Molecular basis for resistance of herpes simplex virus type 1 mutants to the sulfated oligosaccharide inhibitor PI-88

Virology 367 (2007) 244 – 252 www.elsevier.com/locate/yviro

Molecular basis for resistance of herpes simplex virus type 1 mutants to the sulfated oligosaccharide inhibitor PI-88 Maria Ekblad a , Beata Adamiak a,c , Kicki Bergefall a , Hannah Nenonen a , Anette Roth a , Tomas Bergstrom a , Vito Ferro b , Edward Trybala a,⁎ a

Department of Clinical Virology, Göteborg University, Guldhedsgatan 10B, S-413 46 Göteborg, Sweden b Drug Design Group, Progen Industries Ltd, 2806 Ipswich Rd, Darra QLD 4076, Australia c Department of Molecular Virology, University of Gdansk, Kladki 24, 80-822 Gdansk, Poland Received 12 December 2006; returned to author for revision 2 May 2007; accepted 31 May 2007 Available online 2 July 2007

Abstract Herpes simplex virus type 1 variants selected by virus propagation in cultured cells in the presence of the sulfated oligosaccharide PI-88 were analyzed. Many of these variants were substantially resistant to the presence of PI-88 during their initial infection of cells and/or their cell-to-cell spread. Nucleotide sequence analysis revealed that the deletion of amino acids 33–116 of gC but not lack of gC expression provided the virus with selective advantage to infect cells in the presence of PI-88. Purified gC (Δ33–116) was more resistant to PI-88 than unaltered protein in its binding to cells. Alterations that partly contributed to the virus resistance to PI-88 in its cell-to-cell spread activity were amino acid substitutions Q27R in gD and R770W in gB. These results suggest that PI-88 targets several distinct viral glycoproteins during the course of initial virus infection and cell-to-cell spread. © 2007 Elsevier Inc. All rights reserved. Keywords: Herpes simplex virus type 1; PI-88 oligosaccharide; Drug resistance; Cell culture

Introduction The ability of sulfated polysaccharides and other polysulfonated compounds to inhibit infection of cells by herpes simplex virus (HSV) is a long known phenomenon (Vaheri, 1964). These compounds are structurally related to cell surface heparan sulfate (HS) receptor for attachment of HSV to cells (WuDunn and Spear, 1989) and seem to act by competing with HS chains for binding to the virus component(s) (Vaheri, 1964; WuDunn and Spear, 1989). Glycoproteins gC, gB and gD of HSV type 1 (HSV-1) envelope have been reported to interact with heparin/ HS chains (Herold et al., 1991; Tal-Singer et al., 1995; Shukla et al., 1999; Trybala et al., 2000) and these components therefore represent the likely targets for sulfated polysaccharide inhibitors. Indeed, sulfated polysaccharides were shown to inhibit HSV attachment to cells (Vaheri, 1964; WuDunn and Spear, 1989), a step in the HSV–cell interaction mediated by ⁎ Corresponding author. Fax: +46 31 827032. E-mail address: [email protected] (E. Trybala). 0042-6822/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.virol.2007.05.040

glycoproteins gC and/or gB (Herold et al., 1991, 1994). It is unclear whether or not sulfated polysaccharides could interfere with virus entry into cells. Although this step can be initiated by interaction of HSV-1 gD with specific 3-O-sulfated chains of HS (Shukla et al., 1999), this glycoprotein can also interact with TNF-receptor like protein (Montgomery et al., 1996) or nectin (Geraghty et al., 1998) to trigger virus entry into cells. The antiviral potential of sulfated polysaccharides usually increases with increasing sulfation and chain length, however, low molecular weight and extensively sulfated oligosaccharides also exhibit antiviral properties (Witvrouw and De Clercq, 1997). PI-88 is a low molecular weight sulfated oligosaccharide in clinical development as an anticancer drug that also shows anti-HSV (Nyberg et al., 2004), anti-dengue and -encephalitic flavivirus (Lee et al., 2006), and anti-malarial (Adams et al., 2006) activities. Structurally, PI-88 is a mixture of highly sulfated mannose di- to hexasaccharides. Thus, compared with heparin PI-88 is composed of shorter but more extensively sulfated oligosaccharides. In our previous work (Nyberg et al., 2004) we have found that PI-88 reduced the cell-to-cell spread

M. Ekblad et al. / Virology 367 (2007) 244–252

245

of HSV more efficiently than heparin. Importantly, the presence of PI-88 resistant variants in the gC-negative strains of HSV-1 suggested that viral components other than gC could be targeted by PI-88 during the course of viral infection of cells. The development of HSV-1 variants resistant to the sulfated polysaccharides carrageenan (Carlucci et al., 2002) or heparin (Goodman and Engel, 1991; Pertel and Spear, 1996) has been reported, however, except for gK (Pertel and Spear, 1996), nucleotide sequences of genes coding for other relevant viral components were not analyzed. In this work, we attempted to elucidate the molecular basis for the resistance of HSV-1 variants to PI-88. Our study demonstrated that the deletion of amino acids 33–116 of gC but not lack of gC expression provided the virus the ability to infect the cell in the presence of PI-88 while specific alterations in gB or gD partly promoted the PI-88 resistant cell-to-cell spread of HSV-1. Results and discussion PI-88 escape variants of HSV-1 In our previous work we found that, compared with heparin, PI-88 inhibited the initial infection of cultured cells with HSV less efficiently but reduced the cell-to-cell spread of the virus more profoundly (Nyberg et al., 2004). Here we repeated this experiment with different strains of HSV. The effect of PI-88 on the cell-to-cell spread activity of HSV, measured as a reduction in the size of HSV plaques developed in GMK AH1 cells in the presence of PI-88 in an overlay medium, is shown in Table 1. PI-88 reduced the size of plaques developed by all HSV-1 and HSV-2 strains tested. During this investigation we noticed the presence of relatively large PI-88 resistant plaques that occurred in the HSV-1 gC-null/negative strains (Fig. 1). This observation suggesting that alteration in viral component(s) other than gC can be responsible for PI-88 resistant cell-to-cell spread of HSV1 encouraged us to prepare PI-88 escape variants of the virus and to investigate molecular basis for their drug resistance. The gC-proficient HSV-1 KOS strain (not KOS321) was chosen for our attempts to isolate PI-88 resistant variants of the virus. To this end, the plaque purified virus (KOSc) was subjected to ten passages in GMK AH1 cells in the presence of PI-88. The KOSc strain, passaged under similar conditions in the Table 1 Effect of PI-88 oligosaccharides on plaque size of HSV strains in GMK AH1 cells Virus

Average plaque area (mm2, n = 20) PI-88 (100 μg/ml)

Control

HSV-1 17syn+ HSV-1 2762 HSV-1 KOS HSV-1 KOS321 HSV-1 gC−39 (gC-null) HSV-1 MP (gC-negative) HSV-2 333 HSV-2 gCneg1 (gC-negative)

0.058 ± 0.025 0.041 ± 0.018 0.033 ± 0.029 0.043 ± 0.018 0.041 ± 0.043 0.059 ± 0.029 0.028 ± 0.011 0.047 ± 0.018

0.188 ± 0.074 0.154 ± 0.041 0.115 ± 0.077 0.209 ± 0.071 0.162 ± 0.078 0.494 ± 0.157 0.318 ± 0.211 0.478 ± 0.136

a

Statistically significant differences at P values of b0.001.

