Hematopoietic stem/progenitor cells (HSPCs) generate the entire repertoire of immune cells in vertebrates and play a crucial role during infection. Although two copies of CXC motif chemokine receptor 4 (CXCR4) genes are generally identified in teleosts, the function of teleost CXCR4 genes in HSPCs is less known. In this study, we identified two CXCR4 genes from a teleost, ayu (Plecoglossus altivelis), named PaCXCR4a and PaCXCR4b. PaCXCR4b was constitutively expressed in ayu HSPCs, whereas PaCXCR4a was induced by LPS treatment. The stromal-derived factor-1–binding activity of CXCR4b was significantly higher than that of CXCR4a, whereas the LPS-binding activity of CXCR4a was significantly higher than that of CXCR4b in the teleosts ayu, large yellow croaker (Larimichthys crocea), and tiger puffer (Takifugu rubripes). CXCR4a+ HSPCs were mobilized into blood by LPS, whereas CXCR4b+ HSPCs were mobilized by leukocyte cell–derived chemotaxin-2. PaSDF-1 and PaCXCR4b, but not PaCXCR4a, inhibited HSPC proliferation by regulating reactive oxygen species levels. Compared with PaCXCR4b+ HSPCs, PaCXCR4a+ HSPCs preferentially differentiated into myeloid cells in ayu by maintaining high stem cell leukemia expression. These data suggest that the two copies of CXCR4s achieve a division of labor in the regulation of teleost HSPC homeostasis, supporting the concept that subfunctionalization after gene duplication in teleosts may stabilize the immune system.

Hematopoiesis is a complex and well-orchestrated process that produces all lineages of immune cells in vertebrates (1). In the bone marrow niches of mammals, a variety of factors have been reported to be crucial for hematopoietic stem/progenitor cell (HSPC) homeostasis, including proliferation, mobilization, and differentiation (1, 2). Teleosts, as early vertebrates, constitute a highly successful and diverse group, including half of vertebrate species (3). The immune system of teleosts shows many differences compared with that of mammals. For example, teleost-specific genome duplication produces two copies of several crucial immune genes that exhibit subfunctionalization (4, 5). Moreover, HSPC composition has been modified for adaptation to special environments in some teleosts, such as cavefish (Astyanax mexicanus) (R. Peuß, A.C. Box, Y. Wang, S. Chen, J. Krishnan, D. Tsuchiya, B. Slaughter, and N. Rohner, manuscript posted on bioRxiv). However, the teleost-specific regulatory mechanism underlying HSPC homeostasis is still unclear.

Although HSPCs are not mature immune cells and are frequently dormant, it has been suggested that HSPCs participate in the primary response to infections (6, 7). HSPCs can be mobilized by directly sensing infection through TLR (8). In addition to the direct sensing of pathogens, HSPC activity in response to infection is also mediated indirectly by proinflammatory cytokines (9, 10). Furthermore, infections also affect the lymphoid versus myeloid fate choice. LPS binds with TLR4 to induce the rapid generation of myeloid cells, including macrophages (11). In zebrafish, HSPCs can also proliferate and differentiate into required immune cell lineages postinfection (12). HSPCs are mobilized into the peripheral blood from the kidney postinfection in another teleost, ayu (Plecoglossus altivelis) (13). HSPC mobilization is induced by bacterial components such as LPS (14), or host cytokines such as G-CSF and leukocyte cell–derived chemotaxin-2 (LECT2) (15). However, it is necessary to further investigate the regulatory mechanism underlying HSPC self-renewal, mobilization, and differentiation postinfection in teleosts.

CXC chemokines and their receptors play crucial roles in the migration and function regulation of immune cells in both teleosts and mammals (16, 17). Stromal cell–derived factor (SDF)-1 (also known as CXCL12) is a member of the CXC group of chemokines (18). The physiological SDF-1 receptor is CXC motif chemokine receptor 4 (CXCR4), a heptahelical receptor coupled to heterotrimeric GTP-binding proteins (19). SDF-1/CXCR4 is involved in HSPC homeostasis, development, tumor metastasis, metabolism, and HIV entry (2022). CXCR4 normally promotes bone marrow HSPC mobilization, homing, retention, and quiescence (23). Moreover, CXCR4 desensitization affects the lymphoid differentiation of HSPCs (24). CXCR4 genes have also been cloned in a variety of teleosts (16). It has long been known that two copies of CXCR4 genes (CXCR4a and CXCR4b) exist in teleosts, whereas only one single copy of CXCR4 is present in mammals (25). In teleosts, CXCR4b has been found to be involved in embryogenesis and ganglia formation (26), whereas CXCR4a is related to vessel development (27). Furthermore, CXCR4b also participates in neutrophil recruitment in teleosts (28). However, the functions of CXCR4a and CXCR4b in teleost HSPCs are still unclear. Because CXCR4 represents the sole chemokine receptor in HSPCs to mediate migration/chemotaxis (29), investigating its effect on HSPC homeostasis in teleosts is necessary to understand the diverse mechanisms underlying the immune response.

As the most species-rich group of vertebrates, teleosts are useful for finding mammalian immune system paradigms (30). Ayu is an important experimental teleost widely cultured in Japan, China, and Korea, and its genome has been successfully sequenced (National Center for Biotechnology Information Sequence Read Archive accession number SRR8369080). In this study, we found that CXCR4b was highly expressed in ayu whole kidney, whereas CXCR4a was expressed at low levels in the healthy whole kidney and upregulated postinfection. Furthermore, CXCR4a preferred to interact with LPS, whereas CXCR4b preferred to interact with SDF-1. CXCR4a and CXCR4b also showed different capabilities of interacting with LPS and SDF-1 in other teleosts, large yellow croaker (Larimichthys crocea), and tiger puffer (Takifugu rubripes). Considering that the two copies of the CXCR4 gene in teleosts may display subfunctionalization, we further investigated the functions of CXCR4a and CXCR4b in HSPCs.

Ayu, large yellow croaker, and tiger puffer were used in all experiments. Ayu (Zhemin No. 1, weighing 40 ± 5 g each) has undergone seven successive generations of mass selection for fast growth, and the effective population size is ∼4000 fish. After seven generations of mass selection, the growth rate for Zhemin No. 1 is stable. The slow-growth ayu were removed from the population to characterize fast growth. The fish were kept in freshwater tanks at 20°C in a recirculating system using filtered water. From 2017 to 2019, three different year classes of fish were used. During experiments, the ayu feed rate was 3% body weight per day, and the pellet feed mainly consisted of fish meal, wheat flour, soybean meal, fish oil, and vitamin premix. For the HSPC transplantation experiments, the initial body weight of ayu was 40 ± 5 g, whereas the body weight was 82 ± 7 g after 16 wk of HSPC transplantation. The genetic diversity of Zhemin No. 1 has been measured by amplified fragment length polymorphisms (31). The proportion of polymorphic genes in Zhemin No. 1 is 42.74%, which is lower than that in wild ayu (32). Large yellow croakers weighing 100 ± 20 g were purchased from a mariculture farm in Xiangshan County, Ningbo, China and maintained in a flow-through seawater supply at 25°C. Tiger puffers weighing 15 ± 1 g were purchased from Tianzheng (Dalian, China) and kept in recirculating water at 20°C. The fish were held in the laboratory for ≥2 wk, with healthy appearance and normal activity, prior to use in experiments. All fish used in this study were healthy and showed no pathological signs of infection. Two media routinely used to screen for fish health were tryptic soy agar and thiosulfate citrate bile salts sucrose agar. Bacterial CFUs were not detected in the blood of healthy fish. The experimental conditions and procedures were approved by the Ningbo University Institutional Animal Care and Use Committee and were carried out in compliance with the National Institutes of Health’s Guide for the Care and Use of Laboratory Animals.

To inhibit HSPC mobilization in vivo, we neutralized LPS and LECT2 using human bacterial/permeability increasing protein (BPI) peptide and anti-LECT2 Ab. The BPI peptide (85–99 aa) was synthesized by GL Biochem (Shanghai, China) and has been demonstrated to inhibit the binding of LPS to monocytes (33). The anti-LECT2 Ab was prepared by immunizing mice with recombinant LECT2, which was produced in our previous work (34). Ayu were treated i.p. with 0.5 μg of BPI peptide/g body weight or 0.5 μg of anti-LECT2 Ab/g 30 min postinfection.

The cDNA sequences of the ayu CXCR4a (P. altivelis CXCR4a [PaCXCR4a]), CXCR4b (P. altivelis CXCR4b [PaCXCR4b]) and P. altivelis SDF-1 (PaSDF-1) genes were obtained from transcriptome analysis of ayu monocytes/macrophages (MO/MΦs), and transcriptomic data were deposited into the Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/). All accession numbers are listed in Table I. PCR, cloning, and sequencing were used to confirm the authenticity of these sequences. The similarities between the obtained sequences and other known sequences were analyzed using the basic local alignment search tool (http://blast.ncbi.nlm.nih.gov/blast.cgi). Multiple sequence alignments were generated using ClustalW (http://clustalw.ddbj.nig.ac.jp/). Phylogenetic and molecular evolutionary analyses were conducted using Molecular Evolutionary Genetic Analysis 5.0 Program. The sequences used in this study are listed in Table I.

Total RNA was extracted from fish tissues and cells using RNAiso (Takara, Dalian, China). After treatment with DNase I, first-strand cDNA was synthesized using avian myeloblastosis virus reverse transcriptase (Takara), and real-time quantitative PCR (RT-qPCR) was performed on an ABI StepOne Real-Time PCR System (Applied Biosystems, Foster City, CA) using SYBR Premix Ex Taq II (Takara). The specific primer sequences for PaCXCR4a, PaCXCR4b, PaSDF-1, ayu scavenger receptor class B 2a (PaSRB2a), GATA-binding protein 2 (GATA2), runt-related transcription factor 1 (RUNX1), GATA-binding protein 3 (GATA3), reduced paired box 5 (PAX5), early growth response protein 1 (EGR1), PU box-binding protein (PU.1), myeloperoxidase (MPO), stem cell leukemia (SCL), C/EBP-α (C/EBPα), Ikaros, myocyte enhancer factor 2C (Mef2C), and the housekeeping gene 18S rRNA are listed in Table II. Amplifications were carried out in a 25-μl reaction volume containing the sample, primers, and SYBR Premix Ex Taq II. The reaction mixture was incubated in an ABI StepOne Real-Time PCR System (Applied Biosystems) for 300 s at 95°C, followed by 40 amplification cycles of 30 s at 95°C, 30 s at 60°C, and 30 s at 72°C. The mRNA expression of the target genes was normalized to that of 18S rRNA using the 2−ΔΔCT method.

The cells were washed twice in sterile PBS and lyzed in a buffer (20 mM HEPES, 1.5 mM MgCl2, 0.2 mM EDTA, 100 mM NaCl, 0.2 mM DTT, 0.5 mM sodium orthovanadate, and 0.4 mM PMSF [pH 7.4]) containing phosphatase inhibitor (Phosphatase Inhibitor Cocktail; Sigma-Aldrich, St. Louis, MO). The protein concentration in each soluble fraction was measured using the Bradford method (35). For Western blot analysis, the proteins were resolved using SDS-PAGE and transferred to polyvinylidene difluoride membranes. The membranes were blocked for 1 h in a 10% nonfat dry milk solution containing TBS-Tween at 37°C. After a 1.5 h incubation with specific Abs, the membranes were washed and incubated for 1 h with an HRP-labeled secondary Ab (1:5000; Santa Cruz Biotechnology) and visualized using an ECL Western blotting detection system. The intensity of each band obtained by Western blot was analyzed using the National Institutes of Health ImageJ Program. Ayu β-actin was used as the control.

The SCL Abs were from commercial sources (1:400; R&D Systems, Minneapolis, MN). The Abs of PaCXCR4a and PaCXCR4b were prepared using peptides derived from these proteins (1:200, PaCXCR4a: 107–121 aa, and 1:200, PaCXCR4b: 102–115 aa; GL Biochem). These peptides were synthesized to generate the respective mAbs (1 mg/ml; GL Biochem). The PaCXCR4a and PaCXCR4b mAbs generated, named 3G4 and 7D6, are both IgG1 Abs. We cloned the full-length versions of PaCXCR4a, PaCXCR4b, and PaSDF-1 into pcDNA3.1 plasmid (Invitrogen) and then transfected the constructs into HEK293T cells to produce recombinant proteins. For PaCXCR4a and PaCXCR4b detection, HEK293T cells were lyzed for Western blot analysis. For PaSDF-1 detection, the HEK293T supernatant was concentrated in Amicon-Ultra 4 centrifugal filter units with a 3-kDa cutoff (Millipore, Bedford, MA) for Western blot detection. The Abs were validated using Western blot (Supplemental Fig. 1).

SDF-1 protein expression was measured in the whole kidney of ayu by ELISA. Tissue samples tested in this study were homogenized in a buffer used for Western blotting. Microplates (Nunc, Roskilde, Denmark) were coated overnight using tissue samples. Plates were precoated using 50 μg/ml poly-l-lysine (Sigma-Aldrich) to increase protein binding. Blocking of unbound binding sites was performed using 5% teleost gelatin (Sigma-Aldrich). An anti–PaSDF-1 polyclonal Ab was prepared by GL Biochem using a peptide derived from PaSDF-1 (81–98 aa). One hundred microliters of the anti–PaSDF-1 Ab (5 μg/ml) was added to each well, and the plates were then incubated for 1 h and washed three times. One hundred microliters of an HRP-labeled secondary Ab (1:2500 dilution in PBS; Santa Cruz Biotechnology, Santa Cruz, CA) was added to each well, after which the plates were incubated for 1 h and washed three times. Finally, an alkaline phosphatase yellow liquid substrate system for ELISA (Sigma-Aldrich) was used, and the OD was measured at 405 nm.

N-terminal peptides of PaSDF-1 (KPLSLVERCWCRTTASTVPQR), large yellow croaker SDF-1 (PcSDF-1, KPISLVERCYCRSTVNNIPRS), and tiger puffer SDF-1 (TrSDF-1, KPISLVERCWCRSTLNTVPQR) were synthesized for receptor-binding assay and proliferation analysis by GL Biochem. The 21 aa of the N terminus of SDF-1 can activate CXCR4 signaling (36). Receptor-binding assays were performed with CXCR4a-HEK293T or CXCR4b-HEK293T cells using [125I]–PaSDF-1, [125I]-LPS, [125I]–PcSDF-1, and [125I] –TrSDF-1 labeled by the Bolton–Hunter procedure (37). Two days after transfection, cells were washed in PBS and incubated at a total of 5 × 106 cells in binding buffer (50 mM HEPES [pH 7.4], 1 mM CaCl2, 1 mM MgCl2, 0.2% BSA, and 0.1% NaN3) containing 5% dialyzed FBS (Invitrogen) for 1 h at room temperature before being used for the binding experiment. Fifty microliters of cells were incubated with 0.2 nM [125I]-labeled PaSDF-1 or 2 μg/ml [125I]-labeled LPS for 30 min at 24°C, and the reaction was terminated by the addition of 1 ml of ice-cold washing buffer (10 mM HEPES [pH 7.4], 5 mM CaCl2, 1 mM MgCl2, and 0.5 M NaCl). The solution was filtered under a vacuum through a Whatman GF/C glass filter, and the filter was then washed three times with washing buffer. The radioactivity retained on the filter was counted on a γ counter (300 SL; Hidex, Turku, Finland), and the IC50 and Kd values were calculated by nonlinear regression analysis in GraphPad Prism 6 (GraphPad Software, La Jolla, CA). Nonspecific binding was determined in the presence of 1 mg/ml unlabeled LPS or 100 nM unlabeled SDF-1. Specific binding values in the saturation and competitive binding experiments were calculated as the total binding minus nonspecific binding.

