M1-type polarized macrophage contributes to brain damage through CXCR3.2/CXCL11 pathways after RGNNV infection in grouper

ABSTRACT

The vertebrate central nervous system (CNS) is the most complex system of the body. The CNS, especially the brain, is generally regarded as immune-privileged. However, the specialized immune strategies in the brain and how immune cells, specifically macrophages in the brain, respond to virus invasion remain poorly understood. Therefore, this study aimed to examine the potential immune response of macrophages in the brain of orange-spotted groupers (Epinephelus coioides) following red-spotted grouper nervous necrosis virus (RGNNV) infection. We observed that RGNNV induced macrophages to produce an inflammatory response in the brain of orange-spotted grouper, and the macrophages exhibited M1-type polarization after RGNNV infection. In addition, we found RGNNV-induced macrophage M1 polarization via the CXCR3.2- CXCL11 pathway. Furthermore, we observed that RGNNV triggered M1 polarization in macrophages, resulting in substantial proinflammatory cytokine production and subsequent damage to brain tissue. These findings reveal a unique mechanism for brain macrophage polarization, emphasizing their role in contributing to nervous tissue damage following viral infection in the CNS.

KEYWORDS:

• Central nervous system
• RGNNV
• macrophage
• CXCR3.2
• proinflammatory

Introduction

As bilateral animals evolved, their nervous systems underwent centralization, resulting in the formation of the central nervous system (CNS), which includes the spinal cord and brain. The CNS is the most complex system composed of the brain and spinal cord in vertebrates. The CNS controls various bodily functions, encompassing information storage and processing, movement, and responses to environmental stimuli. However, the CNS possesses a limited regenerative capacity, implying that even minor injuries can lead to severe ailments, functional impairment, and, in some cases, death [1,2].

Viral infections of the CNS are rare occurrences and typically self-limiting. However, these infections can cause potentially life-threatening neurological damage. Several virus types, including lymphocytic choriomeningitis virus (LCM), simian immunodeficiency virus, feline immunodeficiency virus (FIV), Zika virus, pseudorabies virus, tick-borne encephalitis virus, and human immunodeficiency virus (HIV), have been reported to cause severe damage to the CNS [3–9]. Recent studies in fish have demonstrated that viruses can induce CNS damage [10,11]. However, these studies primarily focused on the direct brain-damaging effect of the virus or the mechanism of virus invasion, with limited insight into the response of immune cells in the brain to viral intrusion. Therefore, investigating the development of specialized immune strategies in the brain following viral invasion is necessary.

Studying the effects of viral infections on the CNS of mammals and birds presents challenges; however, certain fish species serve as valuable experimental models. The aetiology of viral nervous necrosis (VNN) can be attributed to the nervous necrosis virus (NNV). NNV predominantly damages the nervous system of fish, resulting in viral encephalopathy and substantial economic losses to marine and freshwater fish farming industries worldwide [12,13]. NNV is classified into four clades: barfin flounder nervous necrosis,(BFNNV), tiger puffer nervous necrosis virus,(TPNNV), striped jack nervous necrosis virus,(SJNNV) and red-spotted grouper NNV (RGNNV) [14]. Of them, RGNNV is the most common pathogen associated with VNN in fish. Viral infection induces massive cytoplasmic vacuolization associated with innate immunity [15]. Furthermore, RGNNV infection is known to induce this effect in the retina and brain of infected fish [16]. Infection by several other viruses, including hepatitis C virus, hepatitis A virus, bovine virus, diarrhoea virus, mouse leukaemia virus, and Zika virus, have also been associated with cytoplasmic vacuolization [15,17]. In a previous study, we demonstrated an enrichment of brain macrophages induced by RGNNV [13]. However, the specific function of these neuroimmune cells following viral invasion remains unclear.

CNS-associated macrophages (CAMs) represent a pivotal category of immune cells in vertebrate CNS, fulfilling a crucial role in protecting the CNS against intruders and potential insults [18–20]. Numerous researchers have demonstrated, through in vitro and in vivo studies, that CAMs respond to pathogen-related molecular patterns [21]. Macrophage activation is frequently categorized into classical (M1) or alternative (M2) categories [22]. However, the precise role of CAMs in the immune response following viral infection remains poorly understood. According to Bragg et al., CAMs exhibited proliferation and released cytotoxic agents following FIV infection in cats [23]. This finding suggests that CAMs play a regulatory role in mitigating CNS dysfunction triggered by viral infections [24]. The accumulation of CAMs has been suggested to potentially cause tissue damage in vertebrates [6,25]. In contrast, during LCM infection, toxic T-cells have been observed to eliminate CAMs [26,27]. Nonetheless, the role of CAMs in viral infections of the CNS in these studies remains unclear. Therefore, understanding the function of brain macrophages in viral encephalitis is crucial.

Chemokines are members of a superfamily of chemotactic cytokines that play important roles in cell activation and migration [28–30]. Previous studies have revealed that macrophage activation typically occurs through polarization [19]. Moreover, studies have indicated that CXCR3 influences macrophage polarization in fish and mammals following pathogen infection [31,32]. Our previously study show that there are three CXCR3s (CXCR3.1, CXCR3.2, CXCR3.3) and two CXCL11s (CXCL11.1 and CXCL11.2) in grouper [33]. Also, Lu et al.. (2017) reported that the ligand of CXCR3 in fish was CXCL11 [32]. Although chemokines can initiate macrophage polarization, the precise pathways and underlying molecular mechanisms involved remain unclear.

In groupers (Epinephelus spp.), RGNNV is known to inflict brain damage, resulting in significant losses among natural grouper populations and imposing an economic burden on the aquaculture industry [34]. When groupers are cultured, they can produce large quantities of fry. Based on this evidence, RGNNV and groupers serve as valuable resources for investigating the immune response of brain macrophages during CNS viral infections. Therefore, this study aimed to examine the potential immune response of macrophages in the brain of orange-spotted groupers following RGNNV infection.

