Heparan sulfate-dependent enhancement of henipavirus infection

Abstract

Nipah virus and Hendra virus are emerging, highly pathogenic, zoonotic paramyxoviruses that belong to the genus Henipavirus. They infect humans as well as numerous mammalian species. Both viruses use ephrin-B2 and -B3 as cell entry receptors, and following initial entry into an organism, they are capable of rapid spread throughout the host. We have previously reported that Nipah virus can use another attachment receptor, different from its entry receptors, to bind to nonpermissive circulating leukocytes, thereby promoting viral dissemination within the host. Here, this attachment molecule was identified as heparan sulfate for both Nipah virus and Hendra virus. Cells devoid of heparan sulfate were not able to mediate henipavirus trans-infection and showed reduced permissivity to infection. Virus pseudotyped with Nipah virus glycoproteins bound heparan sulfate and heparin but no other glycosaminoglycans in a surface plasmon resonance assay. Furthermore, heparin was able to inhibit the interaction of the viruses with the heparan sulfate and to block cell-mediated trans-infection of henipaviruses. Moreover, heparin was shown to bind to ephrin-B3 and to restrain infection of permissive cells in vitro. Consequently, treatment with heparin devoid of anticoagulant activity improved the survival of Nipah virus-infected hamsters. Altogether, these results reveal heparan sulfate as a new attachment receptor for henipaviruses and as a potential therapeutic target for the development of novel approaches against these highly lethal infections.

Importance: The Henipavirus genus includes two closely related, highly pathogenic paramyxoviruses, Nipah virus and Hendra virus, which cause elevated morbidity and mortality in animals and humans. Pathogenesis of both Nipah virus and Hendra virus infection is poorly understood, and efficient antiviral treatment is still missing. Here, we identified heparan sulfate as a novel attachment receptor used by both viruses to bind host cells. We demonstrate that heparin was able to inhibit the interaction of the viruses with heparan sulfate and to block cell-mediated trans-infection of henipaviruses. Moreover, heparin also bound to the viral entry receptor and thereby restricted infection of permissive cells in vitro. Consequently, heparin treatment improved survival of Nipah virus-infected hamsters. These results uncover an important role of heparan sulfate in henipavirus infection and open novel perspectives for the development of heparan sulfate-targeting therapeutic approaches for these emerging infections.

FIG 1

trans-Infection with henipaviruses requires the expression of HS. (A) Lymphocyte-mediated trans-infection by NiV and HeV. PBLs were incubated with either NiV or HeV, washed, cultured for 24 h, and then transferred to Vero cell monolayers, which were used for the determination of the viral titers after 4 days of coculture, using infectious center assays. (B) CHO-K1 cells, treated or not with heparinase 3, and three HS-deficient CHO lines, pgsA-745, pgsB-619, and pgsD-677, were incubated with NiV and analyzed for their capacities to transmit infection to susceptible Vero cells in trans. Results are expressed as a percentage of inhibition compared to results with untreated cells ± SD. *, P < 0.05; ***, P < 0.001 (Mann-Whitney U test). (C) SPR analysis of the binding of MLV pseudotyped either with VSV-G (green) or with NiV glycoproteins G and F (red) to HP-activated sensor chip surfaces. (D) SPR analysis of the binding of NiV G and F pseudoparticles to surfaces activated by either HS (blue), DS (green), or HP (red). The binding response, in RU, was recorded as a function of time; results from 1 of 3 experiments are presented.