For in vivo gene knockdown by lentivirus delivery, small interfering RNAs (siRNAs) against PaSRB2a, PaCXCR4a, PaCXCR4b, PaSDF-1, and SCL were predicted by BLOCK-iT RNAi Designer (Invitrogen). Short hairpin RNAs (shRNAs) containing the selected siRNA sequences were designed and are listed in Table II. DNA oligonucleotides for shRNA expression were synthesized by Invitrogen, annealed, and constructed into the pSUPER Vector (Oligoengine, Seattle, WA) downstream of the [1H] promoter as previously described (38). The constructs generated with different siRNAs (1 μg) or control pSUPER (1 μg) along with the overexpression plasmid pcDNA3.1-target genes (1 μg) were cotransfected into HEK293T cells in 12-well plates. The efficiencies of the siRNAs against target genes were determined by RT-qPCR.

The U6 promoter cassette in the lentiviral vector pLB was replaced with the [1H]-siRNA cassette excised from the most effective siRNA constructs to produce lentiviral vectors. Lentiviruses were produced by transient transfection of the packaging cell line HEK293T cells. In brief, lentiviral vectors (20 μg) were cotransfected with the pCMV-dR8.2 dvpr (15 μg) and pCMV–VSV-G (6 μg) packaging vectors into HEK293T cells using FuGENE 6 Transfection Reagent (Promega). The lentiviral supernatant was harvested at 48–72 h posttransfection, concentrated via ultracentrifugation at 25,000 rpm for 90 min at 4°C to be dissolved in 100 μl of PBS, and then purified and concentrated using a Fast-Trap Lentivirus Purification and Concentration Kit (Millipore). Lentiviral titers were determined by transduction analysis of GFP expression in HEK293T cells, and infected cells were examined by fluorescence microscopy (Nikon, Tokyo, Japan).

The silencing efficiencies of the constructed lentiviruses were determined both in vitro and in vivo. In vitro, HEK293T cells were transfected with the pcDNA3.1-target gene. Concentrated lentivirus (10 μl) was added to the culture medium, and 4 μg/ml polybrene (Sigma-Aldrich) was simultaneously added to increase the infection efficiency. The silencing efficiencies of the lentiviruses were detected by RT-qPCR. For the in vivo assay, the lentiviruses (1 × 108 transducing units per fish per day) were repeatedly delivered into fish by injection once a day for 3 d. Total RNA was isolated from the liver and spleen and reverse transcribed into cDNA. RT-qPCR was conducted to evaluate the efficiency of in vivo suppression of the target gene.

A CFU cell (CFU-C) assay was carried out as described previously (13). Briefly, we plated mononuclear cells in 2.5 ml of methylcellulose medium supplemented with 10% conditioned medium obtained from PBMCs in the presence of PHA (Sigma-Aldrich) for 7 d. The numbers of mononuclear cells were 2 × 105 per plate from the whole kidney or 1 × 106 per plate from the blood. After 7 d of culture, the number of colonies per dish was counted.

For CFU-C analysis of PaCXCR4a+ and PaCXCR4b+ whole kidney leukocytes, cells were isolated by magnetic beads. Ayu whole kidney leukocyte–enriched fractions were obtained from the Ficoll medium interface using a Ficoll density gradient (1.077 ± 0.001 g/ml) (Invitrogen, Shanghai, China). After incubating with the Ab against PaCXCR4a (1:200) or PaCXCR4b (1:200), cells were collected by magnetic bead separation (Miltenyi Biotec, Bergisch Gladbach, Germany). In total, 1 × 105, 2 × 105, and 3 × 105 cells per plate were used to determine the CFU-C numbers and calculate the average value in one sample to reduce SE.

For analysis of blood immune cells, blood was taken from ayu that were anesthetized (0.03% [v/v] ethylene glycol monophenyl ether; Sinopharm Chemical Reagent, Shanghai, China) using heparin vacutainers (Becton Dickinson, Franklin Lakes, NJ) and then centrifuged at 1500 rpm for 5 min. Blood plasma was decanted from the blood cells, which were resuspended in 4 ml of PBS, layered using a Ficoll gradient (Invitrogen), and subsequently centrifuged at 400 × g for 25 min. The PBMC fraction at the Ficoll-medium interface was collected. The cells were stained with an Ab for 30 min at 4°C, and the samples were analyzed on a flow cytometer (Gallios; Beckman Coulter Diagnostics). The analysis was performed using Kaluza Software (Beckman Coulter Diagnostics) and FlowJo Software (Tree Star, Ashland, OR). An anti–PaCSF1R Ab was prepared in our previous work (39). mAbs against PaCD4, PaCD8, and PaIgM were prepared using peptides derived from these proteins (PaCD4: 58–70 aa, PaCD8: 25–40 aa, PaIgM: 165–178 aa; GL Biochem). The Abs were labeled with PE using an Ab Labeling Kit (Thermo Fisher Scientific).

For side population (SP) cell isolation, whole kidney leukocytes were stained with 7.5 μg/ml Hoechst 33342 (Sigma-Aldrich) at a concentration of 5 × 107 cells/ml in staining medium (RPMI 1640 containing 2% FBS) for 90 min at 25°C. After Hoechst 33342 staining, whole kidney leukocytes were washed by centrifugation, resuspended in RPMI 1640 containing 2% FBS, and kept on ice until use. Flow cytometry analysis and sorting were performed on a flow cytometer (MACSQuant Analyzer 10; Miltenyi Biotec). The Hoechst dye was excited by a UV laser, and its fluorescence was measured at two wavelengths using a 410/20 (Hoechst blue) bandpass filter and a 575/20 (Hoechst red) bandpass filter. The SP cells were identified by a Hoechst blue versus Hoechst red dot-plot and defined as HSPCs.

Cells were loaded with dihydrorhodamine-123 (DHR; 5 mg/ml; Sigma-Aldrich) for 5 min at 24°C. The cells in the control group were treated with PBS. Reactive oxygen species (ROS) production was quantified via flow cytometry by measuring intracellular rhodamine (Gallios Flow Cytometer; Beckman Coulter Diagnostics), and the ROS content was expressed as the mean fluorescence intensity of treated cells.

Three MHC class I (MHC I) alleles were identified in ayu until now. The frequencies of ayu MHC I alleles are 62% for MhcPlal-UAA*0101, 19% for MhcPlal-UAA*0201, and 21% for MhcPlal-UAA*0301. Ayu were hand-netted, marked by a fin clip, and released. Total RNA was isolated from ayu fin samples for MHC I allele analysis. Ayu were recaptured by hand-netting and assigned into new groups according to the MHC I allele information for MHC-matched HSPC transplantation. We determined the MHC I alleles according to PCR products of the fin cDNA from ayu. The amplification was performed using the primers of MhcPlal-UAA*0101, MhcPlal-UAA*0201, and MhcPlal-UAA*0301 (Table II). HSPCs for transplantation were obtained from ayu whole kidney. Single-cell suspensions of whole kidney were prepared using a 100-μm wire mesh, layered using a Ficoll gradient (GE Healthcare), and subsequently centrifuged at 400 × g for 25 min. The leukocyte fraction in the Ficoll-medium interface was collected and resuspended in PBS. The SP cells were sorted from whole kidney leukocytes using the MoFlo XDP Cell Sorter (Beckman Coulter Diagnostics) and defined as HSPCs. Concentrated lentivirus with GFP was added to the culture medium, and 4 μg/ml polybrene (Sigma-Aldrich) was simultaneously added to increase the infection efficiency. After 48 h of transduction, GFP+ SP cells were sorted for transplantation. The SP cells were injected into an irradiated ayu (25 Gy) via caudal vessels.

Limiting-dilution competitive repopulating unit (CRU) assays were performed as described previously (40). Lethally irradiated host ayu received a mix of limiting quantities of GFP-labeled CXCR4a+ or CXCR4b+ SP cells and 2 × 105 ayu whole kidney leukocytes. After 16 wk, the relative contribution of GFP+ leukocytes to total leukocytes was established. Animals with >1% donor contribution were considered positive for donor cell engraftment. The CRU frequency was calculated using LCALC Software (Stem Cell Technologies).

The CRU frequencies of 1400 CXCR4a+ and 600 CXCR4b+ SP cells were similar to that of 2 × 105 whole kidney leukocytes. For analysis of differentiation capability, 1400 CXCR4a+ SP cells labeled with GFP were transplanted into ayu with 600 unlabeled CXCR4b+ SP cells serving as competitor cells. Moreover, 600 CXCR4b+ SP cells labeled with GFP were transplanted into ayu with 1400 unlabeled CXCR4a+ SP cells serving as competitor cells. The percentages of GFP+ MO/MΦs, neutrophils, T cells, and B cells were analyzed by flow cytometry. The ratios were defined as GFP+ MO/MΦs divided by total MO/MΦs, GFP+ neutrophils divided by total neutrophils, GFP+ CD4 cells divided by total CD4 cells, GFP+ CD8 cells divided by total CD8 cells, and GFP+ IgM cells divided by total IgM cells.

To determine the proliferation status of the cells, DNA synthesis was measured by BrdU incorporation using BD Pharmingen BrdU Flow Kits (BD Biosciences) according to the manufacturer’s protocol. Ayu were given two daily doses of BrdU (6 mg per ayu) by injection (i.p.) for three consecutive days. BrdU+ HSPCs were analyzed 3 h after the last administration.

For proliferation analysis of PaCXCR4a+ and PaCXCR4b+ HSPCs, ayu were transplanted with PaCXCR4a+ and PaCXCR4b+ HSPCs, respectively, and gene knockdown, BrdU injection, and flow cytometry were then performed to measure HSPC proliferation.

Ayu were transplanted with PaCXCR4a+ or PaCXCR4b+ HSPCs, and PaCXCR4a and PaCXCR4b were knocked down using lentivirus vectors encoding shRNA. Ayu were injected weekly with 5-fluorouracil (100 mg/kg body weight) after 16 wk of HSPC transplantation, and the survival rates of the PaCXCR4a and PaCXCR4b knockdown ayu were monitored for 7 wk.

The data represent the mean ± SEM. The biological repeats are indicated by n. Sample size was chosen based on preliminary data and observed effect sizes. Animal experiments were performed by an observer blinded to the experimental conditions. We analyzed the survival curves with the Kaplan–Meier method using SPSS (version 13.0) Software. The remaining data were analyzed by one-way ANOVA. When variances were significantly different (p < 0.05), logarithmic transformation was used to stabilize the variance. If the normal distribution was not valid, statistical significance was evaluated using the Mann–Whitney U test (two-tailed). The single, double, and triple asterisk symbols (*, **, and ***) represent p values < 0.05, 0.01, and 0.001, respectively.

To elucidate the expression and function of CXCR4 proteins in teleosts, we first obtained the PaCXCR4a and PaCXCR4b gene sequences from transcriptome data (GSE73321). Sequence comparisons revealed that PaCXCR4a and PaCXCR4b shared 85.0 and 73.6% amino acid identity, respectively, with the large yellow croaker and cavefish genes. Phylogenetic analysis indicated that the fish CXCR4b genes were grouped together to form a cluster distinct from that of the fish CXCR4a and mammalian CXCR4 clusters (Fig. 1A, Table I). Because HSPCs are known to exist in teleost whole kidney (41), the HSPC function in ayu whole kidney was investigated. After Vibrio anguillarum infection or LPS treatment, PaCXCR4a was dramatically upregulated, whereas PaCXCR4b was slightly upregulated in the whole kidney (Fig. 1B, Table II). The PaCXCR4a protein level was upregulated in ayu HSPCs (SP cells) after LPS treatment, whereas the PaCXCR4b protein level was not changed after LPS treatment (Fig. 1C). We further investigated the surface expression of PaCXCR4a and PaCXCR4b in ayu HSPCs, revealing that the surface expression of PaCXCR4a was increased after LPS treatment (Fig. 1D). Moreover, PaCXCR4a mRNA was expressed in PaCXCR4a+ HSPCs, whereas PaCXCR4b mRNA was expressed in PaCXCR4b+ HSPCs (Fig. 1E).

FIGURE 1.

Two CXCR4 genes were expressed in teleost HSPCs. (A) Phylogenetic (neighbor-joining) analysis of the complete amino acid sequence of teleost and mammalian CXCR4s using the Molecular Evolutionary Genetics Analysis 5.0 Program. Node values represent the percentage bootstrap confidence derived from 1000 replicates. The source sequences are listed in Table I. (B) The mRNA expression of PaCXCR4s in ayu whole kidney following V. anguillarum infection at 1 × 104 CFU per fish or LPS treatment (10 μg/g body weight; n = 5). (C) The protein expression of PaCXCR4a and PaCXCR4b in ayu HSPCs as detected by Western blot. Representative blots of three independent experiments are shown. (D) The expression of PaCXCR4a and PaCXCR4b on the surface of HSPCs following LPS treatment. (E) The mRNA levels of PaCXCR4a and PaCXCR4b in PaCXCR4a+ and PaCXCR4b+ HSPCs. Each bar represents the mean ± SE (n = 5). Data are representative of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 1.

