In vivo importance of heparan sulfate-binding glycoproteins for murid herpesvirus-4 infection

Abstract

Many herpesviruses bind to heparan sulfate (HS). Murid herpesvirus-4 (MuHV-4) does so via its envelope glycoproteins gp70 and gH/gL. MuHV-4 gp150 further regulates an HS-independent interaction to make that HS-dependent too. Cell binding by MuHV-4 virions is consequently strongly HS-dependent. Gp70 and gH/gL show some in vitro redundancy: an antibody-mediated blockade of HS binding by one is well tolerated, whereas a blockade of both severely impairs infection. In order to understand the importance of HS binding for MuHV-4 in vivo, we generated mutants lacking both gL and gp70. As expected, gL(-)gp70(-) MuHV-4 showed very poor cell binding. It infected mice at high dose but not at low dose, indicating defective host entry. But once entry occurred, host colonization, which for MuHV-4 is relatively independent of the infection dose, was remarkably normal. The gL(-)gp70(-) entry deficit was much greater than that of gL(-) or gp70(-) single knockouts. And gp150 disruption, which allows HS-independent cell binding, largely rescued the gL(-)gp70(-) cell binding and host entry deficits. Thus, it appeared that MuHV-4 HS binding is important in vivo, principally for efficient host entry.

Fig. 1.

Generation of MuHV-4 mutants. In order to ablate expression, translational stop codons were inserted into the coding sequences of gL, gp70 and gp150 as indicated. Single gL and gp70 mutants have been described previously. (a) For this study, we combined gL and gp70 mutations, and then made a triple mutant by further disrupting gp150. (b) Viral DNA was analysed by Southern blotting, exploiting the novel EcoRI site introduced during mutagenesis. Wild-type (wt) and gp70⁻ viruses provide comparisons. gL⁻gp70⁻.1 and gL⁻gp70⁻.2 are independent mutants. For gL, there is a 3.1 kb invariant band and a 942 bp wt band that changes to 270 bp+595 bp. For gp70, a 12.7 kb wt band changes to 3.4 kb+8.4 kb. The 3.4 kb band was evident by ethidium bromide staining (data not shown), but not by hybridization because it overlaps the probe by only 115 bp. For gp150, a14.9 kb wt band changes to 11.2 kb+3.8 kb. (c) BHK-21 cells were left uninfected (UI) or infected (2 p.f.u. cell⁻¹) with wt (18 h infection), single mutant (gL⁻, gp70⁻, 18 h infections) or double mutant viruses (gL⁻gp70⁻, 66 h infection), then analysed by flow cytometry for surface expression of virion glycoproteins. (d) BHK-21 cells were exposed to BAC⁺ wt (1 p.f.u. cell⁻¹) or gL⁻gp70⁻ (5 p.f.u. cell⁻¹) viruses plus soluble heparin. Both viruses expressed eGFP from an HCMV IE1 promoter, and infection was quantified by flow cytometry after 24 h. Different virus doses were used to ensure comparable eGFP expression. The difference in susceptibility to heparin was much greater than the difference in virus input.

Fig. 2.

gL⁻gp70⁻ MuHV-4 shows a severe deficit in cell binding. (a) BHK-21 or NMuMG cells were exposed to viruses (1 p.f.u. cell⁻¹, 37 °C) for different times, then washed three times with PBS to remove unbound virions. After 24 h, all cells were analysed by flow cytometry for viral eGFP expression. Each point shows 2×10⁴ cells. (b) NMuMG cells were exposed to wt or mutant viruses (2 h, 4 °C, 5 p.f.u. cell⁻¹), then washed three times with PBS and either fixed immediately (4 °C) or first incubated in complete medium (2 h, 37 °C). All the cells were then stained for gN, an abundant component of the virion envelope. New gN expression is not evident until at least 6 h p.i., so this assay detects only input virus. (c) Virus stocks were compared by immunoblot for gB (mAb MG-4D11), ORF17 (mAb 150-7D1) and gN (mAb 3F7). gB-FL, full-length gB; gB-C, C-terminal cleavage product. ORF17 is auto-cleaved and so appears as a doublet. (d) NMuMG cells were exposed to wt or mutant viruses (2 h, 37 °C, 3 p.f.u. cell⁻¹), then washed three times with PBS and analysed for virion binding by fixation, permeabilization and staining for gB or gN. nil, No virus. Each bar shows 2×10⁴ cells. Fixation/permeabilization was used to make virion detection independent of endocytosis.

Fig. 3.

In vivo replication of gp70⁻ MuHV-4 after intranasal inoculation. Mice were infected intranasally (300 p.f.u.) with wt, gp70⁻ or revertant (gp70⁺) viruses, then analysed for infectious virus in lungs by plaque assay and for latent virus in lymphoid tissue by infectious centre assay. Each point shows the titre of one mouse. gp70⁻ virus titres in lungs and spleens were significantly reduced compared with wt (P<0.001 by Student's t-test). Although the gp70⁻ mediastinal lymph node titres were also lower, the difference did not reach statistical significance (P=0.06).

Fig. 4.

