ITGB4/CD104 mediates zika virus attachment and infection

Abstract

Zika virus (ZIKV) infection can result in a birth defect of the brain called microcephaly and other severe fetal brain defects. ZIKV enters the susceptible host cells by endocytosis, which is mediated by the interaction of the envelope (E) glycoprotein with cellular surface receptor molecules. However, the cellular factors that used by the ZIKV to gain access to host cells remains elusive. Here, we report that the extracellular domain of integrin beta 4 (ITGB4) is an entry factor of ZIKV. ITGB4 mediates ZIKV infection by directly interacting with the E glycoprotein of ZIKV, and ITGB4 knockout hampers the binding and replication of ZIKV to host cells. A functional monoclonal antibody against ITGB4 or the soluble forms of ITGB4 could decrease the binding and infection of ZIKV to permissive cell lines. Importantly, the ITGB4 antibody blocks the infection of ZIKV to mouse placenta, thus protecting the fetuses from ZIKV infection. Together, our study has demonstrated that ZIKV infection involves ITGB4 dependent binding.

Fig. 1. The detection of ITGB4 associates with ZIKV E.

a Hun-7 cell lysates were incubated with ZIKV-E or RuV-E1 proteins and the affinity-isolated proteins were subjected to SDS-PAGE. The red arrowhead indicates the specifically isolated protein using the ZIKV-E protein (n = 3 of independent experiments). b ITGB4-derived tryptic fragments as determined by mass spectrometry. The tryptic fragments corresponding to ITGB4 protein sequence (Gene bank: NM_000213) is indicated in red. c Immunoprecipitation assay showing specific binding between ZIKV-E and the ectodomain of ITGB4. Lysates of 293 cells were incubated with ZIKV-E or RuV-E1 immobilized on anti-Flag M2 monoclonal antibody-conjugated agarose beads. The precipitates were washed by DDM-PBS and eluted with Flag peptide. The eluents were then separated by SDS-PAGE and immunoblotted with the related antibodies (n = 3 of independent experiments). d Immunoprecipitation assay showing specific binding between ZIKV-E and ITGB4/ITGA6 heterodimer in Hun-7 cells. Lysates of Hun-7 cells were incubated with ZIKV-E or RuV-E1 immobilized on anti-FLAG M2 monoclonal antibody-conjugated agarose beads. The precipitates were treated as in (c) (n = 3 of independent experiments). CL: Hun-7 cell Lysates. e ZIKV-E protein interacted with ITGB4. Purified ITGB4, ITGB4/ITGA6 or BSA were coated on 96-well plate overnight and incubated with flag tagged ZIKV-E or RuV-E1 for 3 h at 4°C. ZIKV-E or RuV-E1 was detected by HRP-conjugated anti-Flag antibody. The data were expressed as Mean ± SD of three independent experiments. **p = 0.0057, ****p < 0.0001, two-sided, Student’s t test. f ZIKV virions interacted with the ectodomain of ITGB4. Purified ITGB4, ITGB4/ITGA6 or BSA was coated on 96-well plate overnight and incubated with MR766 or Mock (PBS) for 4 h at 4°C. The bound virus was detected by HRP-conjugated anti-E antibody. The data were expressed as Mean ± SD of three independent experiments. ***p = 0.0002, **p = 0.032, two-sided, Student’s t test. g, h Susceptibility of Vero, Hun-7, A549, LLC-MK₂, hNPCs and HEK-293T cells to ZIKV. Cells were infected with ZIKV MR766 at an MOI = 0.05. 48 h post infection and then visualized using a human antibody that recognizes the structural protein E (green). Cell nuclear were stained with DAPI (blue). n = 3 of independent experiments. All scale bars are 50 μm.

Fig. 2. Soluble ITGB4 and ITGB4/ITGA6 inhibit ZIKV MR766 infection.