Two CXCR4 genes were expressed in teleost HSPCs. (A) Phylogenetic (neighbor-joining) analysis of the complete amino acid sequence of teleost and mammalian CXCR4s using the Molecular Evolutionary Genetics Analysis 5.0 Program. Node values represent the percentage bootstrap confidence derived from 1000 replicates. The source sequences are listed in Table I. (B) The mRNA expression of PaCXCR4s in ayu whole kidney following V. anguillarum infection at 1 × 104 CFU per fish or LPS treatment (10 μg/g body weight; n = 5). (C) The protein expression of PaCXCR4a and PaCXCR4b in ayu HSPCs as detected by Western blot. Representative blots of three independent experiments are shown. (D) The expression of PaCXCR4a and PaCXCR4b on the surface of HSPCs following LPS treatment. (E) The mRNA levels of PaCXCR4a and PaCXCR4b in PaCXCR4a+ and PaCXCR4b+ HSPCs. Each bar represents the mean ± SE (n = 5). Data are representative of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal
Table I.
Sequences used in this study
Accession Number Ensemble IdentifierSpeciesProtein
Latin NameEnglish Name
ENSTRUP00000022640 Takifugu rubripes tiger puffer CXCR4b 
ENSTNIP00000003765 Tetraodon nigroviridis spotted green pufferfish CXCR4b 
ENSGACP00000009641 Gasterosteus aculeatus Three spined stickleback CXCR4b 
XM 010740176 Larimichthys crocea large yellow croaker CXCR4b 
ENSORLP00000025392 Oryzias latipes Japanese rice fish CXCR4b 
ENSONIP00000005554 Oreochromis niloticus Nile tilapia CXCR4b 
XM 020920341 Boleophthalmus pectinirostris giant mudskipper CXCR4b 
ENSGMOP00000010259 Gadus morhua Atlantic cod CXCR4b 
MN148391 Plecoglossus altivelis ayu CXCR4b 
ENSDARP00000061498 Danio rerio zebrafish CXCR4b 
AXF84207 Ctenopharyngodon idella grass carp CXCR4b 
LHQP01018997 Cyprinus carpio common carp CXCR4b 
QPKE01007971 Carassius auratus goldfish CXCR4b 
XP_007232446 Astyanax mexicanus Mexican tetra CXCR4b 
AXF84206 Ctenopharyngodon Idella grass carp CXCR4a 
ENSDARP00000074800 Danio rerio zebrafish CXCR4a 
XP_007230928 Astyanax mexicanus Mexican tetra CXCR4a 
LHQP01016790 Cyprinus carpio common carp CXCR4a 
GBZM01009969 Carassius auratus goldfish CXCR4a 
ENSP00000386884 Homo sapiens human CXCR4 
ENSMUSP00000053489 Mus musculus mouse CXCR4 
GFIX01019416 Gadus morhua Atlantic cod CXCR4a 
MN148390 Plecoglossus altivelis ayu CXCR4a 
ENSTRUP00000026928 Takifugu rubripes tiger puffer CXCR4a 
ENSTNIP00000010726 Tetraodon nigroviridis spotted green pufferfish CXCR4a 
ENSONIP00000019715 Oreochromis niloticus Nile tilapia CXCR4a 
ENSORLP00000014200 Oryzias latipes Japanese ricefish CXCR4a 
ENSLOCP00000001045 Lepisosteus oculatus spotted gar CXCR4a 
ENSGACT000000016399 Gasterosteus aculeatus Three spined stickleback CXCR4a 
JRPU01002392 Larimichthys crocea large yellow croaker CXCR4a 
JACM01036689 Boleophthalmus pectinirostris giant mudskipper CXCR4a 
MN148393 Plecoglossus altivelis ayu SDF-1 
XM_019259616 Larimichthys crocea large yellow croaker SDF-1 
XM_011612317 Takifugu rubripes tiger puffer SDF-1 
KX228907 Plecoglossus altivelis ayu MhcPlal-UAA*0101 
MN628572 Plecoglossus altivelis ayu MhcPlal-UAA*0201 
MN628573 Plecoglossus altivelis ayu MhcPlal-UAA*0301 
Accession Number Ensemble IdentifierSpeciesProtein
Latin NameEnglish Name
ENSTRUP00000022640 Takifugu rubripes tiger puffer CXCR4b 
ENSTNIP00000003765 Tetraodon nigroviridis spotted green pufferfish CXCR4b 
ENSGACP00000009641 Gasterosteus aculeatus Three spined stickleback CXCR4b 
XM 010740176 Larimichthys crocea large yellow croaker CXCR4b 
ENSORLP00000025392 Oryzias latipes Japanese rice fish CXCR4b 
ENSONIP00000005554 Oreochromis niloticus Nile tilapia CXCR4b 
XM 020920341 Boleophthalmus pectinirostris giant mudskipper CXCR4b 
ENSGMOP00000010259 Gadus morhua Atlantic cod CXCR4b 
MN148391 Plecoglossus altivelis ayu CXCR4b 
ENSDARP00000061498 Danio rerio zebrafish CXCR4b 
AXF84207 Ctenopharyngodon idella grass carp CXCR4b 
LHQP01018997 Cyprinus carpio common carp CXCR4b 
QPKE01007971 Carassius auratus goldfish CXCR4b 
XP_007232446 Astyanax mexicanus Mexican tetra CXCR4b 
AXF84206 Ctenopharyngodon Idella grass carp CXCR4a 
ENSDARP00000074800 Danio rerio zebrafish CXCR4a 
XP_007230928 Astyanax mexicanus Mexican tetra CXCR4a 
LHQP01016790 Cyprinus carpio common carp CXCR4a 
GBZM01009969 Carassius auratus goldfish CXCR4a 
ENSP00000386884 Homo sapiens human CXCR4 
ENSMUSP00000053489 Mus musculus mouse CXCR4 
GFIX01019416 Gadus morhua Atlantic cod CXCR4a 
MN148390 Plecoglossus altivelis ayu CXCR4a 
ENSTRUP00000026928 Takifugu rubripes tiger puffer CXCR4a 
ENSTNIP00000010726 Tetraodon nigroviridis spotted green pufferfish CXCR4a 
ENSONIP00000019715 Oreochromis niloticus Nile tilapia CXCR4a 
ENSORLP00000014200 Oryzias latipes Japanese ricefish CXCR4a 
ENSLOCP00000001045 Lepisosteus oculatus spotted gar CXCR4a 
ENSGACT000000016399 Gasterosteus aculeatus Three spined stickleback CXCR4a 
JRPU01002392 Larimichthys crocea large yellow croaker CXCR4a 
JACM01036689 Boleophthalmus pectinirostris giant mudskipper CXCR4a 
MN148393 Plecoglossus altivelis ayu SDF-1 
XM_019259616 Larimichthys crocea large yellow croaker SDF-1 
XM_011612317 Takifugu rubripes tiger puffer SDF-1 
KX228907 Plecoglossus altivelis ayu MhcPlal-UAA*0101 
MN628572 Plecoglossus altivelis ayu MhcPlal-UAA*0201 
MN628573 Plecoglossus altivelis ayu MhcPlal-UAA*0301 
Table II.
Primers used in this study
PrimerGeneAccession NumberNucleotide Sequence (5′→3′)Sequence Information
PaCXCR4aF CXCR4a MN148390 5′-GGAATTCATGTCCTACTATGAGCATATT-3′a Eukaryotic expression 
PaCXCR4aR 5′-CCTCGAGTTAGCTGGAGTGCAGACTAG-3′b 
PaCXCR4bF CXCR4b MN148391 5′-GGAATTCATGGCATACTACGAGGAGAG-3′a Eukaryotic expression 
PaCXCR4bR 5′-CCTCGAGTTAACTGTAGAGAACACTGGAG-3′b 
Pa SDF-1F SDF-1 MN148393 5′-GGAATTCATGGATTTGAAATTATTGGCGTT-3′a Eukaryotic expression 
Pa SDF-1R 5′-CCTCGAGCTAGTTGGCTTGTTTGGATCT-3′b 
PaMhcPlal-UAA*0101F MhcPlal-UAA*0101  5′-CAAGTCCCAAACTTCCCAGA-3′ RT-qPCR 
PaMhcPlal-UAA*0101R 5′-GGTACCCAGCGCTATACCAG-3′ 
PaMhcPlal-UAA*0201F MhcPlal-UAA*0101  5′-TCATTCCGCGACTAACATCA-3′ RT-qPCR 
PaMhcPlal-UAA*0201R 5′-AGCTGCCTTGTTCATCCAGT-3′ 
PaMhcPlal-UAA*0301F MhcPlal-UAA*0101  5′-TCAACATCATTCCGCGACTA-3′ RT-qPCR 
PaMhcPlal-UAA*0301R 5′-GGCTCTCTGGATGTTGCTGT-3′ 
PaCXCR4aF CXCR4a MN148390 5′-CCAGAACTCAGGAAGCAGGA-3′ RT-qPCR 
PaCXCR4aR 5′-ACAGGAAGAAGCAGAGGACC-3′ 
PaCXCR4bF CXCR4b MN148391 5′-GGGCTTCCAGAACAAATGCA-3′ RT-qPCR 
PaCXCR4bR 5′-TGCCAGGTATCGATCCAGAC-3′ 
Pa SDF-1F SDF-1 MN148393 5′-AGTCTGGTGGAGAGGTGCTG-3′ RT-qPCR 
Pa SDF-1R 5′-TATCACTTGGAAGGGGCAGT-3′ 
Pa18S rRNAF 18S rRNA FN646593 5′-GAATGTCTGCCCTATCAACT-3′ RT-qPCR 
Pa18S rRNAR 5′-GATGTGGTAGCCGTTTCT-3′ 
PaSRB2aF SRB2a MH699855 5′-ACTTCTACCAAGCAGACCCC-3′ RT-qPCR 
PaSRB2aR 5′-GGGGAAGATGGTCTGGTTGA-3′ 
PaGATA2F GATA2 KU833214 5′-TGTGCTAACTGCCAGACGAC-3′ RT-qPCR 
PaGATA2R 5′-GGCTCTTTTTGGACTTGCTG-3′ 
PaRunX1F RUNX1 KU833216 5′-CATCCACCACCCTCTCATCT-3′ RT-qPCR 
PaRunX1R 5′-GTCCGTTCTCACCAGCTCTC-3′ 
PaGATA3F GATA3 KU833213 5′-GTGGCTTGAAGGAAGCAAAG-3′ RT-qPCR 
PaGATA3R 5′-CGGGTCTGGAGACACATCTT-3′ 
PaPax5F PAX5 KU833217 5′-CGTGTGTGTGACAACGACAG-3′ RT-qPCR 
PaPax5R 5′-CGCTGATGGAGTAGGAGGAG-3′ 
PaEgr1F EGR1 KU833211 5′-AGCCCAACCCCATCTACTCT-3′ RT-qPCR 
PaEgr1R 5′-AAGCTGGAACTGCACGTCTT-3′ 
PaPu.1F PU.1 KU833215 5′-GAGCTCAGACGAGGACGAAC-3′ RT-qPCR 
PaPu.1R 5′-CCGTTCCTCAACAGGTCAAT-3′ 
PaMPOF MPO MN158724 5′-ATCAACAAGTATCGCACGGC-3′ RT-qPCR 
PaMPOR 5′-TGGGTGAACTCTTGATCGCT-3′ 
PaSCLF SCL MN156535 5′-CGAATGGTGCAGTTGAGTCC-3′ RT-qPCR 
PaSCLR 5′-GAGCAGGTCGACGTTTCATC-3′ 
PaC/EBPαF C/EBPα MN156536 5′-CGTCGACAAGAACAGTAGCG-3′ RT-qPCR 
PaC/EBPαR 5′-GTAATGTGTCCAGTTCCCGC-3′ 
PaIkarosF Ikaros MN158722 5′-TTCGCCATATCAAGCTGCAC-3′ RT-qPCR 
PaIkarosR 5′-CTAGAGAACTGCGCTGCTTG-3′ 
PaMef2CF Mef2C MN158723 5′-CCGAGGACAAATACCGCAAG-3′ RT-qPCR 
PaMef2CR 5′-TAAGGGGCAGTAGGTTGTGG-3′ 
PaSRB2a-siRNA-sense SRB2a MH699855 5′-GATCCCCCCTTGACCTTAATCCGACCACTGGTTTCAAGAGAACCAGTGGTCGGATTAAGGTCAAGGTTTTTA-3′ Lentiviral RNAi 
PaSRB2a-siRNA-antisense 5′-AGCTTAAAAACCTTGACCTTAATCCGACCACTGGTTCTCTTGAAACCAGTGGTCGGATTAAGGTCAAGGGGG-3′ 
PaCXCR4a-siRNA-sense CXCR4a MN148390 5′-GATCCCCGCACGTCATCTACACGGTGAATTCAAGAGATTCACCGTGTAGATGACGTGCTTTTTA-3′ Lentiviral RNAi 
PaCXCR4a-siRNA-antisense 5′-AGCTTAAAAAGCACGTCATCTACACGGTGAATCTCTTGAATTCACCGTGTAGATGACGTGCGGG-3′ 
PaCXCR4b-siRNA-sense CXCR4b MN148391 5′-GATCCCCGGCAGGCAAGGTGATATATGTTTCAAGAGAACATATATCACCTTGCCTGCCTTTTTA-3′ Lentiviral RNAi 
PaCXCR4b-siRNA-antisense 5′-AGCTTAAAAAGGCAGGCAAGGTGATATATGTTCTCTTGAAACATATATCACCTTGCCTGCCGGG-3′ 
PaSDF-1-siRNA-sense SDF-1 MN148393 5′-GATCCCCGCATCCGTGAACTAAGGTTCCTTCAAGAGAGGAACCTTAGTTCACGGATGCTTTTTA-3′ Lentiviral RNAi 
PaSDF-1-siRNA-antisense 5′-AGCTTAAAAAGCATCCGTGAACTAAGGTTCCTCTCTTGAAGGAACCTTAGTTCACGGATGCGGG-3′ 
PaSCL-siRNA-sense SCL MN156535 5′-GATCCCCGGAAGTGCCGGTTATCGAATTTTCAAGAGAAATTCGATAACCGGCACTTCCTTTTTA-3′ Lentiviral RNAi 
PaSCL-siRNA-antisense 5′-AGCTTAAAAAGGAAGTGCCGGTTATCGAATTTCTCTTGAAAATTCGATAACCGGCACTTCCGGG-3′ 
MhcPlal-UAA*0101f MHC KX228907 5′-GGGACACACTCCCTCAAAT-3′ RT-PCR 
MhcPlal-UAA*0101r 5′-CGTAGTAGACCATCTGAACC-3′ 
MhcPlal-UAA*0201f MHC MN628572 5′-GGGACACACTCCCTCAAAT-3′ RT-PCR 
MhcPlal-UAA*0201r 5′-AGCGCGGCTTTAGAATCTC-3′ 
MhcPlal-UAA*0301f MHC MN628573 5′-GGGACACACTCCCTCAAAT-3′ RT-PCR 
MhcPlal-UAA*0301r 5′-GTTACTCATATAGATTCCAGTG-3′ 
PrimerGeneAccession NumberNucleotide Sequence (5′→3′)Sequence Information
PaCXCR4aF CXCR4a MN148390 5′-GGAATTCATGTCCTACTATGAGCATATT-3′a Eukaryotic expression 
PaCXCR4aR 5′-CCTCGAGTTAGCTGGAGTGCAGACTAG-3′b 
PaCXCR4bF CXCR4b MN148391 5′-GGAATTCATGGCATACTACGAGGAGAG-3′a Eukaryotic expression 
PaCXCR4bR 5′-CCTCGAGTTAACTGTAGAGAACACTGGAG-3′b 
Pa SDF-1F SDF-1 MN148393 5′-GGAATTCATGGATTTGAAATTATTGGCGTT-3′a Eukaryotic expression 
Pa SDF-1R 5′-CCTCGAGCTAGTTGGCTTGTTTGGATCT-3′b 
PaMhcPlal-UAA*0101F MhcPlal-UAA*0101  5′-CAAGTCCCAAACTTCCCAGA-3′ RT-qPCR 
PaMhcPlal-UAA*0101R 5′-GGTACCCAGCGCTATACCAG-3′ 
PaMhcPlal-UAA*0201F MhcPlal-UAA*0101  5′-TCATTCCGCGACTAACATCA-3′ RT-qPCR 
PaMhcPlal-UAA*0201R 5′-AGCTGCCTTGTTCATCCAGT-3′ 
PaMhcPlal-UAA*0301F MhcPlal-UAA*0101  5′-TCAACATCATTCCGCGACTA-3′ RT-qPCR 
PaMhcPlal-UAA*0301R 5′-GGCTCTCTGGATGTTGCTGT-3′ 
PaCXCR4aF CXCR4a MN148390 5′-CCAGAACTCAGGAAGCAGGA-3′ RT-qPCR 
PaCXCR4aR 5′-ACAGGAAGAAGCAGAGGACC-3′ 
PaCXCR4bF CXCR4b MN148391 5′-GGGCTTCCAGAACAAATGCA-3′ RT-qPCR 
PaCXCR4bR 5′-TGCCAGGTATCGATCCAGAC-3′ 
Pa SDF-1F SDF-1 MN148393 5′-AGTCTGGTGGAGAGGTGCTG-3′ RT-qPCR 
Pa SDF-1R 5′-TATCACTTGGAAGGGGCAGT-3′ 
Pa18S rRNAF 18S rRNA FN646593 5′-GAATGTCTGCCCTATCAACT-3′ RT-qPCR 
Pa18S rRNAR 5′-GATGTGGTAGCCGTTTCT-3′ 
PaSRB2aF SRB2a MH699855 5′-ACTTCTACCAAGCAGACCCC-3′ RT-qPCR 
PaSRB2aR 5′-GGGGAAGATGGTCTGGTTGA-3′ 
PaGATA2F GATA2 KU833214 5′-TGTGCTAACTGCCAGACGAC-3′ RT-qPCR 
PaGATA2R 5′-GGCTCTTTTTGGACTTGCTG-3′ 
PaRunX1F RUNX1 KU833216 5′-CATCCACCACCCTCTCATCT-3′ RT-qPCR 
PaRunX1R 5′-GTCCGTTCTCACCAGCTCTC-3′ 
PaGATA3F GATA3 KU833213 5′-GTGGCTTGAAGGAAGCAAAG-3′ RT-qPCR 
PaGATA3R 5′-CGGGTCTGGAGACACATCTT-3′ 
PaPax5F PAX5 KU833217 5′-CGTGTGTGTGACAACGACAG-3′ RT-qPCR 
PaPax5R 5′-CGCTGATGGAGTAGGAGGAG-3′ 
PaEgr1F EGR1 KU833211 5′-AGCCCAACCCCATCTACTCT-3′ RT-qPCR 
PaEgr1R 5′-AAGCTGGAACTGCACGTCTT-3′ 
PaPu.1F PU.1 KU833215 5′-GAGCTCAGACGAGGACGAAC-3′ RT-qPCR 
PaPu.1R 5′-CCGTTCCTCAACAGGTCAAT-3′ 
PaMPOF MPO MN158724 5′-ATCAACAAGTATCGCACGGC-3′ RT-qPCR 
PaMPOR 5′-TGGGTGAACTCTTGATCGCT-3′ 
PaSCLF SCL MN156535 5′-CGAATGGTGCAGTTGAGTCC-3′ RT-qPCR 
PaSCLR 5′-GAGCAGGTCGACGTTTCATC-3′ 
PaC/EBPαF C/EBPα MN156536 5′-CGTCGACAAGAACAGTAGCG-3′ RT-qPCR 
PaC/EBPαR 5′-GTAATGTGTCCAGTTCCCGC-3′ 
PaIkarosF Ikaros MN158722 5′-TTCGCCATATCAAGCTGCAC-3′ RT-qPCR 
PaIkarosR 5′-CTAGAGAACTGCGCTGCTTG-3′ 
PaMef2CF Mef2C MN158723 5′-CCGAGGACAAATACCGCAAG-3′ RT-qPCR 
PaMef2CR 5′-TAAGGGGCAGTAGGTTGTGG-3′ 
PaSRB2a-siRNA-sense SRB2a MH699855 5′-GATCCCCCCTTGACCTTAATCCGACCACTGGTTTCAAGAGAACCAGTGGTCGGATTAAGGTCAAGGTTTTTA-3′ Lentiviral RNAi 
PaSRB2a-siRNA-antisense 5′-AGCTTAAAAACCTTGACCTTAATCCGACCACTGGTTCTCTTGAAACCAGTGGTCGGATTAAGGTCAAGGGGG-3′ 
PaCXCR4a-siRNA-sense CXCR4a MN148390 5′-GATCCCCGCACGTCATCTACACGGTGAATTCAAGAGATTCACCGTGTAGATGACGTGCTTTTTA-3′ Lentiviral RNAi 
PaCXCR4a-siRNA-antisense 5′-AGCTTAAAAAGCACGTCATCTACACGGTGAATCTCTTGAATTCACCGTGTAGATGACGTGCGGG-3′ 
PaCXCR4b-siRNA-sense CXCR4b MN148391 5′-GATCCCCGGCAGGCAAGGTGATATATGTTTCAAGAGAACATATATCACCTTGCCTGCCTTTTTA-3′ Lentiviral RNAi 
PaCXCR4b-siRNA-antisense 5′-AGCTTAAAAAGGCAGGCAAGGTGATATATGTTCTCTTGAAACATATATCACCTTGCCTGCCGGG-3′ 
PaSDF-1-siRNA-sense SDF-1 MN148393 5′-GATCCCCGCATCCGTGAACTAAGGTTCCTTCAAGAGAGGAACCTTAGTTCACGGATGCTTTTTA-3′ Lentiviral RNAi 
PaSDF-1-siRNA-antisense 5′-AGCTTAAAAAGCATCCGTGAACTAAGGTTCCTCTCTTGAAGGAACCTTAGTTCACGGATGCGGG-3′ 
PaSCL-siRNA-sense SCL MN156535 5′-GATCCCCGGAAGTGCCGGTTATCGAATTTTCAAGAGAAATTCGATAACCGGCACTTCCTTTTTA-3′ Lentiviral RNAi 
PaSCL-siRNA-antisense 5′-AGCTTAAAAAGGAAGTGCCGGTTATCGAATTTCTCTTGAAAATTCGATAACCGGCACTTCCGGG-3′ 
MhcPlal-UAA*0101f MHC KX228907 5′-GGGACACACTCCCTCAAAT-3′ RT-PCR 
MhcPlal-UAA*0101r 5′-CGTAGTAGACCATCTGAACC-3′ 
MhcPlal-UAA*0201f MHC MN628572 5′-GGGACACACTCCCTCAAAT-3′ RT-PCR 
MhcPlal-UAA*0201r 5′-AGCGCGGCTTTAGAATCTC-3′ 
MhcPlal-UAA*0301f MHC MN628573 5′-GGGACACACTCCCTCAAAT-3′ RT-PCR 
MhcPlal-UAA*0301r 5′-GTTACTCATATAGATTCCAGTG-3′ 
a