% of control 30.5 a 26.3 a 28.7 a 20.6 a 25.4 a 12.0 a 8.9 a 9.8 a

Fig. 1. The PI-88 resistant plaque variants of HSV-1. (A) PI-88 resistant (large) and sensitive (small) plaque variants of HSV-1 gC−39 strain. (B) PI-88 resistant (large and syncytial) and sensitive (small and syncytial) plaque variants of HSV1 MP strain.

absence of PI-88 (KOSc passage 10), served as control material. The virus was exposed to PI-88 throughout its propagation in cells, i.e., during the virus attachment to and entry into the cells and its subsequent spread via released and cell-to-cell transmitted progeny virus. The viral plaque variants (AC variants) that survived the selective pressure from PI-88 were immunostained with monoclonal anti-gC-1 antibody which revealed that 46.6% of them had the gC-negative phenotype. Approximately 1.5% of AC variants formed syncytial plaques of which approximately 50% were gC-negative. To verify the PI-88 resistant phenotype of AC variants as well as to identify possible genotypic alterations, six AC variants were plaque purified and then tested for sensitivity to PI-88 (Table 2). The gC-positive variants (AC1 through AC4) but not the gC-negative (AC5, AC6) ones were 6–12 times more resistant than parental KOSc strain to PI-88 in the plaque number reduction assay, a procedure in which the virus was exposed to PI-88 during initial infection of cells only. To verify the possibility that AC1–AC4 mutants may bind to cells in the presence of PI-88 and then infect them when PI-88 is removed, these variants were retested in the plaque number reduction assay that included the step of washing the cells with the low pH buffer. The results of this assay and the original plaque number

246

M. Ekblad et al. / Virology 367 (2007) 244–252

Table 2 Features of PI-88 escape variants a of HSV-1 Virus strain or variant

Phenotype gC

Syncytium

Plaque number reduction assay b IC50 (μg/ml)

PI-88 sensitivity Plaque number reduction assay, pH 3 c IC50 (μg/ml)

Plaque size reduction assay d % of control area

KOSc KOSc (passage 10) AC1 AC2 AC3 AC4 AC5 AC6 A1 A2 A3 A4 A5 C1 C2 C3 gC−39-S gC−39 R6 MP-S MP R15 gC−39-S+AC3gC e KOSc+AC3gD e KOSc+A1gK e gC−39-S+R6gD e MP-S+R15gB e

+ + + + + + − − − + + − + + + + − − − − + + + − −

− − − − + − +/− − + + + − − − +/− − − − + + − − + − +

8.3 6.0 48 58 N100 47 4.0 4.6 2.8 3.6 5.1 3.0 8.0 7.0 11.0 8.5 1.2 2.4 16 15 30 9 NT 1.6 7

9.1 NT 60 70 N100 42 4.7 5.1 5.1 1.8 5.3 4.2 2.2 2.5 7.0 2.5 NT NT NT NT NT NT NT NT NT

27.5 (0.205) 19.7 (0.094) 53.5 (0.255) 67.7 (0.304) 70.6 (1.750) 50.6 (0.299) 50.0 (0.274) 44.7 (0.250) 14.2 (0.278) 15.0 (0.815) 10.7 (0.338) 20.7 (0.101) 24.9 (0.238) 53.4 (0.306) 58.7 (0.262) 42.9 (0.262) 16.5 (0.068) 57.6 (0.197) 14.0 (0.633) 59.6 (1.801) 20.7 (0.039) 38.1 (0.168) NT 62.5 (0.118) 55.2 (0.980)

a Variants selected for by the PI-88 presence limited to either initial infection of cells (A1–A5) or spread of progeny virus (C1–C3; gC−39 R6; MP R15) or both (AC1–AC6). b Serial PI-88 dilutions were incubated with ∼ 200 PFU of the virus for 10 min prior to and during a 2 h period of infection of GMK AH1 cells at 37 °C. Results are expressed as a concentration of PI-88 that reduced the number of viral plaques by 50%. c Assay performed as per plaque number reduction assay except that the cells were washed with citrate-buffered saline, pH 3.0, at the end of a 2 h period of virus infection of cells. d PI-88 (100 μg/ml) in an overlay methylcellulose medium was added to cells after their infection with the virus and incubated with cells throughout the development of viral plaques. Results are expressed as a percentage of the average area of 20 viral plaques developed in the presence of PI-88 medium relative to mock-treated controls. Shown in parentheses is the average area (in mm2) of 20 viral plaques developed in the absence (control) of PI-88. e Viruses prepared by the molecular transfer of specific gene fragments into DNA of parental strains KOSc, gC−39-S or MP-S.

reduction assay were similar (Table 2) indicating that the AC1– AC4 variants can efficiently infect the cells in the presence of PI-88. In the plaque size reduction assay, both gC-positive and gC-negative AC variants were more resistant to PI-88 than KOSc (Table 2). We sought to identify the alteration(s) in viral variants responsible for PI-88 resistant phenotype based on the reasoning that the viral component(s) involved may provide, in their wild-type form, the potential binding sites for HS and for the HS-mimetic compounds. Nucleotide sequence analysis of genes coding for the known HS-binding proteins gB, gC, gD and the syncytium-modulating protein gK is shown in Table 3. Variants AC1–AC4, which exhibited substantial resistance to PI-88 both in their initial infection of cells and the cell-to-cell spread, all carried the large deletion of amino acids 33–116 of gC (Table 3). Variant AC3, the most resistant to PI-88 (Table 2), had also the Q27P amino acid substitution in gD, and the syncytiuminducing R828H alteration in the endodomain of gB (Table 3). The latter mutation also occurs in HSV-1 HFEMtsB5 (Bzik et al., 1984) and HSZP (Kosovsky et al., 2000) strains. The second most PI-88 resistant variant AC2 had, in addition to gC change,

two mutations in the gB ectodomain. Variants AC5 and AC6 were gC-negative due to the frameshift in the gC gene caused by the deletion of a single nucleotide, and the latter variant also contained the Q27P substitution in gD (Table 3). Both these variants were partly resistant to PI-88 in the plaque size but not the plaque number reduction assay (Table 2). The lack of gC expression is unlikely to contribute to this phenotype since the gC−39 strain (gC-null) was sensitive to PI-88 (Tables 2 and 3). Hence, mutation found in gD (AC6) or in an unsequenced gene (AC5) may account for the PI-88 resistant cell-to-cell spread of these variants (see the next chapter of this section). HSV-1 variants generated under similar experimental conditions as AC variants based on their resistance to other sulfated oligo/polysaccharides, i.e., heparin (Goodman and Engel, 1991) or carrageenan (Carlucci et al., 2002), have been described. Heparin resistant variants of HSV-1 were selected for by five virus passages in rabbit skin cells in the drug presence. Approximately 5% of selected virus formed syncytial plaques and this phenotype was due to the A825V substitution in gB endodomain (Goodman and Engel, 1991; Engel et al., 1993). Likewise, HSV-1 generated by sixteen passages in the presence

