endocytic pathway and that this entry is triggered by the interaction between cellular integrins and the RSD motif present in EHV-1 but not EHV-4 gD (28, 68). We constructed several mutants to shed more light on the process. First, CHO-K1 cells were infected with EGFP-expressing parental EHV-1 and EHV-4 strains as well as the EHV-1_gD4, EHV4_gD1, and EHV-4_gD152D mutants. As shown above, CHO-K1 cells can only be infected with parental EHV-1 and EHV-4_gD1, showing that EHV-1 gD is essential for infection of these cells. We further explored if integrins are essential for EHV-1 entry. EHV-1 gD was shown to contain an RSD amino acid sequence, a motif that mimics the canonical integrin-binding motif RGD, at position 152 (33, 68). Furthermore, ␣V␤5, one of the integrins that can recognize RGD motifs (60), is known to be expressed on the surface of CHO-K1 cells (1, 68). First, the proper expression of ␣V␤5 on CHO-K1 cells using the function-blocking MAb P1F6 was confirmed (Fig. 7a); however, these antibodies failed to block the entry of EHV-1 or EHV-4_gD1 (Fig. 7b). We also confirmed the previous results of Sasaki et al. (61), that anti-MHC class I

Journal of Virology


FIG 4 The role of gD in EHV-1 and EHV-4 cell tropism. CHO-K1 (a), CHO-A (b), CHO-B (c), CHO-C (d), or RK13 (e) cells were infected at an MOI of 0.1 with the engineered EHV recombinants, all of which express EGFP. At 48 h p.i., cells were inspected with a fluorescence microscope (Zeiss Axiovert) and images were taken with a CCD camera. The bar represents 100 ␮m.

antibody H58A failed to block entry of these viruses into CHO-K1 cells (data not shown). Furthermore, our results indicated that, like parental EHV-4, EHV-4_gD152D was unable to infect CHO-K1 cells (Fig. 7c). Next, we quantitated by FACS analysis the binding of parental or mutant viruses to these cells. Binding was assessed in the presence or absence of ␣V␤5 MAb P1F6. We discovered that EHV-1 and EHV-4_gD1 can bind to CHO-K1 cells, while EHV-4, EHV-1_gD4, and EHV-4_gD152D bound only poorly, regardless of whether the integrin MAb was present or not (Fig. 7d). As the latter viruses were unable to enter into CHO-K1 cells, the data suggested that EHV-1 gD is essential for stable virus attachment and entry into CHO-K1 cells but that these processes are independent of gD interaction with ␣V␤5 integrins or equine

MHC-I. The receptor(s) for EHV-1 gD utilized on CHO-K1 for stable binding and initiation of entry still needs to be identified. To further elucidate the role of integrin in virus entry, we determined the expression of different integrins on the surface of different cells. Our results showed that the RGD-binding ␣V␤3 integrin (60) was highly expressed on the surface of Vero and FHK cells (data not shown). However, another integrin, ␣4␤1, which can bind to a different motif (32, 38), was found to be only moderately expressed on the surface of NBL-6 cells. Blocking of integrin receptors, using specific antibodies, on these cells did not affect the entry of parental or mutant viruses (data not shown). Integrins are not essential for virus entry into PBMC. PBMC

FIG 5 gD4 confers an extended cellular host range to EHV-1. CrFK (a) and Vero (b) cells were infected with parental and mutant viruses at an MOI of 0.1. At 48 h p.i., cells were inspected with a fluorescence microscope (Zeiss Axiovert) and images were taken with a CCD camera. The bar represents 100 ␮m.


jvi.asm.org 2037


FIG 6 Infection of various cells by parental and mutant viruses. 293 cells (a), HeLa cells (b), MDBK cells (c), PBMC (d), or EC (e) cells were infected with various viruses. At 48 h p.i., cells were inspected with a fluorescence microscope (Zeiss Axiovert) and images were taken with a CCD camera. The bar represents 100 ␮m.

are highly relevant to EHV-1 pathogenesis as outlined above, whereas EHV-4 infection of leukocyte is reported to be a rare event. Moreover, it was postulated that the RSD integrin motif, present in EHV-1 gD but not EHV-4 gD, is an important determinant for efficient infection of PBMC in vitro (68). In order to address the hypothesis that gD1 but not gD4 can mediate viral entry into PBMC, we first infected PBMC with EHV-1 and EHV-4 strains and analyzed the percentages of infected cells by flow cytometry. Surprisingly, the percentages of infected PBMC, after inoculation with EHV-1 or EHV-4, were nearly identical, ranging from approximately 3% (MOI, 1) to 8% (MOI, 5; Fig. 8a and b). In a second round of experiments, we included all of our mutants in the PBMC infection experiments. Similar to our previous results, the rate of infection was almost the same for all viruses used (Fig. 8c). Furthermore, we used anti-integrin MAb CD51/61 to investigate the role of integrins during entry into PBMC. Flow cytometric analysis showed that, in contrast to CHO-K1 cells, PBMC express ␣V␤3 instead of ␣V␤5 integrin. However, the rate of ␣V␤3 expression was very low (7%) (Fig. 9a) compared to the rate of ␣V␤5 expression on the surface of CHO-K1 cells (25%) (Fig. 7a). Preincubation of PBMC with the ␣V␤3 function-blocking MAb CD51/61 did not affect the rate of infection in PBMC for any of the parental or mutant viruses (Fig. 9b). When similar experiments were performed using anti-MHC class I antibody, H58A, and PT85A, either alone (Fig. 9c) or together with the integrin antibodies CD51/61 and P1F6 (Fig. 9d), no difference in infectivity could be observed. We concluded from our data that PBMC can be infected with similar efficiencies by both EHV-1 and EHV-4 in vitro. Further-

