GCRV-II invades monocytes/macrophages and induces macrophage polarization and apoptosis in tissues to facilitate viral replication and dissemination

ABSTRACT

Grass carp reovirus (GCRV), particularly the highly prevalent type II GCRV (GCRV-II), causes huge losses in the aquaculture industry. However, little is known about the mechanisms by which GCRV-II invades grass carp and further disseminates among tissues. In the present study, monocytes/macrophages (Mo/Mφs) were isolated from the peripheral blood of grass carp and infected with GCRV-II. The results of indirect immunofluorescent microscopy, transmission electron microscopy, real-time quantitative RT-PCR (qRT-PCR), western blot (WB), and flow cytometry analysis collectively demonstrated that GCRV-II invaded Mo/Mφs and replicated in them. Additionally, we observed that GCRV-II induced different types (M1 and M2) of polarization of Mo/Mφs in multiple tissues, especially in the brain, head kidney, and intestine. To assess the impact of different types of polarization on GCRV-II replication, we recombinantly expressed and purified the intact cytokines CiIFN-γ2, CiIL-4/13A, and CiIL-4/13B and successfully induced M1 and M2 type polarization of macrophages using these cytokines through in vitro experiments. qRT-PCR, WB, and flow cytometry analyses showed that M2 macrophages had higher susceptibility to GCRV-II infection than other types of Mo/Mφs. In addition, we found GCRV-II induced apoptosis of Mo/Mφs to facilitate virus replication and dissemination and also detected the presence of GCRV-II virus in plasma. Collectively, our findings indicated that GCRV-II could invade immune cells Mo/Mφs and induce apoptosis and polarization of Mo/Mφs for efficient infection and dissemination, emphasizing the crucial role of Mo/Mφs as a vector for GCRV-II infection.

IMPORTANCE

Type II grass carp reovirus (GCRV) is a prevalent viral strain and causes huge losses in aquaculture. However, the related dissemination pathway and mechanism remain largely unclear. Here, our study focused on phagocytic immune cells, monocytes/macrophages (Mo/Mφs) in blood and tissues, and explored whether GCRV-II can invade Mo/Mφs and replicate and disseminate via Mo/Mφs with their differentiated type M1 and M2 macrophages. Our findings demonstrated that GCRV-II infected Mo/Mφs and replicated in them. Furthermore, GCRV-II infection induces an increased number of M1 and M2 macrophages in grass carp tissues and a higher viral load in M2 macrophages. Furthermore, GCRV-II induced Mo/Mφs apoptosis to release viruses, eventually infecting more cells. Our study identified Mo/Mφs as crucial components in the pathway of GCRV-II dissemination and provides a solid foundation for the development of treatment strategies for GCRV-II infection.

INTRODUCTION

Grass carp reovirus (GCRV), a member of the family Reoviridae, is one of the pathogens for grass carp hemorrhagic disease, causing huge losses in the aquaculture industry (1). As a double-stranded RNA virus, GCRV consists of seven structural proteins and six non-structural proteins encoded by 11 genome segments (2). Based on the complete sequence of RNA-dependent RNA polymerase and the available VP6 core clamp protein gene sequence, GCRV has been classified into three genotypes: genotype I (GCRV-I), genotype II (GCRV-II), and genotype III (GCRV-III) (1). Among them, GCRV-I is the earliest reported type and has been widely investigated (3, 4). At present, GCRV-II is the main epidemic and the most virulent strain type in China (5).

However, little is known about the invasion and dissemination of GCRV-II in grass carp and its target tissues. Our recent study has indicated that the brain of grass carp is the primary target of GCRV-II invasion (6). The pathway of the nostril-olfactory system-brain axis has been identified as a reliable route for GCRV-II propagation to the brain (7). In addition, peripheral blood cells, which can utilize the bloodstream to efficiently reach different tissues in the body, are considered important vehicles for viral infection and tissue dissemination (8, 9). Although GCRV-II has been proven to be infectious in various cell lines such as CIK (Ctenopharyngodon idella kidney) (2), GCO (grass carp ovary) (10), grass carp swimming bladder (11), and grass carp liver cell line 8824 (11), the specific primary blood cells infected by GCRV-II remain to be investigated. Our previous study has indicated that GCRV-II can invade leukocytes but infect no erythrocytes and thrombocytes in grass carp (12). Among leukocytes, monocytes/macrophages (Mo/Mφs) are representative phagocytes in both mammals and teleosts, and their role in viral invasion has been extensively recognized (13). Mo/Mφs have been found to serve as cellular reservoirs for HIV and play a crucial role in persistent HIV infection (14, 15). Similarly, Mo/Mφs have been confirmed to function in persistent EBOV infection in the eyes and brain of surviving rhesus macaques (16, 17). Mo/Mφs have also been identified as target cells for herpes simplex virus 1 (HSV-1) in the corneas of mice (18). These findings collectively highlight the significant involvement of Mo/Mφs in viral invasion.

Indeed, when pathogens invade the host organism, Mo/Mφs in peripheral blood are rapidly recruited to the site of infection to compensate for the absence of tissue-resident macrophages (19). Upon migration into tissues and differentiation into macrophages, monocytes undergo different types of activation depending on the microenvironment, such as classical activation (M1 type) or alternative activation (M2 type) (20). Consequently, the susceptibility of Mo/Mφs to viral infection varies with cell types. Tissue-resident macrophages exhibit higher susceptibility to HIV than Mo/Mφs isolated from blood (21). Generally, macrophages have a longer lifespan and are more resistant to cytopathic effects than monocytes, and thus macrophages support sustained viral infection (14). In mice infected with HSV-1, M1 macrophages exhibit lower susceptibility to HSV-1 infection and replication than M2 or unpolarized macrophages (18). These findings illustrate the inseparable nature among monocytes, M1, and M2 macrophages when the relationship between the virus and Mo/Mφs is investigated. Based on the differences in virus infection efficiency among differently polarized macrophage types, the development of medicines that induce macrophage polarization in the antiviral direction can be an effective method for treating viral diseases. For instance, baicalin, a natural compound, can trigger the polarization of mouse macrophages toward M1, thus significantly inhibiting H1N1 A virus infection (22). Such therapeutic approaches exhibit a great potential to combat grass carp hemorrhagic disease. Therefore, exploring whether GCRV-II infects Mo/Mφs and revealing the differences in GCRV-II infection among different polarization types of macrophages are of great importance.