M. Ekblad et al. / Virology 367 (2007) 244–252 Table 3 Specific alterations detected in the PI-88 escape variants of HSV-1 Virus variant a

Phenotype

Alteration b

gC

Syncytium

gB

gC

gD

gK

AC1

+

−

–

–

–

AC2

+

−

–

–

AC3

+

+

+

−

AC5

−

+/−

–

a155c Q27P c89t A5V –

AC6

−

−

–

A1

−

+

–

A2

+

+

–

A3

+

+

A4

−

−

A5 C1

+ +

− −

c2289t Silent c2289t Silent – –

Δg96-g347 ΔP33-G116 Δg96-g347 ΔP33-G116 Δc264 Frameshift Δc264 Frameshift Δc366 Frameshift g452a R151H –

–

AC4

c158a T23N g820a E244K g2573a R828H –

Δg96-g347 ΔP33-G116 Δg96-g347 ΔP33-G116

C2

+

+/−

–

C3

+

−

–

gC−39 R6

−

−

MP R15

−

+

t348c Silent a1510g T474A c2386t R770W

Δc366 Frameshift – g452a R151H – g752a S251N NT

NT

a155c Q27P – – – – – – c938t P288L – a155g Q27R t529c Silent –

– – – c973a L325I c973a L325I c973a L325I c67t L23F – – g853a A285T c492t Silent –

–

a

Variants selected for by PI-88 presence limited to either initial infection of cells (A1–A5) or spread of progeny virus (C1–C3; gC−39 R6; MP R15) or both (AC1–AC6). b Nucleotide and predicted amino acid alterations are denoted with lowercase and uppercase letters respectively. Nucleotide numbering begins from the first nucleotide of the initiation codon. Amino acid numbering in gC and gK begins from the initiation methionine while that in gD and gB from the first amino acid of the mature protein. The GenBank accession numbers are: EF157309 for AC2 gB, EF157310 for AC3 gB, EF157311 for A3 gB, EF177452 for gC−39-S gB, EF177453 for gC−39 R6 gB, EF177454 for MP-S gB, EF177455 for MP R15 gB, EF157313 for AC1-4 gC, EF157314 for AC5,6 gC, EF157315 for A1 gC, EF157317 for A2,C1 gC, EF157318 for C3 gC, EF157320 for AC3,6 gD, EF157321 for AC4 gD, EF157322 for C2 gD, EF177450 for gC−39-S gD, EF177451 for gC−39 R6 gD, EF157324 for A1–3 gK, EF157325 for C2 gK and EF157326 for C3 gK. All other sequences of AC, A and C variants were identical to those of parental strain KOSc gB (EF157316), KOSc gC (EF157312), KOSc gD (EF157319) and KOSc gK (EF157323).

of carrageenan 1C3 comprised 100% syncytial plaques, and a likely alteration(s) in gB was suggested as responsible for syncytial phenotype. However, possible adaptative alterations in other viral components of these heparin or carrageenan resistant variants of HSV-1 were not investigated. Knowing that some of the AC variants exhibited resistant phenotypes both in their initial infection of cells and the cell-to-

247

cell spread, we asked whether limitation of selective pressure from PI-88 to one of these steps would select for the distinct HSV-1 variants. To this end, the plaque purified virus (KOSc) was subjected to ten passages in GMK AH1 cells, and the presence of PI-88 was limited to the step of initial virus infection of cells (variants A). Analysis of uncloned virus revealed that 61.2% of plaque variants A had the gC-negative phenotype, and 2.3% formed syncytial plaques in GMK AH1 cells. Approximately 0.7% of plaque variants had both gC-negative phenotype and syncytial plaque morphology. Several of these variants were plaque purified and then tested for their sensitivity to PI-88. None of variants A (Table 2) including gC-negative ones (A1, A4) were resistant to PI-88 in both the plaque number and the plaque size reduction assays as their sensitivities to this compound were comparable to or even somewhat greater than that of the parental KOSc or KOSc (passage 10) strain. A possibility that putative PI-88 resistance was lost during plaque purification of variants A is unlikely because the drug resistant AC variants were plaque purified in an identical manner. Moreover, PI-88 at a concentration used in selection experiments (100 μg/ml) decreased the number of plaques developed by parental KOSc and all variants A by approximately 94–98% (data not shown), thus further confirming their PI-88 sensitive phenotype. Genotypic alterations in variants A were mainly detected in gC and gK genes. Of these, variants A1 and A4 were gC-negative due to the frameshift in the gC gene while A2 exhibited a single R151H amino acid substitution in this protein. The syncytium-inducing activity of variants A1–3 was due to the L325I alteration in gK (Table 3). This is a novel syncytiuminducing mutation in gK, confirmed by the marker transfer experiment (Table 2, KOSc+A1gK variant). In the model of HSV-1 gK membrane topology (Foster et al., 2003) this mutation occurs within a transmembrane region that separates domains III and IVof this protein. Note that syncytium-inducing mutants, which as mentioned above account for only 2.3% of all variants A, are overrepresented in Table 3. Therefore the incidence of syncytium-inducing mutations in gK of variants A may in fact be much lower than that suggested by our data. Variant A5 lacking alterations in analyzed genes (Table 3) may have survived selective pressure from PI-88 accidentally or it may contain mutation in an unsequenced gene. These results suggest that when the selective pressure from PI-88 is limited to initial virus infection of cells, a high proportion (61.2%; see above) of the gC-negative variants is, among other mutants, selected for. However, this phenotype is not sufficient to render the virus resistance to PI-88. Pertel and Spear (1996) analyzed variants of HSV-1 generated by virus passage in HEp-2 cells with selective pressure from the sulfated polysaccharide heparin limited to initial virus infection of cells, i.e., under experimental conditions similar to those used for the PI-88-forced selection of variants A. Although the proportions of the gC-negative variants were comparable when heparin (Pertel and Spear, 1996) or PI-88 (this report) was used as a selective drug, the vast majority of heparin variants formed syncytia in cultured cells (Pertel and Spear, 1996) while only a minor fraction of PI-88 selected variants had this phenotype. It is unclear whether this