The underlined nucleotides represent the restriction sites for EcoR I.

b

The underlined nucleotides represent the restriction sites for Xho I.

PaSRB2a is important for LPS internalization, and we knocked down PaSRB2a in ayu HSPCs by shRNA. The shRNA expression efficiency of the lentiviruses detected in HEK293T cells reached >90% using the expression of GFP as an indicator for PaSRB2a expression (Fig. 2A). PaSRB2a shRNA reduced the mRNA levels of PaSRB2a in both the whole kidney and HSPCs (Fig. 2B). PaCXCR4a, PaCXCR4b, and PaSDF-1 were measured in LPS-treated HSPCs after PaSRB2a shRNA incubation. PaCXCR4a mRNA was upregulated in LPS-stimulated HSPCs from the pSUPER-treated ayu (Fig. 2C). In the LPS-treated groups, PaSRB2a shRNA treatment reduced PaCXCR4a mRNA levels in ayu HSPCs (Fig. 2C). To investigate whether the binding of LPS to PaCXCR4a induced auto-upregulation of receptor expression, we used LPS to induce PaCXCR4a expression in ayu first and shRNA treatment was employed to knock down PaSRB2a expression. Hence, PaCXCR4a was expressed in ayu HSPCs, which have low PaSRB2a expression after LPS and PaSRB2a shRNA treatment. We found that PaCXCR4a expression was not changed in HSPCs from ayu with upon secondary LPS stimulation (Fig. 2D). These results did not support that the binding of LPS to PaCXCR4a induce the auto-upregulation of receptor expression. There was no significant change in the expression of PaCXCR4b after PaSRB2a shRNA treatment (Fig. 2E), and PaSRB2a shRNA reduced the LPS-induced inhibition of SDF-1 transcription (Fig. 2F). Moreover, PaSRB2a shRNA also reduced LPS-induced PaCXCR4a protein levels in ayu HSPCs (Fig. 2G). The ELISA results showed that PaSRB2a shRNA reduced the LPS-induced inhibition of PaSDF-1 protein expression (Fig. 2H). These data suggest that LPS regulates the expression of PaCXCR4a and PaSDF-1 in HSPCs.

FIGURE 2.

PaSRB2a mediated the effect of LPS on ayu HSPCs. (A) shRNA expression efficiency of lentiviruses. Scale bar, 100 μm. GFP fluorescence was detected in HEK293T cells by microscopy. (B) PaSRB2a mRNA levels in whole kidney and HSPC after delivery of lentiviruses into ayu. (C) PaCXCR4a mRNA levels after PaSRB2a shRNA treatment in HSPCs from LPS-treated ayu. (D) The binding of LPS with PaCXCR4a did not affect the auto-upregulation of receptor expression. LPS was used to induce PaCXCR4a expression in ayu, and shRNA treatment was then employed to knock down PaSRB2a expression. PaCXCR4a mRNA was detected in HSPCs of ayu after secondary LPS stimulation. (E) PaCXCR4b mRNA levels after PaSRB2a shRNA treatment in HSPCs from LPS-treated ayu. (F) PaSDF-1 mRNA levels after PaSRB2a shRNA treatment in the whole kidney from LPS-treated ayu (n = 5). (G) PaCXCR4a and PaCXCR4b protein levels in HSPCs from LPS-treated ayu. Representative blots of three independent experiments are shown. (H) PaSDF-1 protein levels in whole kidney from LPS-treated ayu (n = 5). Each bar represents the mean ± SE. Data are representative of two independent experiments. **p < 0.01, ***p < 0.001, ##p < 0.01, ###p < 0.001.

FIGURE 2.

PaSRB2a mediated the effect of LPS on ayu HSPCs. (A) shRNA expression efficiency of lentiviruses. Scale bar, 100 μm. GFP fluorescence was detected in HEK293T cells by microscopy. (B) PaSRB2a mRNA levels in whole kidney and HSPC after delivery of lentiviruses into ayu. (C) PaCXCR4a mRNA levels after PaSRB2a shRNA treatment in HSPCs from LPS-treated ayu. (D) The binding of LPS with PaCXCR4a did not affect the auto-upregulation of receptor expression. LPS was used to induce PaCXCR4a expression in ayu, and shRNA treatment was then employed to knock down PaSRB2a expression. PaCXCR4a mRNA was detected in HSPCs of ayu after secondary LPS stimulation. (E) PaCXCR4b mRNA levels after PaSRB2a shRNA treatment in HSPCs from LPS-treated ayu. (F) PaSDF-1 mRNA levels after PaSRB2a shRNA treatment in the whole kidney from LPS-treated ayu (n = 5). (G) PaCXCR4a and PaCXCR4b protein levels in HSPCs from LPS-treated ayu. Representative blots of three independent experiments are shown. (H) PaSDF-1 protein levels in whole kidney from LPS-treated ayu (n = 5). Each bar represents the mean ± SE. Data are representative of two independent experiments. **p < 0.01, ***p < 0.001, ##p < 0.01, ###p < 0.001.

Close modal

A receptor-binding assay showed that the SDF-1–binding activity of PaCXCR4b was significantly higher than that of PaCXCR4a, whereas the LPS-binding activity of PaCXCR4a was significantly higher than that of PaCXCR4b (Fig. 3A). In PaCXCR4a transfected HEK293T cells, increasing [125I]-LPS concentration resulted in a typical saturation curve with Kd value of 10.29 μg/ml (Supplemental Fig. 2A). In a competitive binding assay, unlabeled LPS displaced [125I]-LPS with IC50 values of 4.42 μg/ml (Supplemental Fig. 2B). These data demonstrated that PaCXCR4a had a good affinity for LPS. We also detected LPS binding in PaSRB2a-transfected HEK293T cells (Fig. 3B). In HEK293T cells double-transfected with PaCXCR4a and PaSRB2a, additive LPS binding was detected compared with that in PaCXCR4a-transfected HEK293T cells (Fig. 3B). Moreover, BPI inhibited LPS binding in PaCXCR4a-transfected HEK293T cells, whereas LECT2 did not affect LPS binding in PaCXCR4a-transfected HEK293T cells (Fig. 3C). We further explored the interactions of SDF-1 or LPS in HEK293T cells with PaCXCR4a/PaCXCR4b chimeras (Fig. 3D), revealing that the N terminus of PaCXCR4b was crucial for the interaction with PaSDF-1 (Fig. 3E) and that the extracellular loop 2 of PaCXCR4a was crucial for interaction with LPS (Fig. 3F).

FIGURE 3.

Binding activity analysis of the two PaCXCR4 genes. (A) The differential binding activity of PaCXCR4a or PaCXCR4b with SDF-1 or LPS. The value of PaCXCR4b binding with SDF-1 was set to 1, and the value of PaCXCR4a binding with LPS was set to 1. The HEK293T cells were transfected with PaCXCR4a or PaCXCR4b. (B) LPS-binding activity in PaCXCR4a and PaSRB2a double-transfected HEK293T cells. (C) LPS-binding activity after BPI peptide or LECT2 treatment in PaCXCR4a-transfected HEK293T cells. BPI peptide and LECT2 were incubated with HEK293T cells in all 3 μM. (D) Schematic representations of PaCXCR4a/PaCXCR4b chimeras. (E) The binding activity of SDF-1 with PaCXCR4a/PaCXCR4b chimeras in HEK293T cells. (F) The binding activity of LPS with PaCXCR4a/PaCXCR4b chimeras in HEK293T cells. Each bar represents the mean ± SE (n = 5). Data are representative of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, #p < 0.05.

FIGURE 3.

Binding activity analysis of the two PaCXCR4 genes. (A) The differential binding activity of PaCXCR4a or PaCXCR4b with SDF-1 or LPS. The value of PaCXCR4b binding with SDF-1 was set to 1, and the value of PaCXCR4a binding with LPS was set to 1. The HEK293T cells were transfected with PaCXCR4a or PaCXCR4b. (B) LPS-binding activity in PaCXCR4a and PaSRB2a double-transfected HEK293T cells. (C) LPS-binding activity after BPI peptide or LECT2 treatment in PaCXCR4a-transfected HEK293T cells. BPI peptide and LECT2 were incubated with HEK293T cells in all 3 μM. (D) Schematic representations of PaCXCR4a/PaCXCR4b chimeras. (E) The binding activity of SDF-1 with PaCXCR4a/PaCXCR4b chimeras in HEK293T cells. (F) The binding activity of LPS with PaCXCR4a/PaCXCR4b chimeras in HEK293T cells. Each bar represents the mean ± SE (n = 5). Data are representative of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, #p < 0.05.

Close modal

Because two CXCR4 genes generally exist in teleosts, we further performed receptor-binding assays of the two CXCR4 genes in other teleost species, the large yellow croaker and the tiger puffer. A receptor-binding assay showed that the SDF-1–binding activity of CXCR4b was significantly higher than that of CXCR4a, whereas the LPS-binding activity of CXCR4a was significantly higher than that of CXCR4b in large yellow croaker (Fig. 4A, 4B). The SDF-1–binding activity of CXCR4b was significantly higher than that of CXCR4a, whereas the LPS-binding activity of CXCR4a was significantly higher than that of CXCR4b in tiger puffers (Fig. 4C, 4D). These data suggest that the different binding activities of CXCR4a and CXCR4b are a general phenomenon in teleosts.