2038

jvi.asm.org

more, integrins apparently do not serve as the (only) receptor for EHV-1 or EHV-4 entry into PBMC. In addition, in contrast to other equine cells, we were unable to block entry of EHV-1 or EHV-4 into PBMC with MHC-I antibodies, which indicates that these viruses likely utilize a different receptor(s) to infect PBMC. Virus binding to K562 and K562␣V␤3 cells. To further confirm our finding that EHV-1 and EHV-4 enter cells independently of integrins, we conducted experiments using the K562 and K562␣V␤3 cell lines. The erythroleukemia cell line K562 was chosen because it expresses a very limited number of integrins and no detectable ␣V integrins (12). K562␣V␤3 are stably expressing ␣V␤3 integrins (12), which can bind to the RGD motif. We first confirmed the expression of ␣V␤3 integrins on the surface of K562␣V␤3 cells using anti-␣V␤3 MAb CD51/61 (Fig. 10a). Then, we infected K562 and K562␣V␤3 cells with the parental EHV-1 and EHV-4 strains in the presence or absence of the antibody. Our results showed that both cell types could not be infected with either of the viruses (data not shown). Next, we analyzed the binding of either EHV-1 or EHV-4 to these cells using FACS. Cells were incubated with either EHV-1 or EHV-4 for 2 or 24 h at 4°C and reacted with either EHV-1 gB MAb or EHV-4 polyclonal anti-gD antibody, respectively. Both viruses failed to bind K562 cells, irrespective of whether they expressed ␣V␤3 integrins (Fig. 10b). NBL-6 cells were included as a positive control and allowed binding of both EHV-1 and EHV-4 (Fig. 10b). We concluded from this experiment that neither the RSD motif in EHV-1 gD nor other related motifs in the EHV-1 or EHV-4 envelope are capable of mediating binding to ␣V␤3 integrins and that, therefore, integrins do not play a major role in EHV-1 or EHV-4 tropism.

Journal of Virology


FIG 7 Inhibition of integrins has no effect on EHV-1 infectivity in CHO-K1 cells. (a) CHO-K1 cells in suspension were incubated with anti-␣V␤5, MAb P1F6, for 1 h at RT, followed by incubation with Alexa Fluor 488-labeled goat anti-mouse IgG for 1 h at RT. As controls, cells were incubated with irrelevant MAbs of the same IgG isotype. Integrin expression was determined by flow cytometry. (b) Cells were preincubated with medium or ␣V␤5 function-blocking MAb P1F6, for 1 h at 4°C, followed by infection with L11⌬gp2 or EHV-4gD1 at an MOI of 5 for 2 h at 37°C. At 16 h p.i., cells were detached and the percentage of infected cells was determined by flow cytometry. (c) CHO-K1 cells were infected with EHV-4 or EHV-4gD152D at an MOI of 2. At 48 h p.i., cells were inspected with a fluorescence microscope (Zeiss Axiovert) and images were taken with a CCD camera. The bar represents 100 ␮m. (d) Binding of various viruses to CHO-K1 cells. CHO-K1 cells were incubated with parental and mutant viruses for 2 h at 4°C in the presence of MAb P1F6. Cell surface binding was detected by flow cytometry. All data represent the mean ⫾ SD of three independent experiments.

DISCUSSION

Envelope glycoprotein gD is conserved among almost all alphaherpesviruses and is essential for virus entry if present in a particular virus (14, 19). Members of the gD family are the main alphaherpesviruses receptor binding proteins and can bind to various cellular receptors, including nectin-1 (Hve C), nectin-2 (Hve B), HVEM (Hve A), 3-O-sulfated heparan sulfate, and others to facilitate virus entry (13, 35, 40, 61). However, EHV-1 gD was shown to utilize a unique entry receptor(s) that differs from those used by related viruses (27, 28). The differential use of receptors may allow entry of EHV-1 into a wide range of different cell types. In contrast, EHV-4 appears to be restricted mainly to primary equine cells and little information is available regarding the receptors or receptor-binding glycoproteins that may trigger cellular entry of EHV-4. In the present study, we focused on the contribution of EHV-1 and EHV-4 gD to the infectivity of different cultured cells as well as the ability of gD to bind to various integrins expressed on different cell types. gD plays essential roles in virus entry and cell tropism. From our experiments, we concluded that gD (i) is interchangeable between EHV-1 and EHV-4 without altering the replication kinetics of the respective viruses in conventional equine cell cultures and (ii) determines the host range of both EHV-1 and EHV-4. Inter-