248

M. Ekblad et al. / Virology 367 (2007) 244–252

difference was due to the fact that we used plaque purified HSV1 KOS strain in the selection experiments. More importantly, analysis of heparin resistant virus variants revealed that the absence of gC was sufficient to confer HSV-1 resistance to heparin and that syncytium-inducing mutations in gK augmented these effects by enhancing the lateral spread of HSV-1 in cultured cells (Pertel and Spear, 1996). As mentioned above, we found that although more than 60% of plaque variants had the gC-negative phenotype and one of the mutants carried an alteration in the putative attachment domain of gC (Mardberg et al., 2001), these variants did not appear to be resistant to PI-88. While we have no explanation for this difference, we noticed that PI-88, in contrast to heparin (Pertel and Spear, 1996), inhibited infection of GMK AH1 cells by the gC-negative variants and strain gC−39 more effectively than that of the gCproficient parental strain (Table 2). This suggests that, apart from gC, PI-88 can efficiently target another viral component involved in the initial HSV–cell interaction. Consequently none of variants A appeared to be resistant to PI-88, and possibly a greater number of virus passages or continuous presence of PI88 may be required to select for the drug-resistant variants of HSV-1. We also analyzed variants of HSV-1 KOSc strain generated by virus passage in GMK AH1 cells with the presence of PI-88 limited to the spread of viral infection via released extracellular and cell-to-cell transmitted virus (Table 2, variants C). None of variants C had the gC-negative phenotype, and approximately 0.7% formed syncytia in cultured cells. Variants C1, C2 and C3 exhibited similar sensitivity to PI-88 as the parental KOSc strain in the plaque number reduction assay (Table 2). However, these mutants showed partial resistance to PI-88 in the plaque size reduction assay. Although C1 and C3 variants carried specific amino acid substitutions in gC, none of C variants (Table 3) including those present in the original mixture of the drugselected mutants (see above) appeared to be gC-negative. This suggests that gC is less extensively targeted by PI-88 during the cell-to-cell spread of HSV-1 than the initial infection of cells with free virus. Variant C2 had the P288L substitution in a membraneproximal proline-rich stalk of gD. In the crystal structure of native gD, this region of the protein was found to anchor near the N-terminally located domain that comprises residues L25–Q27 (Krummenacher et al., 2005). It has been proposed that, upon receptor binding, the C-terminal region is released from this “closed” conformation of gD, a change that triggers HSV-1 entry into the cell (Fusco et al., 2005; Krummenacher et al., 2005). Note that mutations detected in variants C occurred in different viral components, and our attempts to verify their importance for the observed phenotype in the marker transfer experiments were unsuccessful (Table 2). Therefore, one cannot exclude that mutation in an unsequenced gene may confer the virus resistance to PI-88 in its cell-to-cell spread. Included in Table 2 are also two plaque variants of HSV-1 gC-negative strains (Fig. 1) that, like variants C, were generated by selective pressure from PI-88 limited to the cell-to-cell spread of the virus, where a single virus passage in cells was sufficient. These variants occurred at frequencies of 9.3% for non-syncytial gC−39 and 2.2% for the syncytium-forming MP strain. The PI-88 resistant variants

gC−39 R6 and MP R15 as well as PI-88 sensitive parental viruses gC−39-S and MP-S were plaque purified to homogeneity and then retested in the plaque size and plaque number reduction assays for their sensitivity to PI-88. As expected, the presence of PI-88 (100 μg/ml) in an overlay medium reduced the plaque area of parental strains gC−39-S or MP-S by N 80% while those of gC−39 R6 or MP R15 by ∼40% (Table 2). Thus, these plaque variants of HSV-1 appeared to be substantially, but not completely, resistant to PI-88. Note also that in the absence of PI-88 these variants produced larger plaques than their respective parental strains (Table 2). Interestingly, in contrast to the results of the plaque size reduction assay, the PI-88selected variants showed either comparable or ∼2-fold higher resistance to PI-88 than parental strains in the plaque number reduction assay (Table 2). These results suggest that the relative PI-88 resistance of variants C as well as plaque variants gC−39 R6 and MP R15 was mainly due to decreased drug sensitivity of their cell-to-cell spread capability rather than the initial infection of cells by these viruses. The gC−39 R6 carried two mutations in gB one of which predicted the T474A amino acid substitution (Table 3). Because alanine at position 474 is also present in gB of the PI-88 sensitive strains KOS (EF157316) and KOS321 (Robbins et al., 1987), this mutation is unlikely to confer the drug resistance of gC−39 R6. This viral variant was also altered in the gD component displaying Q27R amino acid substitution. Since the parental gC−39 strain contains the L25P amino acid substitution as compared to the wild-type gD sequence of KOS strain (Watson et al., 1982), this mutation in gD also occurs in gC−39 R6 variant. Thus, the gC−39 R6 carries the same mutations in the amino-terminal part of gD as HSV-1 ANG strain (Lingen et al., 1995). Because ANG strain was isolated by using a protocol similar to that described for gC− 39 R6 (the presence of natural sulfated oligo/polysaccharides in an agar overlay) (Munk and Ludwig, 1972), it is likely that isolation of viral variants altered at specific viral genes can, to some extent, be related to the presence of sulfated inhibitor during specific stage of viral infection of cells. The MP R15 plaque variant contained the R770W amino acid substitution in the membrane-proximal region of the gB endodomain (Table 3), a change previously identified as a syncytium-inducing mutation (Gage et al., 1993). This variant produced larger syncytial plaques than the parental MP strain. However, it is noteworthy that the size of plaques developed by parental MP strain, which forms syncytia by itself due to alteration in gK (Pogue-Geile and Spear, 1987), was greatly reduced by PI-88 (Table 2) suggesting that specific alteration(s) in gB but not in gK rendered the virus spread in cultured cells resistant to PI-88. Contribution of specific alterations to the PI-88 resistant phenotypes of HSV-1 To investigate importance of specific mutations for the PI-88 resistant phenotype of HSV-1, the DNA fragments comprising part or entire sequence of gB, gC, gD or gK gene were mixed with DNA purified from parental strain KOSc, gC−39-S or MPS and co-transfected into GMK AH1 cells. A list of alterations