a Western blot analysis of the ZIKV in 293 T cells transfected with ZIKV genome RNA. 2 ug purified ZIKV genome RNA was transfected into 293 T cells for 24 h. Then, the supernatant of 293 T cells was transferred into Vero cells (n = 3 of independent experiments). Control: Vero cells infected with ZIKV for 24 h. Mock: Vero-E6 cells. b Viral titers in cell supernatants were determined by TCID₅₀ experiment in Vero-E6 cells. The data were expressed as Mean ± SD of five independent experiments. *P = 0.0204, two-sided, by multiple comparisons by One-way ANOVA test. Control: Vero cells infected with ZIKV for 24 h. 293T-ZIKV: Vero cells incubated with the supernatant of 293 T cells transfected with ZIKV genome RNA. 293 T: Vero cells were incubated with the supernatant of 293 T cells. c Cells were subsequently inoculated for 1 h with the ZIKV MR766-protein mixes on ice followed washing three times with PBS. The virus titer was determined by plaque assay on Vero cells at 48 h after infection. The data were expressed as Mean ± SD of three independent experiments. ****p < 0.0001 (Vero), ****p < 0.0001 (A549), **P = 0.0045 (Hun7), ***P = 0.0002 (Hun7), ****p < 0.0001 (MK₂), **P = 0.0013 (hNPCs), ****p < 0.0001 (hNPCs), two-sided, multiple comparisons by One-way ANOVA test. d Cells were infected with ZIKV as described above. After incubated with virus-protein mixes for 1 h on ice, cells were washed with cold PBS. Total RNA was then extracted and ZIKV RNA levels that represent viral particles bind to cell surface were quantified by RT-PCR assay. The GAPDH was used as internal control gene. The data were expressed as Mean ± SD of three independent experiments. Control: PBS buffer. **P = 0.0038 (Hun7, top), ***P < 0.0003 (Hun7, middle), *P = 0.0241 (Hun7, bottom), **P = 0.0022 (Vero, top), *P = 0.0475 (Vero, bottom), *P = 0.0117 (MK₂), **P = 0.0072 (A549, top), *P = 0.0006 (A549, middle), *P = 0.0072 (A549, bottom), *P = 0.0102 (hNPCs, top), ***P = 0.0002 (hNPCs, middle), **P = 0.0032 (hNPCs, bottom), two-sided, multiple comparisons by One-way ANOVA test. e ZIKV MR766 was preincubated with the indicated concentrations of soluble ITGB4/ITGA6 or BSA. Cells were subsequently inoculated for 1 h with the virus-protein mixes on ice. Cells were washed three times with PBS and the virus titers were determined by plaque assay on Vero cells at 48 h after infection. The data were expressed as Mean ± SD of three independent experiments. ***P = 0.003 (Vero), ****p < 0.0001 (Vero), ***P = 0.0001 (A549), ****p < 0.0001 (A549), **P = 0.0022 (Hun7), ***P = 0.0002 (Hun7), **P = 0.0034 (MK₂), ***p = 0.0007 (MK₂), *P = 0.0257 (hNPCs), **P = 0.0016 (hNPCs), two-sided, multiple comparisons by One-way ANOVA test. f Cells were infected with ZIKV that preincubated with the indicated concentrations of soluble ITGB4/ITGA6 or BSA. Incubated with virus-protein mixes on ice for 1 h, cells were washed three times with cold PBS. Total RNA was then extracted and ZIKV RNA levels were determined by RT-PCR assay. The data were expressed as Mean ± SD of three independent experiments. Control: PBS buffer. **P = 0.0055 (Hun7, top), ***P = 0.0002 (Hun7, middle), **P = 0.0099 (Hun7, bottom), *P = 0.0197 (Vero, top), **P = 0.0017 (Vero, bottom), *P = 0.0227 (MK₂, top), **P = 0.0011 (MK₂, middle), *P = 0.0403 (MK₂, bottom), **P = 0.0061 (A549, top), ***P = 0.0001 (A549, middle), **P = 0.0033 (A549, bottom), **P = 0.0064 (hNPCs, top), ***P = 0.0004 (hNPCs, middle), *P = 0.0232 (hNPCs, bottom), two-sided, multiple comparisons by One-way ANOVA test.

Fig. 3. Inhibition of ZIKV infection by antibody to ITGB4.

