Animals and treatments

Wild-type (WT) BALB/cJ mice were purchased from Vital River Laboratory Animal Technology (Beijing, China). CCR5 KO mice on a BALB/cJ background were constructed by GemPharmatech LLC (Nanjing, China) and showed no overt developmental abnormalities. We bred CCR5 KO mice with BALB/cJ mice to obtain CCR5 heterozygous mice. Male and female CCR5 heterozygous mice were bred to generate CCR5 KO mice and WT littermates. Genotypes of mice were confirmed by polymerase chain reaction. The sequences of the primers were as follows: forward: 5′-CTACTCCCTGGTATTCATC-3′, reverse: 5′-GGCCTGGTCTAGTCTATT-3′. Female CCR5 KO mice and WT littermates aged 6–8 weeks were used for the study. Mice were housed in a specific pathogen-free environment. MHV-3 was purchased from the American Type Culture Collection, and the MHV-3-FHF model was established according to our previous study [21]. In the pharmacological inhibition of CCR5 experiment, mice received maraviroc (1.23 mg per 20 g mouse weight) by oral gavage daily. All animal protocols were approved by the Tongji Hospital of Tongji Medical School Committees on Animal Experimentation.

Cell isolation and quantitation

The mononuclear cells (MNCs) in the liver and peripheral organs, including the blood, spleen and bone marrow (BM), were collected as previously described [22].

Purification of NK cells

An NK Cell Isolation Kit II (Miltenyi Biotec, Germany) was used to purify NK cells according to the manufacturer’s standard protocol. The sorted cells were > 90% pure. The cell viability was > 95%.

Isolation of hepatocytes

Hepatocytes were isolated from fresh liver tissue by a two-step hepatic portal vein perfusion technique as previously described [8].

Real-time PCR assays

TRIzol reagent (Invitrogen, USA) was used to extract total RNA from hepatic NK cells, hepatocytes and liver tissue at 0, 24, 48, and 72 h after MHV-3 infection according to the manufacturer’s standard protocol. Subsequently, the total RNA was reverse transcribed into cDNA using a ReverTra Ace qPCR RT kit (TOYOBO, Japan). Real-time quantitative PCRs were performed using SYBR Green Real-time PCR Master Mix (TOYOBO, Japan). The sequences of the primers used are listed in Additional file 1: Table S1.

Histology and immunohistochemistry

Fresh liver tissues acquired from mice at 0, 24, 48 and 72 h post MHV-3 infection were fixed, embedded and sectioned (4 μm). After haematoxylin and eosin staining, the percentage of necrosis was evaluated by ImageJ software. To detect the expression of MIP-1α, MIP-1β and RANTES, the sections were stained with the following antibodies: MIP-1α (R&D Systems, USA), MIP-1β (R&D Systems, USA), and RANTES (R&D Systems, USA). The sections were observed under a microscope (CX22, OLYMPUS, Japan).

Immunofluorescence

Tissue paraffin sections were dewaxed and subjected to antigen retrieval treatment. The sections were incubated with the following antibodies: anti-mouse CD3 (Abcam, USA), anti-mouse CD49b (Abcam, USA) and anti-mouse CCR5 (eBioscience, USA). The secondary antibodies were HRP-conjugated goat anti-rabbit (SeraCare, USA). Then, the sections were incubated with FITC-Tyramide and Cy3-Tyramide. DAPI (Bioqiandu, China) was used for nuclear staining. The sections were observed under a fluorescence microscope (BX53, OLYMPUS, Japan). The CD3 + cells were considered T cells, and the CD3-CD49b + CCR5 + cells were considered CCR5 + cNK cells.

Gene array

Hepatic NK cells of normal BALB/cJ mice and MHV-3-infected BALB/cJ mice (48 h post infection) were isolated and purified, total RNA was extracted, and a gene array was constructed. The Mouse Genome 430 2.0 Array (Affymetrix, USA) was used according to the manufacturer’s standard protocol.

Liver cytokine measurement

Fresh liver tissues were homogenized on ice in 1 mL of tissue protein extraction reagent (Boster, China) that contained 1% protease inhibitor cocktail (Boster, China). Samples were vortexed, thawed on ice, and centrifuged. The supernatants were collected to analyse MIP-1α (Boster, China), MIP-1β (R&D systems, USA), and RANTES (Boster, China) levels by ELISA, in accordance with the manufacturer’s protocol.

