Christina Begon-Pescia1
Laurent Manchon2
Resul Özbilgiç1Stéphanie Boireau3Anaïs Louis4Simon Georges4
Mai Nguyen-Chi1*- 1LPHI, Université de Montpellier, CNRS, INSERM, Montpellier, France
- 2Independent Researcher, Montpellier, France
- 3Montpellier Resources Imagerie, BioCampus, University of Montpellier, CNRS, INSERM, Montpellier, France
- 4MGX-Montpellier GenomiX, Univ. Montpellier, CNRS, INSERM, Montpellier, France
Introduction: Macrophages participate in wound healing by contributing to host defense and by orchestrating inflammation and repair. They adapt dynamically to the wound microenvironment by adopting diverse polarization states, which are determinant of wound outcome, influencing whether healing is successful or becomes chronic. The zebrafish embryo, widely used for live imaging of immune responses, is a powerful model to study macrophage behavior after injury. However, the transcriptional landscape of polarized macrophages in this model during wound healing remains insufficiently characterized.
Methods: Here, we employed bulk RNA sequencing on FACS-sorted zebrafish macrophages at different times following tail fin wounding in larvae, a robust model for studying sterile inflammation. We analyzed gene expression changes over the time by comparing wounded versus unwounded conditions to characterize macrophage transcriptional programs during wound healing.
Results: Our findings reveal that zebrafish macrophages undergo large transcriptomic changes along different wound healing phases, particularly between 2 and 5 hours post-amputation, suggesting a fast reprogramming leading to different functional states. We further show that, at 2h, macrophages acquire a pro-inflammatory profile with a gene signature closed to M1-like program. At 5h, macrophages express genes involved in immunoregulation and healing, associated with shutoff of pro-inflammatory pathways and the activation of glucose and glycogen metabolism. Finally, we show that macrophage reprogramming becomes deeply attenuated by 29h post injury.
Discussion: Our findings provide a foundation for understanding macrophage polarization in zebrafish, revealing underpinning molecular mechanisms that include both specific and evolutionarily conserved pathways with a potential impact on translational medicine.
1 Introduction
Wound healing is a complex process that involves multiple cell populations interacting in a coordinated manner during successive phases: inflammation, proliferation, re‐epithelialization, and remodeling (1). The restoration of skin integrity is essential not only for protecting the host from pathogens but also for re-establishing barrier function. Despite extensive research, effective therapeutic strategies that accelerate tissue repair while reducing scarring remain limited. A deeper understanding of the roles of various cell populations at play in wound repair is therefore essential to elucidate the mechanisms underlying normal wound closure and to prevent the development of pathological wounds.
Macrophages are among the first responders recruited to the injury site, where they play pivotal roles in both host defense and tissue repair. As key regulators of wound healing, macrophages modulate the onset and resolution of inflammation and orchestrate several processes integral to tissue repair (2). Numerous studies have demonstrated that macrophages are indispensable for proper wound healing and regeneration across a range of species, from mammals (3, 4) to non-mammalian models (5–9).
As the most plastic cells within the hematopoietic system, macrophages adapt to microenvironmental cues through a process known as polarization. This process, driven by specific gene expression programs, enables macrophages to acquire distinct functional phenotypes. In vitro, macrophages were classified into two major groups: the pro-inflammatory M1 (classically activated) and the anti-inflammatory M2 (alternatively activated) (10, 11). At the onset of wound healing, macrophages predominantly adopt an M1 phenotype characterized by the secretion of pro-inflammatory cytokines such as tumor necrosis factor (TNF) and interleukin-6 (IL-6). These M1 macrophages are critical for host defense and the clearance of cellular debris, and they contribute to the stimulation of cell proliferation in the damaged tissue (12). As healing progresses, a phenotypic shift occurs toward an M2 state, evidenced by increased expression of markers such as MRC1. M2 macrophages secrete growth factors most notably transforming growth factor-β1 (TGF-β1) which promote stromal cell proliferation, survival, and migration. In addition, these cells facilitate the resolution of inflammation and mediate reparative processes including angiogenesis, granulation tissue formation, re-epithelialization, and extracellular matrix remodeling through the release of matrix metalloproteases (12, 13). However, this classification has been challenged by in vivo studies (14). In addition single-cell RNA sequencing studies have revealed that macrophage phenotypes during wound healing are more heterogeneous than previously appreciated (15, 16). Importantly, the precise temporal regulation of macrophage polarization is essential for normal healing, and its impairment leads to delay of tissue repair and regeneration in various animal models (6, 8, 17). Despite these advances, the full spectrum of macrophage states in wounds remains incompletely characterized, and it is essential to associate each state with its corresponding functional behavior.
Zebrafish (Danio rerio) is an excellent model for studying wound healing mechanisms due to its remarkable regenerative abilities and its transparent embryos, which permit real-time observation of cell populations (18–20). The strong genetic resemblance to humans further underscores its relevance for biomedical research (21). Recently, there has been increasing focus on the role of the immune system, especially macrophages, in wound repair and regeneration. Zebrafish macrophages, like their mammalian counterparts, contribute to both antimicrobial and pro-inflammatory responses (20, 22–24) as well as to tissue repair (6, 8, 9, 25). Macrophages are crucial in wound healing, including in the caudal fin of both adult and larval zebrafish (8, 9). During the early phases of inflammation, zebrafish macrophages adopt an M1-like phenotype, marked by cytokine production such as tnfa.a, tnfa.b, il1b, and il6. In the later stages, macrophages transition to an M2-like, characterized by the expression of tgf-b1 and ccr2 and cxcr4 (8, 24, 26, 27). Another marker associated with wound macrophage subsets in later stages is arg2, which has been suggested to mediate anti-inflammatory functions (28). Yet, the detailed molecular programs, as well as the timing of macrophage activation during tissue injury in zebrafish are poorly defined. In addition, markers to track distinct wound-macrophage subsets are still lacking in this specie.
Here, we used transcriptomic profiling of macrophages to identify the changes in gene expression that control the molecular mechanisms underlying the macrophage polarization during wound healing in zebrafish. Using a transgenic zebrafish macrophage reporter line, we isolated macrophages by FACS at different time points after caudal fin fold injury and analyzed their transcriptome by bulk RNA-sequencing. We found that the gene expression patterns of macrophages are highly dynamic during the healing process. We identified transcriptomic programs and biological pathways that distinguish the acute phase of inflammation, transition to resolution and remodeling phase.
2 Materials and methods
2.1 Ethics statement
All animal experiments were conducted by well-trained and authorized staff in compliance with the European Union guidelines for handling of laboratory animals (http://ec.europa.eu/environment/chemicals/labanimals/indexen.htm) and the ARRIVE guidelines and were approved by the Comité d’Ethique pour l’Expérimentation Animale Languedoc Roussillon under reference APAFIS #36309–2022040114222432 V2. Breeding and maintenance of adult fish were performed at the ZEFIX-LPHI (CNRS, University of Montpellier, Montpellier, France; license number CEEA-LR-B34–172–37). Developmental stages used were 3 dpf and 4 dpf. When required, specimens were euthanized using an anesthetic overdose of buffered tricaine (500 mg/L).
2.2 Fish husbandry
Zebrafish (Danio rerio) maintenance and husbandry were performed at the fish facility of the University of Montpellier as previously described (29). Males and females were kept in 3.5-L polycarbonate tank (maximum 22 fish per tank) connected to a recirculating system (Tecniplast), in the following conditions: 4 ‰ salinity/400 conductivity, temperature of 27.5°C and a 12:12-h light: dark cycle. The fish were fed twice a day. Embryos were obtained from pairs of adult fish by natural spawning at 28°C in tank water. Embryos and larvae were staged according to Kimmel et al. (30) in zebrafish medium in 10-cm petri dishes. The zebrafish embryo medium was changed every day. Strains used were transgenic lines: Tg (mfap4:mCherry-F) ump6Tg, referred as Tg (mfap4:mCherry-F) (31) to specifically label the macrophages with a farnesylated mCherry protein and AB strain, referred as WT, was used as a control for FACS.