In this study, we found that GCRV-II infected peripheral blood Mo/Mφs in grass carp, and it induced apoptosis of Mo/Mφs and polarization of tissue macrophages. The in vitro macrophage polarization assays indicated that M2-type macrophages exhibited a higher susceptibility to GCRV-II than other Mo/Mφs. Inducing apoptosis was another way for GCRV-II to enhance its replication and dissemination. Taken together, GCRV-II invades monocytes/macrophages and induces their polarization and apoptosis to facilitate viral replication and dissemination. The high susceptibility of M2 macrophages to GCRV-II provides a theoretical foundation for the development of strategies for controlling hemorrhagic disease of grass carp.

MATERIALS AND METHODS

Cell isolation and culture

Grass carp were acquired from a fish farm in Huanggang City (Hubei Province, China). Two groups of fish were used: one weighing 500 ± 25 g for the isolation of Mo/Mφs and another weighing 30 ± 5 g for the virus challenge experiments in vivo. Prior to the experiments, the fish were acclimatized for a period of 2 weeks. The isolation of peripheral blood and head kidney monocytes/macrophages (Mo/Mφs) was performed with slight modifications based on previously described methods (23, 24). Briefly, aseptic blood samples were collected in phosphate-buffered saline (PBS) containing 1% heparin wetting, and the head kidney was removed aseptically and passed through a 100 µm mesh in PBS. Preliminary Mo/Mφs were obtained through density gradient centrifugation using 34%/51% Percoll (Pharmacia, Sweden), followed by purification through adherence. Both Mo/Mφs and CIK cells were cultured in L-15 medium (Life Technologies, USA) supplemented with 10% fetal bovine serum (FBS) and 200 IU/mL penicillin-streptomycin (Amresco, USA). The cells were maintained in an incubator at 28°C with a humidified atmosphere of 5% CO₂. The morphology and purity of the isolated cells were observed using Giemsa staining and transmission electron microscopy (TEM).

Viral infection

For the viral challenge experiment, Mo/Mφs and CIK cells were adjusted to 1 × 10⁶ cells/mL. After being washed three times with PBS, the pretreated cells were infected with GCRV-097 at a multiplicity of infection (MOI) of 1, following the previously described protocol (12). The cells were collected at 0, 12, 24, 36, 48, and 60 h after GCRV-II infection. Grass carp were challenged with GCRV-097 (1 × 10⁶ TCID₅₀/mL, 4 µL/g) as described previously (25). The control animals were injected with an equal volume of PBS. Six individuals were sacrificed, and tissues were harvested at 0 and 60 h post-infection.

RNA isolation, RT-PCR, and qRT-PCR

Total RNAs were extracted with Trizol (Simgen, China) and transcribed into cDNA using the ABScript III RT Master Mix (ABclonal, USA). RT-PCR was performed using gene-specific primers (Table S1), and the PCR products were subjected to gel electrophoresis. qRT-PCR was performed using the BioEasy Master Mix (Hangzhou Bioer Technology Co., Ltd., China) on a Roche LightCycler 480 system, and EF1α and 18s rRNA were employed as an internal control gene for cDNA normalization (26). The primer sequences used in the study are listed in Table S1.

Western blot analysis

Cells were lysed using 4× SDS loading buffer and denatured at 95°C for 10 minutes. Protein samples were separated by SDS-PAGE gels and subsequently transferred to nitrocellulose membranes (Millipore, USA). Western blot (WB) detection of VP4 was performed using a mouse anti-VP4 antibody (1:1,000) with a goat anti-mouse IgG-HRP antibody (1:3,000, AS066, ABclonal) as the secondary antibody. The anti-VP4 mouse polyclonal antibody was prepared and conserved by our lab (27). Other antibodies used in the study included β-actin mouse mAb (1:10,000, AC004, ABclonal), arginase-2 rabbit pAb (1:500, A6355, ABclonal), iNOS rabbit mAb (1:500, A3774, ABclonal), Bcl-2 rabbit pAb (1:5,000, A0208, ABclonal), HRP rabbit anti-goat IgG (1:5,000, AS029, ABclonal), p-STAT5 rabbit mAb (1:2,000, AP0758, ABclonal), p-ERK rabbit pAb (1:500, AP0472, ABclonal), and p-mTOR rabbit pAb (1:500, AP0094, ABclonal). Protein bands were visualized using a chemiluminescent ECL substrate (GE Healthcare, USA).

Indirect immunofluorescent microscopy

In cellular experiments, cells were plated on glass coverslips in 24-well plates and infected with GCRV-097 at an MOI of 1 for 48 h. After being washed three times with PBS, the mock and infected cells on slides were fixed with 4% paraformaldehyde for 10 minutes and permeabilized with 0.1% Triton X-100 for 10 minutes at room temperature. Next, the pretreated cells were blocked with 2% BSA at 37°C for 2 h. After being washed three times with PBST, the slides were incubated with mouse anti-VP4 antibodies (1:1,000) for 2 h, followed by incubation with FITC-conjugated goat anti-rabbit IgG (1:200, AS011, ABclonal) for 45 minutes at room temperature. The nuclei were stained with 1 µg/mL Hoechst 33342 at room temperature for 10 minutes. The images were taken by OlyVIA FV1000.