Materials and methods

Animals

Orange-spotted grouper (Epinephelus coioides) with an average body length of 3.24 ± 0.35 cm and a body weight of 3.66 ± 0.35 g were used in our experiments. The fish were reared in filtered seawater. The fish were fed a commercial grouper diet twice daily. To maintain water quality, the entire water volume (100%) was replaced daily. All animal experiments were conducted in accordance with the guidelines and approval was obtained from the appropriate Animal Research and Ethics Committees of South China Agricultural University (SHAUQ202211).

Primary culture of macrophages and primary brain cells

Grouper primary brain macrophages were isolated as previously described [35,36]. Briefly, the brains were aseptically removed and placed in a Petri dish containing an incomplete medium. Subsequently, the brains were homogenized in the same medium, transferred to a 15-mL tube, and suspended using a sterile Pasteur pipette for 1 min. The resulting cellular suspension was then transferred to a new 15-mL tube containing a Percoll gradient and centrifuged at 400 × g for 30 min. The brain cells were partially enriched for macrophages through separation on a 31–45% discontinuous Percoll gradient. The macrophages were collected at the 31–45% Percoll interface and then washed twice with L-15 medium (Gibco, USA) containing 0.1% FCS, 1% P/S, and 20 U ml⁻¹ heparin. The macrophages were cultured at 28 ºC in an L-15 medium (Gibco, USA) containing 10% FBS, 5% grouper serum, and 1% penicillin/streptomycin (P/S).

For the primary brain cell culture, the brain was aseptically removed and washed three times with Leibovitz’s L-15 medium (Gibco, USA). Following that, it was digested for 30 min with 1 mL of 0.25% trypsin solution (Gibco). Subsequently, the digested mixture was percolated through a 100-mesh screen, and the resulting filtrate was centrifuged at 180×g for 10 min to collect the precipitate. The precipitate was maintained at 28ºC and cultured in 3 mL of L-15 medium (Gibco, USA) containing 20% foetal bovine serum (FBS), 0.5% of 1 M N-2-Hydroxyethylpiperazine-N’-2 ethanesulfonic acid (HEPES), and 20 ng/mL basic fibroblast growth factor (bFGF).

Viruses

For viral stock propagation, a line of grouper spleen (GS) cells line was cultured in Leibovitz’s L-15 medium (Gibco, USA) containing 10% foetal bovine serum (Gibco) at 28°C. The cells were then inoculated with RGNNV for 2 h, washed three times, and supplemented with Leibovitz’s L-15 medium (Gibco, USA) and 10% FBS. After 48–60 h, the virus-containing cellular supernatant was collected and stored at −80°C until use. The RGNNV titre was 107 TCID50/ml.

RNA extraction and RT-qPCR

Total RNA was extracted from each sample using Trizol Reagent (Invitrogen, USA). The Transcriptor First Strand cDNA Synthesis Kit (Roche, Basel, Switzerland) was utilized to reverse transcribe total RNA and synthesize first-strand cDNA. A Roche LightCycler 480 real-time PCR system was employed for the PCR analysis. Each assay was conducted in triplicate with the following cycling conditions: 1 min for activation at 95°C, 15 s at 95°C, 15 s at 60°C, and 45 s at 72°C, totalling 40 cycles. The β-actin gene served as the internal control. The triplicate Ct values were calculated using the comparative Ct (ΔΔCt) method. The data are expressed as mean ± standard deviation. Table S1 lists the primers used in this study. They have been described previously.

Fluorescence in situ hybridization

The RGNNV capsid protein (CP) gene and the sense and antisense riboprobes of the macrophage marker gene mannose receptor 1 (Mrc1) were synthesized using a DIG RNA Labeling Kit (Roche Diagnostics, Germany). Table SI provides the CP and Mrc1 sequences. RNA FISH was performed as follows: The grouper brains were fixed in buffered 4% paraformaldehyde and then transferred to 30% sucrose solution until the samples sank. The samples were embedded in an optimal cutting temperature compound (Sakura, USA). The tissue was sliced 5 μm into sections at −20°C, covered, and incubated for 1 day at 42°C, as follows. First, prehybridized tissue sections were treated with a hybridization buffer containing 150 ng DIG-labelled CP and Mrc1, DIG-labelled sense only, antisense CP, or Mrc1 riboprobe only. Slides were incubated for 16 h at 55°C. After hybridization, sections were washed in 1 × saline-sodium citrate (SSC) and 0.1 × SSC for 1 h at 55°C. Finally, the sections were mounted using Fluoroshield and 6-diamidino-2-pheny-lindole (DAPI; Sigma-Aldrich, USA). Fluorescence signals from the FISH were captured using a Zeiss confocal microscope (Oberkochen, Germany).

Western blot

The CXCR3.2 antibody was prepared utilizing a peptide obtained from the grouper CXCR3.2 protein (aa 17–29; DNITWSPETSKSL). Western blotting was used to confirm the specificity of the anti-CXCR3.2 antibody (Figure S3). For western blot analysis, cells or brain tissue samples were lysed using Pierce IP Lysis Buffer (Thermo Scientific, USA) and protease inhibitors were added. The proteins were separated on a 12% SDS-PAGE gel before being transferred onto polyvinylidene fluoride membranes. Following 1 h of blocking in 5% bovine serum albumin in Tris-buffered saline with 0.1% Tween-20 (TBS-T), membranes were incubated for 16–18 h at 4°C with anti-CXCR3.2 (1:1000) and anti-actin (1:5000). Following a TBS-T wash, the membranes were incubated for 1 h at 37°C with secondary antibodies (anti-mouse or anti-rabbit; Proteintech) and developed using enhanced chemiluminescence reagents (Millipore, USA). The Chemiluminescence signals were captured using GBOX-CHEMI-XT4- E (Syngene, UK).