FIG 2

Heparin inhibits the interaction between henipavirus and HS. (A) PBLs or CHO-K1 cells were incubated for 30 min with the indicated doses of heparin before being put into contact with NiV. (B) PBLs were treated for 30 min with 0.5 mg/ml of heparin before contact with either NiV or HeV (pretreatment) or after 1 h of incubation with the virus (posttreatment). trans-Infection of Vero cells was than determined as described for Fig. 1. Results are expressed as a percentage of inhibition compared to results in untreated cells, ± the SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001 (Mann-Whitney U test). (C) Analysis of NiV binding to CHO-K1 and HS-deficient CHO-pgsA-745 cells, pretreated or not with heparin (0.5 mg/ml). Cells were put into contact with NiV for 24 h, and the number of viral RNA (N gene) copies was determined by RT-qPCR. (D) SPR analysis of NiV pseudoparticle binding to an HP-activated surface in the absence or presence of soluble heparin at 1 µg/ml or 10 µg/ml. (E) SPR analysis of the binding of NiV pseudoparticles to HP-activated sensor chips, after preincubation with either PBS or 10 µg/ml of soluble heparin, CS-A, CS-C, or DS. (F) SPR analysis of the binding of NiV pseudoparticles, expressing either NiV-G or NiV-F protein, to HP-activated sensor chips. The binding response, in RU, was recorded as a function of time after removal of the background provided by nonpseudotyped particles. Results are presented as the averages of 3 separate experiments (± standard errors of the means) and reflect statistically significant binding of NiV-G to HP. **, P < 0.01 (one-sample t test).

FIG 3

HS plays a role in henipavirus infection. (A) Vero cells were either treated with heparinase 3 or left untreated prior to NiV infection. Titration was performed 3 days later in a plaque assay. (B and C) Vero cells were treated with increasing concentrations of sodium chlorate for 48 h and then infected with either NiV (B) or HeV (C); titration was performed 3 days later in a plaque assay. Results are expressed as a percentage of results for nontreated controls from triplicate cultures, ± the SD. *, P < 0.05; **, P < 0.01; ***, P < 0.01 (Mann-Whitney U test).

FIG 4

Heparin inhibits infection and limits viral binding to EFN-B2 and -B3. (A) Vero cells were treated with heparin for 30 min before infection with NiV or HeV. Results are expressed as a percentage relative to results in nontreated controls. (B) Vero cells were treated with increasing concentrations of heparin for 30 min before infection with NiV; alternatively, virus was incubated with heparin before contact with cells. Viral titration was performed via a plaque assay, and results are expressed as a percentage of the results in nontreated controls. (C) NiV infection of CHO-pgsA-745–EFN-B2 or –EFN-B3 cells, pretreated with heparin (0.5 mg/ml; 30 min, 37°C) and analyzed in a plaque assay. Results are expressed as the mean percentage relative to results in nontreated controls ± the SD from 8 different experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.01 (Mann-Whitney U test). (D) SPR analysis of binding of CHO-pgsA-745, CHO-pgsA-745–EFN-B2, and CHO-pgsA-745–EFN-B3 to heparin. Results are presented as means ± SD of 4 independent experiments. (E) Quantification results for EFN-B2 and EFN-B3 mRNA expression levels in stably transfected CHO-pgsA-745 cells via RT-qPCR. (F) SPR analysis of the binding of soluble EFN-B3 (0 to 250 nM, as indicated), injected over HP-activated sensor chips; results from triplicate experiments were analyzed. The binding response, in RU, was recorded as a function of time.

FIG 5

Antiviral effect of PO-heparin. (A) In vitro comparison of the inhibitory effects of heparin and PO-heparin (0.5 mg/ml) on the trans-infection ability of leukocytes treated before contact with NiV. The NiV titer was measured in an infectious center assay, and results are expressed as the percentage of inhibition compared to results in untreated cells. (B) Groups of 5 hamsters were either left untreated or treated daily by subcutaneous injections of PO-heparin (10 mg/kg) for 12 days. Animals were infected intraperitoneally with 500 LD₅₀s of NiV on the first day of treatment and followed for 3 weeks. The results are expressed as the percentage of surviving animals in each group. Survival was significantly increased in the group of treated animals. *, P = 0.017 (Mantel-Cox test).

FIG 6

Schematic presentation of possible implications of heparan sulfate and heparin in henipavirus infection. (A) NiV and HeV interact with their entry receptors, EFN-B2 and -B3, and with HS. While the first interaction is important for virus infection in cis, the second leads to infection in trans and may facilitate virus dissemination in the host. In addition, HS may help the virus to reach its entry receptors and accumulate on the cell surface and/or stabilize the interaction with EFN (B2 and B3). (B) Heparin binds henipavirus G-protein as well as ephrin receptors and may thus displace the virus from the cell surface and prevent it from reaching its entry receptors. Consequently, heparin inhibits infection in trans and restrains direct infection in cis, respectively.