Transwell migration assay

Transwell migration assays were performed using 6.5 mm Transwell inserts with a 5 µm pore size (Corning, USA) within individual wells of 24-well ultra-low attachment plates (Corning, USA). The wells of the 24-well ultra-low attachment plates were individually filled with 0.6 mL (6 × 106 cells/well) of normal hepatocyte suspension, 0.6 mL (6 × 106 cells/well) of MHV-3-infected hepatocyte (24 h post MHV-3 infection) suspension, 0.6 mL (6 × 106 cells/well) of MHV-3-infected hepatocyte (24 h post MHV-3 infection) suspension with anti-MIP-1α/MIP-1β/RANTES neutralizing antibodies (10 μg/mL, 6 µg/mL, 20 µg/mL, respectively, all from R&D systems, USA), and 0.6 mL RPMI 1640 medium. Rat IgG1 Isotype Control (R&D systems, USA) and Rat IgG2A Isotype Control (R&D systems, USA) were used as isotype controls. Normal NK cells and MHV-3-infected NK cells (24 h post MHV-3 infection) isolated from the spleen with or without anti-CCR5 neutralizing antibodies (BD, USA) were added in 100 µL (1 × 105 cells/well) into the upper chamber, in contact with the medium in the wells. Purified anti-mouse CD191 (CCR1) antibody (Biolegend, USA) and rat IgG2c kappa isotype control (Novus, USA) were used as controls. Then, the plates were incubated for 8 h. NK cells were collected from the bottom chambers after 8 h of incubation. The density of cells was evaluated by a cell counter. Total cells were calculated via density and volume. A FACS assay was used to confirm the proportion of NK cells, and then we calculated the total number of migrated NK cells.

Flow cytometry analysis

Cells were preincubated with Mouse BD Fc Block (clone 2.4G2, BD Biosciences) before staining. To identify rNK and cNK cells, cells were incubated with antibodies against surface markers: APC-Cy7-conjugated anti-mouse CD3 antibody (clone 17A2), PerCP-Cy5.5-conjugated anti-mouse NKP46 antibody (clone 29A1.4), APC-conjugated anti-mouse CD49a antibody (clone HMα1), FITC-conjugated anti-mouse CD49b antibody (clone HMα2), and PE-conjugated anti-mouse CCR5 antibody (clone HEK/1/85a). APC/Cy7-conjugated Rat IgG2b, κ Isotype Ctrl Antibody (clone RTK4530), PerCP/Cy5.5-conjugated Rat IgG2a, κ Isotype Ctrl Antibody (clone RTK2758), APC-conjugated Armenian Hamster IgG Isotype Ctrl Antibody (clone HTK888), FITC-conjugated Armenian Hamster IgG Isotype Ctrl Antibody (clone HTK888), PE-conjugated Rat IgG2a, κ Isotype Ctrl Antibody (clone RTK2758). All antibodies were purchased from BioLegend (California, USA). Cells were finally subjected to flow cytometry with a BD LSR II flow cytometer (BD Biosciences). The CD3-NKP46 + cells were considered NK cells. Within the CD3-NKP46 + gate, the CD49a + CD49b− cells were considered rNK cells, and the CD49a-CD49b + CCR5 + cells were considered CCR5 + cNK cells.

Adoptive NK cell transfer

Isolated and purified splenic NK cells were resuspended in PBS, and 5 × 104 cells were intravenously injected into CCR5 KO mice at 0 h, 24 h, and 48 h post MHV-3 infection.

Statistical analysis

Data gained from experiments are shown as the mean ± SD. To compare differences between multiple groups, one-way ANOVA with post hoc Bonferroni correction (GraphPad Prism 4.02; GraphPad Software) was employed. Group differences in cell migration and liver damage were analysed using two-way ANOVA, with the factors being treatment and genotype, followed by Bonferroni post hoc correction (GraphPad Prism 4.02; GraphPad Software). The log-rank test was used to analyse the survival rate (GraphPad Prism 4.02; GraphPad Software). Statistical significance was identified when P < 0.05.