Recombinant protein preparation

CHO cells were used to induce the expression of the recombinant proteins CXCL11.1 and CXCL11.2, as outlined in a previous study [37]. Specific primers for CXCL11.1 and CXCL11.2 were designed using PrimerPremier 5.0 and subsequently utilized for PCR amplification (Table S1). CXCL11.1 and CXCL11.2 were cloned into the matching pcDL-SRa296 expression plasmids. Subsequently, the expression plasmids were transfected into CHO cells and maintained in a 10% FBS medium. The recombinant proteins in the cell supernatants were purified using an endotoxin removal column (Pierce, USA). The resulting endotoxin levels were 0.1 EU/mg.

In vitro assays

To analyse macrophage polarization and primary brain cells, the isolated macrophages and primary brain cells were seeded in 12-well plates. RGNNV was introduced to the cells following 48 h of culture. The viral titre of RGNNV was 10⁷ times the 50% tissue culture infective dose (TCID₅₀)/ml. After 24 h, the cells were washed, and total RNA was extracted for RT-qPCR analysis, as described in section 2.4. Additionally, cell lysates for arginase activity analysis and supernatants for inducible NO synthase (iNOS) activity analysis were collected. The specific procedures of arginase and iNOS activity analyses are described in section 2.12 below.

For CXCR3.2 screening, macrophages and primary brain cells were isolated and seeded in a 12-well plate. Following 48 h, RGNNV was added to the cells. After 24 h, the cells were collected, and total RNA was extracted. Subsequently, first-strand cDNA was synthesized. RT-qPCR was used to analyse cxcr3.1, cxcr3.2, and CXCR3.3 expression in control and RGNNV-infected macrophages and primary brain cells.

During the time-course experiment, isolated macrophages were seeded in a 12-well plate, and after 48 h, RGNNV was introduced to the cells. Total RNA and protein were extracted at 6, 12, 24, 36, and 48 h following RGNNV infection. The cells were collected for RNA or protein extraction. The CXCR3.2 gene and protein expression were examined through RT-qPCR and western blotting (the exact experiments for western blotting are detailed in section 2.6).

CXCR3 affects macrophage polarization in fish and mammals [31,32]. Our previous study revealed the presence of three CXCR3s (CXCR3.1, CXCR3.2, and CXCR3.3) in groupers [33]. Lu et al.. (2017) reported that the ligand of CXCR3 in fish is CXCL11 [32]. In our previous study, we identified two CXCL11s (CXCL11.1 and CXCL11.2) in groupers [33]. The sequences of CXCR3.2, CXCL11.1, and CXCL11.2 were presented in our previous study [33]. To assess RNAi performance, small interfering RNAs (siRNAs) were synthesized by GenePharma (Shanghai, China) using CXCR3.2, CXCL11.1, and CXCL11.2 sequences as the basis (Table S2) [33]. To verify the knockdown efficacy, CXCR3.2, CXCL11.1, and CXCL11.2 were inserted into the pcDNA3.1–3×FLAG vector. Subsequently, we utilized RT-qPCR and western blot analysis to evaluate siRNA knockdown efficacy. GS cells were transfected with 0.72 μg/mL of CXCR3.2, CXCL11.1, CXCL11.2, or negative control siRNAs using Lipofectamine 3000 (Invitrogen). At 36 h post transfection, the GS cells were collected for RT-qPCR analysis. The siRNA knockdown efficacy was assessed based on the mRNA levels of CXCR3.2, CXCL11.1, and CXCL11.2 (Fig S4). The detailed experiments are outlined in section 2.4.

For antagonist analysis, we used the CXCR3.2 antagonist NBI-74330 (Patent WO02083143, USA) at a dose of 1 pM to knockout CXCR3.2. Macrophages were seeded in 12-well plates, cultured for 12 h, and treated with PBS, RGNNV, or NBI-74330 + RGNNV. Cells were collected 24 h after treatment, and the mRNA levels of IL-1β and TNFα, and arginase and iNOS activity were assessed. The detailed experiments are described in sections 2.4 and 2.12.

To evaluate the effects of RNAi with chemokine ligands and recombinant protein treatment, macrophages were seeded in 12-well plates. Subsequently, we treated macrophages with siCXCL12a and siCXCL11.1, siCXCL11.2, or siCXCL11.1 + siCXCL11.2 using the Lipofectamine 3000 transfection reagent. After a 24-h incubation, RGNNV was introduced to the cells, which were then cultured for an additional 24 h before collection for subsequent use. We also treated macrophages with recombinant CXCL11.1 or CXCL11.2 and measured arginase and iNOS activities after 24 h.

To determine whether macrophage polarization can induce proinflammatory reactions in primary brain cells, we isolated macrophages and primary brain cells, seeding them in separate six-well plates. After 48 h, RGNNV was introduced to the macrophages. After 2 h, we collected the macrophage supernatant (containing RGNNV) through centrifugation. Subsequently, we added PBS, RGNNV, RGNNV infection macrophage supernatant, or RGNNV+NBI-74330-treated macrophage supernatant to the primary brain cell plate. After 4, 8, 12, 24, and 48 h, we measured IL-1β and TNFα mRNA levels and iNOS activity. The RT-qPCR experiments are detailed in section 2.4, and the specific measurements of iNOS and arginase activity are outlined in section 2.12 below.

In vivo assays

For CXCR3.2 screening, two groups of orange-spotted groupers (PBS and RGNNV) were cultured for 1 week before the experiment. Healthy fry was subjected to a 24-h fast before being injected intracranially with 3 µL of PBS or 10⁷ TCID₅₀/mL of RGNNV using the injection method previously described [38]. The fish were sacrificed 48 h post injection, and their brains were separated for total RNA extraction for RT-qPCR analysis. Additional fish were sacrificed at 6, 12, 24, 36, and 48 h post injection, and their brains were separated for subsequent RT-qPCR analysis.