Immunohistochemistry of frozen sections

In in vivo experiments, grass carp were challenged with GCRV-097 (1 × 10⁶ TCID₅₀/mL, 4 µL/g) for 60 h. Subsequently, the head kidney, brain, and intestines were collected and fixed in 4% paraformaldehyde for 24 h, then encased in OTC (Sakura, USA), frozen overnight at −80°C, and subsequently sliced in freezing microtome (Leica, Germany). The frozen tissue sections were stained with rabbit anti-CSF-1R antibodies (1:100), arginase-2 rabbit pAb (1:500, A6355, ABclonal), or iNOS rabbit mAb (1:500, A3774, ABclonal), following the instructions of the TSA Fluorescence Double Staining Kit (RK05902, ABclonal). The anti-CSF-1R rabbit polyclonal antibody was prepared and conserved by our lab (28). The images were taken by OlyVIA FV1000.

Transmission electron microscopy

To observe the virus particles in the cells, Mo/Mφs and CIK cells that were infected with GCRV-097 were fixed with 2.5% glutaraldehyde for 24 h. Ultrathin sections of the cells were prepared following the previously described method (12). The images of the ultrathin sections were captured using an HT-7700 Transmission Electron Microscope (Hitachi, Japan).

Flow cytometry analyses

Flow cytometry analysis was performed to determine the infection rates of GCRV-II. After GCRV-097 infection, Mo/Mφs, CIK cells, M1, and M2 macrophages were collected into centrifuge tubes by trypsinization. The collected cells were then fixed with paraformaldehyde for 10 minutes and permeabilized by incubating in Triton X-100 for 10 minutes at room temperature. Next, the cells were blocked with 2% BSA for 2 h and incubated with anti-VP4 antibodies (1:500) for 2 h. Subsequently, FITC-conjugated goat anti-mouse IgG (1:50, ABclonal, USA) was added and incubated for 1 h. Finally, the cells were analyzed using flow cytometry (Beckman Coulter, USA), counting 10,000 cells per sample.

Recombinant expression and purification of intact CiIFN-γ2, CiIL-4/13A, and CiIL-4/13B

Recombinant expression and purification of CiIFN-γ2, CiIL-4/13A, and CiIL-4/13B were performed as described previously with slight modifications (29–31). Briefly, the sequences encoding CiIFN-γ2 were subcloned into the pGEX-4T1 vector (Novagen), while the sequences encoding CiIL-4/13A and CiIL-4/13B were subcloned into the pEHISTEVb vector (Novagen). The resultant plasmids were verified by sequencing and transformed into the Escherichia coli BL21 (DE3) cells (Novagen). The recombinant expression proteins were produced after induction with IPTG and purified by desalting column (17-5087-01, GE Healthcare). Subsequently, CiIFN-γ2, CiIL-4/13A, and CiIL-4/13B were cleaved by enterokinase (2 U/mg protein; PE001-01A, Novoprotein Scientific) at 37°C overnight, respectively. The recombinant proteins were then separated by HiTrapTM SP HP (10275571, GE Healthcare) and dialyzed to buffer D [20 mM Tris (pH 7.4)]. Finally, the recombinant proteins were concentrated using Amicon Ultra centrifugal filters (UFC900324, MilliporeSigma, Germany) and stored at −80°C. The purity and size of CiIFN-γ2, CiIL-4/13A, and CiIL-4/13B were confirmed by SDS-PAGE.

Activation of macrophages in vitro

Mo/Mφs were polarized as described previously with slight modifications (30, 31). Briefly, Mo/Mφs were stimulated with recombinant IFN-γ2 (350 ng/mL) and LPS (30 µg/mL) for 12 h to M1 type or CiIL-4/13A (200 ng/mL) and CiIL-4/13B (200 ng/mL) to M2 type. Cells were harvested and total RNAs were isolated. The marker of M1 polarization (high expression of IL-1β, CXCR 3.1, and iNOS) and phenotype of M2 macrophages (high level of IL-10, CXCR 3.2, and arginase-2) were detected by qRT-PCR. The activity of iNOS was detected according to the instructions of Nitric Oxide Synthase Assay Kit (S0025, Beyotime, China). The activity of arginase was detected using the Fish Arginase ELISA kit (MM92666801, Meimian, China).

Apoptosis detection

Cells were prepared in 6-well plates and infected with GCRV-097 at an MOI of 1. After being washed three times with PBS, cells were collected into a centrifuge tube by trypsinization, followed by staining with Annexin V-FITC/PI Apoptosis Detection Kit (MedChemExpress, USA), and analyzed by flow cytometry. To further detect the occurrence of apoptosis, cells were cultured in confocal dishes and subsequently infected with GCRV-097 at an MOI of 1. Following three washes with PBS, cells were stained with 1 µg/mL Hoechst 33258 at room temperature for 10 minutes. Imaging was performed using the OlyVIA FV1000 system.

Inhibition and activation of apoptosis

BH3 hydrochloride (HY-P2343, MedChemExpress, USA) and Ilexsaponin A (HY-N2638, MedChemExpress, USA) were used as agonist or antagonist in apoptosis experiments, respectively. In in vitro experiments, cells were plated in 24-well plates and incubated with 200 ng/mL Ilexsaponin A and 200 ng/mL BH3 hydrochloride for 3 h as described previously, respectively (32–34). Subsequently, the treated cells were infected with GCRV-097 at an MOI of 1. Cell samples were collected at 60 h post-viral infection, and changes in viral load were verified through qRT-PCR and WB. Apoptosis rates were assessed using flow cytometry.