estingly, gD1 was able to considerably expand the host range of EHV-4 and provide the ability to infect cells that were not accessible through gD4. On the other hand, although gD4 was associated with a narrow host range, it similarly endowed EHV-1 with the ability to infect cells that were originally resistant to infection. Two of the most indicative cell lines used in this study were RK13 and Vero cells. It has long been known that RK13 cells are permissive for EHV-1, yet these cells are highly resistant to EHV-4 infection (22, 75). On the other hand, EHV-1 is known to be severely impaired with respect to entry into Vero cells while EHV-4 can infect these cells efficiently (27, 80). By infecting these cells with parental EHV-1 and EHV-4 as well as the EHV-1_gD4 and EHV4_gD1 viruses, we observed that the cytopathic effect (CPE) induced by the mutant viruses was not identical to that induced by parental viruses, which had a cluster-like morphology and only minimal signs of syncytia. Our results are consistent with the previous finding of Whalley and coworkers, who showed that morphological changes upon EHV-4 infection in RK13 cells constitutively expressing gD1 were different from those induced by EHV-1 (75). As gB, gH/L, and gK also control the process of entry, syncytium formation, and egress, one may expect this difference in CPE morphology (25, 50, 74). As with other alphaherpesviruses, we confirmed that efficient

jvi.asm.org 2039


FIG 8 Infection of equine PBMC with recombinant viruses. PBMC were incubated with EHV-1 or EHV-4 at an MOI of 1 (a) or 5 (b) for 1 h at 37°C. After 48 h, the percentage of infected cells was determined by flow cytometry. In another experiment, PBMC were infected with all mutant viruses (c). The data represent the mean ⫾ SD of at least three independent experiments.

EHV-1 and EHV-4 entry was mainly dependent on gD, which can bind to several cell surface receptors. As stated before (27), we also proved that EHV-1 can efficiently enter and replicate in CHO-K1, CHO-A, CHO-B, and CHO-C cells. On the other hand, our results show that EHV-4 gD cannot bind to any of the wellestablished herpesvirus entry receptors HveA, HveB, and HveC or even the unique receptor on the surface of CHO-K1 cells. However, EHV-4 gD allows virus entry into Vero cells, which is one of the cell lines highly permissive for HSV-1, but the cellular receptors, which interact with gD1 or gD4 in the case of CHO-K1 or Vero cells, respectively, still need to be elucidated. Since equine cell lines such as FHK and NBL-6 cells are permissive for both EHV-1 and EHV-4, both viruses may use the same receptor in the case of these cells. Recently, two independent studies have identified equine MHC class I as a cellular entry receptor for EHV-1 (40, 61). Equine MHC-I acts as a gD receptor for EHV-1 entry into equine cells, including NBL-6 and EC

2040

jvi.asm.org

(CNS). Our own data also show that EHV-4 also utilizes equine MHC-I to infect different equine cells, including FHK, NBL6, and EC cells (unpublished data). The crystal structure of gD1 or gD4 has not been determined yet. However, on the basis of overall similarity, one may derive some information from their HSV-1 counterpart. X-ray structural analysis revealed that gD residues 7 to 32 are required for binding of gD to different cell receptors (15, 20, 45, 79). The EHV-1 and EHV-4 gD homologues have approximately 77% amino acid identity with each other and 20 and 25%, respectively, with HSV gD (66, 76). The alignment supports the notion of conservation of tertiary structures across the gD family, with positional identity of six cysteine residues and other sequence motifs. Therefore, it is likely that regions similar to those in HSV-1 gD are involved in receptor binding of EHV-1 and EHV-4 gD (24, 73). The N-terminal region is among the most variable between the gDs of EHV-1 and EHV-4 and may therefore be implicated in the differences in host cell tropism. In order for gD to initiate entry, the glycoprotein must interact with other viral glycoproteins. Previous studies have shown that the C-terminal region of HSV-1 gD (encompassing residues 260 to 310) is required to trigger fusion (18, 42). In addition, the C terminus partially masks receptor-binding sites and prevents access to key residues located in the N-terminal part. Upon receptor binding, the C terminus is pushed aside, and this movement is necessary to activate the fusion machinery represented by gB and gH/gL (18). As it seems likely that the mechanism of gD triggering fusion is shared among alphaherpesviruses (30), EHV-1 or EHV-4 gD may bind to the cell surface receptor and still be able to activate the fusion machinery of the other virus. This theoretical possibility is supported by the observation that gD1 and gD4 differ in only 5 amino acids in the respective region encompassing residues 260 to 310. gD interacts with cell surface proteins independently of integrins. One of the main objectives of this work was to study the role of gD in determining the host range of EHV-1. Therefore, we included many cell lines and primary cells, including PBMC, in our study. Surprisingly, we found that both EHV-1 and EHV-4 can infect PBMC with the same efficiency when using the same MOI. This directed our attention toward the role of integrins in EHV-1, but not EHV-4, entry into PBMC as stated before (68). It has been shown that EHV-1 can enter certain cell types, mainly CHO-K1 and PBMC, via endocytosis upon binding of the RSD motif of gD1 to integrins ␣V␤3 and ␣V␤5 (68), although, in the case of HSV-1, there was no effect on plaque formation in epithelial cells by using an RGD peptide or monoclonal antibodies to the human ␤1 or ␤4 integrin subunit (37). More recently, several lines of evidence have shown that HSV-1 gH/gL bind to cells independently of ␣V␤3 integrin (31). Using different mutants and revertants, we asked whether gD is a ligand to ␣V␤3 and ␣V␤5 integrins. We observed that function-blocking antibodies against ␣V␤5 and ␣V␤3 had no effect on productive EHV-1 or EHV-4 infection in CHO-K1 cells and PBMC. While ␣V␤5 integrins are moderately expressed on the surface of CHO-K1 cells, ␣V␤3 integrins are poorly expressed on the surface of PBMC. This may indicate that ␣V␤3 integrins may not be implicated in virus entry. Furthermore, we observed that EHV-1 and EHV-4 failed to infect or bind to K562 cells, irrespective of whether they were negative or positive for ␣V␤3 integrins. We could demonstrate by mutational analysis that EHV-