a A surface plasmon resonance assay (SPR) characterizing the binding between ITGB4/ITGA6 and 13H10 antibody. The 13H10 antibody were immobilized on the chip at about 300 response units. Concentrations of ITGB4/ITGA6 was used to flow over the chip surface. 13H10 was ITGB4 specific antibody. b Cells were treated with various concentrations of 13H10 or isotype antibody for 1 h and infected with ZIKV with MOI = 0.05. Virus titers were determined by plaque assay on Vero cells at 48 h after infection. The values in the graph represent the mean ± SD of n = 3 independent experiments. ***P = 0.0082 (Vero), ****p < 0.0001 (Vero), **P = 0.0052 (A549), ***p = 0.0005 (A549), ***P = 0.0002 (Hun7), ****p < 0.0001 (Hun7), ****p < 0.0001 (MK₂), **P = 0.0046 (hNPCs), ***P = 0.0002 (hNPCs), two-sided, multiple comparisons by One-way ANOVA test. c Cells were preincubated with isotype antibody or indicated concentrations of 13H10 antibody for 1 h and then challenged with ZIKV with MOI = 10 at 4 °C for 1 h. Cells were washed, and total RNA was extracted. ZIKV RNA was quantified by RT-PCR. The results were expressed as ZIKV RNA levels relative to the expression of the GAPDH internal control gene. The data were expressed as Mean ± SD of three independent experiments. ****P < 0.0001 (Hun7, top), ***P = 0.0001 (Hun7, middle), **P = 0.0042 (Hun7, bottom), ***P = 0.0002 (Vero, top), **P = 0.0019 (Vero, middle), *P = 0.0192 (Vero, bottom), ****P < 0.0001 (A549, top), ****P < 0.0001 (A549, middle), **P = 0.0029 (A549, bottom), ***P = 0.0002 (MK₂, top), **P = 0.0025 (MK₂, middle), *P = 0.0216 (MK₂, bottom), ***P = 0.0003 (hNPCs, top), **P = 0.0060 (hNPCs, middle), *P = 0.0169 (hNPCs, bottom), two-sided, multiple comparisons by One-way ANOVA test.

Fig. 4. ITGB4 confers susceptibility of ZIKV to HEK-293T cells.

a HEK-293T cells were transfected with plasmids encoding human ITGB4 or ITGA6 for 48 h and infected with ZIKV as described above. Cells were subsequently inoculated for 1 h with the virus-protein mixes on ice. Cells were washed three times with PBS and the virus titers were determined by plaque assay on Vero cells at 48 h after infection. The values in the graph represent the mean ± SD of three independent experiments. ****P < 0.0001 (SZ_SMGC-1, top), **P = 0.0012 (MR766, top), ***P = 0.0007 (MR766, bottom), two-sided, multiple comparisons by One-way ANOVA test. b HEK-293T cells were transfected with plasmids encoding human ITGB4 for 48 h, and then infected with ZIKV that preincubated with the indicated concentrations of soluble ITGB4, ITGB4/ITGA6 or BSA. After incubated with 1 h with virus-protein mixes on ice, cells were washed three times with cold PBS. ZIKV RNA levels were determined by RT-PCR assay. The data were expressed as Mean ± SD of three independent experiments. **P = 0.0032 (left, top), ****P < 0.0001 (left, bottom), **P = 0.0014 (right, top), ****P < 0.0001 (right, bottom), two-sided, multiple comparisons by One-way ANOVA test. c Effects of ITGB4 antibody on ZIKV SZ_SMGC-1 binding. HEK-293Tcells were transfected with plasmids encoding human ITGB4 were treated with the indicated amount of 13H10 antibody or an isotype antibody before incubation with ZIKV. 1 h after incubation on ice, cells were washed, and total RNA was extracted. ZIKV RNA was quantified by RT-PCR. The data were expressed as Mean ± SD of three independent experiments. ***p = 0.0003, *p = 0.0129, two-sided, multiple comparisons by One-way ANOVA test.

Fig. 5. ITGB4 knockout reduced ZIKV infection.

a The expression of ITGB4 on cell surface of A549 (orange line) and A549^ITGB4−/− (red line) was monitored by flow cytometry. Cyan line represents A549^ITGB4−/− cells stained with normal mouse IgG. b RT-PCR analysis showing the binding of ZIKV to A549 and A549^ITGB4−/−cells. Cells were incubated with ZIKV SZ_SMGC-1 or ZIKV MR766 with MOI = 10 at a multiplicity of infection at 4°C for 45 min. Cells were washed, and total RNA was extracted. ZIKV RNA was quantified by RT-PCR. The data were expressed as Mean ± SD of five independent experiments. ****p < 0.0001, two-sided, Student’s t test. c Cells were infected with ZIKV with MOI = 0.05. Virus titers were determined by plaque assay on Vero cells at 48 h after infection. The values in the graph represent the mean ± SD of three independent experiments. ***P = 0.002(SZ_SMGC-1), ***P = 0.001(MR766), two-sided, Student’s t test. d Cells were infected with ZIKV as described above. The percentages of infected cells were measured by flow cytometry 48 h post infection. Mock represents cells with no ZIKV infection. e The complementation expression of ITGB4 in A549^ITGB4−/− cells restored ZIKV infectivity. A549^ITGB4−/− cells were transfected with plasmids expressing ITGB4, ITGA6 or empty vector. 24 h post infection, cells were infected with ZIKV MR766 at MOI of 0.05 for 48 h. ZIKV positive cells were counted by immunofluorescence assay. The data were expressed as Mean ± SD of three independent experiments. **p = 0.0024, two-sided, Student’s t test. f Viral titers in A549 and A549^ITGB4−/− cells transfected with ZIKV genome RNA determined by TCID₅₀ experiment in Vero-E6 cells. 2 ug purified ZIKV genome RNA was transfected into A549 and A549^ITGB4−/− cells by using Lipofectamine 3000 for 24 h. And then, the supernatant of A549 cells was transferred into Vero-E6 cells. The data were expressed as Mean ± SD of six independent experiments and multiple comparisons by One-way ANOVA test. ND: Vero cells incubated with the supernatant of A549^ITGB4−/− cells.