M. Ekblad et al. / Virology 367 (2007) 244–252

that were successfully transferred into DNA of parental virus is shown in Table 2. Glycoprotein C gene of PI-88 resistant variants AC1–AC4 that contains a large deletion of 252 nucleotides was successfully transferred into the background of gC-null strain gC−39-S. Analysis of the resulting mutant gC− 39-S+AC3gC (Table 2) revealed that this alteration in gC conferred the virus resistance to the presence of PI-88 during initial infection of cells but not during the cell-to-cell spread of HSV-1 (Table 2). The deleted segment of gC gene codes for amino acids 33 through 116 at the amino terminal part of gC protein. This region has been reported to promote HSV-1 attachment to cells (Tal-Singer et al., 1995). Due to the presence of numerous O-and N-glycosylation sites, this fragment of gC forms a mucin-like structure, and it would be of interest to determine how PI-88 could interact with such moiety. It is also noteworthy that variants with this specific deletion in gC but not the gC-negative ones appeared to be resistant to PI-88 (Table 2). To clarify this issue, purified radiolabeled virions of AC1 (Δ33–116), AC5 (gC-negative) or KOSc were tested for their ability to attach to cells in the presence of PI-88 (Fig. 2A). AC1 variant was less efficiently than AC5 or KOSc inhibited by PI88. Note that similar tendencies were observed when these viruses were examined in the plaque number reduction assay (Table 2). Likewise, truncated gC purified from variants AC2 or AC3 was less efficiently than gC of KOSc strain inhibited by PI-88 in the cell-binding assay (Fig. 2B). One explanation of these data could be that absence of entire gC may activate/ expose a redundant attachment domain or PI-88 sensitive site in another viral protein while the presence of truncated gC may block this domain or inefficiently compete with it for binding to PI-88. Generation of additional PI-88 resistant variants of HSV1 gC-negative viruses may help to identify this putative domain. In contrast to the AC1–4 variants where alteration in sole viral component (gC) conferred resistance to PI-88 in their initial infection of cells, many mutants resistant to PI-88 in the cell-tocell spread activity (AC1–6; C1–3) contained occasional changes in gB, gC, gD and/or gK. This suggests that specific alterations in different viral components can, to some extent, contribute to the virus resistance in the cell-to-cell spread or that a mutation in an unsequenced gene may confer resistance. The Q27R alteration, identified in gD of variant gC−39 R6, and successfully transferred into gC−39-S+R6gD mutant, rendered its cell-to-cell spread but not initial infection of cells relatively resistant to PI-88 (Table 2). However similar alteration (Q27P) found in AC3 and AC6 variants and transferred into the gCproficient KOSc strain caused only partial resistance to PI-88 in the cell-to-cell spread activity (Table 2) suggesting that presence of gC may influence functions of gD. It was reported that the heparin resistant variants of HSV-1, which as mentioned above were altered in gC and gK, exhibited partial resistance to gDmediated interference (Pertel and Spear, 1997). HSV-1 gD can interact among other cellular receptors also with nectin, a component of adherens junctions. It was reported that monoclonal anti-nectin-1 antibody reduced cell-to-cell spread of HSV1 (Cocchi et al., 2000) further confirming essential role of gD in this process. The Q27R, Q27P or L25P amino acid substitutions in gD are known to confer both resistance of superinfecting

249

Fig. 2. The effect of PI-88 on binding to cells of purified HSV-1 virions or HSV1 glycoproteins. PI-88 at different concentrations was incubated with purified 3 H-thymidine labeled virions of HSV-1 KOSc, AC1 or AC5 variant (A) or isolated glycoproteins gC (B) or gB (C) of KOSc, AC2 or AC3 variant for 15 min prior to and during 1–2 h period of virus or glycoprotein adsorption to GMK AH1 cells. The results are shown as a percentage of attached viral cpm, or absorbance of cell-bound viral glycoprotein detected with PI-88 treated samples relative to mock-treated controls. Values shown are means of four determinations from two separate experiments.

HSV-1 to gD-mediated interference (Campadelli-Fiume et al., 1990; Dean et al., 1994) and HSV-1 entry into the cells via the nectin-2 receptor (Warner et al., 1998; Lopez et al., 2000; Yoon et al., 2003). Interestingly ANG strain, which contains L25P and Q27R alterations in gD, formed larger plaques in cells expressing nectin-2 than in cells producing nectin-1 (Delboy et al., 2006). This suggests that viruses with these alterations in gD may exhibit enhanced entry and cell-to-cell spread activities thus escaping selective pressure from PI-88 inhibitor. The L25P and

250

M. Ekblad et al. / Virology 367 (2007) 244–252

Q27P/R substitutions are also known to abrogate functional binding of gD to 3-O-sulfated forms of HS (Yoon et al., 2003) suggesting that PI-88 can interfere with these interactions. However, it is uncertain whether resistance of HSV-1 cell-to-cell spread activity to PI-88 was due to usage of nectin-2 receptor or abrogated virus interaction with 3-O-sulfated HS. The reason for this is that the L25P substitution that alters receptor usage by HSV-1 gD is present in the parental, PI-88 sensitive gC−39-S strain (Table 2). Although this mutation had a less profound effect on usage of gD receptors than did the Q27P/R substitution (Yoon et al., 2003), its presence in the PI-88 sensitive virus suggests that other explanations of the drug resistance phenomenon cannot be excluded. We managed to transfer the R770W alteration, identified in the gB component of MP R15 plaque variant, into parental strain MP-S (Table 2; MP-S+R15gB variant). This specific alteration in gB was required to escape the selective pressure from PI-88 during the cell-to-cell spread of HSV-1 (Table 2). We also purified gB components from two PI-88 resistant variants (AC2 and AC3) and investigated their binding to cells in the presence of PI-88 (Fig. 2C). Although gB components of both AC2 and AC3 were somewhat less efficiently than KOSc gB inhibited by PI-88, it is uncertain whether this little difference was due to decreased affinity of mutated gB species for PI-88. Mutations in the gB endodomain may alter expression of this protein at the cell surface (Ruel et al., 2006) or induce distant conformational change(s) in the gB ectodomain thus modulating its interaction with cells. The recently reported crystal structure of HSV-1 gB (Heldwein et al., 2006) reveals that the E244K alteration (AC2) occurs in domain I of this protein that structurally resembles the PH domain found in some phospholipid-binding cytoplasmic proteins. It is also noteworthy that PI-88 selected for both syncytium-inducing and non-syncytial variants of HSV-1 suggesting that the cell–cell fusion activity of the virus is not a required feature to escape the antiviral effects of PI-88. In conclusion, through analysis of PI-88 escape variants we have found that this oligosaccharide inhibitor targeted two distinct events which occur during initial HSV-1 infection of cells and its cell-to-cell spread. Deletion of the mucin-like region of viral gC conferred HSV-1 resistance to PI-88 in its initial infection of cells while specific alteration in gB and gD partly contributed to the PI-88 resistant cell-to-cell spread of the virus. Given the fact that we have exposed the virus to strong PI88 selective pressure to generate these variants, and that the gCnegative or syncytium-forming HSV-1 is only rarely isolated from patients, PI-88, like other sulfated polysaccharides (Witvrouw and De Clercq, 1997), could be regarded as a slow and inefficient inducer of viral resistance. Materials and methods Cells, viruses and PI-88 oligosaccharide African green monkey kidney (GMK AH1) cells (Gunalp, 1965) were cultivated in Eagle's minimum essential medium (EMEM) supplemented with 2% fetal calf serum, 0.05% Primaton RL substance (Kraft Inc., Norwich, CT, USA),