2.3 Zebrafish caudal fin fold amputation and cell dissociation
Larvae were anaesthetized in zebrafish medium supplemented with 200 μg/ml tricaine (ethyl 3-aminobenzoate methanesulfonate, MS-222 sigma # A5040) before any manipulation. Caudal fin fold amputation was performed at 3 dpf for all experiments as previously described (8). Briefly, fin folds were transected with a sterile scalpel, posterior to muscle and notochord. Larvae were transferred in their medium at 28°C for 2, 5 and 29 hours and they correspond to 2 hours post-amputation (hpA), 5 hpA and 29 hpA groups. 360 Tg (mfap4:mCherry-F) larvae were either amputated (CUT) or not amputated as a control (UNcut), then, whole larvae were crushed on cell strainer 70-μm mesh in cold working buffer (0.9X DPBS with 2% FBSi and 2 mM EDTA). Isolated cells were washed, filtered through a cell strainer 40-μm mesh and transferred into 5 ml round bottom polystyrene FACS tubes as previously described (29). A total of four replicates for each condition (CUT ou UNcut) within each group (2 hpA, 5 hpA and 29 hpA) were conducted.
2.4 FACS and mRNA isolation from sorted macrophage cells
Following cell dissociation from amputated and non-amputated larvae, mCherry-F positive cells were sorted using a FACSAria™ IIu (BD Biosciences) as previously described (29). To minimize RNA degradation and material loss, the mCherry-F positive and mCherry-F negative sorted cells were separately collected directly into RNase/DNase-free 1.5 ml microfuge tubes containing 100 μl RLT/β-ME Qiagen’s lysis buffer and kept on ice for immediate lysis. A set of 24 total RNA extractions from the mCherry-F positive sorted macrophages were performed using RNeasy® Micro Kit from Qiagen according to manufacturer’s instructions. Between 30–000 and 97–000 mCherry-F positive cells were sorted, yielding a total RNA amount ranging from 5 to 50 ng per sample. All RNA integrity number (RIN) obtained were above 7.
2.5 Illumina sequencing and differential gene expression analysis on FACS-sorted macrophages
The sequencing data generated by this study have been deposited in the NCBI Gene Expression Omnibus (GEO) under the accession number GSE234476. We generated the RNA-Seq libraries with the Illumina DNA prep kit from 5 to 10 ng input RNA. The cDNA synthesis and amplification were performed using the SMART-Seq® V4 Ultra Low kit (Takara Bio). The libraries have been sequenced using a 100-nucleotides single-read SP flow cell lane using the 2-channel Sequence By Synthesis (SBS) method. The libraries were validated by quantification of the DNA concentration and fragment length using the Fragment Analyzer (Standard Sensitivity NGS kit, Agilent Technologies, Santa Clara, CA, USA) and qPCR (Roche LightCycler®480). FeatureCounts (v2.0.3) was used to count the number of reads for each sample, which ranged between 30 and 48 million at 2 hpA, between 35 and 51 million at 5 hpA and between 35 and 45 million at 29 hpA. RNA-sequencing raw reads were then aligned to the Danio rerio reference genome (Ensembl, GRCz11) using TopHat2 software (32). The “raw read count” data sets were analyzed using the Bioconductor package DESeq2 (v1.38.2) (33, 34) to identify the differentially expressed genes (DEGs) between CUT and UNcut conditions at different times post-amputation (2, 5, 29 hpA) with cut off Log2FC = |1| and adjusted p value ≤ 0.05 (according to the FDR method from Benjamini-Hochberg (35) as the significance cut-off to identify DEGs). A second analysis was performed at 29 hpA including all DEGs (no cut off, adjusted p value ≤ 0.05), since at this time point most DEGs have a Log2FC <1.
2.6 RNA sequencing bioinformatics data analysis, GO and pathway analysis
The transcriptome profiling of macrophages during wound healing was analyzed using Bioconductor (34) in R (v4.1.2) with the following packages: ggvenn, ggplot2, EnhancedVolcano, pheatmap (R Core Team: https://www.R-project.org/). Venn diagrams were generated using ggvenn to illustrate the overlap between different sets of DEGs and identify common genes across the time course. A scatter-plot was created using ggplot2 to depict the fold-change correlation between all common genes identified in the Venn diagram. To visualize gene expression changes at different time points, we constructed Volcano plots, using EnhancedVolcano. Additionally, hierarchical clustering heatmaps were generated using pheatmap to display the expression patterns of the top 50 most up-regulated and down-regulated genes, comparing CUT conditions to UNcut conditions within each group. Finally, to functionally categorize the differentially expressed genes, we performed Gene Ontology (GO) analysis to determine their biological roles based on the biological process, the molecular function enrichment and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway annotations. This analysis was conducted using ShinyGO v0.77 (36) and SRplot (37).
2.7 RNA-sequencing validation by quantitative real time PCR assays and statistical analysis
Validation of RNA-sequencing data was assessed using quantitative Real Time PCR (RT-qPCR). We selected several differentially expressed genes, including both up-regulated and down-regulated genes for each group. Primers were designed with Primer 3 Plus and Primer-BLAST online tool (https://www.primer3plus.com/index.html and https://www.ncbi.nlm.nih.gov/tools/primer-blast/). Zebrafish ef1a (eukaryotic translation elongation factor 1 alpha 1, like 1) was used as the reference gene for normalization to determine the relative gene expression levels. Primer sequences and amplicon size for the following zebrafish genes was listed in Supplementary Table 1: il-1b, cxcl8 (il-8), tnfa.a, il6, cxcl18b, cxcr4b, tlr8a, tlr2, sox4b, s1pr4, timp2b, arg2, f5, mmp9, c1qb, dicp1.1, tnfa.b, p2rx7, sulf2b. Each primer set was tested for amplification efficiency before use. For cDNA synthesis, 10 ng of total RNA was reverse-transcribed using Oligo (dT)18 primers and M-MLV reverse transcriptase (Invitrogen™ # 28025-013) following the manufacturer’s recommendations. RT-qPCR was performed using a LightCycler® 480 (Roche) with SYBER Green (Meridian Bioscience® SensIFAST™ SYBR® No-ROX Mix # BIO-98050), according to the manufacturer’s protocol. All experiments were performed in both biological and technical replicates, ranging from 3 to 7 biological replicates per time point. Samples were run in triplicate and data were analyzed using LightCycler® 480 software (version 1.5.1). Zebrafish ef1a (eukaryotic translation elongation factor 1 alpha 1, like 1) was used as the reference housekeeping gene for normalization and the relative gene expression levels were calculated using the 2–ΔCt method (target/reference). The data analysis was performed using GraphPad Prism 8 software (version 8.3.0). Results were visualized as scatter plots with individual values, bar means and standard error mean (SEM). Statistical significance was evaluated using Mann-Whitney U tests (unpaired, non-parametric), with a significance threshold set at p-value < 0.05.
2.8 Fold change correlation between RNA-sequencing and quantitative real time PCR data
To evaluate the correlation between RNA-sequencing data and RT-qPCR results, we assessed a Spearman’s rho correlation coefficient of the fold change. Several DEGs per time point were taken randomly from the RNA-Seq datasets. Their fold change was then compared with that of the RT-qPCR results.
2.9 Imaging of living zebrafish larvae
Larvae were anesthetized in 200-μg/ml tricaine (Ethyl 3-aminobenzoate 646 methanesulfonate, MS-222 Sigma), positioned in 35 mm glass-bottom dishes (FluoDish, World Precision Instruments, UK), immobilized in 1% low melting point agarose (Sigma) and covered with 2 ml of embryo water containing tricaine. For imaging, confocal microscopy was performed on Zeiss LSM880 FastAiryscan, using 20X/0.8 objective. The wavelength was 561 nm for excitation. The images were taken in a sequential mode by line. The 3D files generated by multi-scan acquisitions were processed by Image J software. Brightness and contrast were adjusted for maximal visibility.