In in vivo experiments, grass carp were intraperitoneally injected with 10 µg/g BH3 hydrochloride or 10 µg/g of A Ilexsaponin (PBS as control) for 3 h as described previously (32–34) and then intraperitoneally injected with GCRV-097 (1 × 10⁶ TCID₅₀/mL, 4 µL/g) for 60 h. Six individuals were sacrificed, and tissues were harvested at 60 h post-viral infection. qRT-PCR was used to detect changes in viral loads in tissues.

Statistical analyses

Statistical analyses and presentation graphics were carried out using the GraphPad Prism 8.0 software. Results were presented as mean ± standard deviation (SD) for at least three independent experiments. The P-value was analyzed by Student’s t test or one-way analysis of variance with Dunnett’s post hoc test. For all tests, P values below 0.05 were regarded as being significantly different (*P < 0.05 and **P < 0.01).

RESULTS

GCRV-II infects peripheral blood Mo/Mφs in grass carp

To obtain relatively pure cells, Mo/Mφs were isolated from blood by Percoll density gradient centrifugation. The morphology of resulting cells was investigated by Giemsa staining and TEM. The cells were round or oval with eccentric nuclei, which was consistent with the description of typical Mo/Mφs (Fig. 1A and B). Since CD11b and CSF-1R are commonly used as surface marker genes of Mo/Mφs (28, 35), we further examined the expression of multiple marker genes of Mo/Mφs and peripheral blood leukocytes (PBLs) using RT-PCR. As expected, Mo/Mφs expressed CD11b and CSF-1R, but did not express other no-Mo/Mφ surface marker genes, which indicated relatively high purity of our isolated Mo/Mφs (Fig. 1C).

Fig 1.

Mo/Mφ infection by GCRV-II in grass carp. (A) Morphology of Mo/Mφs after Giemsa staining under light microscopy. Scale bar = 20 µm. (B) Morphology of Mo/Mφs observed by TEM. N, nucleus; Scale bar = 2 μm. (C) Identification of cell purity. The specific marker genes were detected by RT-PCR in PBLs, Mo/Mφs, and no template control (N.T.C.). (D and E) Viral RNAs and protein determined by RT-PCR (D) and WB (E) after 48-h challenge of CIK cells and Mo/Mφs with GCRV-097 (MOI = 1). β-actin was employed as an internal control. (F) Subcellular localization of GCRV-II in CIK cells and Mo/Mφs under confocal microscopy. CIK cells and Mo/Mφs were challenged with GCRV-097 (MOI = 1) for 48 h. Blue represents the nucleus, and green indicates GCRV-II. (G) Mo/Mφs at indicated time points post-GCRV-097 challenge (MOI = 1) observed by TEM. The arrows point at the rupture of the cytomembrane. Scale bar = 2 μm.

Viruses in the organism typically require mediators to approach and infect blood cells (36). In this study, we observed GCRV-II virus in infected grass carp plasma, indicating that the presence of GCRV-II virus in plasma was a prerequisite for the infection of Mo/Mφs by GCRV-II in the grass carp (Fig. S1). Structural proteins VP4 (major outer capsid protein) and VP56 (fibrin) were used as the markers of GCRV-II in this study (37). To determine whether GCRV-II infected peripheral blood Mo/Mφs, WB and RT-PCR assays were performed on Mo/Mφs and CIK cells from mock or infected group. The results showed that mRNA and protein expressions of VP4 were detected in peripheral blood Mo/Mφs post-GCRV infection (Fig. 1D and E). The data of immunofluorescence showed the presence of VP4 protein in infected Mo/Mφs, which further confirmed the ability of GCRV-II to infect Mo/Mφs (Fig. 1F). In addition, TEM results revealed that cell membrane ruptured at 24 and 48 h post-GCRV-II infection but no syncytia (hallmark of cytopathic effects) formed, (Fig. 1G). At 24 h post-GCRV-II infection, virus particles with a diameter of 70–80 nm were scattered in the cytoplasm (Fig. 2A and C). At 48 h post-GCRV-II infection, an increased number of GCRV-II virus-like particles were observed in Mo/Mφs. These virus particles appeared to be enveloped by distinct membrane structures and exhibited a less dense interior, showing vacuolation, and they were notably different from the viral inclusion bodies lacking obvious membrane encapsulation in CIK cells (Fig. 2B and D). Additionally, substantial lipid droplets were observed in Mo/Mφs at 24 h after GCRV-II infection (Fig. 2C). Taken together, these results suggested that the peripheral blood Mo/Mφs were the target cells for GCRV-II invasion in grass carp.

Fig 2.

TEM observation of CIK cells and Mo/Mφs post-GCRV-097 (MOI = 1) challenge for the indicated time. N, nucleus; LD, lipid droplets; and GCRV, GCRV-II particles. (A) TEM observation of mock Mo/Mφs. Scale bar = 2 μm. (B) GCRV-II-infected CIK cells under TEM. Scale bar = 200 nm. (C and D) GCRV-infected Mo/Mφs for 24 h (C) or 48 h (D) under TEM. Scale bar = 2 μm. The right panel is the magnification of TEM images. Scale bar = 200 nm.

GCRV-II replication increases in Mo/Mφs

To assess the replication of GCRV-II in peripheral blood Mo/Mφs, qRT-PCR and WB assays were performed. The qRT-PCR results showed a significant upregulation in the expression levels of GCRV-II VP4 and VP56 mRNA post-GCRV-II infection, particularly at 48 and 60 h (Fig. 3A and B). Consistent with the results of qRT-PCR, WB analysis demonstrated an increase in VP4 protein expression at 48 and 60 h after GCRV-II infection (Fig. 3C). Furthermore, flow cytometry assay indicated a significantly higher percentage of FITC-labeled GCRV-II-positive Mo/Mφs at 48 and 60 h post-infection than the control group (Fig. 3D through G). Interestingly, the GCRV-II positivity rate of Mo/Mφs was significantly lower than that of CIK cells at 60 h post-infection (Fig. 3E and G). Collectively, these results indicated that GCRV-II was capable of replicating in Mo/Mφs, with a notable increase in replication efficiency at 48 and 60 h post-infection.