Journal of Virology


FIG 9 Integrins have no role in EHV-1 entry into equine PBMC. (a) PBMC were incubated with anti-␣V␤3 MAb CD51/61 or ␣V␤5 MAb P1F6 for 1 h at RT, followed by incubation with Alexa Fluor 488-labeled goat anti-mouse IgG for 1 h at RT. As controls, cells were incubated with irrelevant MAbs of the same IgG isotype. Integrin expression was determined by flow cytometry. Cells were preincubated with ␣V␤3 function-blocking MAb CD51/61 (b), anti-equine MHC-I MAb H58A (c), or both MHC-I and integrin antibodies (d) for 1 h at 4°C, followed by infection with recombinant viruses at an MOI of 2 for 2 h at 37°C. At 48 h p.i., cells were washed and the percentage of infected cells was determined by flow cytometry. The rate of infection of parental viruses was set to 100%. All data represent the mean ⫾ SD of three independent experiments.

4_gD152D, a mutant specifying an RSD motif in gD, failed to infect and/or bind to CHO-K1 and that the function-blocking ␣V␤5 had no effect on productive EHV-1 or EHV-4_gD1 infection in CHO-K1 cells. Furthermore, blocking of ␣V␤3 integrin receptors on the surface of Vero cells using specific antibodies did not affect


the entry of EHV-4_gD152D. In addition, PBMC were infectible at virtually identical levels by parental EHV-1, EHV-4, and mutant viruses (EHV-1_gD4, EHV-4_gD1, and EHV-4_gD152D) in the presence or absence of integrin antibodies. This is in contrast to the earlier results (68) where anti-integrin antibodies were shown

jvi.asm.org 2041


FIG 10 Binding of EHV-1 to K562 cells. (a) K562 or K562␣V␤3 cells were incubated with anti-␣V␤3 MAbs for 1 h at RT, followed by incubation with Alexa Fluor 488-labeled goat anti-mouse IgG for 1 h at RT. As controls, cells were incubated with irrelevant MAbs of the same IgG isotype. Integrin expression was determined by flow cytometry. (b) K562, K562␣V␤3, or NBL-6 cells were incubated with EHV-1 or EHV-4 virus for 2 h at 4°C in the presence of ␣V␤5 function-blocking antibodies. Cell surface binding was detected by flow cytometry. All data represent the mean ⫾ SD of three independent experiments.

to have a significant effect on productive EHV-1 infection in CHO-K1 cells and PBMC. However, the low expression levels of ␣V␤3 integrin on the surface of PBMC or ␣V␤5 on CHO-K1 cells reported here, compared to the extremely high expression of MHC-I, may indicate that integrins may not be “convenient” cellular receptors for EHV-1 or EHV-4 entry. Therefore, it is possible that at least the PBMC used here would have different activation levels and likely higher MHC-I levels. Our results are, however, in agreement with those obtained before with HSV-1, where a recombinant virus in which the potential integrin-binding motif RGD present in gH was mutated entered cells as efficiently as the wild type (29). Also in another study, Gianni and coworkers state that the presence of ␣V␤3 integrins is not a requirement for gH/gL binding to different cells (31). Based on the data obtained in this study, we conclude that the presence of integrins is not necessary for EHV-1 or EHV-4 cellular entry. Consistent with this interpretation, expression of ␣V␤3 integrins in K562 cells that are negative for both subunits did not confer the ability to bind gD. While anti-MHC-I antibodies could block the entry of EHV-1 and EHV-4 into equine cells (data not shown), a function-blocking ␣V␤3 integrin MAb failed to reduce the infection of integrinpositive cells. To further assess the role of integrins in the entry of EHV-1 into PBMC, two other mutant viruses generated in our laboratory, EHV-1_gH4 and EHV-4_gH1, were used to infect PBMC in the presence or absence of ␣4␤1 or ␣4␤7 antibodies. EHV-1 gH is known to harbor an integrin-binding motif (LDI) (32, 38), which may mediate entry into PBMC through the respective integrins. Again, and very similar to the results reported here, PBMC were infected at similar rates by all parental and mutant