100 U/ml of penicillin and 100 μg/ml of streptomycin. The virus strains and mutants used were HSV-1 KOS (ATCC, VR1493), HSV-1 KOS321, a plaque purified isolate of wild-type strain KOS (Holland et al., 1983), HSV-1 KOS gC-null strain gC−39 (Holland et al., 1984), HSV-1 17syn+ (Brown et al., 1973), HSV-1 2762 (Bergstrom and Lycke, 1990), HSV-1 gCnegative and syncytium-inducing strain MP (Hoggan and Roizman, 1959), HSV-2 strain 333 (Duff and Rapp, 1971) and its gC-negative derivative designated HSV-2 gCneg1 (Trybala et al., 2000). PI-88 was prepared as described previously (Yu et al., 2002). Preparation of PI-88 escape variants of HSV-1 HSV-1 strain KOS, which had been subjected to three rounds of plaque purification, was used. The virus has been serially (ten times) passaged in GMK AH1 cells in the presence of 100 μg/ml of PI-88 according to three different protocols. (i) The virus was incubated with PI-88 for 15 min prior to and during a 2 h period of infection of GMK AH1 cells at 37 °C. The cells were then washed with EMEM and incubated in fresh EMEM without PI88 until complete cytopathic effect (CPE) was observed. (ii) The virus infection of cells took place at 37 °C for 2 h in the absence of PI-88. The cells were then washed with EMEM and incubated in EMEM supplemented with PI-88 until the development of complete CPE. (iii) The virus was incubated with PI-88 for 15 min prior to and during a 2 h period of infection of GMK AH1 cells at 37 °C. The cells were then washed with EMEM and incubated in EMEM supplemented with PI-88 until the development of CPE. After each passage the infectious culture medium and infected cells were harvested and subjected to one freeze–thaw cycle. Following centrifugation for 10 min at 1000×g, the supernatant medium was diluted in EMEM and used for a subsequent passage. After the final passage, several plaque variants from each group were subjected to three rounds of plaque purification in the absence of PI-88. Viral plaque assays The plaque number reduction assay was carried out as described previously (Ekblad et al., 2006; Nyberg et al., 2004). For the plaque size reduction assay, monolayer cultures of densely growing GMK AH1 cells were used. After a 2 h period of virus infection of cells at 37 °C, the serum-free overlay medium, composed of 1% methylcellulose solution in EMEM and PI-88 (100 μg/ml), was added and incubated with cells throughout the entire period of the development of viral plaques. The images of 20 neighboring plaques were captured using a DC300 digital camera (Leica, Heerbrugg, Switzerland) attached to a Diavert microscope (Leitz-Wetzlar, Germany). Plaque area was measured using IM500 image software (Leica, Cambridge, UK). Nucleotide sequencing The preparation of HSV DNA and nucleotide sequencing of the coding regions of gB, gC, gD and gK gene were performed

M. Ekblad et al. / Virology 367 (2007) 244–252

as described previously for the gG gene of HSV-2 (Liljeqvist et al., 1999). The primers to be used in amplification and sequencing reactions were designed based on the published sequence of these genes in strain 17 (NC001806). The positions of sense primers for HSV-1 gB gene were− 71–(− 52), 406–423, 773–790, 1116–1134, 1493–1512, 1731–1750, 2167–2184 and 2513–2530, and antisense primers 472–453, 870–849, 1186–1165, 1587–1570, 1899–1882, 2273–2256, 2587–2568 and 2854–2837. The positions of sense primers for HSV-1 gD gene were − 63–(− 42), 330–350, 622–639 and 901–919, and antisense primers 430–408, 727–708, 1020–1002 and 1335– 1317. The positions of sense primers for HSV-1 gC gene were − 152–(− 132), 12–33, 318–338, 591–613, 813–833 and 1122–1144, and antisense primers 135–117, 392–373, 702– 681, 991–968, 1305–1283 and 1659–1639. The positions of sense primers for HSV-1 gK were − 95–(− 78), 307–327 and 600–618, and antisense primers 395–375, 706–688 and 1057– 1039. All sequences were related to the original gene sequence of cloned HSV-1 KOS strain (KOSc) from which the viral variants were selected. In comparison with the published sequences of HSV-1 KOS gB (AF311740), gC (J02216), gD (J02217) and gK (U14422), KOSc possesses three nucleotide changes in gB (cg938/9gc, a1544g), five in gC (t9g, g40a, a899t, t1056c, c1070t), one in gD (a10g) and no alteration in gK gene. Marker transfer assay HSV-1 DNA was purified as described by Mettenleiter et al. (1988). Specific sequences of viral DNA were amplified by PCR using Pfu Turbo polymerase (Stratagene) and flanking primers 1116–1134 and 2854–2837 for gB, − 222–(− 202) and 1880–1900 for gC, − 63–(− 43) and 1335–1317 for gD and − 95–(− 78) and 1057–1039 for gK sequences. One microgram of quantities of purified viral DNA and amplified viral sequence in serum- and antibiotic-free EMEM was mixed with lipofectamine 2000 reagent and incubated for 20 min at room temperature. The mixture was added to a one-day-old culture of GMK AH1 cells that reached ∼ 90% confluence at the day of transfection. After incubation of cells for 5 h at 37 °C, the medium was removed and fresh EMEM was added. In some experiments EMEM was supplemented with 2% fetal calf serum and PI-88 (100 μg/ml). Binding of purified viral particles and viral glycoproteins to cells Methyl-3H-thymidine labeled extracellular virions of HSV-1 were purified by centrifugation through a three-step discontinuous sucrose gradient while HSV-1 glycoproteins gB and gC were isolated from lysates of extracellular virions and virusinfected cells by immunoaffinity chromatography as previously described (Trybala et al., 2000). The binding or radiolabeled HSV-1 virions to GMK AH1 cells in the presence of PI-88 inhibitor was assayed as previously described (Nyberg et al., 2004) except for that PI-88 at specific concentrations in EMEM was incubated with purified radiolabeled virus for 15 min prior