2.10 Data and code availability
The RNA-sequencing raw data from this study have been submitted and archived to the NCBI Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/geo/) under sample accession number GSE234476. For the transcriptome of macrophages activated by Salmonella enterica infection, raw data is available under accession number: GSE224985.
3 Results
3.1 Comparative analysis of zebrafish macrophage transcriptomes reveals distinct profiles across wound healing
To study wound healing, we used the well-established model of caudal fin fold amputation in 3 dpf zebrafish larvae. In order to accurately determine the acute inflammatory phase and resolution phase, we performed a global analysis of pro-inflammatory marker expression in the whole larvae at various times after wounding: 0, 1, 2, 3, 5, 6 and 24 hours post-amputation. The induction kinetics of il1b, il8, tnfa.b (here named tnfb), nos2a mRNAs were thus monitored by RT-qPCR in amputated (CUT) versus non-amputated (UNcut) conditions (Supplementary Figure S1). An increase in il1b and tnfb was observed in the amputated condition. Notably, il1b and tnfb reached their highest expression levels between 1 and 2 hours post-amputation. In contrast, nos2a and il8 were detected but not significantly modulated (Supplementary Figure S1). These results indicate that in whole tissue larva, the acute inflammatory phase occurs within the first few hours following injury, with a clear decline by 5–6 hpA, marking the end of acute phase. Therefore, we decided to analyze the transcriptomic profile of zebrafish macrophages during wound healing at 3 different time points: the 2 hpA time point, corresponding to the acute phase of inflammation, and the 5 hpA time point, corresponding to the end of the acute phase, were chosen based on the kinetic of expression of pro-inflammatory cytokines (il1b and tnfb). The 29 hpA time point was selected based on three criteria: 1/the return of pro-inflammatory cytokine levels to baseline, 2/the transition in cell morphology toward a mesenchymal shape accompanied by a circular mode of migration (18), and 3/the engagement of the tissue environment in regenerative processes, corresponding to a time after the formation of the blastema, a transient proliferative structure required for regeneration (i.e., beyond 24 hpa) (38). To molecularly characterize activated macrophages during this process, we performed a caudal fin fold amputation on macrophage reporter larvae, Tg (mfap4: mCherry-F) (Supplementary Figure S2, Figure 1A) and macrophages from whole larvae were sorted by FACS based on their fluorescence at 2, 5 and 29 hpA (Supplementary Figure S3). Pure populations of macrophages were obtained as previously shown (39) (and Supplementary Figure S4) and gene expression analysis was performed using bulk RNA-sequencing. All the time points were performed with four biological replicates. To identify differentially expressed genes (DEGs) at each time point, we compared the amputated (CUT) and non-amputated (UNcut) conditions using an adjusted p-value ≤ 0.05 as the significance threshold. We found that 1,895 genes were significantly differentially expressed at 2 hpA (Supplementary Table 2), while 4,216 genes were significantly differentially expressed at 5 hpA (Supplementary Table 3), and 45 genes at 29 hpA (Supplementary Table 4). Considering the most modulated genes (cut off Log2FC = |1|), the volcano plot at 2 hpA, showed 599 genes were up-regulated and 224 down-regulated, while at 5 hpA, 414 genes were up-regulated and 609 down-regulated (Figures 1B, C). At 29 hpA, most DEGs have a Log2FC <1, therefore we analyzed all DEGs (no cutoff) and found that 17 genes were down-regulated and 28 up-regulated (Figure 1D), indicating that macrophages in CUT condition are very similar to that of UNcut condition. Next, we compared the macrophage transcriptomes at 2 and 5 hpA. Interestingly, the Venn diagram in Figure 1E showed 141 DEGs were shared. However, many DEGs remained specific to each healing stage (682 genes for 2 hpA and 882 genes for 5 hpA).
Figure 1. Volcano plots of RNA-sequencing data at 2, 5, and 29 hours post-amputation. (A) Experimental workflow for the collection and isolation of zebrafish macrophage cells. Tg (mfap4:mCherry-F) larvae were either amputated (CUT) or not amputated as a control (UNcut), then, whole larvae were collected and dissociated at indicated time points, followed by FACS to isolate fluorescent cells. After quality control, samples were used for sequencing and downstream analysis. (B-D) Volcano plots illustrating gene expression changes at different time points following injury. Each dot represents a gene. The x-axis shows the log2 fold change (log2FC) in gene expression between the amputated (cut) and control (UNcut) conditions. The vertical dashed lines mark the fold change threshold (log2FC |1|) used for 2 and 5 hpA; no fold change cutoff was applied at 29 hpA. The y-axis shows the -log10 of the Benjamini-Hochberg adjusted p-value (with a significance threshold of p < 0.05), indicated by the horizontal dashed line. Colors denote up-regulated and down-regulated genes. (B) At 2 hpA, 599 genes are up-regulated and 224 genes are down-regulated, with the remaining genes unchanged. (C) At 5 hpA, 414 genes are up-regulated and 609 genes are down-regulated, with the rest unchanged. (D) At 29 hpA, 28 genes are up-regulated and 17 genes are down-regulated. (E) The Venn diagram illustrates the number of up-regulated (Up) and down-regulated (Down) DEGs at 2 hours post-amputation (hpA; green) and 5 hpA (brown). Each hemisphere represents the DEGs identified at each time point, with the overlapping region indicating shared genes across both conditions. A total of 823 genes were significantly regulated at 2 hpA, 1,023 at 5 hpA, with 141 DEGs shared between the two time points.
Collectively, these results show that zebrafish macrophages underwent large transcriptomic changes along different wound healing phases, and particularly between 2 hpA and 5 hpA, suggesting that reprogramming is rather fast and leads to different functional states.
3.2 Gene expression changes observed by RNA sequencing corroborate the results of real time PCR
For validating RNA-sequencing data, several DEGs per time points were randomly selected and analyzed by RT-qPCR. For 2 hpA, we studied cxcr4b, sox4b, tlr8a and tlr2 (down-regulated) and il-1b, cxcl8a, il-6, tnfa.a and cxcl18b (up-regulated). We found that all the genes that were down-regulated in sequencing data, were also down-regulated in RT-qPCR data, with the exception of tlr2, which followed the same down-regulated trend, but was not significant (Supplementary Figure S5A, Figure 2A). Similarly, all the genes that were up-regulated in sequencing data were also up-regulated in RT-qPCR data (Supplementary Figure S5B, Figure 2A). The Spearman’s rho correlation coefficient of the log2 fold change was evaluated and showed that results from RNA-sequencing data and RT-qPCR were similar (rho = 0.8, p = 0.014) (Figure 2B).
Figure 2. Correlation of fold changes between RT-qPCR and RNA-seq data. Bar plots comparing expression profiles of selected genes at each time point, as measured by RNA-seq (magenta) and RT-qPCR (blue). For RT-qPCR, gene expression was normalized to the ef1a reference gene. Results are presented as mean Log2 fold change values (A, C, E). Correlation of gene expression fold changes between RNA-seq and RT-qPCR was assessed using Spearman’s rho correlation coefficient. Several genes were sampled per time point for comparison. (B, D, F) Log2 fold change correlation analyses between RNA-seq and RT-qPCR at 2, 5, and 29 hours post-amputation (hpA), respectively. Log2 fold change values were used for both methods. Spearman’s correlation results: at 2 hpA: rho = 0.8, p = 0.014; at 5 hpA: rho = 0.79, p = 0.01; at 29 hpA: rho = 0.83, p = 0.003).