2042

jvi.asm.org

viruses (L. Zajic, W. Azab, and N. Osterrieder, unpublished observations). Interestingly, our results showed that PBMC were infected at similar rates and with comparable kinetics by all viruses, including the mutant viruses. Previous studies on EHV-4 showed that viral genomic DNA can be initially detected in PBMC of infected foals. However, viral DNA loads were low and lasted only for a short period of time (56). In contrast, in the case of EHV-1, high levels of cell-associated viremia were detected for prolonged periods (3, 51). In another study, it was reported that EHV-4 is incapable of efficiently infecting PBMC in vitro (53). However, one has to carefully interpret these results, as the EHV-4 strain used in that study was a mutant in which the viral gM gene was absent. The resulting virus mutant was shown to exhibit a significant defect in viral replication (80). From our experiments and the existing literature, one may hypothesize that the difference in pathogenic potential between EHV-1 and EHV-4 may be caused by the decreased ability of EHV-4 to manipulate mononuclear cells in vivo but not by reduced infectivity. Taken together, we provide evidence here of the following. (i) gD of EHV-1 confers an extended cellular host range to EHV-4, and we consider it likely that it represents an important virulence factor. Therefore, it will be critical to relate our findings in vitro to cells and tissues in vivo. However, while EHV-1 can infect mice and cause respiratory disease, in vivo studies of EHV-4 pathogenesis have been restricted owing to the lack of suitable small-animal models for this virus (6, 10, 54). (ii) Integrins apparently do not play a decisive role in the entry of EHV-1 or EHV-4 into PBMC or any other cells tested in this study. However, we cannot rule out

Journal of Virology


the possibility that, in specific cells, integrins may play a role in EHV-1 or EHV-4 entry or postentry steps, particularly in signaling cascades. (iii) PBMC can be infected with similar efficiency by both EHV-1 and EHV-4 in vitro. Our findings therefore raise the question of why EHV-4 is obviously unable to establish a systemic infection in the natural host. Finally, in contrast to other equine cells, we were unable to block entry of EHV-1 or EHV-4 into PBMC with MHC-I antibodies, which indicates that these viruses utilize a different receptor(s) to infect PBMC and that the alternative receptor is the one preferably used in these cells. ACKNOWLEDGMENTS This work was supported by a grant from the Alexander von Humboldt Foundation to W.A. and by unrestricted funds made available to N.O. by Freie Universität Berlin.

REFERENCES 1. Ahrens IG, et al. 2006. Evidence for a differential functional regulation of the two beta(3)-integrins alpha(V)beta(3) and alpha(IIb)beta(3). Exp. Cell Res. 312:925–937. 2. Akula SM, Pramod NP, Wang FZ, Chandran B. 2002. Integrin alpha3beta1 (CD 49c/29) is a cellular receptor for Kaposi’s sarcomaassociated herpesvirus (KSHV/HHV-8) entry into the target cells. Cell 108:407– 419. 3. Allen GP, Breathnach CC. 2006. Quantification by real-time PCR of the magnitude and duration of leucocyte-associated viraemia in horses infected with neuropathogenic vs. non-neuropathogenic strains of EHV-1. Equine Vet. J. 38:252–257. 4. Allen GP, Bryans JT. 1986. Molecular epizootiology, pathogenesis, and prophylaxis of equine herpesvirus-1 infections. Prog. Vet. Microbiol. Immunol. 2:78 –144. 5. Allen GP, Yeargan MR. 1987. Use of lambda gt11 and monoclonal antibodies to map the genes for the six major glycoproteins of equine herpesvirus 1. J. Virol. 61:2454 –2461. 6. Awan AR, Chong YC, Field HJ. 1990. The pathogenesis of equine herpesvirus type 1 in the mouse: a new model for studying host responses to the infection. J. Gen. Virol. 71(Pt 5):1131–1140. 7. Azab W, et al. 2009. Cloning of the genome of equine herpesvirus 4 strain TH20p as an infectious bacterial artificial chromosome. Arch. Virol. 154: 833– 842. 8. Azab W, et al. 2010. Characterization of a thymidine kinase-deficient mutant of equine herpesvirus 4 and in vitro susceptibility of the virus to antiviral agents. Antiviral Res. 85:389 –395. 9. Azab W, et al. 2010. Glycoprotein C of equine herpesvirus 4 plays a role in viral binding to cell surface heparan sulfate. Virus Res. 151:1–9. 10. Azmi M, Field HJ. 1993. Interactions between equine herpesvirus type 1 and equine herpesvirus type 4: T cell responses in a murine infection model. J. Gen. Virol. 74:2339 –2345. 11. Bergelson JM, Shepley MP, Chan BM, Hemler ME, Finberg RW. 1992. Identification of the integrin VLA-2 as a receptor for echovirus 1. Science 255:1718 –1720. 12. Blystone SD, Graham IL, Lindberg FP, Brown EJ. 1994. Integrin alpha v beta 3 differentially regulates adhesive and phagocytic functions of the fibronectin receptor alpha 5 beta 1. J. Cell Biol. 127:1129 –1137. 13. Campadelli-Fiume G, et al. 2007. The multipartite system that mediates entry of herpes simplex virus into the cell. Rev. Med. Virol. 17:313–326. 14. Campadelli-Fiume G, Menotti L. 2007. Entry of alphaherpesviruses into the cell, p 93–111. In Arvin A, Campadelli-Fiume G, Mocarski E, Moore PS, Roizman B, Whitley R, Yamanishi K (ed), Human herpesviruses: biology, therapy, and immunoprophylaxis. Cambridge University Press, Cambridge, United Kingdom. 15. Carfi A, et al. 2001. Herpes simplex virus glycoprotein D bound to the human receptor HveA. Mol. Cell 8:169 –179. 16. Chesnokova LS, Nishimura SL, Hutt-Fletcher LM. 2009. Fusion of epithelial cells by Epstein-Barr virus proteins is triggered by binding of viral glycoproteins gHgL to integrins alphavbeta6 or alphavbeta8. Proc. Natl. Acad. Sci. U. S. A. 106:20464 –20469. 17. Clark EA, Brugge JS. 1995. Integrins and signal transduction pathways: the road taken. Science 268:233–239.