Fig 3.

GCRV-II replication in Mo/Mφs. (A and B) Relative mRNA expression levels of GCRV-II VP4 (A) and VP56 (B) in Mo/Mφs after challenge with GCRV-097 (MOI = 1) for the indicated time, determined by qRT-PCR. The relative mRNA expression level was normalized to that of the EF1a gene (n = 3). (C) Expression of GCRV-II VP4 protein in Mo/Mφs after challenge with GCRV-097 (MOI = 1) for the indicated time, determined by WB. (D–F) Percentages of GCRV-II-positive cells, determined by flow cytometry assay. (D) Mouse IgG group was used as a negative control. (E) Percentages of GCRV-II-positive cells in CIK cells after 60-h challenge with GCRV-097 (MOI = 1). (F) Percentages of GCRV-II-positive cells in Mo/Mφs after challenge with GCRV-097 (MOI = 1) for the indicated time. (G) Percentages of GCRV-II-positive cells from three separate experiments (*P < 0.05 and **P < 0.01).

GCRV-II induces tissue macrophage polarization

Upon entering tissues, peripheral blood Mo/Mφs rapidly differentiate into different types of macrophages, with M1 and M2 being the typical differentiation types (20). iNOS and arginase-2 commonly serve as markers for M1 and M2 macrophages, respectively (38), and IFN-γ and IL-4/13 are upstream regulators of these markers (39, 40). To investigate the effects of GCRV-II invasion into grass carp on tissue macrophage polarization, we examined the mRNA expression levels of these marker genes in various tissues of grass carp at 60 h post-infection. The qRT-PCR results showed a significant upregulation of iNOS expression in the brain, head kidney, spleen, and intestine in the GCRV-II-infected group, relative to the PBS group. Additionally, the expression of arginase-2 was significantly increased in the brain, head kidney, and heart (Fig. 4A). Moreover, IFN-γ2 expression was significantly elevated in the eyes, brain, head kidney, spleen, and intestine, and IL-4/13A expression was also increased in the eyes, brain, head kidney, heart, and intestine, whereas IL-4/13B expression was upregulated only in the brain (Fig. 4B). These results showed significant mRNA expression changes of polarization-associated factors in multiple tissues, especially in the brain, head kidney, and intestine.

Fig 4.

GCRV-II-induced tissue macrophage polarization. (A) Relative mRNA expression levels of iNOS and arginase-2, the marker genes for M1 and M2 macrophages, respectively, in different tissues after 60-h challenge with GCRV-097, determined by qRT-PCR (n = 6). (B) Relative mRNA expression levels of IFN-γ2, IL-4/13A, and IL-4/13B, the marker genes inducing polarization toward M1 (IFN-γ2) and M2 (IL-4/13A and IL-4/13B), respectively, in different tissues after being challenged with GCRV-097 for 60 h, determined by qRT-PCR. The mRNA relative expression level was normalized to that of the 18s rRNA gene (n = 6). (C) Antibody specificity validation of iNOS (131 kDa) and arginase-2 (39 kDa). (D and E) Expression of iNOS and arginase-2 proteins in head kidney macrophages (D) and peripheral blood Mo/Mφs (E), determined by WB. The Mo/Mφs were isolated from the head kidney and peripheral blood of grass carp at 60 h after injection with PBS or GCRV-097 (1 × 10⁶ TCID50/mL, 4 μL/g). (F and G) Expression levels of iNOS and arginase-2 proteins in head kidney macrophages (F) and blood Mo/Mφs (G), determined by flow cytometry assay. (H and I) Relative fluorescence intensity of iNOS and arginase-2 in brain Mo/Mφs (H) and intestine Mo/Mφs (I), measured by ImageJ software, corresponding to the results of immunohistochemistry in Fig. S2. Brain and intestine were collected from the infected and control groups (n = 3) (*P < 0.05 and **P < 0.01).

The changes in mRNA expression of polarization-associated factors in tissues imply alterations in the number of polarized macrophages in tissues. Our previous studies have found that the brain, head kidney, and intestine are the tissues with the high GCRV-II viral loads or the severe lesions (6, 7). In this study, peripheral blood Mo/Mφs were found to be target cells of GCRV-II. Based on these findings, we further explored whether polarization of Mo/Mφs occurred in blood, head kidney, brain, and intestine. We isolated head kidney macrophages and blood Mo/Mφs from the GCRV-II-infected grass carp. The specificity of the iNOS and arginase-2 antibodies was first verified (Fig. 4C). WB results showed a significantly higher expression of both iNOS and arginase-2 in the head kidney macrophages in the GCRV-infected group than in the uninfected group, but no significant difference was observed in blood Mo/Mφs (Fig. 4D and E), which was further confirmed by flow cytometry assay results (Fig. 4F and G). Such different results between blood Mo/Mφs and head kidney macrophages might be due to the undifferentiated state of Mo/Mφs in peripheral blood. Since it was difficult to isolate macrophages from the brain and intestine due to these tissue structure characteristics, we had to investigate the changes in these tissue macrophages by measuring the fluorescence intensity of iNOS and arginase-2 around CSF-1R-labeled macrophages from frozen sections through immunohistochemistry assay (Fig. 4H and I; Fig. S2). The immunohistochemistry assay results showed that both iNOS and arginase-2 proteins were significantly upregulated in macrophages in the brain and intestine, indicating that GCRV-II induced the tissue macrophage polarization in vivo (Fig. 4H and I; Fig. S2). However, our in vitro experiments showed that GCRV-II infection resulted in no expression change of polarization-associated factors, implying that GCRV-II could not induce Mo/Mφs polarization (Fig. S3) This might be due to the fact that Mo/Mφs alone cannot produce polarization-stimulating factors in response to pathogen invasion. Actually, IFN‐γ2 is primarily generated by natural killer (NK) cells, natural killer T cells, CD8, and CD4 Th1 effector T cells; IL-4/13A and IL-4/13B are mainly produced by T lymphocytes, NK cells, mast cells, basophils, normal and malignant B cells (41, 42). Overall, these results indicated a significant increase in the number of M1 or M2 macrophages after GCRV-II infection, especially in the brain, head kidney, and intestine, suggesting that GCRV-II could induce polarization of tissue macrophages.