251

to and during 2 h period of virus attachment to cells at 37 °C. The binding of purified HSV-1 glycoproteins gB and gC to GMK AH1 cells in the presence of specific concentrations of PI-88 was assayed by an ELISA-based procedure as previously described (Nyberg et al., 2004). Acknowledgments This work was supported by grants from the Mizutani Foundation for Glycoscience, the Swedish Research Council, the Sahlgren's University Hospital Läkarutbildningsavtal and the Scandinavian Society for Antimicrobial Chemotherapy. References Adams, T., Freeman, C., Schwartz-Albiez, R., Ferro, V., Parish, C.R., Andrews, K.T., 2006. Inhibition of Plasmodium falciparum growth in vitro and adhesion to chondroitin-4-sulfate by the heparan sulfate mimetic PI-88 and other sulfated oligosaccharides. Antimicrob. Agents Chemother. 50, 2850–28522. Bergstrom, T., Lycke, E., 1990. Neuroinvasion by herpes simplex virus. An in vitro model for characterization of neurovirulent strains. J. Gen. Virol. 71, 405–410. Brown, S.M., Ritchie, D.A., Subak-Sharpe, J.H., 1973. Genetic studies with herpes simplex virus type 1. The isolation of temperature-sensitive mutants, their arrangement into complementation groups and recombination analysis leading to a linkage map. J. Gen. Virol. 18, 329–346. Bzik, D.J., Fox, B.A., DeLuca, N.A., Person, S., 1984. Nucleotide sequence of a region of the herpes simplex virus type 1 gB glycoprotein gene: mutations affecting rate of virus entry and cell fusion. Virology 137, 185–190. Campadelli-Fiume, G., Qi, S., Avitabile, E., Foa-Tomasi, L., Brandimarti, R., Roizman, B., 1990. Glycoprotein D of herpes simplex virus encodes a domain which precludes penetration of cells expressing the glycoprotein by superinfecting herpes simplex virus. J. Virol. 64, 6070–6079. Carlucci, M.J., Scolaro, L.A., Damonte, E.B., 2002. Herpes simplex virus type 1 variants arising after selection with an antiviral carrageenan: lack of correlation between drug susceptibility and syn phenotype. J. Med. Virol. 68, 92–98. Cocchi, F., Menotti, L., Dubreuil, P., Lopez, M., Campadelli-Fiume, G., 2000. Cell-to-cell spread of wild-type herpes simplex virus type 1, but not of syncytial strains, is mediated by the immunoglobulin-like receptors that mediate virion entry, nectin1 (PRR1/HveC/HIgR) and nectin2 (PRR2/ HveB). J. Virol. 74, 3909–3917. Dean, H.J., Terhune, S.S., Shieh, M.T., Susmarski, N., Spear, P.G., 1994. Single amino acid substitutions in gD of herpes simplex virus 1 confer resistance to gD-mediated interference and cause cell-type-dependent alterations in infectivity. Virology 199, 67–80. Delboy, M.G., Patterson, J.L., Hollander, A.M., Nicola, A.V., 2006. Nectin-2mediated entry of a syncytial strain of herpes simplex virus via pHindependent fusion with the plasma membrane of Chinese hamster ovary cells. Virol. J. 3, 105. Duff, R., Rapp, F., 1971. Oncogenic transformation of hamster cells after exposure to herpes simplex virus type 2. Nat. New Biol. 233, 48–50. Ekblad, M., Bergstrom, T., Banwell, M.G., Bonnet, M., Renner, J., Ferro, V., Trybala, E., 2006. Anti-herpes simplex virus activities of two novel disulphated cyclitols. Antivir. Chem. Chemother. 17, 97–106. Engel, J.P., Boyer, E.P., Goodman, J.L., 1993. Two novel single amino acid syncytial mutations in the carboxy terminus of glycoprotein B of herpes simplex virus type 1 confer a unique pathogenic phenotype. Virology 192, 112–120. Foster, T.P., Alvarez, X., Kousoulas, K.G., 2003. Plasma membrane topology of syncytial domains of herpes simplex virus type 1 glycoprotein K (gK): the UL20 protein enables cell surface localization of gK but not gK-mediated cell-to-cell fusion. J. Virol. 77, 499–510. Fusco, D., Forghieri, C., Campadelli-Fiume, G., 2005. The pro-fusion domain of

252

M. Ekblad et al. / Virology 367 (2007) 244–252

herpes simplex virus glycoprotein D (gD) interacts with the gD N terminus and is displaced by soluble forms of viral receptors. Proc. Natl. Acad. Sci. U. S.A. 102, 9323–9328. Gage, P.J., Levine, M., Glorioso, J.C., 1993. Syncytium-inducing mutations localize to two discrete regions within the cytoplasmic domain of herpes simplex virus type 1 glycoprotein B. J. Virol. 67, 2191–2201. Geraghty, R.J., Krummenacher, C., Cohen, G.H., Eisenberg, R.J., Spear, P.G., 1998. Entry of alphaherpesviruses mediated by poliovirus receptor-related protein 1 and poliovirus receptor. Science 280, 1618–1620. Goodman, J.L., Engel, J.P., 1991. Altered pathogenesis in herpes simplex virus type 1 infection due to a syncytial mutation mapping to the carboxy terminus of glycoprotein B. J. Virol. 65, 1770–1778. Gunalp, A., 1965. Growth and cytopathic effect of rubella virus in a line of Green Monkey kidney cells. Proc. Soc. Exp. Biol. Med. 118, 185–190. Heldwein, E.E., Lou, H., Bender, F.C., Cohen, G.H., Eisenberg, R.J., Harrison, S.C., 2006. Crystal structure of glycoprotein B from herpes simplex virus 1. Science 313, 217–220. Herold, B.C., WuDunn, D., Soltys, N., Spear, P.G., 1991. Glycoprotein C of herpes simplex virus type 1 plays a principal role in the adsorption of virus to cells and in infectivity. J. Virol. 65, 1090–1098. Herold, B.C., WuDunn, D., Visalli, R.J., Sumarski, N., Brandt, C., Spear, P.G., 1994. Glycoprotein C-independent binding of herpes simplex virus to cells requires cell surface heparan sulfate and glycoprotein B. J. Gen. Virol. 75, 1211–1222. Hoggan, M.D., Roizman, B., 1959. The isolation and properties of a variant of herpes simplex producing multinucleated giant cells in monolayer cultures in the presence of antibody. Am. J. Hyg. 70, 208–219. Holland, T.C., Marlin, S.D., Levine, M., Glorioso, J., 1983. Antigenic variants of herpes simplex virus selected with glycoprotein-specific monoclonal antibodies. J. Virol. 45, 672–682. Holland, T.C., Homma, F.L., Marlin, S.D., Levine, M., Glorioso, J., 1984. Herpes simplex virus type 1 glycoprotein C-negative mutants exhibit multiple phenotypes, including secretion of truncated glycoproteins. J. Virol. 52, 566–574. Kosovsky, J., Vojvodova, A., Oravcova, I., Kudelova, M., Matis, J., Rajcani, J., 2000. Herpes simplex virus 1 (HSV-1) strain HSZP glycoprotein B gene: comparison of mutations among strains differing in virulence. Virus Genes 20, 27–33. Krummenacher, C., Supekar, V.M., Whitbeck, J.C., Lazear, E., Connolly, S.A., Eisenberg, R.J., Cohen, G.H., Wiley, D.C., Carfi, A., 2005. Structure of unliganded HSV gD reveals a mechanism for receptor-mediated activation of virus entry. EMBO J. 24, 4144–4153. Lee, E., Pavy, M., Young, N., Freeman, C., Lobigs, M., 2006. Antiviral effect of the heparan sulfate mimetic, PI-88, against dengue and encephalitic flaviviruses. Antivir. Res. 69, 31–38. Liljeqvist, J.-A., Svennerholm, B., Bergstrom, T., 1999. Herpes simplex virus type 2 glycoprotein G-negative clinical isolates are generated by single frameshift mutations. J. Virol. 73, 9796–9802. Lingen, M., Seck, T., Dehoust, U., Weise, K., Falke, D., 1995. Amino acid substitutions in the glycoprotein D mediate the ability of fusion from without-positive herpes simplex virus type 1 strains to penetrate at 4 degrees and in the presence of soluble glycoprotein D. Intervirology 38, 283–289. Lopez, M., Cocchi, F., Menotti, L., Avitabile, E., Dubreuil, P., CampadelliFiume, G., 2000. Nectin2alpha (PRR2alpha or HveB) and nectin2delta are low-efficiency mediators for entry of herpes simplex virus mutants carrying the Leu25Pro substitution in glycoprotein D. J. Virol. 74, 1267–1274. Mardberg, K., Trybala, E., Glorioso, J.C., Bergstrom, T., 2001. Mutational analysis of the major heparan sulfate-binding domain of herpes simplex virus type 1 glycoprotein C. J. Gen. Virol. 82, 1941–1950. Mettenleiter, T.C., Lomniczi, B., Sugg, N., Schreurs, C., Ben-Porat, T., 1988.