For the in vivo assessment of CXCR3.2 antagonistic effect on grouper, healthy fish were subjected to a 24-h starvation before being injected intracranially with 3 µL of PBS, RGNNV, or RGNNV + NBI-74330. The fish were sacrificed 48 h post injection, and their brains were isolated for total RNA extraction for subsequent RT-qPCR analysis. Furthermore, the brains of five fish from each group were isolated and fixed in Bouin’s solution for histological analysis, as described in section 2.13. Daily mortality was recorded for 14 days.

For the in vivo assessment of the effects of RGNNV-treated macrophage supernatant and RGNNV + NBI-74330 treated macrophage supernatant on groupers, fish were cultured for 1 week before the experiment. Healthy fish were subjected to a 24-h starvation before being injected intracranially with 3 µL of PBS, RGNNV, RGNNV infection macrophage supernatant, or RGNNV+NBI-74330 treatment macrophage supernatant. The fish were sacrificed at 12, 24, and 48 h post injection, and their brains were separated for subsequent RT-qPCR analysis. The specific details of RT-qPCR experiments are presented in section 2.4. After 48 h post injection, five fish from each group were sacrificed and their brains were fixed in Bouin’s fluid for histological analysis, as described in section 2.13. Daily mortality was recorded for 14 days.

Chemotaxis

Polyornithincoated Nucleopore® PVP-free polycarbonate filters (Corning, Acton, MA, USA) with a 5-mm pore size were used to separate the upper and lower wells. The chamber was then incubated for 4 h at 37°C in a water-saturated atmosphere of 95% air and 5% CO2. Following incubation, non-migrating cells were scraped from the upper surface of the filter. The sorted macrophages were then seeded in the upper chambers of Transwell plates with a pore size of 5 mm and medium containing CXCL11.1 or CXCL11.2 chemokine, while the control medium was introduced to the bottom chambers. After 4 h of incubation at 24ºC, the macrophages that had migrated through the Transwell membrane to the bottom chambers were collected and counted using flow cytometry. Briefly, cells were incubated with 2 uL of anti-CXCR3.2 IgG (1:200) for 30 min at 4ºC. After washing, the cells were incubated with the secondary goat anti-rabbit IgG antibody (1:500) (CAS No. A21428; Invitrogen) for 30 min at 4ºC. The cells were suspended in 0.2 mL of PBS after unbound antibodies were removed, and a Gallios Flow Cytometer (Beckman Coulter, Miami, FL, USA) was used for analysis. At least 10,000 events within the microglial gates were recorded. The data were subsequently analysed using Alza software (Beckman Coulter, Pasadena, CA, USA).

Immunofluorescence assay

Immunofluorescence assays were performed as previously described [13]. Macrophages or GS cells were seeded onto coverslips in six-well plates and infected with the virus. After 6, 12, 24, 36, and 48 h of infection, macrophages or GS cells were fixed. The cells were then incubated with anti-CXCR3.2 (1:200) or anti-CP (1:300) at 37°C for 1 h after a PBS wash. The source specificity of the CP antibody has been previously reported [39]. Subsequently, the cells were incubated with a goat anti-rabbit IgG secondary antibody (1:500) (CAS No. A21428; Invitrogen) for 30 min at 37ºC. Finally, the cells were stained with DAPI (1 mg/ml) and observed under a fluorescence microscope.

iNOS and arginase activity assay

Nitric oxide (NO) generation was indirectly assessed by measuring stable end-product NO accumulation in the cell culture supernatant using the Griess reaction, as previously described (Kleinbongard et al., 2022). Equivalent volumes of the Griess reagent were mixed with the supernatant or sodium nitrate standards. The samples were incubated at 24°C for 10 min, and absorbance at 540 nm was measured. Arginase activity was assessed in cell lysates as previously described [40] and expressed as micromoles of urea per milligram of protein. The cells underwent lysis using Triton X-100. After 30 min on a shaker, 100 μL of Tris·HCl was introduced into 100 μL of lysate, and the enzyme was activated by heating for 10 min at 55°C. Arginine hydrolysis was achieved by incubating the lysates. Upon cessation of the reaction, the urea concentration was measured at 550 nm following the addition of 40 μL of α-isonitrosopropiophenone, followed by heating at 100°C for 30 min. One unit of enzyme activity is defined as the amount of enzyme that catalyzes the formation of 1 μmol·urea/min.

Brain histology

The brains were dissected from fish in different groups. The in vivo assessment of CXCR3.2 antagonistic effect on groupers included the PBS, RGNNV, and RGNNV + NBI-74330 groups. Healthy fish were subjected to starvation for 24 h before being injected intracranially with 3 µL of PBS, RGNNV, or RGNNV + NBI-74330. After 48 h post injection, five fish from each group were sacrificed, and their brains were isolated and fixed in Bouin’s fluid for histological analyses.

The in vivo test for assessing the effects of macrophage supernatant on groupers included PBS, RGNNV, RGNNV infection macrophage supernatant, and RGNNV+NBI-74330 treatment macrophage supernatant. Healthy fish were subjected to a 24-h starvation before being injected intracranially with 3 µL of PBS, RGNNV, RGNNV infection macrophage supernatant, or RGNNV+NBI-74330 treatment macrophage supernatant. After 48 h post injection, five fish from each group were sacrificed, and their brains were isolated and fixed in Bouin’s fluid for histological analysis. Subsequently, the brains were dehydrated and embedded in paraffin. Finally, the tissue was cut into 5-μm sections and subjected to haematoxylin and eosin staining for analysis.

TUNEL staining

TUNEL analysis was performed, as previously described [41]. Brain apoptosis was assessed using the In Situ Cell Death Detection Kit, POD (Roche Molecular Biochemicals), according to the manufacturer’s instructions. Following TUNEL staining, haematoxylin was used to counterstain the tissues. Images were captured using a Nikon optical microscope (Nikon, Tokyo, Japan).