18. Cocchi F, et al. 2004. The soluble ectodomain of herpes simplex virus gD contains a membrane-proximal pro-fusion domain and suffices to mediate virus entry. Proc. Natl. Acad. Sci. U. S. A. 101:7445–7450. 19. Connolly SA, Jackson JO, Jardetzky TS, Longnecker R. 2011. Fusing structure and function: a structural view of the herpesvirus entry machinery. Nat. Rev. Microbiol. 9:369 –381. 20. Connolly SA, et al. 2003. Structure-based mutagenesis of herpes simplex virus glycoprotein D defines three critical regions at the gD-HveA/HVEM binding interface. J. Virol. 77:8127– 8140. 21. Crabb BS, Studdert MJ. 1995b. Equine herpesviruses 4 (equine rhinopneumonitis virus) and 1 (equine abortion virus). Adv. Virus Res. 45:153–190. 22. Csellner H, et al. 2000. EHV-1 glycoprotein D (EHV-1 gD) is required for virus entry and cell-cell fusion, and an EHV-1 gD deletion mutant induces a protective immune response in mice. Arch. Virol. 145:2371–2385. 23. Davison AJ, et al. 2009. The order Herpesvirales. Arch. Virol. 154: 171–177. 24. Flowers CC, O’Callaghan DJ. 1992. Equine herpesvirus 1 glycoprotein D: mapping of the transcript and a neutralization epitope. J. Virol. 66:6451– 6460. 25. Foster TP, Kousoulas KG. 1999. Genetic analysis of the role of herpes simplex virus type 1 glycoprotein K in infectious virus production and egress. J. Virol. 73:8457– 8468. 26. Fox G, et al. 1989. The cell attachment site on foot-and-mouth disease virus includes the amino acid sequence RGD (arginine-glycine-aspartic acid). J. Gen. Virol. 70(Pt 3):625– 637. 27. Frampton AR, Jr, et al. 2005. Equine herpesvirus 1 utilizes a novel herpesvirus entry receptor. J. Virol. 79:3169 –3173. 28. Frampton AR, Jr, et al. 2007. Equine herpesvirus 1 enters cells by two different pathways, and infection requires the activation of the cellular kinase ROCK1. J. Virol. 81:10879 –10889. 29. Galdiero M, et al. 1997. Site-directed and linker insertion mutagenesis of herpes simplex virus type 1 glycoprotein H. J. Virol. 71:2163–2170. 30. Geraghty RJ, Krummenacher C, Cohen GH, Eisenberg RJ, Spear PG. 1998. Entry of alphaherpesviruses mediated by poliovirus receptor-related protein 1 and poliovirus receptor. Science 280:1618 –1620. 31. Gianni T, et al. 2010. Herpes simplex virus glycoproteins H/L bind to cells independently of ␣V␤3 integrin and inhibit virus entry, and their constitutive expression restricts infection. J. Virol. 84:4013– 4025. 32. Graham KL, et al. 2005. Rotaviruses interact with alpha4beta7 and alpha4beta1 integrins by binding the same integrin domains as natural ligands. J. Gen. Virol. 86:3397–3408. 33. Hamidpour M, Behrendt M, Griffiths B, Partridge L, Lindsey N. 2006. The isolation and characterisation of antiplatelet antibodies. Eur. J. Haematol. 76:331–338. 34. Hasebe R, et al. 2009. Infectious entry of equine herpesvirus-1 into host cells through different endocytic pathways. Virology 393:198 –209. 35. Heldwein EE, Krummenacher C. 2008. Entry of herpesviruses into mammalian cells. Cell. Mol. Life Sci. 65:1653–1668. 36. Hynes RO. 2002. Integrins: bidirectional, allosteric signaling machines. Cell 110:673– 687. 37. Israel BA, Schultz KT, Murphy CJ. 1998. Lack of detectable interaction between HSV-1 and integrins or tachykinins. Intervirology 41:132–134. 38. Komoriya A, et al. 1991. The minimal essential sequence for a major cell type-specific adhesion site (CS1) within the alternatively spliced type III connecting segment domain of fibronectin is leucine-aspartic acid-valine. J. Biol. Chem. 266:15075–15079. 39. Krummenacher C, et al. 2004. Comparative usage of herpesvirus entry mediator A and nectin-1 by laboratory strains and clinical isolates of herpes simplex virus. Virology 322:286 –299. 40. Kurtz BM, Singletary LB, Kelly SD, Frampton AR, Jr. 2010. Equus caballus major histocompatibility complex class I is an entry receptor for equine herpesvirus type 1. J. Virol. 84:9027–9034. 41. Kydd JH, Smith KC, Hannant D, Livesay GJ, Mumford JA. 1994. Distribution of equid herpesvirus-1 (EHV-1) in respiratory tract associated lymphoid tissue: implications for cellular immunity. Equine Vet. J. 26:470 – 473. 42. Lazear E, et al. 2008. Engineered disulfide bonds in herpes simplex virus type 1 gD separate receptor binding from fusion initiation and viral entry. J. Virol. 82:700 –709. 43. MacEachern KE, Smith GL, Nolan AM. 1997. Methods for the isolation, culture and characterisation of equine pulmonary artery endothelial cells. Res. Vet. Sci. 62:147–152.