M2 macrophages exhibit higher susceptibility to GCRV-II

To determine whether the replication of GCRV-II differed in the Mo/Mφs, M1, and M2 macrophages, we first established an in vitro macrophage polarization model. Classically activated M1 macrophages are induced by LPS and IFN-γ2, while alternatively activated M2 macrophages are induced by IL-4 and IL-13 (43). In this study, recombinant proteins CiIFN-γ2 (17.6 kDa), CiIL-4/13A (17.2 kDa), and CiIL-4/13B (17.1 kDa) were produced in a prokaryotic expression system and were confirmed by WB analysis (Fig. 5A). The addition of these recombinant proteins into Mo/Mφs resulted in the upregulation of the phosphorylation levels of STAT5, ERK, and mTOR, indicating the high biological activities of these recombinant proteins (Fig. 5B and C). Next, co-stimulation with IFN-γ2 and LPS polarized Mo/Mφs into M1 macrophages, while co-stimulation with IL-4/13A and IL-4/13B polarized Mo/Mφs into M2 macrophages. M1 macrophages from IFN-γ2 and LPS co-stimulation exhibited elevated expression of IL-1β, CXCR 3.1, and iNOS, accompanied by increased iNOS activity (Fig. 5D and E). M2 macrophages from IL-4/13A and IL-4/13B co-stimulation displayed increased expression of IL-10, CXCR 3.2, and arginase-2, along with enhanced arginase-2 activity (Fig. 5F and G). These results indicated the successful construction of the M1 and M2 macrophage polarization models.

Fig 5.

Polarization of M1 and M2 macrophages in vitro. (A) SDS-PAGE analysis of purified recombinant intact CiIFN-γ2 (17.6 kDa), CiIL-4/13A (17.2 kDa), and CiIL-4/13B (17.1 kDa) proteins. (B and C) Measurement of CiIFN-γ2, CiIL-4/13A, and CiIL-4/13B protein activity by determining the phosphorylation levels of STAT5, ERK, and mTOR. (C) Measurement of phosphorylation levels of STAT5, ERK, and mTOR in three separate experiments by determining the gray value of the WB bands of CiIFN-γ2, CiIL-4/13A, and CiIL-4/13B proteins using the ImageJ software. (D and E) Verification of M1 macrophage polarization by mRNA expression levels of IL-1β, CXCR 3.1, and iNOS (D), as well as enzyme activity of iNOS (E). (F and G) Verification of M2 macrophage polarization by mRNA expression levels of IL-10, CXCR 3.2, and arginase-2 (F), as well as arginase-2 enzyme activity (G). Relative mRNA expression level was normalized to that of EF1a (n = 3) (*P < 0.05 and **P < 0.01).

Our in vitro experiments showed that GCRV-II infection induced no significant changes in the macrophage polarization type, thus ruling out the direct effect of GCRV-II on Mo/Mφs polarization (Fig. S3). However, GCRV-II resulted in macrophage polarization in vivo (Fig. 4). The different experiment results between in vivo and in vitro might be attributed to the fact that GCRV-II requires the cytokines released from other cells rather than from Mo/Mφs to perform the polarization function. Then, the polarized macrophages and Mo/Mφs (unpolarized) were infected with GCRV-097, as described above, and the replication of the virus was analyzed using qRT-PCR and WB assays at 24 or 48 h post-infection. The results showed that mRNA expression levels of VP4 and VP56 were significantly higher in M2 macrophages than in M1 or unpolarized macrophages at all time points (Fig. 6A and B). In contrast, the VP4 mRNA expression in the M1 macrophages was similar to that in unpolarized Mo/Mφs (Fig. 6A and B). The WB analysis showed similar results, indicating that the expression of VP4 protein was higher in M2 macrophages than in M1 and unpolarized macrophages at 48 h post-infection (Fig. 6C). Furthermore, flow cytometry analysis further supported these results that the percentage of FITC-labeled GCRV-II-positive cells was significantly higher in M2 macrophages than in M1 and Mo/Mφs at all time points, while the percentage of GCRV-II-positive cells was not significantly different between M1 and unstimulated cells (Fig. 6D and E). In summary, these results suggested that M2 macrophages exhibited higher susceptibility to GCRV-II infection than M1 macrophages and unpolarized Mo/Mφs.

Fig 6.

High susceptibility of M2 macrophages to GCRV-II infection. (A and B) Relative mRNA expression levels of GCRV-II VP4 (A) and VP56 (B) in Mo/Mφs, M1, and M2 macrophages after being challenged with GCRV-097 (MOI = 1) for the indicated time, determined by qRT-PCR. The relative mRNA expression level was normalized to that of the EF1a gene (n = 3) (*P < 0.05 and **P < 0.01). (C) Expression levels of GCRV-II VP4 protein in Mo/Mφs, M1, and M2 macrophages after being challenged with GCRV-097 (MOI = 1) for 48 h, determined by WB. (D) Percentage of GCRV-II-positive cells in Mo/Mφs, M1, and M2 macrophages after being challenged with GCRV-097 (MOI = 1) for 24 and 48 h, determined by flow cytometry assay. Non-specific fluorescence was eliminated through mouse IgG. (E) Percentage of GCRV-II-positive cells from three separate experiments.