Statistical analyses

All data are expressed as mean ± SD. Statistical analyses were performed using the SPSS software (version 17.0). One- or two-way analysis of variance was used, with statistical significance set at *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. GraphPad Prism 6 was utilized for all statistical analyses.

Results

RGNNV induces macrophage polarization

To confirm the presence of macrophages in grouper brain tissue, we initially performed fluorescent in situ hybridization (FISH) to assess the abundance of macrophage cells in the brain. Macrophages were identified based on the expression of a macrophage marker gene, mannose receptor C-type 1 (Mrc1). Most of the macrophages were located in the membrane surrounding the brain (Figure 1a). Subsequently, we isolated these macrophages and performed a primary culture. Figure 1b shows the macrophages isolated from brain tissue through density gradient centrifugation and subsequent culture.

Figure 1. RGNNV induces polarization of grouper macrophages (a) Localization of macrophages in the grouper brain as indicated by in situ hybridization. Macrophages were located on the membrane surrounding the brain marketed by Mrc1 (green). Scale bars was 100 μm. (b) Primary cultured macrophages. (c) iNOS activities in RGNNV infected and uninfected macrophages (n = 4). (d) arginase activities in RGNNV infected and uninfected macrophages. (e) mRNA levels of cytokines in primary cultured grouper macrophages, include IL-1β, TNFα, IL-10, and TGF-β (n = 4). (f) mRNA levels of cytokines in vivo, include IL-1β, TNFα, IL-10, and TGF-β (n = 5). **p < 0.01, ***p < 0.001, ****p < 0.0001.

Next, we examined whether RGNNV induces macrophage polarization. Initially, we assessed tumour necrosis factor alpha (TNFα), interleukin (IL)-1β, IL-10, and transforming growth factor beta (TGF-β) expressions in macrophages following RGNNV infection. The macrophages exhibited high expression of TNFα and IL-1β, along with heightened NO synthase (iNOS) activity, while TGF-β and IL-10 expressions remained unchanged. Further, arginase activity was downregulated following RGNNV exposure (Figure 1c–e). Upon examining in vivo gene expression after RGNNV infection, a significant increase was observed in the expression of IL-1β and TNFα, while the levels of IL-10 and TGF-β remained unchanged (Figure 1f). These findings suggest that RGNNV infection-induced macrophage polarization towards M.

CXCR3.2 plays important roles in macrophage polarization

Initially, we used RT-qPCR to assess the mRNA expression of these genes in brain tissues and macrophages following RGNNV exposure. Among the three CXCR3s, only CXCR3.2 exhibited a significant increase in expression in macrophages after viral infection (Figure 2a); this heightened expression was not observed in brain cells except in macrophages (Figure S1a). Similarly, CXCR3.2 expression levels in the brain tissue significantly increased after the RGNNV challenge, while those of CXCR3.1 and CXCR3.3 exhibited no significant difference between the PBS and RGNNV-exposed groups (Figure 2b). Further, the expressions of CXCL11.1 and CXCL11.2 were measured in the brain and macrophages, which were found to be significantly increased following viral challenge (Figure S2 a & b). However, this increased expression was not observed in brain cells except in macrophages (Figure S1b). Therefore, we propose that CXCR3.2, CXCL11.1, and CXCL11.2 serve as indicators of macrophage function after RGNNV infection.

Figure 2. mRNA expression of CXCR3.2 in macrophages and brain tissues (a) mRNA expression of CXCR3s in macrophages after RGNNV challenge (n = 4). (b) mRNA expression of CXCR3s in vivo after RGNNV challenge (n = 5). (c) The expression of CXCR3.2 at 6 h, 12 h, 24 h, 36 h, and 48 h after RGNNV treatment in macrophages (n = 4). (d) protein levels of CXCR3.2 were determined by Western blotting in grouper macrophages at 6 h, 12 h, 24 h, 36 h, and 48 h following RGNNV infection. (e) The immunofluorescence shows CXCR3.2 (green) in macrophages at 6 h, 12 h, 24 h, 36 h, and 48 h after RGNNV treatment in macrophages, Scale bar, 20 μm. (f) The expression of CXCR3.2 at 6 h, 12 h, 24 h, 36 h, and 48 h after RGNNV treatment in grouper (n = 5). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Next, we measured CXCR3.2 mRNA expression levels in macrophages at various time points following RGNNV infection. Compared with those in the control group, macrophages exhibited a significant increase in CXCR3.2 mRNA levels after RGNNV infection, with the peak expression observed at 6 h post infection. However, the CXCR3.2 mRNA level decreased at 12, 24, 36, and 48 h, compared with the levels observed at 6 h following RGNNV infection (Figure 2c).

The specificity of anti-CXCR3.2 IgG was validated using western blotting (Figure S3) before assessing the expression levels of CXCR3.2 proteins in macrophages. Following RGNNV infection, the expression of CXCR3.2 significantly increased compared with that in the control group. However, a gradual decrease in CXCR3.2 protein levels in macrophages was observed at 12, 24, 36, and 48 h after RGNNV infection compared with that at 6 h (Figure 2d). Immunofluorescence assay revealed that exposure of macrophages to RGNNV resulted in decreased positive fluorescence signals of CXCR3.2 at 12, 24, 36, and 48 h compared with that at 6 h (Figure 2e). Furthermore, we assessed CXCR3.2 expression in grouper brain tissue following RGNNV challenge and observed a significant increase in its expression compared with that in the control group. However, this enhancement gradually diminished at 12, 24, 36, and 48 h compared with that at 6 h following RGNNV infection (Figure 2f). These findings indicate that RGNNV infection led to increased CXCR3.2 expression in macrophages; however, this enhancement slowly diminished with prolonged RGNNV infection time.