A comparable analysis was performed at 5 hpA, with randomly selected DEGs: c1qb, dicp1.1, tnfb, p2rx7 and sulf2b (down-regulated genes) and s1pr4, timp2b, arg2, f5 and mmp9 (up-regulated genes). Genes that were down-regulated in sequencing data, were also down-regulated in RT-qPCR data, with the exception of tnfb which followed the same down-regulated trend, but was not significant (Supplementary Figure S5C, Figure 2C). Some genes that were up-regulated in sequencing data, were also up-regulated in RT-qPCR data with the exception of mmp9 and f5 which followed the same up-regulated trend, but were not significant (Supplementary Figure S5D, Figure 2C). When compared, the log2 values of the fold change from sequencing data and RT-qPCR data gave a Spearman’s rho, rho = 0.79, p = 0.01 (Figure 2D). All of this suggests an acceptable level of concordance between RNA-sequencing and RT-qPCR results.
At 29 hpA, RT-qPCR analysis with randomly selected DEGs revealed that cd74a, sumo3b, dut, fen1 and rps17 showed a down-regulated trend (Supplementary Figure S5E, Figure 2E). Similarly, mmp9, padi2, tom1/2, rab7b and arhgap1 showed an up-regulated trend (Supplementary Figure S5F, Figure 2E). In addition, the Spearman’s rho correlation coefficient of the log2 fold change was evaluated and showed that results from RNA-sequencing data and RT-qPCR had an acceptable level of concordance (rho = 0.83, p = 0.003) (Figure 2F). All of this suggests an overall good level of concordance between RNA-sequencing and RT-qPCR results for 2 hpA, 5 hpA and 29 hpA time points. These data confirm that macrophages modulate their profile during the first steps of wound healing.
3.3 Two-hour-post-amputation macrophages display an acute pro-inflammatory profile
Zebrafish macrophages were previously shown to express pro-inflammatory makers after an immune challenge (6, 27), however the exact molecular program driving M1-like activation during wound healing in this species is incompletely understood. At 2 hpA, we detected pro-inflammatory genes among the top 50 up-regulated genes including tnfa, il6, il11, cxcl18b, cxcl8a and cxcl11.1 (Figure 3A, Supplementary Figure 6, Supplementary Table 5A).
Figure 3. Transcriptome profiling of macrophages at 2 hpA in zebrafish larvae. (A). Heat-map of differentially expressed genes. The 50 most strongly up- and down-regulated genes are displayed. For each gene, normalized read counts were Z-score–transformed across all samples (columns = individual biological replicates). Color scale: dark brown = high expression; light yellow = low expression. Clusters enriched in up-regulated genes are marked with red outlines; clusters enriched in down-regulated genes are marked in blue. Samples from amputated larvae (CUT) are labelled with green squares, and control samples (UNcut) with purple squares. Hierarchical clustering was performed with Ward’s minimum-variance method on log2 fold-change values. (B, C) Bubble plots of Gene Ontology enrichment. Biological-process (B) and molecular-function (C). Terms enriched among the up-regulated DEGs are shown. Dot size represents the number of genes in each term; dot color encodes significance (FDR): red = higher significance (lower p-value), green = lower significance (higher p-value). The x-axis shows fold enrichment (gene ratio); the y-axis lists enriched GO terms. (D) GO chord plot of Kyoto Encyclopedia of Genes and Genomes (KEGG)-associated terms. Genes are connected to their associated pathways by ribbons. Genes are ordered by log2 fold change (red = strong up-regulation; blue = strong down-regulation).
Up-regulated genes at 2 hpA were classified according the gene ontology (GO) terms and the KEGG pathways. GO terms related to biological processes such as “inflammatory response”, “defense response” and “response to cytokine” were enriched from up-regulated genes (Figure 3B). In the molecular functions, the enriched GO terms were: “cytokine activity” (il1b, il12bb, il10, il6, il34, tnfa), “molecular function regulator” (socs1a, socs3a, socs3b, noxo1a, wnt4b, serpine1, stim1b), “transmembrane signaling receptor activity” (adora2b, ptger2a, lifrb, tlr5b, cxcr3.3, il4r.1, il4r.2, il6r) and “signaling receptor activity” (tnfrsf9a, tnfrsf18), indicating activation of pro-inflammatory response (Figure 3C). Further, GO term linked to “chemotaxis”, “leukocyte migration” and “locomotion” were significantly enriched, alongside molecular function categories such as “chemokine activity” (cxcl18b, cxcl11.6, cxcl11.1, cxcl8a) and “CCR chemokine receptor” (ccl38a.5, ccl35.1, ccl35.2, ccl34a.3, ccl34a.4 ccl39.6) (Figures 3B, C, Supplementary Table 6). KEGG analysis revealed enrichment in 7 key inflammatory pathways, with the most prominent being the “c-type lectin receptor signaling” (ptgs2a, bcl3, bcl10, irf1b). Additional enriched pathways included “RIG-I-like receptor signaling” (nfkbiaa, nfkbiab), “NOD-like receptor signaling” (tnfaip3, traf3), “MAPK signaling” (mknk1, mknk2a), and “Toll-like receptor signaling” (tlr5b, cd40) (Figure 3D, Supplementary Figure 7, Supplementary Table 7). In short, at 2 hpA, macrophages enhance the transcription of genes clearly associated to immune cell activity, acute inflammation and defense mechanisms.
Among the top 50 down-regulated genes at 2 hpA, we identified several involved in the regulation of inflammation, including tyk2 (40), sesn1 (41) and cd59, a known inhibitor of complement-mediated cell lysis. Notably this list also includes the cxcr4b gene, which was shown to plays pleiotropic roles in wound repair and regeneration (42) (Figure 3A, Supplementary Table 5B). These results reveal that macrophages first respond to the wound by adopting an acute pro-inflammatory program. While some genes were previously described in zebrafish activated wound macrophages, we identified novel genes underpinning early healing phase.
M1 macrophage signature in zebrafish is still poorly characterized and defining authentic markers for M1 could provide novel and improved tools to distinguish M1 macrophages in this specie. We hypothesized that the pro-inflammatory M1 signature would be conserved across various inflammatory conditions (sterile inflammation or bacterial infection). To test this, we compared the transcriptome of macrophages activated by a wound at 2 hpA with that of macrophages activated by Salmonella enterica infection at 4 hours post-infection, as we previously showed the pro-inflammatory profile of these macrophages (39). We found that a portion of DEGs were shared (Figure 4): 383 genes that were commonly up-regulated and 83 genes that were down-regulated in wound and infection (Figures 4B, C, Supplementary Table 8). The Spearman’s rho correlation coefficient of the log2 fold change was evaluated and showed that common DEGs were similarly modulated during early wound healing and early Salmonella infection (Figure 4A). To functionally annotate these shared genes, we performed a GO enrichment analysis. KEGG pathways such as “RIG-I-like receptor signaling”, “NOD-like receptor signaling” and “C-type lectin receptor signaling” were enriched from shared up-regulated genes (Supplementary Figure 8). In addition, the enriched GO terms were: “cytokine-mediated signaling pathways”, “cytokine activity”, “response to cytokines”, “Defense response” and “leukocyte migration” (Supplementary Figure 8). Thus, wound-activated macrophages at 2 hpA exhibited a pronounced M1-like core gene expression signature related to inflammatory pathways and cytokine activity that may represent a common transcriptional program in response to danger.
Figure 4. Correlation of gene expression fold changes between early wound healing and early Salmonella infection. (A) The scatter plot displays the correlation of log2 fold changes for genes that are commonly up- or down-regulated in both early wound healing and early Salmonella infection. Fold change correlation was calculated using Spearman’s correlation coefficient, based on 383 commonly up-regulated and 83 commonly down-regulated differentially expressed genes (DEGs). The y-axis represents the log2 fold change (log2 FC) in the early wound healing condition, while the x-axis shows the log2 fold change during early Salmonella infection. The pink line indicates the linear regression fit. Dashed blue lines mark the fold change thresholds (log2FC|1|) for both conditions. Each dot represents one gene, with color-coding indicating the regulation direction: pink for up-regulated and blue for down-regulated DEGs. Spearman’s correlation: rho = 0.73, p < 0.001. (B, C) The Venn diagram illustrates the number of down-regulated (B) and up-regulated (C) DEGs in both early wound healing and early Salmonella infection. Each hemisphere represents the DEGs identified in each condition, with the overlapping region indicating shared genes across both conditions.