In vivo infection with gL⁻gp70⁻ MuHV-4. (a) Mice were infected intranasally (300 p.f.u.) with wt or gL⁻gp70⁻ MuHV-4, then analysed for infectious virus in lungs by plaque assay and for latent virus in lymphoid tissue by infectious centre assay. Each point shows the titre of one mouse. gL⁻gp70⁻ titres were significantly lower than wt in lungs at day 6 (P<0.001 by Student's t-test) but not in lymphoid tissue. (b) gL⁻gp70⁻ viruses from the individual mouse lungs in (a) (L1-L5) were propagated in BHK-21 cells for 7 days then analysed for glycoprotein expression by flow cytometry. Cells were scored as stained or not, based on a gate excluding >99 % of uninfected cells. Uninfected (UI) and wt virus-infected (wt) BHK-21 cells provided controls. Each bar shows 2×10⁴ cells. For equivalent gN expression, gp150 expression by the gL⁻gp70⁻ knockouts was significantly reduced compared with wt (P<10⁻⁵ by χ² test).

Fig. 5.

In vitro growth of gL⁻gp70⁻ MuHV-4. (a) BHK-21 cells were infected (0.01 p.f.u. cell⁻¹) with eGFP-expressing forms of each virus, then monitored by flow cytometry of eGFP expression. Each point shows 2×10⁴ cells. (b) gL⁻gp70⁻ infected cells from (a) were then analysed for viral glycoprotein expression. Cells uninfected (UI) or infected overnight with wt MuHV-4 (1 p.f.u. cell⁻¹) provided controls. For both the gL⁻gp70⁻ mutants, gp150 expression was equivalent to gN expression. (c) BHK-21 cells were exposed to virus stocks normalized by immunoblot (2 h, 37 °C). The equivalent infectivity for wt was 3 p.f.u. (100 %) or 0.6 p.f.u. (20 %). The cells were washed three times with PBS to remove unbound virions, fixed, permeabilized and stained for gN or gB. The horizontal dashed lines show the fluorescence of uninfected cells analysed in parallel. Each bar shows the result for 2×10⁴ cells. Uptake of the gL⁻gp70⁻gp150⁻ mutant was significantly higher than the gl⁻gp70⁻ mutant even with 1/5 the input (P<10⁻⁵ by Student's t-test).

Fig. 6.

In vivo infection with gL⁻gp70⁻gp150⁺ MuHV-4. (a) The gL⁻gp70⁻.2 mutant was analysed for gp150 content by immunoblot with mAbs that recognize gp150 residues 152–269 (T1A1) or 108–151 (T4G2) (Gillet et al., 2007d). gN and the ORF17 capsid component provided loading controls. The 62 kDa band in the T4G2+3F7 immunoblot of the gL⁻gp70⁻ stock is non-specific staining of bovine albumin. Because gL⁻gp70⁻ stocks had lower titres than wt, more albumin was present at comparable infectivity. (b) Mice were infected intranasally (300 p.f.u.) with wt or gL⁻gp70⁻ viruses analysed in (a), then titrated for infectious virus in lungs by plaque assay and for latent virus in lymphoid tissue by infectious centre assay. Each point shows the titre of one mouse. The horizontal dashed lines show lower limits of detection. gL⁻gp70⁻, but not gL⁻gp70⁻gp150⁻ virus titres were significantly lower than wt in all sites (P<0.01 by Student's t-test). (c) DNA was extracted from the spleens in (b) and viral genome copy numbers determined by real-time PCR. Each bar shows the average of triplicate reactions. The horizontal dashed line shows the average of three uninfected controls. (d) gL⁻gp70⁻ viruses were recovered from infected lungs by propagation in BHK-21 cells for 14 days, then analysed for virion glycoprotein expression by flow cytometry. BHK-21 cells either uninfected or infected overnight with wt MuHV-4 (1 p.f.u. cell⁻¹, 18 h), provided staining controls. nil, Secondary antibody only. Each bar shows 2×10⁴ cells. mAbs T4G2, T1A1 and BH-6H2 recognize different gp150 epitopes.

Fig. 7.

Low dose in vivo infection with gL⁻ and gp70⁻ viruses. (a) The gL⁻gp70⁻.2 mutant was compared with wt MuHV-4 by intranasal inoculation of mice with 1–1000 p.f.u. Twelve mice per group for each dose were scored as infected or not at 16 days p.i. by infectious centre assay of spleens and by ELISA for virus-specific serum IgG, with concordant results. The gL⁻gp70⁻ infection rate was significantly lower than wt at 1–10 p.f.u. (P<0.005 by Fisher's exact test), but not at 100 p.f.u. (P=0.1). (b) Virus knockouts were compared with wt for infectivity (12 mice per group) after intranasal inoculation. The gL⁻REV and gL⁻gp70⁻gp150⁻ mutants were derived from the gL⁻gp70⁻ mutant. Their infectivities were equivalent to wt. The gL⁻ and gL⁻gp70⁻ infection rates were both significantly less than wt at 1 p.f.u. (P<0.05 by Fisher's exact test). Only the gL⁻gp70⁻ infection rate was significantly lower at 10 p.f.u. (P<0.02). At 1 p.f.u. the gL⁻gp70⁻ infection rate was significantly lower than that of the gL⁻ (P<0.01).