To determine whether CXCR3.2 affects macrophage polarization in fish, we used siRNA to inhibit CXCR3.2 expression. The efficacy of the siRNAs in inducing knockdown was confirmed through RT-qPCR analysis (Figure S4a). During the CXCR3.2 siRNA experiment, we observed a significant increase in the mRNA expression levels of IL-1β and TNFα mRNA in macrophages in the RGNNV exposure group only. However, in the siCXCR3.2+RGNNV group, the expression of IL-1β and TNFα was comparable to that in the PBS group and significantly decreased compared with that in the RGNNV infection group only (Figure 3a). We also observed that treating RGNNV-infected macrophages with the CXCR3 antagonist NBI-74330 resulted in significantly lower expression levels of IL-1β and TNFα than those in the RGNNV infection group only (Figure 3b). To confirm the effect of CXCR3.2 on macrophage polarization, we replicated the same experiment in vivo. We observed that the expression of IL-1β and TNFα in the NBI-74330 + RGNNV treatment group was significantly higher than that in the PBS group but lower than that in the RGNNV infection group (Figure 3c). We also observed that RGNNV increased iNOS activity; nonetheless, this increase was inhibited by NBI-74330 (Figure 3d). Similarly, the arginase activity decreased following RGNNV infection. However, this decrease was also inhibited by NBI-74330 (Figure 3e). These findings suggest that CXCR3.2 plays an essential role in macrophage polarization. Additionally, they indicate the presence of other cells in the grouper brain, aside from macrophages, capable of expressing IL-1β and TNFα at high levels following RGNNV infection.

Figure 3. CXCR3.2 regulate the polarization of macrophages. (a) the expression of IL-1β and TNFα in macrophages after treated with PBS, RGNNV, and siCXCR3.2+RGNNV (n = 4). (b) the expression of IL-1β and TNFα in macrophages after treated with PBS, RGNNV, or NBI-74330+RGNNV (n = 4). (c) the expression of IL-1β and TNFα in orange-spotted grouper after treated with PBS, RGNNV, or NBI-74330+RGNNV (n = 5). (d and e) iNOS and arginase activity in macrophages after PBS, RGNNV, or NBI-74330+RGNNV treatment (n = 4). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Identifying CXCR3.2 ligands involved in regulating macrophage polarization

First, we utilized chemotaxis assays to identify specific ligands for CXCR3.2. The induction of CXCR3.2 was observed with CXCL11.1 and CXCL11.2, indicating the presence of macrophages (Figure 4a). Maximal cell migration was observed at 50 and 10 nM of CXCL11.1 and CXCL11.2, respectively. Recombinant CXCL11.1 and CXCL11.2 proteins were prepared to explore extracellular signals of CXCR3.2. CXCL11.1 and CXCL11.2 upregulated the iNOS activity of resting macrophages but did not affect arginase activity (Figure 4b,c). To further investigate whether CXCL11.1 and CXCL11.2 affect macrophage polarization, siRNAs were utilized to suppress CXCL11.1 and CXCL11.2 expression. The knockdown efficacies of CXCL11.1 and CXCL11.2 were confirmed through RT-qPCR (Figure S3 b,c). CXCL11.1 or CXCL11.2 silencing alone did not result in a significant decrease in RGNNV-induced TNFα and IL-1β high expression. However, when CXCL11.1 and CXCL11.2 were silenced together, RGNNV infection resulted in a significant decrease in TNFα and IL-1β expression (Figure 4d–i). These findings indicate that CXCL11.1 and CXCL11.2 interact with CXCR3.2, thereby regulating macrophage polarization induced by RGNNV.

Figure 4. CXCL11s mediate macrophage chemotaxis and polarization. (a) Effect of CXCL11s on the chemotaxis ability of resting macrophages (n = 4). (b) Effect of CXCL11s on iNOS activity in resting macrophages. The macrophages were treated with 50 nM CXCL11.1 and 10 nM CXCL11.2 (n = 5). (c) Effect of CXCL11s on arginase activity in resting macrophages. The macrophages were treated with 50 nM CXCL11.1 and 10 nM CXCL11.2 (n = 5). (d and e) the expression of TNFα and IL-1β after PBS, RGNNV or siCXCL11.1+RGNNV treated macrophages (n = 4). (f and g) the expression of TNFα and IL-1β after PBS, RGNNV or siCXCL11.2+RGNNV treated macrophages (n = 4). (h and i) the expression of TNFα and IL-1β after PBS, RGNNV or siCXCL11.1+ siCXCL11.2+RGNNV treated macrophages (n = 4). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Effects of macrophage damage brain through CXCR3.2 in grouper

To determine whether CXCR3.2 can affect viral replication and brain tissue damage in fish, we used NBI-74330 to inhibit their function and subsequently measured viral gene (CP and RdRp) expression levels and assessed brain tissue damage. RT-qPCR analysis revealed that inhibiting CXCR3.2 resulted in CP and RdRp expression levels comparable to those in the RGNNV-only infection group (Figure 5a,b). FISH analysis of the CP genome revealed that inhibiting CXCR3.2 resulted in a CP-positive signal similar to that in the RGNNV-only infection group (Figure 5c). Brain histology analysis revealed that inhibiting macrophage polarization with NBI-74330 led to a reduction in the number of vacuoles in the brain compared with that in the RGNNV-only group (Figure 5d). The cumulative mortality of groupers injected with RGNNV only was approximately 90%, while RGNNV + NBI-74330 reduced it to ~ 70% by day 7 post infection (Figure 5e). These findings suggest that CXCR3.2 has no effect on RGNNV replication in the orange-spotted grouper. However, inhibiting CXCR3.2, which reduces macrophage polarization, can mitigate brain tissue damage and cumulative mortality.

Figure 5. CXCR3.2 have effect on virus replication and brain tissue damage in fish. (a and b) the expression of virus gene (CP and RdRp) in PBS, RGNNV, and NBI-74330+RGNNV treatment grouper (n = 5). (c) Representative RGNNV fluorescent situ hybridization (FISH) staining of brain sections from fish injected with the PBS, RGNNV and NBI-74330+rgnnv (n = 3). Scale bar: 50 µm. (d) the brain histology when injected with RGNNV and RGNNV+NBI-74330. The black arrowhead indicates vacuoles. Scale bar: 25 µm. (e) Values indicate the cumulative mortality in each group of the orange-spotted groupers during the 14-day experimental period after different injection treatments (PBS, RGNNV, or RGNNV+ NBI-74330).