Fig. 6. ITGB4 specific antibody protects against lethality and viral burden induced by ZIKV infection.

a, b Four to five weeks old WT male mice (n = 12 per group from three independent experiments) were intraperitoneal treated with anti-Ifnar1 mAb (2.5 mg each mouse). 18 h later, mice were treated with 1 mg of isotype-control mAb or 13H10 followed by subcutaneous infection with 10⁴ PFU of ZIKV-SZ_SMGC-1. Mouse weight loss (a) and survival (b) was observed and recorded daily until 16 d.p.i. Data were expressed as Mean ± SD, two-sided, **P = 0.0016, log-rank test). c Viremia measurements at days 1 through 4 after infection with ZIKV as determined by plaque assay. The data were expressed as Mean ± SD of four independent experiments. **P(1) = 0.0047, ***P(2) = 0.0002, **P(3) = 0.0012, ***P(4) = 0.0003, two-sided, Student’s t test. d and e Four to five weeks old WT male mice were treated with anti-Ifnar1 mAb. 24 h later, mice were treated with 1 mg of isotype-control mAb or 13H10 followed by subcutaneous inoculation with 10⁴ PFU of ZIKV-SZ_SMGC−1. At day +3, viral RNA in serum (d) brain (e) was measured by RT-PCR. n = 7 mice per group (d) and n = 9 mice per group (e). The data were expressed as Mean ± SD of three independent experiments. ***P = 0.0003 (d) ***P = 0.0001 (e) two-sided, Student’s t test. Dashed lines indicate the limit of detection of the assay.

Fig. 7. ITGB4 specific antibody protects against placental.

a Scheme of antibody injection and ZIKV challenge of pregnant dams. Eight to nine weeks-old pregnant WT dams (n = 12 per group) were treated with anti-Ifnar1 (2.5 mg each mouse) at E4.5 or E5.5. 18 h later, mice were treated with a single 1 mg dose of isotype control antibody or 13H10 by intraperitoneal injection followed by subcutaneous inoculation with 10⁴ PFU of ZIKV-SZ_SMGC−1. Mice were euthanized on E13.5. b ZIKV RNA levels in maternal serum (n = 10), maternal brain (n = 12), fetal heads (n = 14), and placenta (n = 11). The dashed line indicates the limit of detection of the qRT-PCR assay. The data were expressed as Mean ± SD of three biological replicates. **P = 0.0031 (serum), **P = 0.0091 (brain), **P = 0.0025 (head), ****P < 0.0001 (placenta), two-sided, Student’s t test. c Measurement of thickness at indicated areas of placentas from ZIKV infected dams after isotype control or 13H10 antibody treatment. The data were expressed as Mean ± SD of three independent experiments, n = 8 in all the measurement but n = 7 in placenta junctional thickness measurement. **P = 0.0060 (labyrinth area), ****P < 0.0001 (labyrinth thickness), **P = 0.0025 (placenta area), **P = 0.0093 (junctional thickness), **P < 0.0001 (junctional area), two-sided, multiple comparisons by One-way ANOVA test.

Fig. 8. ITGB4 specific antibody protects against fetal Infection.

a RNA in situ hybridization staining of placentas at E13.5. The images are representative of several independent placentas from multiple dams. Black arrows indicate cells positive for ZIKV RNA. (n = 3 of independent experiments) b Measurements of fetus body size at E13.5. Each dot represents data from eight individual placenta. The data were expressed as Mean ± SD of three biological replicates. ***P = 0.0005, two-sided, Multiple comparisons by One-way ANOVA test. c The percentage of fetuses survival on E13.5 (n = 17 for 13H10, n = 14 for isotype control antibody). Data were representative of 3 independent experiments. ****p < 0.0001, two-sided, Student’s t test. d Representative images of E13.5 uterus recovered from ZIKV-infected dams after isotype control or 13H10 antibody treatment. Most fetuses carried by isotype control antibody treated dams died in utero and had undergone resorption, leaving only the residual abnormal placenta.