3.4 Five-hour-post-amputation macrophages display an immunoregulation and pro-reparative signature associated with glucose metabolism
Beyond pro-inflammatory macrophages, other wound-associated macrophage subtypes remain poorly characterized. We identified the arginase 2 (arg2) gene, recently shown as an anti-inflammatory maker for macrophages and neutrophils during wound healing and mycobacteria infection (28), was among the 50 up-regulated genes at 5 hpA. We also identified itln1, a lectin that reduces apoptosis and oxidative stress, previously described as anti-inflammatory (43). Among the top up-regulated genes was illr4, a macrophage-inducible C-type lectin that promotes macrophage activation and inflammatory cell infiltration during sterile inflammation (44) (Figure 5A, Supplementary Table 9). Although not present in the top list, genes involved in lipid mediator class switching, such as ptger2a and alox5ap were also up-regulated.
Figure 5. Transcriptome profiling of macrophages at 5 hpA in zebrafish larvae. (A) Heat-map of differentially expressed genes. The 50 most strongly up- and down-regulated genes are displayed. For each gene, normalized read counts were Z-score-transformed across all samples (columns = individual biological replicates). Color scale: dark brown = high expression; light yellow = low expression. Clusters enriched in up-regulated genes are marked with red outlines; clusters enriched in down-regulated genes are marked in blue. Samples from amputated larvae (CUT) are labelled with green squares, and control samples (UNcut) with purple squares. Hierarchical clustering was performed with Ward’s minimum-variance method on log2 fold-change values. (B, D) Bubble plots of Gene Ontology enrichment. Biological-process (B) and molecular-function (D) terms enriched among the up-regulated DEGs are shown. Dot size represents the number of genes in each term; dot color encodes significance (FDR): red = higher significance (lower p-value), green = lower significance (higher p-value). The x-axis shows fold enrichment (gene ratio); the y-axis lists enriched GO terms. (C) GO chord plot of Kyoto Encyclopedia of Genes and Genomes (KEGG)-associated terms. Genes are connected to their associated pathways by ribbons. Genes are ordered by log2 fold change (red = strong up-regulation; blue = strong down-regulation).
Overrepresentation analysis was carried out on the DEGs using gene ontology (GO) and the KEGG pathways analysis. GO analysis at 5 hpA showed little overlap with processes detected at 2 hpA. At this later stage, up-regulated genes were mapped in glucose metabolism-related categories: “polysaccharide biosynthetic/metabolic process”, “glycogen biosynthetic/metabolic process” and “cellular carbohydrate biosynthetic process” (Figure 5B). KEGG pathway enrichment further confirmed a shift toward metabolic pathways, particularly: “glycolysis” (aldocb, acss2, eno1a, hk1, idha, pgam1a, dlat, gpia) and “starch and sucrose metabolism” including glycogenesis (gys1, pgm2) and glycogenolysis (pygb and gyg1b) (Figure 5C, Supplementary Table 10). These results suggest that 5hpA-macrophages rely primarily on glucose-fueled glycolysis and the glycogenesis-glycogenolysis cycle to meet energy demand and support their functional activities.
Among molecular functions, up-regulated DEGs were enriched GO terms “signaling receptor activity” (tnfsf13b, gpia, pyyb, adm2a, rln3a, eda, and ccl34a.3) (Figure 5D). This category also featured the expression of key homeostatic signaling regulators such as interleukins il6, il10 and penka (45). Of particular interest in the context of tissue repair, we identified up-regulated genes associated with the GO term “positive regulation of cell population proliferation,” including growth factors such as fgf6a, hbegfa, hbegfb, and vegfab (Figures 5B, D). These factors, recognized as pro-resolving markers, have been implicated in initiating tissue repair by promoting fibroblast and keratinocyte proliferation, supporting vasculogenesis, and epithelialization (1). In addition, enriched GO terms linked to “chaperone cofactor-dependent protein refolding” and “de novo posttranslational protein folding” were enriched, highlighting molecular chaperones involved in protein quality control during tissue repair (Figures 5B, D). Notably, several members of the heat shock protein family, including hspa9, hspa8b, hsp70.1 and hsp70.2 (46, 47), were among the up-regulated genes. Interestingly, genes known to be involved in wound healing, such as annexin (anxa2a, anxa1a, anxa6, anxa13) (48), were identified within the enriched GO term “calcium ion binding”. (Figure 5D, Supplementary Table 11).
In addition, the KEGG analysis revealed the “Neuroactive ligand-receptor interaction” pathway, which encode GPCRs or ligands that bind to GPCRs. Some of these genes are involved in immune response modulation (adora2b, penka, ptger2a, adm2a, ltb4r2b, lepr and s1pr4) (Figure 5C, Supplementary Table 10).
By contrast, GO analysis of down-regulated DEGs at 5 hpA revealed enrichment in terms related to immune modulation, including “regulation of immune response,” “regulation of defense response,” and “positive regulation of the MAPK cascade” (Supplementary Table 12). It is noteworthy that many genes associated with the acute pro-inflammatory phase (2hpA) presented in Supplementary Table 2A are not differentially expressed at 5 hpA (Supplementary Table 3A). Genes such as il1b, il12bb, tnfa, cxcl8a, cxcl18b, ccl34a.4, ccl38a.5, and traf2b, which were up-regulated at 2hpA, showed no significant changes at 5 hpA. These findings suggest that by this time point, macrophages begin to downregulate genes associated with the acute inflammatory response, indicating a transition toward the resolution phase and progression of wound healing.
Overall, these findings suggest that by 5 hpA, macrophages undergo a metabolic shift towards glucose and glycogen metabolism while regulating the acute pro-inflammatory phase and transcribing genes to promote repair.
3.5 Some DEGs are shared between times 2 hpA and 5 hpA
Several differentially expressed genes (DEGs) were common between 2 hpA and 5 hpA (Figure 1D). We analyzed the correlation of log2 fold change between all commonly regulated genes at 2 and 5 hpA using the Spearman log2 fold change correlation coefficient (Figure 6). We found an acceptable relationship between the log2 fold change of both conditions (2 hpA and 5 hpA) (rho = 0.7 with a p-value < 0.001), indicating that a part of the genes display similar modulations during first steps of wound healing. However, some genes subset exhibiting rapid modulation during early wound healing. In particular, 14 DEGs were up-regulated at 2 hpA and subsequently down-regulated by 5 hpA (Figure 1D, Supplementary Table 13A). These genes encode key regulators of the host immune response that induce pro-inflammatory cytokines and interferon transcription. For instance, nfkb2, nfkbib, and zfand5a serve as central activators of pro-inflammatory response, while signaling factors such as traf3, fosl1a, mknk2a, tagapa, and irf1b further modulate this response. Additionally, chemokine ccl35.2 and the receptor crfb15 were identified, suggesting that the pro-inflammatory pathway is rapidly down-regulated to optimize healing and minimize tissue damage.
Figure 6. Correlation of gene expression fold changes between 2hpA and 5hpA. The scatter plot depicts the correlation of log2 fold changes in gene expression between the 2 hpA and 5 hpA conditions for the 141 shared DEGs. Each point represents a gene. The x-axis shows the log2 fold change at 5 hpA, and the y-axis shows the log2 fold change at 2 hpA. The dashed vertical and horizontal lines indicate the fold change threshold (log2FC |1|) for both time points. The color-coding differentiates up- and down-regulated genes unique to each time point (shades of green for 2 hpA, shades of brown for 5 hpA), commonly regulated genes, anti-correlated genes (dark red), and non-significant (ns) genes (grey). The pink line represents the linear regression. Fold change correlation was assessed using Spearman’s correlation coefficient (rho = 0.7, p < 0.001).