FIGURE 4.

Binding activity analysis of CXCR4s from large yellow croaker and tiger puffer. (A and B) The interaction of large yellow croaker CXCR4s with SDF-1 or LPS. (C and D) The interaction of tiger puffer CXCR4s with SDF-1 or LPS. The value of CXCR4b interacting with SDF-1 was set to 1, and the value of CXCR4a interacting with LPS was set to 1. Each bar represents the mean ± SE (n = 5). Data are representative of two independent experiments. ***p < 0.001.

FIGURE 4.

Binding activity analysis of CXCR4s from large yellow croaker and tiger puffer. (A and B) The interaction of large yellow croaker CXCR4s with SDF-1 or LPS. (C and D) The interaction of tiger puffer CXCR4s with SDF-1 or LPS. The value of CXCR4b interacting with SDF-1 was set to 1, and the value of CXCR4a interacting with LPS was set to 1. Each bar represents the mean ± SE (n = 5). Data are representative of two independent experiments. ***p < 0.001.

Close modal

Postinfection, HSPC homeostasis is affected by many factors, including bacterial components and host-derived cytokines, in mammals. We further investigated whether the HSPC number in the blood is regulated by infection. Escherichia coli infection resulted in a dramatic increase in the HSPC number in the blood of ayu (Fig. 5A), whereas the HSPC number was decreased in the blood of infected ayu 1 d after treatment with the BPI peptide, which neutralizes LPS (Fig. 5A). Moreover, infection in teleosts upregulates the expression of the cytokine LECT2 (42), which regulates HSPC homeostasis in mammals (15). We further investigated whether the HSPC number in the blood of infected ayu was regulated by LECT2. The HSPC number was decreased in the blood of infected ayu 3 d after treatment with an anti-LECT2 Ab (Fig. 5B). We further investigated the effect of V. anguillarum infection on the HSPC number in blood after LPS neutralization or LECT2 blockage. V. anguillarum infection resulted in an increase in the HSPC number in ayu blood (Fig. 5C). The HSPC number was decreased in the blood of infected ayu 3 d after treatment with the BPI peptide (Fig. 5C). After V. anguillarum infection, the HSPC number in the blood was decreased at 4 d after treatment with an anti-LECT2 Ab (Fig. 5D).

FIGURE 5.

LPS and LECT2 participate in HSPC mobilization in ayu. (A) The number of CFU-Cs in one million WBCs after BPI peptide treatment in E. coli–infected ayu. Ayu were injected i.p. with 1 × 108 CFU of E. coli. (B) The number of CFU-Cs in one million WBCs after anti-LECT2 Ab treatment in E. coli–infected ayu. (C) The number of CFU-Cs in blood after BPI peptide treatment in V. anguillarum–infected ayu. Ayu were injected i.p. with 1 × 103 CFU of V. anguillarum. (D) The number of CFU-Cs in blood in V. anguillarum–infected ayu after anti-LECT2 Ab treatment. Each bar represents the mean ± SE, n = 5. Data are representative of two independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 5.

LPS and LECT2 participate in HSPC mobilization in ayu. (A) The number of CFU-Cs in one million WBCs after BPI peptide treatment in E. coli–infected ayu. Ayu were injected i.p. with 1 × 108 CFU of E. coli. (B) The number of CFU-Cs in one million WBCs after anti-LECT2 Ab treatment in E. coli–infected ayu. (C) The number of CFU-Cs in blood after BPI peptide treatment in V. anguillarum–infected ayu. Ayu were injected i.p. with 1 × 103 CFU of V. anguillarum. (D) The number of CFU-Cs in blood in V. anguillarum–infected ayu after anti-LECT2 Ab treatment. Each bar represents the mean ± SE, n = 5. Data are representative of two independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

The transfection efficiency of the lentiviruses was detected using the expression of GFP as an indicator for HSPCs. The GFP-labeled HSPCs were transplanted into ayu with 2 × 105 whole kidney leukocytes, and GFP+ immune cells in ayu were differentiated from transplanted GFP-labeled HSPCs. The positive rates of GFP+ immune cells maintained within the ayu graft are shown in Fig. 6A. The similar frequencies of MO/MΦs, neutrophils, T cells, and B cells were found after GFP-labeled HSPC transplantation, suggesting that GFP labeling did not change the differentiation of immune cells. The number of hematopoietic stem cells (HSCs) were further estimated by measuring the whole kidney repopulating cells, known as the CRU, the frequency of which was measured using a competitive limiting-dilution repopulation assay (Fig. 6B). The frequency of CRUs was 14.1 and 35.7 per ml in the blood of LPS-treated and LECT2-treated ayu, respectively (Fig. 6C, 6D), suggesting that LECT2 mobilized more HSCs into blood compared with LPS. Moreover, the CFU-C number was increased in the blood of LPS- or LECT2-treated ayu (Fig. 6E). The CRU frequency in the blood of LECT2-treated ayu was 2.5-fold higher than that of LPS-treated ayu, whereas the number of CFU-Cs in the blood of LECT2-treated ayu was 1.3-fold higher than that of LPS-treated ayu. Hence, we further investigated the CRU frequency of PaCXCR4a+ and PaCXCR4b+ HSPCs. The CRU frequency was 1 per 331 in PaCXCR4a+ HSPCs from ayu whole kidney and 1 per 144 in PaCXCR4b+ HSPCs from ayu (Fig. 6F), suggesting that the numbers of HSCs in PaCXCR4b+ SP cells was higher compared with PaCXCR4a+ SP cells. PaCXCR4a+ and PaCXCR4b+ HSPCs therefore possess different reconstitution capabilities.

FIGURE 6.

The number of HSPCs in the blood of LPS- and LECT2-treated ayu. (A) GFP+ immune cells are maintained within the ayu graft. Scale bar, 50 μm. GFP fluorescence was detected in HEK293T cells by microscopy. GFP+ leukocytes were detected in the blood after transplantation with HSPCs labeled with GFP. (B) The protocol used to detect the CRU of ayu HSPCs. (C and D) The frequency of CRU in the blood of LPS- or LECT2-treated ayu. (E) The number of CFU-Cs in the blood of LPS- or LECT2-treated ayu. (F) The frequency of CRU within each group of ayu competitively transplanted with PaCXCR4a+ or PaCXCR4b+ HSPCs at each dose (n = 15 each point). Horizontal dashed line, 37% of recipient mice failed to be engrafted. Vertical dashed lines, various CRU frequencies for each treatment. ***p < 0.001, #p < 0.05.

FIGURE 6.

The number of HSPCs in the blood of LPS- and LECT2-treated ayu. (A) GFP+ immune cells are maintained within the ayu graft. Scale bar, 50 μm. GFP fluorescence was detected in HEK293T cells by microscopy. GFP+ leukocytes were detected in the blood after transplantation with HSPCs labeled with GFP. (B) The protocol used to detect the CRU of ayu HSPCs. (C and D) The frequency of CRU in the blood of LPS- or LECT2-treated ayu. (E) The number of CFU-Cs in the blood of LPS- or LECT2-treated ayu. (F) The frequency of CRU within each group of ayu competitively transplanted with PaCXCR4a+ or PaCXCR4b+ HSPCs at each dose (n = 15 each point). Horizontal dashed line, 37% of recipient mice failed to be engrafted. Vertical dashed lines, various CRU frequencies for each treatment. ***p < 0.001, #p < 0.05.

Close modal

We further investigated whether the SDF-1/PaCXCR4 axis participated in LPS- or LECT2-induced HSPC mobilization. Although the number of CFU-Cs was negligible in PaCXCR4a+ cells that were isolated by magnetic beads from the whole kidney and blood of healthy ayu, the number of CFU-Cs in PaCXCR4a+ cells from both whole kidney and blood were detectable in LPS-treated ayu (Fig. 7A, 7B). The number of CFU-Cs in PaCXCR4b+ cells from blood was not increased in LPS-treated ayu compared with that of PBS-treated ayu (Fig. 7C, 7D). The experimental protocol to analyze the effect of the SDF-1/CXCR4 axis on LPS-induced HSPC mobilization is shown in Fig. 7E. Although the number of CFU-Cs was not changed in blood after PBS treatment in PaCXCR4a shRNA-treated ayu, PaCXCR4a shRNA reduced the LPS-induced promotion of the number of CFU-Cs in ayu blood (Fig. 7F). However, PaCXCR4b and PaSDF-1 shRNA treatment led to an increase in the number of CFU-Cs in the blood of LPS-treated ayu (Fig. 7G, 7H). Because BPI affected the binding of LPS with PaCXCR4a and PaSRB2a affected PaCXCR4a expression, we further investigated the effect of BPI and PaSRB2a on HSPC mobilization, revealing that BPI peptide and PaSRB2a shRNA treatment reduced the LPS-induced HSPC mobilization (Fig. 7I, 7J). Hence, PaCXCR4a, BPI, and PaSRB2 affected LPS-induced HSPC mobilization in ayu.

FIGURE 7.

LPS induced HSPC mobilization via the SDF-1/CXCR4 axis in ayu. (A and B) The number of CFU-Cs per 106 cells after LPS treatment among PaCXCR4a+ cells in the whole kidney. (B) The number of CFU-Cs per 106 cells after LPS treatment among PaCXCR4a+ cells in blood. (C) The number of CFU-Cs per 106 cells after LPS treatment among PaCXCR4b+ cells in the whole kidney. (D) The number of CFU-Cs per 106 cells after LPS treatment among PaCXCR4b+ cells in blood. (E) The protocol used to evaluate the effect of LPS on HSPC mobilization. (F) The number of CFU-Cs per 106 cells of whole kidney and blood after LPS treatment in PaCXCR4a knockdown ayu. (G) The number of CFU-Cs per 106 whole kidney and blood cells after LPS treatment in the PaCXCR4b knockdown ayu. (H) The number of CFU-Cs per 106 whole kidney and blood cells after LPS treatment in PaSDF-1 knockdown ayu. (I) The number of CFU-Cs per 106 whole kidney and blood cells after LPS treatment in BPI peptide treated ayu. (J) The number of CFU-Cs per 106 whole kidney and blood cells after LPS treatment in PaSRB2a knockdown ayu. Each bar represents the mean ± SE (n = 5). Data are representative of three independent experiments. **p < 0.01, ***p < 0.001, #p < 0.05, ##p < 0.01, ###p < 0.001.

FIGURE 7.

LPS induced HSPC mobilization via the SDF-1/CXCR4 axis in ayu. (A and B) The number of CFU-Cs per 106 cells after LPS treatment among PaCXCR4a+ cells in the whole kidney. (B) The number of CFU-Cs per 106 cells after LPS treatment among PaCXCR4a+ cells in blood. (C) The number of CFU-Cs per 106 cells after LPS treatment among PaCXCR4b+ cells in the whole kidney. (D) The number of CFU-Cs per 106 cells after LPS treatment among PaCXCR4b+ cells in blood. (E) The protocol used to evaluate the effect of LPS on HSPC mobilization. (F) The number of CFU-Cs per 106 cells of whole kidney and blood after LPS treatment in PaCXCR4a knockdown ayu. (G) The number of CFU-Cs per 106 whole kidney and blood cells after LPS treatment in the PaCXCR4b knockdown ayu. (H) The number of CFU-Cs per 106 whole kidney and blood cells after LPS treatment in PaSDF-1 knockdown ayu. (I) The number of CFU-Cs per 106 whole kidney and blood cells after LPS treatment in BPI peptide treated ayu. (J) The number of CFU-Cs per 106 whole kidney and blood cells after LPS treatment in PaSRB2a knockdown ayu. Each bar represents the mean ± SE (n = 5). Data are representative of three independent experiments. **p < 0.01, ***p < 0.001, #p < 0.05, ##p < 0.01, ###p < 0.001.

Close modal

LECT2, a multifunctional cytokine, can induce HSPC mobilization. We further explored whether the SDF-1/PaCXCR4 axis mediated LPS-induced HSPC mobilization. The number of CFU-Cs from PaCXCR4b+ cells in both the whole kidney and blood increased significantly after LECT2 treatment, especially in the blood (Fig. 8A). No CFU-Cs were detected in PaCXCR4a+ cells from the whole kidney and blood after BSA or LECT2 treatment. PaCXCR4b shRNA increased HSPC mobilization in the BSA-treated ayu (Fig. 8B). LECT2 also increased HSPC mobilization in the pSUPER-treated ayu (Fig. 8B). However, LECT2 did not increase HSPC mobilization in the PaCXCR4b shRNA–treated ayu (Fig. 8B). Furthermore, LECT2 also did not increase HSPC mobilization in the PaSDF-1 shRNA–treated ayu (Fig. 8C). Hence, PaCXCR4b and PaSDF-1 mediated LECT2-induced HSPC mobilization.

FIGURE 8.

LECT2 induced HSPC mobilization via the SDF-1/CXCR4 axis in ayu. (A) The number of CFU-Cs per 106 cells among PaCXCR4b+ cells after LPS treatment. (B) The number of CFU-Cs per 106 whole kidney and blood cells after LPS treatment in PaCXCR4b knockdown ayu. (C) The number of CFU-Cs per 106 cells after LPS treatment in the PaSDF-1 knockdown ayu. Each bar represents the mean ± SE (n = 5). Data are representative of three independent experiments. **p < 0.01, ***p < 0.001, ##p < 0.01, ###p < 0.001.

FIGURE 8.

LECT2 induced HSPC mobilization via the SDF-1/CXCR4 axis in ayu. (A) The number of CFU-Cs per 106 cells among PaCXCR4b+ cells after LPS treatment. (B) The number of CFU-Cs per 106 whole kidney and blood cells after LPS treatment in PaCXCR4b knockdown ayu. (C) The number of CFU-Cs per 106 cells after LPS treatment in the PaSDF-1 knockdown ayu. Each bar represents the mean ± SE (n = 5). Data are representative of three independent experiments. **p < 0.01, ***p < 0.001, ##p < 0.01, ###p < 0.001.

Close modal

We further evaluated the proliferation capacity of HSPCs treated with PaCXCR4a or PaCXCR4b shRNA. BrdU can be incorporated into the newly synthesized DNA in place of thymidine and is commonly used in cell proliferation studies, and the percentage of BrdU+ HSPCs in the whole kidney was markedly increased in the PaCXCR4b+ HSPCs from the PaCXCR4b shRNA–treated ayu (Fig. 9A), suggesting that PaCXCR4b knockdown increased HSPC proliferation. We further investigated the roles of PaCXCR4a and PaCXCR4b in hematological failure. First, we distinguished the fish that died of hematological failure rather than from infection. The CFU-C number was significantly decreased in the blood of the ayu that were dying from hematological failure, whereas the CFU-C number was significantly increased in the blood of V. anguillarum–infected fish (Fig. 9B). There was no significant change in the PaCXCR4a+ HSPCs from the PaCXCR4a shRNA–treated ayu (Fig. 9C). In accordance with the enhanced proliferation of HSPCs in PaCXCR4b shRNA–treated ayu, PaCXCR4b shRNA–treated ayu died more readily from hematological failure after depletion of cycling HSPCs by weekly challenge with the cell cycle cytotoxic agent 5-fluorouracil (Fig. 9D). Furthermore, PaSDF-1 inhibited HSPC proliferation in the pSUPER-treated ayu but not in the PaCXCR4b shRNA–treated ayu (Fig. 9E). Hence, the SDF-1/CXCR4b axis inhibited HSPC proliferation in ayu.