jvi.asm.org 2043


44. Maeda K, et al. 2007. Establishment of a novel equine cell line for isolation and propagation of equine herpesviruses. J. Vet. Med. Sci. 69:989 –991. 45. Manoj S, Jogger CR, Myscofski D, Yoon M, Spear PG. 2004. Mutations in herpes simplex virus glycoprotein D that prevent cell entry via nectins and alter cell tropism. Proc. Natl. Acad. Sci. U. S. A. 101:12414 –12421. 46. Matsumura T, Sugiura T, Imagawa H, Fukunaga Y, Kamada M. 1992. Epizootiological aspects of type 1 and type 4 equine herpesvirus infections among horse populations. J. Vet. Med. Sci. 54:207–211. 47. Montgomery RI, Warner MS, Lum BJ, Spear PG. 1996. Herpes simplex virus-1 entry into cells mediated by a novel member of the TNF/NGF receptor family. Cell 87:427– 436. 48. Neubauer A, Beer M, Brandmuller C, Kaaden OR, Osterrieder N. 1997. Equine herpesvirus 1 mutants devoid of glycoprotein B or M are apathogenic for mice but induce protection against challenge infection. Virology 239:36 – 45. 49. Neubauer A, Braun B, Brandmuller C, Kaaden OR, Osterrieder N. 1997. Analysis of the contributions of the equine herpesvirus 1 glycoprotein gB homolog to virus entry and direct cell-to-cell spread. Virology 227:281–294. 50. Neubauer A, Osterrieder N. 2004. Equine herpesvirus type 1 (EHV-1) glycoprotein K is required for efficient cell-to-cell spread and virus egress. Virology 329:18 –32. 51. Nugent J, et al. 2006. Analysis of equid herpesvirus 1 strain variation reveals a point mutation of the DNA polymerase strongly associated with neuropathogenic versus nonneuropathogenic disease outbreaks. J. Virol. 80:4047– 4060. 52. Osterrieder N. 1999. Construction and characterization of an equine herpesvirus 1 glycoprotein C negative mutant. Virus Res. 59:165–177. 53. Osterrieder N, Van de Walle GR. 2010. Pathogenic potential of equine alphaherpesviruses: the importance of the mononuclear cell compartment in disease outcome. Vet. Microbiol. 143:21–28. 54. Palfi V, Christensen LS. 1995. Analyses of restriction fragment patterns (RFPs) and pathogenicity in baby mice of equine herpesvirus 1 and 4 (EHV-1 and EHV-4) strains circulating in Danish horses. Vet. Microbiol. 47:199 –204. 55. Patel JR, Heldens J. 2005. Equine herpesviruses 1 (EHV-1) and 4 (EHV4)— epidemiology, disease and immunoprophylaxis: a brief review. Vet. J. 170:14 –23. 56. Pusterla N, et al. 2005. Equine herpesvirus-4 kinetics in peripheral blood leukocytes and nasopharyngeal secretions in foals using quantitative realtime TaqMan PCR. J. Vet. Diagn. Invest. 17:578 –581. 57. Roivainen M, et al. 1994. Entry of coxsackievirus A9 into host cells: specific interactions with alpha v beta 3 integrin, the vitronectin receptor. Virology 203:357–365. 58. Roizman B. 1996. Herpesviridae, p 2221–2230. In Field BN, Knipe DM, Howley PM, Channock RM, Melnick JL, Monath TP, Roizman B, Straus SE (ed), Virology, third ed. Lippincott-Raven, Philadelphia, PA. 59. Rudolph J, O’Callaghan DJ, Osterrieder N. 2002. Cloning of the genomes of equine herpesvirus type 1 (EHV-1) strains KyA and racL11 as bacterial artificial chromosomes (BAC). J. Vet. Med. B Infect. Dis. Vet. Public Health 49:31–36. 60. Ruoslahti E, Pierschbacher MD. 1987. New perspectives in cell adhesion: RGD and integrins. Science 238:491– 497. 61. Sasaki M, et al. 2011. Equine major histocompatibility complex class I molecules act as entry receptors that bind to equine herpesvirus-1 glycoprotein D. Genes Cells 16:343–357.