Macrophage polarization can promote grouper brain damage

To confirm whether macrophage polarization induces brain damage, we collected the supernatant from RGNNV-infected macrophages and introduced it to primary brain cells to observe their polarization (Figure 6a). Upon adding the supernatant to primary brain cells, the expression levels of IL-1β and TNFα were significantly higher than that in the control and RGNNV only infection groups from 4 to 48 h. Furthermore, after introducing the supernatant, the expression of IL-1β and TNFα increased rapidly within 4 h, while it took 12 h for a significant increase to occur in the RGNNV-only group (Figure 6b,c). Moreover, the iNOS activity of the supernatant-added group was significantly higher than that of the RGNNV-only infection group (Figure 6d). Subsequently, we collected the supernatant from RGNNV+NBI-74330 treated macrophages and added it to primary brain cells to observe their polarization (Figure 6e). Upon adding the supernatant to primary brain cells, IL-1β and TNFα expression levels were significantly higher than those of the control group and consistent with those in the RGNNV-only infection group from 4 to 48 h (Figure 6f,g). Additionally, the iNOS activity of the supernatant-added group was significantly higher than that of the control group; however, it was consistent with that of the RGNNV-only infection group (Figure 6h). These findings indicate that macrophages secrete substances that markedly accelerate primary brain cell polarization.

Figure 6. The supernatant of macrophages (after RGNNV infection or RGNNV+NBI-74330 treatment) affect brain polarization. (a) Protocol for collected supernatant from RGNNV-infected macrophages and added it to primary brain cells. (b and c) Expression levels of TNFα and IL-1β after PBS, RGNNV or RGNNV infection macrophages supernatant treatment primary brain cells at 4 h, 8 h, 12 h, 24 h and 48 h (n = 4). (d) iNOS activity in PBS, RGNNV or RGNNV infection macrophages supernatant treatment primary brain cells at 4 h, 8 h, 12 h, 24 h and 48 h. (e) Protocol for collected supernatant from RGNNV+NBI-74330 treatment macrophages and added it to primary brain cells. (f and g) Expression levels of TNFα and IL-1β after PBS, RGNNV or RGNNV+NBI-74330 treatment macrophages supernatant treatment primary brain cells at 4 h, 8 h, 12 h, 24 h and 48 h (n = 4). (h) iNOS activity in PBS, RGNNV or RGNNV+NBI-74330 treatment macrophages supernatant treatment primary brain cells at 4 h, 8 h, 12 h, 24 h and 48 h.

Subsequently, we examined the effects of the supernatants from these two treatment strategies on grouper brains in vivo. We collected the supernatants and injected them into the grouper brain. Subsequently, the survival rate and cytokine expression were observed (Figure 7a). No significant change was observed in the expression of virus genes CP and RdRp in grouper brain following intracranial injection of the macrophage supernatant or RGNNV (Figure 7b,c). However, IL-1β and TNFα expression levels in the brain following the injection were significantly higher than those after RGNNV injection (Figure 7d,e). Histological analysis results revealed that the number of vacuoles in the supernatant injection group was significantly higher than that in the RGNNV injection group (Figure 7f). Furthermore, positive TUNEL staining was observed in the RGNNV treatment group, with a higher number of TUNEL-positive cells in the macrophage supernatant treatment group. Nevertheless, no positive signals were observed in the control group (Figure S5 a–c). The final mortality rate in the supernatant group was comparable to that in the RGNNV injection group. However, the peak mortality in the supernatant injection group occurred within 2–4 days after injection, whereas in the RGNNV injection group, it occurred 4–7 days following injection (Figure 7g). Similarly, the expression of virus genes CP and RdRp in the brain of grouper did not differ significantly following intracranial injection of the RGNNV+NBI-74330 macrophage supernatant or RGNNV (Figure 7h,i). Moreover, IL-1β and TNFα expression in the brain following intracranial injection of the RGNNV+NBI-74330 supernatant was comparable to that after RGNNV injection (Figure 7j,k). Histological examination revealed that the number of vacuoles in the supernatant injection group was comparable to that in the RGNNV injection group (Figure 7l). Furthermore, the final mortality rate in the supernatant group was comparable to that in the RGNNV injection group (Figure 7n).

Figure 7. The supernatant of macrophages on the grouper brain in vivo. (a) Protocol for collected supernatant from RGNNV infection or RGNNV+NBI-74330 treatment macrophages and injected it to grouper. (b and c) Expression levels of virus gene CP and RdRp after L15, RGNNV or RGNNV infection macrophages supernatant treatment at 12 h, 24 h and 48 h (n = 5). (d and e) Expression levels of virus gene IL-1β and TNFα after L15, RGNNV or RGNNV infection macrophages supernatant treatment at 12 h, 24 h and 48 h (n = 5). (f) the brain histology when injected with L15, RGNNV or RGNNV infection macrophages supernatant. The black arrowhead indicates vacuoles. Scale bar: 25 µm. (g) Values indicate the cumulative mortality in each group of the orange-spotted groupers during the 14-day experimental period after different injection treatments (L15, RGNNV or RGNNV infection macrophages supernatant). (h and i) Expression levels of virus gene CP and RdRp after L15, RGNNV or RGNNV+NBI-74330 treatment macrophages supernatant treatment at 12 h, 24 h and 48 h (n = 5). (j and k) Expression levels of virus gene IL-1β and TNFα after L15, RGNNV or RGNNV+NBI-74330 macrophages treatment supernatant treatment at 12 h, 24 h and 48 h (n = 5). (l) the brain histology when injected with L15, RGNNV or RGNNV+NBI-74330 macrophages treatment supernatant. The black arrowhead indicates vacuoles. Scale bar: 25 µm. (n) Values indicate the cumulative mortality in each group of the orange-spotted groupers during the 14-day experimental period after different injection treatments (L15, RGNNV or RGNNV+NBI-74330 macrophages treatment supernatant).