GCRV-II induces Mo/Mφs apoptosis to enhance self-replication

GCRV-II does not cause cytopathic effect in CIK and GCO cells (44). Interestingly, we detected GCRV-II mRNA in the plasma of grass carp at 60 h post-infection (Fig. S1). Therefore, we further examined how it was released from the infected cells. Programmed cell death is often seen as a pathway for viral release (45–47), and thus, we investigated the apoptosis of Mo/Mφs post-GCRV-II infection. Flow cytometry results demonstrated that GCRV-II promoted apoptosis of Mo/Mφs at 60 h post-infection (Fig. 7A and B). Additionally, we also examined the expression levels of Caspase-2, Caspase-7, Bcl2, and Mcl-1 genes associated with apoptosis (47). Both qRT-PCR and WB analyses showed that the expressions of genes promoting apoptosis (Caspase-2 and Caspase-7) were upregulated, while those of genes inhibiting apoptosis (Bcl2 and Mcl-1) were downregulated (Fig. 7C and D). To verify this result, we performed Hoechst 33258 staining. The staining results showed chromatin condensation and nuclear fragmentation in Mo/Mφs at 60 h post-GCRV-II infection, which represented typical apoptosis (Fig. 7E). These results indicated that GCRV-II induced apoptosis of Mo/Mφs.

Fig 7.

GCRV-II induces Mo/Mφs apoptosis to enhance self-replication. (A) Apoptosis rate of mock and infected Mo/Mφs, determined by flow cytometry assay. (B) Apoptosis rate of Mo/Mφs in three independent experiments. (C) Bcl-2 protein expression levels in Mo/Mφs at 60-h GCRV-097 (MOI = 1) challenge, determined by WB. (D) mRNA expression levels of Caspase-2, Caspase-7, Bcl2, and Mcl-1 in Mo/Mφs at 60-h post-GCRV-097 challenge (MOI = 1), determined by qRT-PCR. The relative mRNA level was normalized to that of the EF1a gene (n = 3). (E) Hoechst 33258 staining of Mo/Mφs at 60-h post-GCRV-097 challenge (MOI = 1). Scale bar = 10 μm. White arrows point to typical apoptotic symptoms of chromatin condensation, and red arrows point to typical apoptotic symptoms of nuclear fragmentation. (F) Incubation of Mo/Mφs with BH3 hydrochloride (BH3.h) (agonist) or Ilexsaponin A (antagonist) for 3 h and subsequent infection with GCRV-II (MOI = 1) for 60 h. The apoptosis rates of treated Mo/Mφs were measured by flow cytometry. (G and H) GCRV-II replication in Mo/Mφs, determined by qRT-PCR (G) and WB (H). Mo/Mφs were incubated with equal volumes of PBS, 200 ng/mL BH3.h (agonist), or 200 ng/mL Ilexsaponin A (antagonist) for 3 h and then infected with GCRV-II (MOI = 1) for 60 h. The relative mRNA expression level was normalized to that of the EF1a gene (n = 3). (I) GCRV-II replication in the brain, head kidney, intestine, and blood from different treatment groups, determined by qRT-PCR. Grass carp were injected intraperitoneally with equal volume of PBS, 10 µg/g BH3.h (agonist), or 10 µg/g Ilexsaponin A (antagonist) for 3 h and then infected with GCRV-097 (1 × 10⁶ TCID50/mL, 4 µL/g) for 60 h. The relative mRNA expression level was normalized to that of the 18s rRNA gene (n = 6) (*P < 0.05 and **P < 0.01).

To further explore the effect of apoptosis on GCRV replication, we treated Mo/Mφs for 3 h by adding agonist BH3 hydrochloride (BH3.h) or antagonist Ilexsaponin A to Mo/Mφs before infection with GCRV-II (Fig. 7F). The results showed that the addition of agonist or antagonist successfully promoted or inhibited the GCRV-II-induced apoptosis. The results of qRT-PCR and WB showed that the GCRV-II replication in Mo/Mφs was significantly enhanced after apoptosis promotion, but it was weakened after apoptosis inhibition (Fig. 7G and H). Furthermore, we validated these results in vivo. We found that GCRV-II replication in the brain, head kidney, intestine, and blood was significantly enhanced in the group injected with agonist BH3.h, but it was significantly weakened in the group injected with antagonist Ilexsaponin A, relative to that in the PBS group (Fig. 7I). These results indicated that GCRV-II could induce Mo/Mφs apoptosis to enhance self-replication and dissemination.

DISCUSSION

As obligate intracellular parasites, viruses require vectors to complete their life cycle and spread inside the host (48). Mo/Mφs have been reported as effective vectors of infection and replication for many viruses such as HIV-1, Ebola virus (EBOV), and African swine fever virus (ASFV). These viruses show high infectivity for Mo/Mφs regardless of their genotypes (49), although Mo/Mφs have been found to have certain viral clearance (38). Therefore, the particular role of Mo/Mφs during viral infection may vary with the types of viruses and infected host. In this study, we found that GCRV-II infected Mo/Mφs in the peripheral blood of grass carp and replicated in Mo/Mφs, indicating that Mo/Mφs could serve as vectors for GCRV-II replication and dissemination. Moreover, excessive replication of the virus leads to rapid death of infected cells, which is detrimental to the spread of the virus. Interestingly, we found that the replication of GCRV-II was significantly weaker in Mo/Mφs than in CIK cells under the same infection pattern, indicating that Mo/Mφs potentially functioned as a reservoir for GCRV-II replication just as they did for HIV infection (14). This makes us more inclined to consider Mo/Mφs as vectors for GCRV-II transmission rather than as sites for large-scale virus replication.