Of the differentially expressed genes (DEGs), 104 up-regulated at 2 hpA remained elevated at 5 hpA (Figure 1D). Notably, several showed a nearly fivefold increase at 5 hpA (Supplementary Table 13B). They include genes are involved in mitochondrial oxidative metabolism such as, ncf2 and cybrd1, underlying the mitochondria’s key role in wound healing. Genes like, s1pr4 (immunoregulation), prdx1 (oxidative stress protection), and dgkaa, associated with immunoregulatory functions in macrophages (49), showed enhanced expression at 5 hpA. Similarly, cdaa, linked to the pyrimidine salvage pathway essential for nucleotide balance (50), and slc25a15a, a mitochondrial carrier involved in urea metabolism and nitric oxide clearance (51) exhibited ~7-fold increases. Moreover, genes coding for wound healing effectors (timp2b, arg2, and tp53i11b), and enzymes remodeling extracellular matrix exhibited an average 1.5-fold increase at 5 hpA (Supplementary Table 13B).
Taken together, this finding suggests that in wound-macrophages, the early activation of healing mechanisms during acute inflammation phase sets the stage for subsequent resolution and tissue repair.
3.6 Macrophage reprogramming becomes attenuated during the later stages of healing
By contrast to earlier stages, at 29 hpA, limited number of DEGs (N = 45) were identified (Supplementary Table 4A). Among them, 28 were found up-regulated. Seven genes remained consistently regulated across all three time points (2 hpA, 5 hpA, and 29 hpA), albeit with variable fold changes (Supplementary Table 13C). They included genes encoding extracellular matrix remodeling factors (mmp9 and mmp13), which play critical roles in maintaining tissue homeostasis and facilitating epithelial repair during wound healing. The sustained expression of these genes suggests a prolonged role in wound repair and may represent a general signature of wound-associated macrophages.
Among the up-regulated genes at 29hpA time point, we identified genes associated with immune regulation (pdlim1, chia.6, padi2) and host defense (npsn, nfkb2, mpx, cp). Further several up- regulated genes were involved in intracellular trafficking (bicdl2l, tom1l2, snx11, sacm1la), cytoskeletal dynamics (scinlb, bcas3, arhgap1) and autophagy/mitophagy (prkcda, ulk1a, rab7b, fnip1) (Supplementary Table 4A), suggesting that 29-hpA macrophages display functions relying on membrane trafficking and degradative pathways such as efferocytosis and fine-tuning signaling pathways by recycling of surface receptors. Further increased levels of mitophagy related genes suggest a metabolic reprogramming linked to mitochondria. Importantly at 29 hpA, 22 genes (9 up-regulated and 13 down-regulated genes) were unique to this time point, representing a signature of macrophage associated to late healing (Supplementary Tables 4B, C).
This finding demonstrates that by 29 hours post-injury, the reprogramming of macrophages is deeply attenuated. It also reveals a distinct wound-healing signature marked by the non-inflammation, the matrix remodeling and membrane trafficking, supportive of tissue repair.
4 Discussion
The full transcriptome of zebrafish macrophages associated to wound healing has not been mapped in detail. Here, we report the first comprehensive analysis of transcriptional profiles of macrophages during wound healing in zebrafish. Bulk RNA sequencing of purified zebrafish macrophages at different times following fin fold injury allowed us to characterize the highly dynamic changes in gene expression occurring in macrophage reflecting shifts in phenotype over the first hours (Figure 7).
Figure 7. Diagram representing macrophage reprogramming during zebrafish wound healing. Macrophage profiles in terms of cell activity, signaling, cell physiology and metabolism and global function.
Specifically, we show that 2 hpA-macrophages display a strong pro-inflammatory phenotype characterized by the expression of genes involved in immune cell activation, acute inflammation and defense response. Indeed, based on their mRNA profile, 2 hpA-macrophages may secrete variety of chemokines and inflammatory cytokines, such as il1b, il12bb, il6 and tnfa, respond to them by using chemotaxis and by robustly activating inflammatory signaling pathways. Key members of pro-inflammatory pathways including nfkbia/b, nfkb2, tlr5b and traf3 are upregulated. In mammals, TRAF3 plays different roles in the inflammatory signaling pathways mediated by TNFR, TLR, and RLR by participating in different protein complexes, including the direct activation IKK complex in NF-κB signaling pathway (52) and actuation of MAPK pathway. All these inflammatory pathways were found enriched in the zebrafish 2 hpA-macrophages. We detected expression of ptgs2a (also known as cox2), a key enzyme in eicosanoid synthesis. In addition, the chemokine cxcl11.1 and its receptor cxcr3.3 were identified. Cxcl11 has been reported as a macrophage polarization marker during mycobacterial infection (23) while cxcr3.3 regulates macrophage trafficking by antagonizing the cxcl11/cxcr3.2 axis (53). Consistent with previous findings, cxcl8a, cxcl18b, and the receptor ccl38a.5 are up-regulated in 2 hpA macrophages; these chemokines are known to induce neutrophil chemotaxis in zebrafish (54, 55). While 2 hpA-macrophages upregulate pro-inflammatory genes early in response to injury, several genes that inhibit these pathways are also induced (e.g., il10, arg2), most likely to exert negative feedback and limit excessive inflammation. Finally, while we confirmed conserved pro-inflammatory markers in M1-like zebrafish macrophages (56, 57), our findings also reveal a broader set of markers, signaling pathways, and effectors.
In this study, we show that 5-hpA-macrophages downregulate many of the inflammatory genes elevated at 2hpA, including nfkb2, nfkbib, zfand5a, traf3, fosl1a, mknk2a and irf1b. Concomitantly, they strongly up-regulate genes modulating inflammation such as adora2b and lepr in line with a transition period from acute to resolution-phase. In addition, 5 hpA-macrophages exhibit expression of growth factor like fgf6a, hbegfa/b and vegfab and activation of cell proliferation. While Fibroblast growth factor (Fgf) signaling is essential for fin regeneration in zebrafish, the specific role of fgf6a has not yet been studied in this specie (58). In mammals, FGF6 enhance muscle regeneration (59, 60). Further hbegf and vegf family members are pro-regenerative macrophage makers, however their exact role in fish fin fold healing is unknown. Five-hpA-macrophages also express key genes involved in the production of lipid class switch and pro-resolving lipid mediators such as alox5ap, alox12, alox5a. These findings align with zebrafish studies showing that Alox12 upregulation during wound healing promotes anti-inflammatory lipid signaling (61). Importantly these macrophages show phenotypic shifts at the transcriptomic level that are associated with tissue repair, highlighted by expression of annexins which are Ca2+-regulated membrane binding proteins responsive to cellular stress. This suggests that 5 hpA macrophages are transitioning from an inflammatory to a reparative state, displaying a mixed phenotype with features of both M1- and M2-like macrophages.
At 29 hpA, macrophages exhibit a non-inflammatory, reparative phenotype. They markedly up-regulate extracellular-matrix-remodeling genes, notably mmp9 and mmp13a, metalloproteinases previously shown to be essential for zebrafish regeneration (62, 63). Comparable ECM remodelling-focused programs have been described in other injury models including zebrafish fin repair and murine and human cardiac injury, where macrophages shape scar formation and matrix composition (64, 65). Beyond matrix genes, 29 hpA macrophages increase transcripts linked to actin remodeling, intracellular trafficking, and autophagy, consistent with active efferocytosis and receptor recycling. The relatively small set of differentially expressed genes at this late stage suggests that major transcriptional reprogramming has subsided, reflecting a transition toward stabilization rather than continued plasticity.