FIGURE 9.

Proliferation analysis of PaCXCR4a+ and PaCXCR4b+ HSPCs. (A) Percentage of BrdU+ HSPCs in the whole kidney of PaCXCR4a+ or PaCXCR4b+ HSPC-transplanted ayu. Proliferation was determined by BrdU incorporation assay (n = 5). Each bar represents the mean ± SE. (B) The number of CFU-Cs per 106 blood cells in ayu after 5-FU or V. anguillarum treatment. (C) The survival rate of pSUPER- and PaCXCR4a shRNA–treated ayu. (D) The survival rate of PaCXCR4b shRNA–treated ayu. (E) Percentage of BrdU+ HSPCs after PaCXCR4b shRNA or PaSDF-1 treatment. Ayu were injected weekly with 5-FU (100 mg/kg body weight). For infection, ayu were injected i.p. with 1 × 104 CFU of V. anguillarum (n = 25 in each group). *p < 0.05, ***p < 0.001.

FIGURE 9.

Proliferation analysis of PaCXCR4a+ and PaCXCR4b+ HSPCs. (A) Percentage of BrdU+ HSPCs in the whole kidney of PaCXCR4a+ or PaCXCR4b+ HSPC-transplanted ayu. Proliferation was determined by BrdU incorporation assay (n = 5). Each bar represents the mean ± SE. (B) The number of CFU-Cs per 106 blood cells in ayu after 5-FU or V. anguillarum treatment. (C) The survival rate of pSUPER- and PaCXCR4a shRNA–treated ayu. (D) The survival rate of PaCXCR4b shRNA–treated ayu. (E) Percentage of BrdU+ HSPCs after PaCXCR4b shRNA or PaSDF-1 treatment. Ayu were injected weekly with 5-FU (100 mg/kg body weight). For infection, ayu were injected i.p. with 1 × 104 CFU of V. anguillarum (n = 25 in each group). *p < 0.05, ***p < 0.001.

Close modal

To analyze ROS production, cells were stained with DHR, and the fluorescence emitted by the oxidized DHR was measured by flow cytometry. ROS levels were lower in PaCXCR4b+ HSPCs than in PaCXCR4a+ HSPCs (Fig. 10A). Because ROS have been found to regulate HSPC proliferation (43), we further investigated whether ROS mediated the proliferation of PaCXCR4a+ and PaCXCR4b+ HSPCs. Briefly, PaCXCR4a+ or PaCXCR4b+ HSPCs were isolated from infected ayu. PaCXCR4a expression in PaCXCR4a+ HSPCs and PaCXCR4b expression in PaCXCR4b+ HSPCs were knocked down by shRNA treatment. ROS were detected in HSPCs after PaCXCR4a+ or PaCXCR4b+ HSPCs transplanted into ayu (Fig. 10B). Although PaCXCR4a shRNA treatment did not affect the ROS levels in PaCXCR4a+ HSPCs (Fig. 10C), PaCXCR4b shRNA treatment increased the ROS levels in PaCXCR4b+ HSPCs (Fig. 10D). PaCXCR4a shRNA treatment did not affect HSPC proliferation, which was blocked by the ROS inhibitor N-acetylcysteine (NAC) (Fig. 10E). PaCXCR4b shRNA treatment enhanced HSPC proliferation, whereas NAC blocked the PaCXCR4b effect on HSPC proliferation (Fig. 10F). PaSDF-1 treatment decreased the ROS levels of PaCXCR4b+ HSPCs but not those of PaCXCR4a+ HSPCs (Fig. 10G, 10H). PaSDF-1 treatment did not affect the ROS levels in PaCXCR4b shRNA–treated HSPCs (Fig. 10H). These data suggested that ROS mediated CXCR4b knockdown–induced HSPC proliferation.

FIGURE 10.

PaCXCR4b mediated HSPC proliferation through ROS signaling. (A) ROS levels in PaCXCR4a+ and PaCXCR4b+ HSPCs. (B) The protocol used to evaluate the effect of ROS on the proliferation of PaCXCR4a+ and PaCXCR4b+ HSPCs. (C) ROS levels were not changed in HSPCs after PaCXCR4a knockdown. (D) ROS levels were upregulated in HSPCs after PaCXCR4b knockdown. (E) Proliferation of PaCXCR4a knockdown HSPCs after treatment with NAC, an inhibitor of ROS. (F) Proliferation of PaCXCR4b knockdown HSPCs after treatment with NAC. (G) PaSDF-1 did not affect the ROS content in HSPCs. (H) PaSDF-1 decreased the ROS content in HSPCs. Each bar represents the mean ± SE (n = 5). Data are representative of three independent experiments. *p < 0.05, ***p < 0.001.

FIGURE 10.

PaCXCR4b mediated HSPC proliferation through ROS signaling. (A) ROS levels in PaCXCR4a+ and PaCXCR4b+ HSPCs. (B) The protocol used to evaluate the effect of ROS on the proliferation of PaCXCR4a+ and PaCXCR4b+ HSPCs. (C) ROS levels were not changed in HSPCs after PaCXCR4a knockdown. (D) ROS levels were upregulated in HSPCs after PaCXCR4b knockdown. (E) Proliferation of PaCXCR4a knockdown HSPCs after treatment with NAC, an inhibitor of ROS. (F) Proliferation of PaCXCR4b knockdown HSPCs after treatment with NAC. (G) PaSDF-1 did not affect the ROS content in HSPCs. (H) PaSDF-1 decreased the ROS content in HSPCs. Each bar represents the mean ± SE (n = 5). Data are representative of three independent experiments. *p < 0.05, ***p < 0.001.

Close modal

We further investigated the differentiation capability of PaCXCR4a+ HSPCs. The mRNA levels of GATA2, RUNX1 (mainly expressed in HSPCs), GATA3, PAX5 (mainly expressed in lymphocytes), EGR1, PU.1, and MPO (mainly expressed in myeloid cells) were measured in CFU-Cs from PaCXCR4a+ and PaCXCR4b+ cells. The mRNA expression of GATA2 and RUNX1 was downregulated after stimulation with PMA in both PaCXCR4a+ and PaCXCR4b+ HSPCs (Fig. 11A, 11B). The expression levels of GATA3 and PAX5 in PaCXCR4b+ CFU-Cs were similar to those in PaCXCR4a+ cells (Fig. 11C, 11D). The expression levels of EGR1, PU.1, and MPO were higher in PaCXCR4a+ CFU-Cs than in PaCXCR4b+ cells (Fig. 11E–G).

FIGURE 11.

The expression of lineage markers in CFU-Cs from PaCXCR4a+ and PaCXCR4b+ HSPCs. (A and B) mRNA expression of GATA2 and RUNX1 in CFU-Cs from PaCXCR4a+ or PaCXCR4b+ HSPCs. (C and D) mRNA expression of GATA3 and PAX5 in CFU-Cs from PaCXCR4a+ or PaCXCR4b+ HSPCs. (E and F) mRNA expression of EGR1 and PU.1 in CFU-Cs from PaCXCR4a+ or PaCXCR4b+ HSPCs. (G) mRNA expression of MPO in CFU-Cs from PaCXCR4a+ or PaCXCR4b+ HSPCs. Target gene transcripts were normalized to 18S rRNA transcripts. Each bar represents the mean ± SE (n = 5). Data are representative of two independent experiments. ***p < 0.001, #p < 0.05, ##p < 0.01, ###p < 0.001.

FIGURE 11.

The expression of lineage markers in CFU-Cs from PaCXCR4a+ and PaCXCR4b+ HSPCs. (A and B) mRNA expression of GATA2 and RUNX1 in CFU-Cs from PaCXCR4a+ or PaCXCR4b+ HSPCs. (C and D) mRNA expression of GATA3 and PAX5 in CFU-Cs from PaCXCR4a+ or PaCXCR4b+ HSPCs. (E and F) mRNA expression of EGR1 and PU.1 in CFU-Cs from PaCXCR4a+ or PaCXCR4b+ HSPCs. (G) mRNA expression of MPO in CFU-Cs from PaCXCR4a+ or PaCXCR4b+ HSPCs. Target gene transcripts were normalized to 18S rRNA transcripts. Each bar represents the mean ± SE (n = 5). Data are representative of two independent experiments. ***p < 0.001, #p < 0.05, ##p < 0.01, ###p < 0.001.

Close modal

We further explored the differentiation capability of PaCXCR4a+ and PaCXCR4b+ HSPCs by transplantation. GFP-labeled PaCXCR4a+ or PaCXCR4b+ HSPCs were transplanted into ayu. The GFP+ cell ratio of MO/MΦs and neutrophils after transplantation with PaCXCR4a+ HSPCs was significantly higher than that after transplantation with PaCXCR4b+ HSPCs (Fig. 12A, 12B). The GFP+ cell ratio of CD4+, CD8+, and IgM+ cells after transplantation with PaCXCR4a+ HSPCs was similar to that after transplantation with PaCXCR4b+ HSPCs (Fig. 12C–E).

FIGURE 12.

PaCXCR4a+ and PaCXCR4b+ HSPCs displayed distinct differentiation patterns. (A and B) The GFP+ cell ratio of MO/MΦs and neutrophils in ayu after PaCXCR4a+ or PaCXCR4b+ HSPC transplantation. (CE) The GFP+ cell ratio of T and B cells in ayu after PaCXCR4a+ or PaCXCR4b+ HSPC transplantation. Each bar represents the mean ± SE (n = 5). Data are representative of three independent experiments. ***p < 0.001.

FIGURE 12.

PaCXCR4a+ and PaCXCR4b+ HSPCs displayed distinct differentiation patterns. (A and B) The GFP+ cell ratio of MO/MΦs and neutrophils in ayu after PaCXCR4a+ or PaCXCR4b+ HSPC transplantation. (CE) The GFP+ cell ratio of T and B cells in ayu after PaCXCR4a+ or PaCXCR4b+ HSPC transplantation. Each bar represents the mean ± SE (n = 5). Data are representative of three independent experiments. ***p < 0.001.

Close modal

Transcription factors play important roles in the differentiation capability of HSPCs. In PaCXCR4a+ HSPCs, the mRNA expression of the transcription factors SCL and C/EBPα was significantly higher than that of in PaCXCR4b+ HSPCs (Fig. 13A, 13B). The mRNA expression levels of GATA3, Ikaros, and Mef2C in PaCXCR4b+ HSPCs were significantly higher than those in PaCXCR4a+ HSPCs (Fig. 13C–E). Because SCL expression in PaCXCR4a+ HSPCs was 3.25-fold greater than that in PaCXCR4b+ HSPCs, we focused on SCL as an example to investigate the effect of PaCXCR4 on SCL expression. In isolated PaCXCR4a+ HSPCs, SCL expression was downregulated after PaCXCR4a shRNA treatment (Fig. 13F). PaCXCR4b shRNA treatment did not affect SCL expression in HSPCs (Fig. 13G). Moreover, LPS also increased SCL expression in HSPCs (Fig. 13H). In the SCL knockdown ayu, the mRNA expression of PaCXCR4a and PaCXCR4b in the ayu whole kidney was not changed (Fig. 13I, 13J). The SCL protein levels were downregulated in HSPCs after PaCXCR4a shRNA treatment (Fig. 13K). Furthermore, LPS increased SCL mRNA expression in HSPCs of the pSUPER-treated ayu but not in those of the PaCXCR4a shRNA–treated ayu (Fig. 13L).

FIGURE 13.

PaCXCR4a mediated SCL upregulation. (AE) The mRNA expression of transcription factors in PaCXCR4a+ or PaCXCR4b+ HSPCs. (F and G) SCL mRNA expression in PaCXCR4a or PaCXCR4b knockdown ayu. (H) SCL mRNA expression was upregulated after LPS treatment. (I and J) PaCXCR4a and PaCXCR4b mRNA expression in the whole kidney of SCL knockdown ayu. (K) The effect of PaCXCR4a shRNA on SCL protein levels. (L) LPS did not change the SCL expression after PaCXCR4a shRNA treatment. Each bar represents the mean ± SE (n = 5). Data are representative of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 13.

PaCXCR4a mediated SCL upregulation. (AE) The mRNA expression of transcription factors in PaCXCR4a+ or PaCXCR4b+ HSPCs. (F and G) SCL mRNA expression in PaCXCR4a or PaCXCR4b knockdown ayu. (H) SCL mRNA expression was upregulated after LPS treatment. (I and J) PaCXCR4a and PaCXCR4b mRNA expression in the whole kidney of SCL knockdown ayu. (K) The effect of PaCXCR4a shRNA on SCL protein levels. (L) LPS did not change the SCL expression after PaCXCR4a shRNA treatment. Each bar represents the mean ± SE (n = 5). Data are representative of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

The effect of SCL on the differentiation capability of HSPCs was further investigated. The GFP+ percentages of MO/MΦs and neutrophils were decreased in the blood of ayu after transplantation with SCL shRNA–treated HSPCs (Fig. 14A, 14B). The GFP+ percentages of T and B cells were not changed in ayu after transplantation with SCL shRNA–treated HSPCs (Fig. 14C–E). Furthermore, SCL shRNA treatment did not change the mRNA expression of GATA3 and PAX5 in CFU-Cs from HSPCs (Fig. 14F, 14G). SCL shRNA treatment decreased the mRNA expression of EGR1, PU.1, and MPO in CFU-Cs from HSPCs (Figs. 14H–J, 15).

FIGURE 14.

SCL mediated HSPC differentiation. (AE) The ratio of GFP+ immune cells in ayu after the transplantation of pSUPER- or SCL shRNA–treated ayu HSPCs. (FJ) The expression of lineage markers in CFU-Cs from pSUPER- or SCL shRNA–treated ayu HSPCs. Each bar represents the mean ± SE (n = 5). Data are representative of three independent experiments. **p < 0.01, ***p < 0.001.

FIGURE 14.

SCL mediated HSPC differentiation. (AE) The ratio of GFP+ immune cells in ayu after the transplantation of pSUPER- or SCL shRNA–treated ayu HSPCs. (FJ) The expression of lineage markers in CFU-Cs from pSUPER- or SCL shRNA–treated ayu HSPCs. Each bar represents the mean ± SE (n = 5). Data are representative of three independent experiments. **p < 0.01, ***p < 0.001.

Close modal
FIGURE 15.

Teleost CXCR4s mediate HSPC proliferation, mobilization, and differentiation. CXCR4b suppresses ROS levels in HSPCs, whereas CXCR4a does not affect ROS levels in HSPCs. LPS preferentially mobilizes CXCR4a+ HSPCs into the blood, whereas LECT2 preferentially mobilizes CXCR4b+ HSPCs into the blood. CXCR4a+ HSPCs preferentially differentiate into myeloid cells.

FIGURE 15.

Teleost CXCR4s mediate HSPC proliferation, mobilization, and differentiation. CXCR4b suppresses ROS levels in HSPCs, whereas CXCR4a does not affect ROS levels in HSPCs. LPS preferentially mobilizes CXCR4a+ HSPCs into the blood, whereas LECT2 preferentially mobilizes CXCR4b+ HSPCs into the blood. CXCR4a+ HSPCs preferentially differentiate into myeloid cells.