We found that M2 macrophages exhibited a higher susceptibility to GCRV-II replication than M1 macrophages or unpolarized Mo/Mφs. There are two potential reasons for this finding. First, Mo/Mφs in peripheral blood may not be mature enough to promote efficient virus replication, which has been supported in several previous reports that Mo/Mφs are generally resistant to HIV invasion but become permissive when they mature into macrophages (21, 50). Second, polarized M1 macrophages are typically considered as antiviral, and they combat viral infection through multiple antiviral strategies. M1 macrophages can produce nitric oxide to destroy viral ribonucleotide reductase activity, thus suppressing the replication of the vaccinia virus (51). Moreover, M1 macrophages can also release proinflammatory cytokines such as IL-1β, IL-6, and TNF-α (52), which exert antiviral activities directly or indirectly. The above two reasons could explain why M2 macrophages exhibited a higher susceptibility to GCRV-II in our study. Similar findings have been reported in one previous study that the replication level of HSV-1 was significantly higher in infected mice with overexpressed M2 macrophages than in wild-type and M2-deficient mice (38). Importantly, circulating Mo/Mφs in peripheral blood can be recruited into different tissues and differentiate into several types of macrophages, such as microglia, Kupffer cells, histocytes, and osteoclasts (53), which further polarize into M1 and M2 macrophages by cytokines. Consistently, we also found an increase in the number of M1 and M2 macrophages in grass carp tissues after GCRV-II infection. It is the preference of GCRV-II for M2-type macrophages that allows Mo/Mφs to avoid rapid cell death due to massive viral replication during the blood circulation process, thus ensuring efficient virus replication after GCRV-II entry into the target tissues. This is supported by our results that no macrophage polarization was observed in peripheral blood after viral infection. Therefore, GCRV-II exploits Mo/Mφs to safely arrive at the target tissues and induces the tissue macrophages’ polarization to facilitate viral replication, which confirms our speculation that Mo/Mφs might act as vehicles for GCRV-II dissemination.

Viruses have evolved diverse strategies to enhance their replication, such as actively altering the polarization of macrophages. Generally, pathogenic virus strains suppress the antiviral responses of M1 macrophages, while attenuated virus strains induce Mo/Mφs to adopt an M2-type polarization (54). An adequate equilibrium between M1 and M2 macrophages maintains the homeostatic milieu, whereas disequilibrium can lead to disease (55). In this study, we found that after GCRV-II infection, the number of polarized M1 and M2 macrophages was significantly increased in various tissues of grass carp, especially in the brain, head kidney, and intestine. This disruption of the M1/M2 macrophage equilibrium in these tissues might explain our previous findings that these sites exhibit high GCRV-II replication or severe hemorrhagic symptoms (6). These results indicated that GCRV-II could increase its pathogenicity by altering the polarized microenvironment in grass carp. Our data showed that GCRV-II failed to independently alter the polarization type of Mo/Mφs, which might be due to the inability of infected Mo/Mφs alone to produce the relevant cytokines regulating polarization (Fig. S3). It has been reported that viruses can actively induce apoptosis, thus causing the breakdown of infected cells eventually enhancing their replication and transmission (56). Virus replication can also be enhanced in macrophages by phagocytosing apoptotic cells (57). In this study, we found that GCRV-II induced apoptosis of Mo/Mφs, and that promotion and inhibition of apoptosis significantly increased and decreased viral replication, respectively, both in vivo and in vitro. This indicates that apoptosis is another strategy by which GCRV-II enhances its own replication. This strategy is particularly important for the release of GCRV-II from infected cells and its in vivo dissemination, in the case of the failure to induce a cytopathic effect, which is supported by the observation of the presence of GCRV-II virus in the plasma of infected grass carp. In addition, we found substantial lipid droplets in Mo/Mφs at 24 h post-infection with GCRV-II, indicating that GCRV-II induced the formation of lipid droplets in Mo/Mφs. Accumulation of lipid droplets during apoptosis serves as sites or scaffolds for GCRV-I replication and assembly (58, 59). Induction of lipid droplet formation may be the third strategy for GCRV-II to increase its infectivity in Mo/Mφs. However, whether the lipid droplets promote GCRV-II replication remains unknown, but this is definitely a fascinating question.

In conclusion, GCRV-II infects Mo/Mφs, exploits them to safely reach target tissues, and achieves massive replication by inducing apoptosis of Mo/Mφs and polarization of Mo/Mφs toward the preferred M2 phenotype. These results highlight the crucial role of Mo/Mφs in GCRV-II efficient infection and dissemination. Furthermore, the GCRV-II-induced disturbance of macrophage polarization equilibrium may lead to severe disease, which underscores the importance of maintaining macrophage polarization balance for the health of grass carp. Correspondingly, promoting polarization of macrophages toward the antiviral M1 phenotype may be the potential therapeutic strategy for grass carp hemorrhagic disease.

ETHICS APPROVAL

All the experiments were performed in accordance with the guidelines of the Laboratory Animal Center of Huazhong Agricultural University. The protocols were approved by the Ethical Committee on Animal Research at Huazhong Agricultural University (HZAUFI-2021-0035). All efforts were made to minimize animal suffering.

DATA AVAILABILITY

The data sets supporting the conclusions of this article are included within the article.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/jvi.01469-23.

Supplemental material. jvi.01469-23-s0001.doc.

Fig. S1 to S3; Table S1.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.