Interestingly the discovery of immunometabolic features at 5 hpA, provides insights into macrophage functional metabolic reprogramming during wound healing in zebrafish. Macrophage metabolism plays a key role in determining polarization states in mammals (66, 67). Pro-inflammatory macrophages are typically driven by glycolysis in response to bacterial and inflammatory stimuli, whereas anti-inflammatory (M2-like) macrophages preferentially utilize oxidative phosphorylation (OXPHOS) and the TCA cycle. It is widely accepted that metabolic reprogramming precedes and supports the resolution of inflammation (66, 67). Strikingly, our data reveal a metabolic shift in macrophages at 5 hpA, characterized by the upregulation of genes involved in glycolysis, glycogenolysis, gluconeogenesis, and activation of the pentose phosphate pathway (PPP). Although transcripts associated with the TCA cycle were present, we initially observed no strong enrichment of genes related to fatty acid oxidation or oxidative phosphorylation (OXPHOS). Given this unexpected result, we re-analyzed all the up-regulated DEGs including with moderate fold changes (log2FC > 0), and identified several genes linked to OXPHOS (including ndufs5, uqcrfs1, ndufv2, ndufa10, atp6v1ba, atp5pf, uqcrc2a, ndufab1b, ndufs3, cox5aa, cox6a1, cox6b1, cox6b2, cox7c, cox7b, cox10, cox4i1, cox11, cox17, atp5mc3a, ndufb4, atp5f1d, atp6v0a1a, atp6v1c1a, ndufs1, ndufb10, uqcrq, sdhdb, ppa1b). This suggests that while glucose metabolism is the predominant energy source at this stage, a modest OXPHOS signature is also emerging. These findings highlight the metabolic complexity of macrophages in vivo, where multiple pathways likely interact and compensate to support functional demands.
This metabolic profile contrasts with that observed during skeletal muscle regeneration in mammals, where macrophages progressively shift toward TCA cycle activity as they transition from pro- to anti-inflammatory states (68). Similarly, studies using myocardial infarction (MI) models have shown that macrophages undergo a metabolic shift toward oxidative phosphorylation during the post-MI remodeling phase (69). In contrast, our findings align with previous zebrafish studies reporting increased glycolysis in macrophages after injury (70), a process shown to be essential for regenerative function (71). This suggests that during the transition phase in zebrafish, macrophages may rely on glucose metabolism to support their function. Interestingly, 5-hpA-macrophages display enrichment in gene of glycogen metabolism. Glycogen, a glucose-based polymer, functions as a key carbohydrate reserve within cells, readily mobilized to meet energy demands. Glycogen synthase (gys1) and glycogen phosphorylase (pygb) that are respectively involved in glycogenesis and glycogenolysis, were found upregulated at this stage, suggesting that macrophages specifically engage in glycogen cycling for energy production and functional activity. The presence of glycogen stores in human and rodent myeloid cells was reported several decades ago (72). A recent study showed that in mice, M1 macrophages actively synthesize and degrade glycogen (73). Similarly, human M1 and tumor-associated macrophages store glycogen and perform gluconeogenesis. This glycogen-centered metabolic program may support cytokine production and phagocytosis under low-glucose conditions. Further functional investigations are needed to determine whether zebrafish macrophages adapt to physiological stress through metabolic specialization, potentially by cycling between gluconeogenesis and glycogenesis, to maintain energy homeostasis.
Interestingly, several genes identified as DEGs in the present dataset have been previously identified as neutrophils markers or neutrophil subset makers in zebrafish (74, 75) such as mpx, mmp9, mmp13a and padi2. This observation can be explained by two options: 1. mfap4+ cells include a small number of neutrophils as well; 2. macrophages in general or a subpopulation of them can express these neutrophil marker genes at low level. Because mfap4 is specifically expressed by macrophages (76) and the Tg(mfap4:mCherryF) line specifically labels macrophages, with no overlap with Tg(mpx:GFP) line (Supplementary Figure 9), we consider the first scenario unlikely. Some of these genes were previously shown to be expressed in macrophages at very low levels (23). For instance, mmp9 has been shown to be expressed by a subset of macrophages in zebrafish (77), as well as in human and murine macrophages (78, 79), thereby corroborating our findings. This suggests that macrophage express some neutrophil markers at very low levels.
One intriguing question is when appear the resolutive macrophages? Interestingly, we observed that genes associated with tissue repair and resolution, such as timp2b, begin to be expressed as early as 2 hours post-injury. This suggests that a subset of macrophages with reparative or pro-resolving functions may be activated very early, in parallel with those driving acute inflammation. We previously demonstrated, macrophages at the wound site include both tnf-a.a -positive and -negative subsets, suggesting the coexistence of functionally distinct populations (8). Rather than a strictly sequential transition from pro-inflammatory to anti-inflammatory states, our data support a model in which heterogeneous macrophage populations with distinct functional programs are recruited or activated simultaneously. Functional heterogeneity has been previously described for both larval (80) and adult macrophages (81) in the zebrafish. Furthermore resolutive/pro-healing 5-hpA-macrophages display features characteristic of both M2 and M1 phenotypes rendering the traditional M1/M2 nomenclature inadequate for describing wound-associated macrophages in zebrafish.
Our study demonstrates that zebrafish macrophages exhibit dynamic phenotypic plasticity in response to the changing microenvironment of a wound. These phenotypes are not static but are tightly regulated over time, with the majority of transcriptional reprogramming occurring within a relatively short window and showing marked attenuation by 29 hpA. While previous transcriptomic studies in zebrafish focusing on macrophage program during injury at specific time point (70, 80) they did not fully capture the transition from pro-inflammatory activation to resolution, our study provides the first comprehensive description of both the pro-inflammatory response and the onset of resolution.
Additional transcriptomic and metabolic analyses at intermediate time points between 6 and 29 hpA will be essential to uncover potential metabolic transitions and signaling pathways involved in intermediate stages of wound healing. It is worth noting that our macrophage isolation protocol, which employed enzyme-free dissociation and FACS purification, was designed to minimize artificial activation under steady-state conditions (29). Although few transcriptional changes were observed at 29 hpA, it remains likely that macrophages exhibit altered protein expression at this stage that are not captured by bulk transcriptomic, underscoring the importance of complementary proteomic approaches in future studies. Additionally, the use of bulk RNA sequencing, while offering a comprehensive overview of gene expression, inherently masks cellular heterogeneity. In future, analysis of wound-associated macrophages using single cell approaches will further improve our understanding of macrophage heterogeneity during wounding in zebrafish. Beside these limitations, we were able to identify bona fide makers for polarized macrophages across wound healing in zebrafish. Finally, we note that data from various injury paradigms, organs and species suggest that the molecular pathways discovered in this study are conserved across multiple biological contexts.
Data availability statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.
Ethics statement
The animal study was approved by Comité d'Ethique pour l'Expérimentation Animale Languedoc Roussillon under reference APAFIS #36309-2022040114222432 V2. This study was conducted in accordance with the local legislation, the European Union guidelines for handling of laboratory animals (http://ec.europa.eu/environment/chemicals/labanimals/indexen.htm) and institutional requirements. Breeding and maintenance of adult fish were performed at the ZEFIX-LPHI (CNRS, University of Montpellier, Montpellier, France; license number CEEA-LR-B34–172–37).
Author contributions
CB-P: Writing – original draft, Data curation, Conceptualization, Investigation, Validation, Formal Analysis, Visualization. LM: Visualization, Formal Analysis, Writing – review & editing. RÖ: Investigation, Writing – review & editing. SB: Visualization, Formal analysis, Writing – review & editing. AL: Data curation, Formal analysis, Writing – review & editing. SG: Investigation, Writing – review & editing. MN-C: Writing – original draft, Investigation, Conceptualization, Supervision, Funding acquisition, Writing – review & editing.