Close modal

Chemokine receptors have been identified in diverse teleosts, with great differences between teleosts and mammals (16). A variety of teleost species possess two copies of the CXCR4 gene (16), whereas only one CXCR4 gene exists in mammals. In teleosts, CXCR4s have been found to be involved in development and neutrophil recruitment (2628), and we further investigated the functions of CXCR4a and CXCR4b in the hematopoietic system of teleosts. First, we found that CXCR4a and CXCR4b had different binding activities for LPS and SDF-1. The different binding activities of CXCR4a and CXCR4b also existed in other teleosts, large yellow croaker and tiger puffer. Moreover, CXCR4a+ and CXCR4b+ HSPCs displayed significant differences in HSPC homeostasis. To our knowledge, this report is the first study to illustrate the subfunctionalization of CXCR4a and CXCR4b in the hematopoietic system of teleosts. The gene duplication of the immune system may contribute to different immune regulation in teleosts compared with that in mammals (44, 45). Considering the well-known importance of CXCR4 in mammalian HSPC homeostasis (46), the subfunctionalization of CXCR4a and CXCR4b in teleosts is helpful for understanding the evolution of the hematopoietic system in vertebrates.

Peripheral inflammation or infection will affect the hematopoietic microenvironment to regulate HSPC mobilization in vertebrates (6). We first found that PaCXCR4b knockdown increased HSPCs in ayu blood and decreased HSPCs in ayu whole kidney. Hence, CXCR4b participates in HSPC mobilization from kidney into blood in teleosts. In mammals, CXCR4 also plays a major role in HSPC mobilization and retention in the hematopoietic microenvironment (47). Teleost CXCR4b may perform a similar function with mammalian CXCR4 in HSPC mobilization. In mammals, LPS induces HSPC mobilization through its receptor TLR4 and NOD-like receptors (8). Infection and inflammation also upregulate the expression of host-derived cytokines such as G-CSF, which is widely known to mediate HSPC mobilization (48). In this study, we found that LPS regulated HSPC mobilization through CXCR4a, whereas LECT2 regulated HSPC mobilization through CXCR4b and SDF-1 in teleosts. These results suggest that HSPC mobilization is regulated by both pathogen components and host cytokines in vertebrates. In teleosts, G-CSF also plays an important role in HSPC proliferation (49). However, G-CSF mobilization of HSPCs has not been observed in teleosts (13). Postinfection, the multifunctional cytokine LECT2 is downregulated in mammals (50), whereas it is dramatically upregulated in teleosts (42). LECT2, not G-CSF, induces HSPC mobilization in teleosts, suggesting that the mechanisms of infection-induced HSPC mobilization are different between teleost and mammals. Although teleost CXCR4b plays a similar role with mammalian CXCR4 in HSPC mobilization, cytokine-mediated HSPC mobilization is diverse in vertebrates.

We first found that CXCR4a preferentially interacted with LPS in teleosts. Although mammalian CXCR4 can interact with LPS (51), LPS affects HSPC homeostasis by interacting with TLR4 (14). However, TLR4 does not mediate LPS-induced HSPC mobilization (14). In this study, we further found that LPS also induced HSPC mobilization through CXCR4a. Our data did not support that CXCR4b participated in LPS-induced HSPC mobilization in ayu. Moreover, CXCR4a+ HSPCs exhibited a lower CRU frequency but a higher proliferation capability than CXCR4b+ HSPCs. Hence, we found specific LPS-mobilized HSPCs marked by CXCR4a, suggesting that teleosts produce bacterial infection–derived HSPCs. In our previous work, we identified SRB2a as the late-phase allergic skin reaction in teleost macrophages, including those from ayu (52). In this study, we found that both PaCXCR4a and PaSRB2a were expressed in the HSPCs of ayu as LPS receptors. BPI, PaSRB2a, and PaCXCR4a affected LPS-induced HSPC mobilization in ayu. Although PaSRB2a regulated PaCXCR4a expression after binding with LPS, PaCXCR4a did not affect the auto-upregulation of receptor expression after binding with LPS. Hence, PaSRB2a and PaCXCR4a have different functions as LPS receptors in HSPCs of ayu.

We further found that CXCR4b preferred to interact with SDF-1, which is the sole endogenous chemokine ligand for the CXCR4 of mammalian HSPCs (29). Furthermore, CXCR4b+ HSPCs could not be significantly mobilized after LPS treatment in teleosts. CXCR4b+ HSPC mobilization was induced by the endogenous cytokine LECT2 after 5 d of treatment. Moreover, CXCR4b also inhibited the ROS level in HSPCs to prevent their proliferation in ayu. CXCR4b may prevent HSPC exhaustion at the early stage of bacterial infection because CXCR4b has a low affinity for LPS and inhibits HSPC proliferation. LPS plays an important role in infection to induce HSPC exhaustion through the TLR4 downstream adapters TRIF and MyD88 (53). HSPC exhaustion during infection is detrimental to the replenishment of immune cells (14). Hence, the fact that CXCR4b+ HSPCs do not mobilize after LPS treatment keep the immune system stable.

As primary vertebrate species, teleosts possess both innate immune and adaptive immune systems (54). The innate immune system of teleosts has more diverse immune molecules than that of mammals, including lectins, complement, and NK cell receptors (30, 55). However, the adaptive immune system of teleosts has fewer Ig types than that of mammals (56, 57), a lack of Ab class-switch recombination, and low Ab affinity maturation (30). Hence, innate immunity may play a more important role as a first response to infection in teleosts. In this study, we found that LPS-mobilized CXCR4a+ HSPCs preferentially differentiated into myeloid cells, macrophages, and neutrophils, which are the main cell types of innate immunity. Myeloid cells play an important role in sepsis (58). Hence, the LPS-induced mobilization of CXCR4a+ HSPCs to preferentially produce myeloid cells is beneficial to resolving inflammation. Moreover, we identified the transcription factor SCL as the downstream signaling regulator of CXCR4a in ayu HSPCs. SCL has been found to direct HSC fate in mammals (59). We do not exclude the possibility that other transcription factors may also mediate the differentiation of CXCR4a+ HSPCs. Because SCL shows the highest differential expression between CXCR4a+ and CXCR4b+ HSPCs, SCL may be the main transcription factor that regulates HSPC differentiation.

The immune systems between teleosts and mammals are different in a variety of aspects, including phagocyte types (6062). Because HSPCs give rise to different lineages of immune cells, investigating HSPC homeostasis regulation will contribute to better understanding the evolution of the vertebrate immune system. The isolation and characterization of HSPCs have been performed in several teleosts, including zebrafish (42), carp (63), goldfish (64, 65), and ayu (13). Transplantation is a crucial method to investigate HSPC function in both mammals and teleosts. In zebrafish, several approaches have been used to improve engraftment rates, including the application of sublethal irradiation (66), MHC-matched donors (67), and embryonic recipients (68). In this study, we developed a transplantation method using GFP-labeled HSPCs in MHC-matched ayu after irradiation, and no host-versus-graft reactions were observed. In principle, this strategy can be used in other teleosts to investigate HSPC function by GFP labeling in MHC-matched animals for transplantation.

In summary, our study reveals that the chemokine receptors CXCR4a and CXCR4b mediate HSPC homeostasis in teleosts (Fig. 15). CXCR4a preferentially binds with LPS, whereas CXCR4b preferentially binds with SDF-1. CXCR4a+ HSPCs are prone to differentiate into innate immune cells. Our study illustrates that the two CXCR4 genes in teleosts may stabilize the HSPC pool postinfection, which also supports the concept that LPS tolerance in teleosts results from CXCR4b+ HSPCs not mobilizing into the blood after LPS treatment.

This work was supported by the Program for the Natural Science Foundation of China (31972821; 31772876; 41776151), the Zhejiang Provincial Natural Science Foundation of China (LR18C040001; LZ18C190001), the Scientific Innovation Team Project of Ningbo (2015C110018) and the K.C. Wong Magna Fund in Ningbo University.

The sequences presented in this article have been submitted to GenBank (https://www.ncbi.nlm.nih.gov/genbank/) under accession numbers MN148390, MN148391, MN148393, MN158722, MN158723, MN158724, MN156535, and MN156536, MN628572, MN628573.

The online version of this article contains supplemental material.

Abbreviations used in this article:

BPI

bacterial/permeability increasing protein

CFU-C

CFU cell

CRU

competitive repopulating unit

DHR

dihydrorhodamine-123

EGR1

early growth response protein 1

GATA2

GATA-binding protein 2

GATA3

GATA-binding protein 3

HSC

hematopoietic stem cell

HSPC

hematopoietic stem/progenitor cell

LECT2

leukocyte cell–derived chemotaxin-2

Mef2C

myocyte enhancer factor 2C

MHC I

MHC class I

MO/MΦ

monocyte/macrophage

MPO

myeloperoxidase

NAC

N-acetylcysteine

PaCXCR4a

P. altivelis CXCR4a

PaCXCR4b

P. altivelis CXCR4b

PaSRB2a

ayu scavenger receptor class B 2a

PAX5

reduced paired box 5

PU.1

PU box-binding protein

ROS

reactive oxygen species

RT-qPCR

real-time quantitative PCR

RUNX1

runt-related transcription factor 1

SCL

stem cell leukemia

SDF

stromal cell–derived factor

shRNA

short hairpin RNA

siRNA

small interfering RNA

SP

side population.