Host cell-specific growth advantage of pseudorabies virus with a deletion in the genome sequences encoding a structural glycoprotein. J. Virol. 62, 12–19. Montgomery, R.I., Warner, M.S., Lum, B.J., Spear, P.G., 1996. Herpes simplex virus-1 entry into cells mediated by a novel member of the TNF/NGF receptor family. Cell 87, 427–436. Munk, K., Ludwig, G., 1972. Properties of plaque variants of herpes virus hominis strains of genital origin. Arch. Gesamte Virusforsch. 37, 308–315. Nyberg, K., Ekblad, M., Bergstrom, T., Freeman, C., Parish, C.R., Ferro, V., Trybala, E., 2004. The low molecular weight heparan sulfate-mimetic, PI88, inhibits cell-to-cell spread of herpes simplex virus. Antivir. Res. 63, 15–24. Pertel, P.E., Spear, P.G., 1996. Modified entry and syncytium formation by herpes simplex virus type 1 mutants selected for resistance to heparin inhibition. Virology 226, 22–33. Pertel, P.E., Spear, P.G., 1997. Partial resistance to gD-mediated interference conferred by mutations affecting herpes simplex virus type 1 gC and gK. J. Virol. 71, 8024–8028. Pogue-Geile, K.L., Spear, P.G., 1987. The single base pair substitution responsible for the Syn phenotype of herpes simplex virus type 1, strain MP. Virology 157, 67–74. Robbins, A.K., Dorney, D.J., Wathen, M.W., Whealy, M.E., Gold, C., Watson, R.J., Holland, L.E., Weed, S.D., Levine, M., Glorioso, J.C., Enquist, L.W., 1987. The pseudorabies virus gII gene is closely related to the gB glycoprotein gene of herpes simplex virus. J. Virol. 61, 2691–2701. Ruel, N., Zago, A., Spear, P.G., 2006. Alanine substitution of conserved residues in the cytoplasmic tail of herpes simplex virus gB can enhance or abolish cell fusion activity and viral entry. Virology 346, 229–237. Shukla, D., Liu, J., Blaiklock, P., Shworak, N.W., Bai, X., Esko, J.D., Cohen, G.H., Eisenberg, R.J., Rosenberg, R.D., Spear, P.G., 1999. A novel role for 3-O-sulfated heparan sulfate in herpes simplex virus entry. Cell 99, 13–22. Tal-Singer, R., Peng, C., Ponce De Leon, M., Abrams, W.R., Banfield, B.W., Tufaro, F., Cohen, G.H., Eisenberg, R.J., 1995. Interaction of herpes simplex virus glycoprotein gC with mammalian cell surface molecules. J. Virol. 69, 4471–4483. Trybala, E., Liljeqvist, J.-A., Svennerholm, B., Bergstrom, T., 2000. Herpes simplex virus types 1 and 2 differ in their interaction with heparan sulfate. J. Virol. 74, 9106–9114. Vaheri, A., 1964. Heparin and related polyionic substances as virus inhibitors. Acta Pathol. Microbiol. Scand., Suppl. 171, 1–98. Warner, M.S., Geraghty, R.J., Martinez, W.M., Montgomery, R.I., Whitbeck, J.C., Xu, R., Eisenberg, R.J., Cohen, G.H., Spear, P.G., 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. Watson, R.J., Weis, J.H., Salstrom, J.S., Enquist, L.W., 1982. Herpes simplex virus type-1 glycoprotein D gene: nucleotide sequence and expression in Escherichia coli. Science 218, 381–384. Witvrouw, M., De Clercq, E., 1997. Sulfated polysaccharides extracted from sea algae as potential antiviral drugs. Gen. Pharmacol. 29, 497–511. WuDunn, D., Spear, P.G., 1989. Initial interaction of herpes simplex virus with cells is binding to heparan sulfate. J. Virol. 63, 52–58. Yoon, M., Zago, A., Shukla, D., Spear, P.G., 2003. Mutations in the N termini of herpes simplex virus type 1 and 2 gDs alter functional interactions with the entry/fusion receptors HVEM, nectin-2, and 3-O-sulfated heparan sulfate but not with nectin-1. J. Virol. 77, 9221–9231. Yu, G., Gunay, N.S., Linhardt, R.J., Toida, T., Fareed, J., Hoppensteadt, D.A., Shadid, H., Ferro, V., Li, C., Fewings, K., Palermo, M.C., Podger, D., 2002. Preparation and anticoagulant activity of the phosphosulfomannan PI-88. Eur. J. Med. Chem. 37, 783–791.