2044

jvi.asm.org

62. Shukla D, et al. 1999. A novel role for 3-O-sulfated heparan sulfate in herpes simplex virus 1 entry. Cell 99:13–22. 63. Smith KC, Mumford JA, Lakhani K. 1996. A comparison of equid herpesvirus-1 (EHV-1) vascular lesions in the early versus late pregnant equine uterus. J. Comp. Pathol. 114:231–247. 64. Spear PG. 2004. Herpes simplex virus: receptors and ligands for cell entry. Cell Microbiol. 6:401– 410. 65. Takada Y, Ye X, and Simon S. 2007. The integrins. Genome Biol. 8:215. 66. Telford EA, Watson MS, Perry J, Cullinane AA, Davison AJ. 1998. The DNA sequence of equine herpesvirus-4. J. Gen. Virol. 79:1197–1203. 67. Tischer BK, von Einem J, Kaufer B, Osterrieder N. 2006. Two-step red-mediated recombination for versatile high-efficiency markerless DNA manipulation in Escherichia coli. Biotechniques 40:191–197. 68. Van de Walle GR, Peters ST, VanderVen BC, O’Callaghan DJ, and Osterrieder N. 2008. Equine herpesvirus 1 entry via endocytosis is facilitated by alphaV integrins and an RSD motif in glycoprotein D. J. Virol. 82:11859 –11868. 69. Verheyen K, et al. 1998. Possible case of EHV-4 ataxia in warmblood mare. Vet. Rec. 143:456. 70. von Einem J, Smith PM, Van de Walle GR, O’Callaghan DJ, Osterrieder N. 2007. In vitro and in vivo characterization of equine herpesvirus type 1 (EHV-1) mutants devoid of the viral chemokine-binding glycoprotein G (gG). Virology 362:151–162. 71. Wang D, Shenk T. 2005. Human cytomegalovirus virion protein complex required for epithelial and endothelial cell tropism. Proc. Natl. Acad. Sci. U. S. A. 102:18153–18158. 72. Warner MS, et al. 1998. A cell surface protein with herpesvirus entry activity (HveB) confers susceptibility to infection by mutants of herpes simplex virus type 1, herpes simplex virus type 2, and pseudorabies virus. Virology 246:179 –189. 73. Wellington JE, Gooley AA, Love DN, Whalley JM. 1996. N-terminal sequence analysis of equine herpesvirus 1 glycoproteins D and B and evidence for internal cleavage of the gene 71 product. J. Gen. Virol. 77(Pt 1): 75– 82. 74. Wellington JE, Love DN, Whalley JM. 1996. Evidence for involvement of equine herpesvirus 1 glycoprotein B in cell-cell fusion. Arch. Virol. 141: 167–175. 75. Whalley JM, et al. 2007. Host cell tropism of equine herpesviruses: glycoprotein D of EHV-1 enables EHV-4 to infect a non-permissive cell line. Arch. Virol. 152:717–725. 76. Whalley M, et al. 1991. Identification and comparative sequence analysis of a gene in equine herpesvirus 1 with homology to the herpes simplex virus glycoprotein D gene. Virus Genes 5:313–325. 77. Wickham TJ, Mathias P, Cheresh DA, Nemerow GR. 1993. Integrins alpha v beta 3 and alpha v beta 5 promote adenovirus internalization but not virus attachment. Cell 73:309 –319. 78. Wilson WD. 1997. Equine herpesvirus 1 myeloencephalopathy. Vet. Clin. North Am. Equine Pract. 13:53–72. 79. Yoon M, Spear PG. 2004. Random mutagenesis of the gene encoding a viral ligand for multiple cell entry receptors to obtain viral mutants altered for receptor usage. Proc. Natl. Acad. Sci. U. S. A. 101:17252–17257. 80. Ziegler C, Just FT, Lischewski A, Elbers K, Neubauer A. 2005. A glycoprotein M-deleted equid herpesvirus 4 is severely impaired in virus egress and cell-to-cell spread. J. Gen. Virol. 86:11–21.

Journal of Virology