1
Laurenti
,
E.
,
B.
Göttgens
.
2018
.
From haematopoietic stem cells to complex differentiation landscapes.
Nature
553
:
418
426
.
2
Lucas
,
D.
2019
.
Leukocyte trafficking and regulation of murine hematopoietic stem cells and their niches.
Front. Immunol.
10
:
387
.
3
Nelson
,
J. S.
2006
.
Fishes of the World
, 4th Ed.
John Wiley & Sons
,
Hoboken, NJ
, p.
601
.
4
Sandve
,
S. R.
,
R. V.
Rohlfs
,
T. R.
Hvidsten
.
2018
.
Subfunctionalization versus neofunctionalization after whole-genome duplication.
Nat. Genet.
50
:
908
909
.
5
Lu
,
X. J.
,
Q.
Chen
,
Y. J.
Rong
,
F.
Chen
,
J.
Chen
.
2017
.
CXCR3.1 and CXCR3.2 differentially contribute to macrophage polarization in teleost fish.
J. Immunol.
198
:
4692
4706
.
6
Chavakis
,
T.
,
I.
Mitroulis
,
G.
Hajishengallis
.
2019
.
Hematopoietic progenitor cells as integrative hubs for adaptation to and fine-tuning of inflammation.
Nat. Immunol.
20
:
802
811
.
7
King
,
K. Y.
2018
.
Stem cell regulation during chronic infection.
Blood
132
(
Suppl. 1
): SCI-32.
8
Burberry
,
A.
,
M. Y.
Zeng
,
L.
Ding
,
I.
Wicks
,
N.
Inohara
,
S. J.
Morrison
,
G.
Núñez
.
2014
.
Infection mobilizes hematopoietic stem cells through cooperative NOD-like receptor and toll-like receptor signaling.
Cell Host Microbe
15
:
779
791
.
9
Boettcher
,
S.
,
M. G.
Manz
.
2017
.
Regulation of inflammation- and infection-driven hematopoiesis.
Trends Immunol.
38
:
345
357
.
10
Clapes
,
T.
,
S.
Lefkopoulos
,
E.
Trompouki
.
2016
.
Stress and non-stress roles of inflammatory signals during HSC emergence and maintenance.
Front. Immunol.
7
:
487
.
11
Megías
,
J.
,
A.
Yáñez
,
S.
Moriano
,
J. E.
O’Connor
,
D.
Gozalbo
,
M. L.
Gil
.
2012
.
Direct toll-like receptor-mediated stimulation of hematopoietic stem and progenitor cells occurs in vivo and promotes differentiation toward macrophages.
Stem Cells
30
:
1486
1495
.
12
Hall
,
C. J.
,
M. V.
Flores
,
S. H.
Oehlers
,
L. E.
Sanderson
,
E. Y.
Lam
,
K. E.
Crosier
,
P. S.
Crosier
.
2012
.
Infection-responsive expansion of the hematopoietic stem and progenitor cell compartment in zebrafish is dependent upon inducible nitric oxide.
Cell Stem Cell
10
:
198
209
.
13
Lu
,
X. J.
,
Q.
Chen
,
Y. J.
Rong
,
J.
Chen
.
2016
.
Mobilisation and dysfunction of haematopoietic stem/progenitor cells after Listonella anguillarum infection in ayu, Plecoglossus altivelis.
Sci. Rep.
6
:
28082
.
14
Takizawa
,
H.
,
K.
Fritsch
,
L. V.
Kovtonyuk
,
Y.
Saito
,
C.
Yakkala
,
K.
Jacobs
,
A. K.
Ahuja
,
M.
Lopes
,
A.
Hausmann
,
W. D.
Hardt
, et al
.
2017
.
Pathogen-induced TLR4-TRIF innate immune signaling in hematopoietic stem cells promotes proliferation but reduces competitive fitness.
Cell Stem Cell
21
:
225
240.e5
.
15
Lu
,
X. J.
,
Q.
Chen
,
Y. J.
Rong
,
G. J.
Yang
,
C. H.
Li
,
N. Y.
Xu
,
C. H.
Yu
,
H. Y.
Wang
,
S.
Zhang
,
Y. H.
Shi
,
J.
Chen
.
2016
.
LECT2 drives haematopoietic stem cell expansion and mobilization via regulating the macrophages and osteolineage cells.
Nat. Commun.
7
:
12719
.
16
Bird
,
S.
,
C.
Tafalla
.
2015
.
Teleost chemokines and their receptors.
Biology (Basel)
4
:
756
784
.
17
Zou
,
J.
,
A. K.
Redmond
,
Z.
Qi
,
H.
Dooley
,
C. J.
Secombes
.
2015
.
The CXC chemokine receptors of fish: insights into CXCR evolution in the vertebrates.
Gen. Comp. Endocrinol.
215
:
117
131
.
18
Nagasawa
,
T.
,
H.
Kikutani
,
T.
Kishimoto
.
1994
.
Molecular cloning and structure of a pre-B-cell growth-stimulating factor.
Proc. Natl. Acad. Sci. USA
91
:
2305
2309
.
19
Nie
,
Y.
,
Y. C.
Han
,
Y. R.
Zou
.
2008
.
CXCR4 is required for the quiescence of primitive hematopoietic cells.
J. Exp. Med.
205
:
777
783
.
20
Yakulov
,
T. A.
,
A. P.
Todkar
,
K.
Slanchev
,
J.
Wiegel
,
A.
Bona
,
M.
Groß
,
A.
Scholz
,
I.
Hess
,
A.
Wurditsch
,
F.
Grahammer
, et al
.
2018
.
CXCL12 and MYC control energy metabolism to support adaptive responses after kidney injury.
Nat. Commun.
9
:
3660
.
21
Seleit
,
A.
,
I.
Krämer
,
E.
Ambrosio
,
N.
Dross
,
U.
Engel
,
L.
Centanin
.
2017
.
Sequential organogenesis sets two parallel sensory lines in medaka.
Development
144
:
687
697
.
22
Janssens
,
R.
,
S.
Struyf
,
P.
Proost
.
2018
.
The unique structural and functional features of CXCL12.
Cell. Mol. Immunol.
15
:
299
311
.
23
Sugiyama
,
T.
,
H.
Kohara
,
M.
Noda
,
T.
Nagasawa
.
2006
.
Maintenance of the hematopoietic stem cell pool by CXCL12-CXCR4 chemokine signaling in bone marrow stromal cell niches.
Immunity
25
:
977
988
.
24
Freitas
,
C.
,
M.
Wittner
,
J.
Nguyen
,
V.
Rondeau
,
V.
Biajoux
,
M. L.
Aknin
,
F.
Gaudin
,
S.
Beaussant-Cohen
,
Y.
Bertrand
,
C.
Bellanné-Chantelot
, et al
.
2017
.
Lymphoid differentiation of hematopoietic stem cells requires efficient Cxcr4 desensitization.
J. Exp. Med.
214
:
2023
2040
.
25
Chong
,
S. W.
,
A.
Emelyanov
,
Z.
Gong
,
V.
Korzh
.
2001
.
Expression pattern of two zebrafish genes, cxcr4a and cxcr4b.
Mech. Dev.
109
:
347
354
.
26
Doitsidou
,
M.
,
M.
Reichman-Fried
,
J.
Stebler
,
M.
Köprunner
,
J.
Dörries
,
D.
Meyer
,
C. V.
Esguerra
,
T.
Leung
,
E.
Raz
.
2002
.
Guidance of primordial germ cell migration by the chemokine SDF-1.
Cell
111
:
647
659
.
27
Packham
,
I. M.
,
C.
Gray
,
P. R.
Heath
,
P. G.
Hellewell
,
P. W.
Ingham
,
D. C.
Crossman
,
M.
Milo
,
T. J.
Chico
.
2009
.
Microarray profiling reveals CXCR4a is downregulated by blood flow in vivo and mediates collateral formation in zebrafish embryos.
Physiol. Genomics
38
:
319
327
.
28
Paredes-Zúñiga
,
S.
,
R. A.
Morales
,
S.
Muñoz-Sánchez
,
C.
Muñoz-Montecinos
,
M.
Parada
,
K.
Tapia
,
C.
Rubilar
,
M. L.
Allende
,
O. A.
Peña
.
2017
.
CXCL12a/CXCR4b acts to retain neutrophils in caudal hematopoietic tissue and to antagonize recruitment to an injury site in the zebrafish larva.
Immunogenetics
69
:
341
349
.
29
Wright
,
D. E.
,
E. P.
Bowman
,
A. J.
Wagers
,
E. C.
Butcher
,
I. L.
Weissman
.
2002
.
Hematopoietic stem cells are uniquely selective in their migratory response to chemokines.
J. Exp. Med.
195
:
1145
1154
.
30
Sunyer
,
J. O.
2013
.
Fishing for mammalian paradigms in the teleost immune system.
Nat. Immunol.
14
:
320
326
.
31
Li
,
M. Y.
,
L.
Miao
,
J.
Chen
.
2016
.
Breeding of a new variety of ayu (Plecoglossus altivlis) named “Zhemin No.1” with fast growth and few abnormality characters.
J. Agric. Biotechnol.
24
:
1392
1397
.
32
Seki
,
S.
,
J. J.
Agresti
,
G. A. E.
Gall
,
N.
Taniguchi
,
B.
May
.
1999
.
AFLP analysis of genetic diversity in three populations of ayu Plecoglossus altivelis.
Fish. Sci.
65
:
888
892
.
33
de Haas
,
C. J.
,
R.
van der Zee
,
B.
Benaissa-Trouw
,
K. P.
van Kessel
,
J.
Verhoef
,
J. A.
van Strijp
.
1999
.
Lipopolysaccharide (LPS)-binding synthetic peptides derived from serum amyloid P component neutralize LPS.
Infect. Immun.
67
:
2790
2796
.
34
Zhang
,
R. C.
,
J.
Chen
,
C. H.
Li
,
X. J.
Lu
,
Y. H.
Shi
.
2011
.
Prokaryotic expression, purification, and refolding of leukocyte cell-derived chemotaxin 2 and its effect on gene expression of head kidney-derived macrophages of a teleost fish, ayu (Plecoglossus altivelis).
Fish Shellfish Immunol.
31
:
911
918
.
35
Kruger
,
N. J.
2009
.
The Bradford method for protein quantitation.
In
The Protein Protocols Handbook
, 2nd Ed.
J. M.
Walker
, eds.
Humana Press
,
Totowa, NJ
, p.
15
21
.
36
Filippo
,
T. R.
,
L. T.
Galindo
,
G. F.
Barnabe
,
C. B.
Ariza
,
L. E.
Mello
,
M. A.
Juliano
,
L.
Juliano
,
M. A.
Porcionatto
.
2013
.
CXCL12 N-terminal end is sufficient to induce chemotaxis and proliferation of neural stem/progenitor cells.
Stem Cell Res. (Amst.)
11
:
913
925
.
37
Clark-Lewis
,
I.
,
I.
Mattioli
,
J. H.
Gong
,
P.
Loetscher
.
2003
.
Structure-function relationship between the human chemokine receptor CXCR3 and its ligands.
J. Biol. Chem.
278
:
289
295
.
38
Zhu
,
L. Y.
,
A. F.
Lin
,
T.
Shao
,
L.
Nie
,
W. R.
Dong
,
L. X.
Xiang
,
J. Z.
Shao
.
2014
.
B cells in teleost fish act as pivotal initiating APCs in priming adaptive immunity: an evolutionary perspective on the origin of the B-1 cell subset and B7 molecules.
J. Immunol.
192
:
2699
2714
.
39
Chen
,
Q.
,
X. J.
Lu
,
M. Y.
Li
,
J.
Chen
.
2016
.
Molecular cloning, pathologically-correlated expression and functional characterization of the colonystimulating factor 1 receptor (CSF-1R) gene from a teleost, Plecoglossus altivelis.
Zool. Res.
37
:
96
102
.
40
Szilvassy
,
S. J.
,
R. K.
Humphries
,
P. M.
Lansdorp
,
A. C.
Eaves
,
C. J.
Eaves
.
1990
.
Quantitative assay for totipotent reconstituting hematopoietic stem cells by a competitive repopulation strategy.
Proc. Natl. Acad. Sci. USA
87
:
8736
8740
.
41
Wattrus
,
S. J.
,
L. I.
Zon
.
2018
.
Stem cell safe harbor: the hematopoietic stem cell niche in zebrafish.
Blood Adv.
2
:
3063
3069
.
42
Chen
,
J.
,
X. J.
Lu
,
H. Y.
Yang
,
Y. H.
Shi
.
2010
.
An interaction between a C-type lectin receptor and leukocyte cell-derived chemotaxin 2 of ayu, Plecoglossus altivelis.
Fish Shellfish Immunol.
28
:
245
248
.
43
Ludin
,
A.
,
S.
Gur-Cohen
,
K.
Golan
,
K. B.
Kaufmann
,
T.
Itkin
,
C.
Medaglia
,
X. J.
Lu
,
G.
Ledergor
,
O.
Kollet
,
T.
Lapidot
.
2014
.
Reactive oxygen species regulate hematopoietic stem cell self-renewal, migration and development, as well as their bone marrow microenvironment.
Antioxid. Redox Signal.
21
:
1605
1619
.
44
Wang
,
T.
,
F.
Liu
,
G.
Tian
,
C. J.
Secombes
,
T.
Wang
.
2019
.
Lineage/species-specific expansion of the Mx gene family in teleosts: differential expression and modulation of nine Mx genes in rainbow trout Oncorhynchus mykiss.
Fish Shellfish Immunol.
90
:
413
430
.
45
Wang
,
T.
,
Y.
Hu
,
E.
Wangkahart
,
F.
Liu
,
A.
Wang
,
E.
Zahran
,
K. R.
Maisey
,
M.
Liu
,
Q.
Xu
,
M.
Imarai
,
C. J.
Secombes
.
2018
.
Interleukin (IL)-2 is a key regulator of T helper 1 and T helper 2 cytokine expression in fish: functional characterization of two divergent IL2 paralogs in salmonids.
Front. Immunol.
9
:
1683
.
46
Wei
,
Q.
,
P. S.
Frenette
.
2018
.
Niches for hematopoietic stem cells and their progeny.
Immunity
48
:
632
648
.
47
Karpova
,
D.
,
J. K.
Ritchey
,
M. S.
Holt
,
G.
Abou-Ezzi
,
D.
Monlish
,
L.
Batoon
,
S.
Millard
,
G.
Spohn
,
E.
Wiercinska
,
E.
Chendamarai
, et al
.
2017
.
Continuous blockade of CXCR4 results in dramatic mobilization and expansion of hematopoietic stem and progenitor cells.
Blood
129
:
2939
2949
.
48
de Kruijf
,
E. F. M.
,
W. E.
Fibbe
,
M.
van Pel
.
2019
.
Cytokine-induced hematopoietic stem and progenitor cell mobilization: unraveling interactions between stem cells and their niche.
Ann. N. Y. Acad. Sci.
DOI: 10.1111/nyas.14059.
49
Stachura
,
D. L.
,
O.
Svoboda
,
C. A.
Campbell
,
R.
Espín-Palazón
,
R. P.
Lau
,
L. I.
Zon
,
P.
Bartůněk
,
D.
Traver
.
2013
.
The zebrafish granulocyte colony-stimulating factors (Gcsfs): 2 paralogous cytokines and their roles in hematopoietic development and maintenance.
Blood
122
:
3918
3928
.
50
Lu
,
X. J.
,
J.
Chen
,
C. H.
Yu
,
Y. H.
Shi
,
Y. Q.
He
,
R. C.
Zhang
,
Z. A.
Huang
,
J. N.
Lv
,
S.
Zhang
,
L.
Xu
.
2013
.
LECT2 protects mice against bacterial sepsis by activating macrophages via the CD209a receptor.
J. Exp. Med.
210
:
5
13
.
51
Triantafilou
,
K.
,
M.
Triantafilou
,
R. L.
Dedrick
.
2001
.
A CD14-independent LPS receptor cluster.
Nat. Immunol.
2
:
338
345
.
52
Lu
,
X. J.
,
Y. J.
Ning
,
H.
Liu
,
L.
Nie
,
J.
Chen
.
2018
.
A novel lipopolysaccharide recognition mechanism mediated by internalization in teleost macrophages.
Front. Immunol.
9
:
2758
.
53
Zhang
,
H.
,
S.
Rodriguez
,
L.
Wang
,
S.
Wang
,
H.
Serezani
,
R.
Kapur
,
A. A.
Cardoso
,
N.
Carlesso
.
2016
.
Sepsis induces hematopoietic stem cell exhaustion and myelosuppression through distinct contributions of TRIF and MYD88.
Stem Cell Reports
6
:
940
956
.
54
Star
,
B.
,
A. J.
Nederbragt
,
S.
Jentoft
,
U.
Grimholt
,
M.
Malmstrøm
,
T. F.
Gregers
,
T. B.
Rounge
,
J.
Paulsen
,
M. H.
Solbakken
,
A.
Sharma
, et al
.
2011
.
The genome sequence of Atlantic cod reveals a unique immune system.
Nature
477
:
207
210
.
55
Brinchmann
,
M. F.
,
D. M.
Patel
,
N.
Pinto
,
M. H.
Iversen
.
2018
.
Functional aspects of fish mucosal lectins-interaction with non-self.
Molecules
23
: E1119.
56
Rajan
,
B.
,
G.
Løkka
,
E. O.
Koppang
,
L.
Austbø
.
2017
.
Passive immunization of farmed fish.
J. Immunol.
198
:
4195
4202
.
57
Xu
,
Z.
,
F.
Takizawa
,
D.
Parra
,
D.
Gómez
,
L.
von Gersdorff Jørgensen
,
S. E.
LaPatra
,
J. O.
Sunyer
.
2016
.
Mucosal immunoglobulins at respiratory surfaces mark an ancient association that predates the emergence of tetrapods.
Nat. Commun.
7
:
10728
.
58
Delano
,
M. J.
,
P. A.
Ward
.
2016
.
Sepsis-induced immune dysfunction: can immune therapies reduce mortality?
J. Clin. Invest.
126
:
23
31
.
59
Kunisato
,
A.
,
S.
Chiba
,
T.
Saito
,
K.
Kumano
,
E.
Nakagami-Yamaguchi
,
T.
Yamaguchi
,
H.
Hirai
.
2004
.
Stem cell leukemia protein directs hematopoietic stem cell fate.
Blood
103
:
3336
3341
.
60
Li
,
J.
,
D. R.
Barreda
,
Y. A.
Zhang
,
H.
Boshra
,
A. E.
Gelman
,
S.
Lapatra
,
L.
Tort
,
J. O.
Sunyer
.
2006
.
B lymphocytes from early vertebrates have potent phagocytic and microbicidal abilities.
Nat. Immunol.
7
:
1116
1124
.
61
Nagasawa
,
T.
,
C.
Nakayasu
,
A. M.
Rieger
,
D. R.
Barreda
,
T.
Somamoto
,
M.
Nakao
.
2014
.
Phagocytosis by thrombocytes is a conserved innate immune mechanism in lower vertebrates.
Front. Immunol.
5
:
445
.
62
Lu
,
X. J.
,
J.
Chen
.
2019
.
Specific function and modulation of teleost monocytes/macrophages: polarization and phagocytosis.
Zool. Res.
40
:
146
150
.
63
Kobayashi
,
I.
,
F.
Katakura
,
T.
Moritomo
.
2016
.
Isolation and characterization of hematopoietic stem cells in teleost fish.
Dev. Comp. Immunol.
58
:
86
94
.
64
Katzenback
,
B. A.
,
F.
Katakura
,
M.
Belosevic
.
2016
.
Goldfish (Carassius auratus L.) as a model system to study the growth factors, receptors and transcription factors that govern myelopoiesis in fish.
Dev. Comp. Immunol.
58
:
68
85
.
65
Hanington
,
P. C.
,
D. R.
Barreda
,
M.
Belosevic
.
2006
.
A novel hematopoietic granulin induces proliferation of goldfish (Carassius auratus L.) macrophages.
J. Biol. Chem.
281
:
9963
9970
.
66
Traver
,
D.
,
A.
Winzeler
,
H. M.
Stern
,
E. A.
Mayhall
,
D. M.
Langenau
,
J. L.
Kutok
,
A. T.
Look
,
L. I.
Zon
.
2004
.
Effects of lethal irradiation in zebrafish and rescue by hematopoietic cell transplantation.
Blood
104
:
1298
1305
.
67
de Jong
,
J. L.
,
L. I.
Zon
.
2012
.
Histocompatibility and hematopoietic transplantation in the zebrafish.
Adv. Hematol.
2012
: 282318.
68
Traver
,
D.
,
B. H.
Paw
,
K. D.
Poss
,
W. T.
Penberthy
,
S.
Lin
,
L. I.
Zon
.
2003
.
Transplantation and in vivo imaging of multilineage engraftment in zebrafish bloodless mutants.
Nat. Immunol.
4
:
1238
1246
.

The authors have no financial conflicts of interest.

Supplementary data