These findings suggest that macrophages secrete substances that significantly accelerate grouper brain tissue damage, primarily through the accelerated release of proinflammatory factors.

Discussion

The immune system plays a crucial role in responding to infections and injuries. Peripheral immune cell activation typically leads to tissue leukocyte infiltration; however, this phenomenon is notably absent in the CNS. CAMs are immune cells that are important for maintaining CNS health [20]. In the present study, we observed a significant role of macrophages in the immune response in the brains of orange-spotted groupers. Macrophages are considered the key cellular defence mechanisms against viral infections in the CNS. Given their strategic location at the surface or entry sites of the CNS, macrophages are considered the primary targets for viral infection in the CNS. As demonstrated in a previous study, macrophages often serve as sentinel immune cells, establishing the first line of defence to prevent the spread of the virus into the CNS parenchyma [42]. In mammals, brain macrophages play a crucial role in the immune response against viral infections. For example, macrophage numbers significantly increased following HIV infection, and this augmentation is closely associated with disease progression [43]. Macrophages in the brain have also been observed to release toxic substances in cats infected with the FIV [23], suggesting that macrophages are crucial in mediating CNS dysfunction following viral brain infection [24]. In fish, it is plausible that brain macrophages play an important role in CNS viral infections.

Previous studies in mammals have demonstrated that brain macrophages respond to viral infection in various ways, mainly by assuming anti- or proinflammatory roles or serving as potential virus reservoirs [44]. However, the response of brain macrophages to viral invasion via the CNS remains unclear in fish. Our findings suggest that RGNNV infection in the grouper brain induces M1 polarization of macrophages. Fish macrophages have been demonstrated to respond to pathogen invasion by producing cytokines via polarization [32]. These findings underscore the significance of macrophage polarization in the resistance of fish to pathogen invasion in the brain. However, the underlying mechanism by which polarization occurs remains unclear.

Although the main factors regulating macrophage polarization remain elusive, some studies have suggested a pivotal role for chemokines [32,45]. The CXCR3 ligand, CXCL11, regulates macrophage migration and activation in mice [46,47]. The number of CXC chemokine genes in teleost exhibits variations among species [48]. Teleost possess at least two CXCL11 homologs, while zebrafish possess several CXCL11 homologs [49,50]. In our previous study, three CXCR3 and two CXCL11 homologs were identified in groupers. Exploration of the role of chemokines in macrophage polarization revealed an elevation in CXCR3.2 in grouper macrophages following RGNNV infection. The observed pattern of CXCR3.2 indicates a potentially crucial role in regulating macrophage function. Furthermore, this study revealed that RGNNV infection induced CXCL11 chemotaxis and promoted the M1 phenotype in macrophages via interacting with CXCR3.2. Another study involving grass carp and spotted green pufferfish demonstrated the regulation of macrophage polarization by CXCR3s [32]. This suggests that macrophage polarization induction by CXCR3s is conserved in teleost fish.

Brain macrophages, as the primary organ-specific macrophages in the CNS, play a crucial role during steady state and perturbation [18]. In this study, we found that upon RGNNV invasion, these cells did not possess the capacity to directly store or eliminate the virus. However, RGNNV infection triggered the M1-type macrophage polarization. Moreover, inhibiting macrophage polarization diminishes proinflammatory factor levels generated in the brain significantly. These findings suggest that brain injury may predominantly arise from macrophage polarization and proinflammatory factor production. However, following the in vitro stimulation of macrophages with RGNNV, the collection of supernatant, and subsequent stimulation of primary brain cells or injected into fish, we observed that the supernatant induced a more pronounced inflammation response in primary brain cells and caused more severe damage to brain tissue than RGNNV treatment alone. These findings imply that brain injury and fish mortality may primarily stem from macrophage polarization, triggering an excessive release of proinflammatory factors that damage the brain tissue.

In summary, we elucidated the underlying mechanism through which RGNNV directly induces macrophages to polarize towards M1 via the CXCR3.2-CXCL11 pathway, producing proinflammatory factors that destroy brain tissue. Our study explore the long-term consequences of macrophage polarization and pro-inflammatory cytokine production.

Author’ contributions

Qing Wang: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Resources, Writing – original draft. Weiqi Qin: Conceptualization, Funding acquisition, Project administration, Validation, Writing – review & editing. Huihong Zhao: Data curation, Writing – review & editing. Kaishan Liang: Formal analysis, Methodology, Writing – original draft. Minlin Zhang: Formal analysis, Investigation, Methodology, Software. Jiantao Liang: Investigation. Xiaoling Zuo: Formal analysis, Investigation, Software, Validation. Xianze Jia: Methodology. Jinhong Shan: Methodology. Songyong Gan: Methodology. Ding Liu: Methodology. Zongyang Li: Investigation, Methodology. Jie Yu: Software. Zijie Xuan: Methodology. Liyuan Luo: Methodology.

Ethics statement

All relevant ethical safeguards were met for animal experimentation, and all animal experiments in this study were approved by the Animal Care and Use Committee of the College of Marine Sciences, South China Agricultural University.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors without undue reservation to any qualified researcher. The raw data in DOI: 10.17632/f5gd8tsgy9.1.

Supplemental data

Supplemental data for this article can be accessed online at https://doi.org/10.1080/21505594.2024.2355971

Additional information

Funding

This work was supported by the grants from the National Natural Science Foundation of China [42176103, U20A20102], the National Key Research and Development Program of China [2022YFD2400502], the Guangdong Provincial Natural Science Foundation [2022A1515012505], The talent team tender grant of Zhanjiang marine equipment and biology [2021E05035].