Funding
The author(s) declared that financial support was received for this work and/or its publication. This work has received funding from the European Union’s Horizon 2020 research and innovation programme by the grant Marie Skłodowska-Curie Training Network INFLANET, grant agreement n°955576 and from the French Agence Nationale de la Recherche (ANR-19-CE15-0005-01, MacrophageDynamics). This work was also supported by a grant from Region Occitanie (REPERE « INFLANET », grant n° 21018327).
Acknowledgments
This work was undertaken with support from the fish facility of ZEFIX-LPHI (University of Montpellier), Catherine Gonzalez (LPHI, CNRS) and the MGX-Montpellier GenomiX platform and the qPCR Haut Debit platform of the University of Montpellier. We are grateful to Georges Lutfalla for his experimental help and secure funding, Clara Cayron for her experimental help. MGX acknowledges financial support from France Génomique National infrastructure, funded as part of “Investissement d’Avenir” program managed by Agence Nationale pour la Recherche (contract ANR-10-INBS-09). We acknowledge the imaging facility MRI, member of the national infrastructure France-BioImaging (https://ror.org/01y7vt929) supported by the French National Research Agency (ANR-24-INBS-0005 FBI BIOGEN).
Conflict of interest
The authors declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fimmu.2025.1721612/full#supplementary-material
References
1. Fernández-Guarino M, Hernández-Bule ML, and Bacci S. Cellular and molecular processes in wound healing. Biomedicines. (2023) 11:2526. doi: 10.3390/biomedicines11092526
2. Wynn TA and Vannella KM. Macrophages in tissue repair, regeneration, and fibrosis. Immunity. (2016) 44:450–62. doi: 10.1016/j.immuni.2016.02.015
3. Arnold L, Henry A, Poron F, Baba-Amer Y, Van Rooijen N, Plonquet A, et al. Inflammatory monocytes recruited after skeletal muscle injury switch into antiinflammatory macrophages to support myogenesis. J Exp Med. (2007) 204:1057–69. Available online at. doi: 10.1084/jem.20070075
4. Goren I, Allmann N, Yogev N, Schürmann C, Linke A, Holdener M, et al. A transgenic mouse model of inducible macrophage depletion: effects of diphtheria toxin-driven lysozyme M-specific cell lineage ablation on wound inflammatory, angiogenic, and contractive processes. Am J Pathol. (2009) 175:132–47. doi: 10.2353/ajpath.2009.081002
5. Hasegawa T, Hall CJ, Crosier PS, Abe G, Kawakami K, Kudo A, et al. Transient inflammatory response mediated by interleukin-1β is required for proper regeneration in zebrafish fin fold. Elife. (2017) 6:e22716. doi: 10.7554/eLife.22716
6. Miskolci V, Squirrell J, Rindy J, Vincent W, Sauer JD, Gibson A, et al. Distinct inflammatory and wound healing responses to complex caudal fin injuries of larval zebrafish. Elife. (2019) 8:e45976. doi: 10.7554/eLife.45976
7. Godwin JW, Pinto AR, and Rosenthal NA. Macrophages are required for adult salamander limb regeneration. Proc Natl Acad Sci U.S.A. (2013) 110:9415–20. doi: 10.1073/pnas.1300290110
8. Nguyen-Chi M, Laplace-Builhé B, Travnickova J, Luz-Crawford P, Tejedor G, Lutfalla G, et al. TNF signaling and macrophages govern fin regeneration in zebrafish larvae. Cell Death Dis. (2017) 8:e2979. doi: 10.1038/cddis.2017.374
9. Petrie TA, Strand NS, Tsung-Yang C, Rabinowitz JS, and Moon RT. Macrophages modulate adult zebrafish tail fin regeneration. Dev. (2014) 141:2581–91. doi: 10.1242/dev.098459
10. Murray PJ, Allen JE, Biswas SK, Fisher EA, Gilroy DW, Goerdt S, et al. Macrophage activation and polarization: nomenclature and experimental guidelines. Immunity. (2014) 41:14–20. doi: 10.1016/j.immuni.2014.06.008
11. Sica A and Mantovani A. Macrophage plasticity and polarization: In vivo veritas. J Clin Invest. (2012) 122:787–95. doi: 10.1172/JCI59643
12. Daley JM, Brancato SK, Thomay AA, Reichner JS, and Albina JE. The phenotype of murine wound macrophages. J Leukoc Biol. (2009) 87:59–67. doi: 10.1189/jlb.0409236
13. Lucas T, Waisman A, Ranjan R, Roes J, Krieg T, Müller W, et al. Differential roles of macrophages in diverse phases of skin repair. J Immunol. (2010) 184:3964–77. doi: 10.4049/jimmunol.0903356
14. Mosser DM and Edwards JP. Exploring the full spectrum of macrophage activation. Nat Rev Immunol. (2008) 8:958–69. doi: 10.1038/nri2448
15. Chen CJ, Kajita H, Takaya K, Aramaki-Hattori N, Sakai S, Asou T, et al. Single-cell RNA-seq analysis reveals cellular functional heterogeneity in dermis between fibrotic and regenerative wound healing fates. Front Immunol. (2022) 13:875407. doi: 10.3389/fimmu.2022.875407
16. Pang J, Maienschein-Cline M, and Koh TJ. Monocyte/macrophage heterogeneity during skin wound healing in mice. J Immunol. (2022) 209:1999–2011. doi: 10.4049/jimmunol.2200365
17. Klinkert K, Whelan D, Clover AJP, Leblond AL, Kumar AHS, and Caplice NM. Selective M2 macrophage depletion leads to prolonged inflammation in surgical wounds. Eur Surg Res. (2017) 58:109–20. doi: 10.1159/000451078
18. Sipka T, Park SA, Ozbilgic R, Balas L, Durand T, Mikula K, et al. Macrophages undergo a behavioural switch during wound healing in zebrafish. Free Radic Biol Med. (2022) 192:200–12. doi: 10.1016/j.freeradbiomed.2022.09.021
19. Beffagna G. Zebrafish as a smart model to understand regeneration after heart injury: how fish could help humans. Front Cardiovasc Med. (2019) 6:107. doi: 10.3389/fcvm.2019.00107
20. Speirs ZC, Loynes CA, Mathiessen H, Elks PM, Renshaw SA, and Jørgensen L von G. What can we learn about fish neutrophil and macrophage response to immune challenge from studies in zebrafish. Fish Shellfish Immunol. (2024) 148:109490. doi: 10.1016/j.fsi.2024.109490
21. Xie Y, Meijer AH, and Schaaf MJM. Modeling inflammation in zebrafish for the development of anti-inflammatory drugs. Front Cell Dev Biol. (2021) 8:620984. doi: 10.3389/fcell.2020.620984
22. Herbomel P, Thisse B, and Thisse C. Ontogeny and behaviour of early macrophages in the zebrafish embryo. Development. (1999) 126:3735–45. doi: 10.1242/dev.126.17.3735
23. Rougeot J, Torraca V, Zakrzewska A, Kanwal Z, Jansen HJ, Sommer F, et al. RNAseq profiling of leukocyte populations in zebrafish larvae reveals a cxcl11 chemokine gene as a marker of macrophage polarization during mycobacterial infection. Front Immunol. (2019) 10:832. doi: 10.3389/fimmu.2019.00832
24. Lewis A and Elks PM. Hypoxia induces macrophage tnfa expression via cyclooxygenase and prostaglandin E2 in vivo. Front Immunol. (2019) 10:2321. doi: 10.3389/fimmu.2019.02321
25. Sanz-Morejón A, García-Redondo AB, Reuter H, Marques IJ, Bates T, Galardi-Castilla M, et al. Wilms tumor 1b expression defines a pro-regenerative macrophage subtype and is required for organ regeneration in the zebrafish. Cell Rep. (2019) 28:1296–306.e6. doi: 10.1016/j.celrep.2019.06.091
26. Leiba J, Özbilgiç R, Hernández L, Demou M, Lutfalla G, Yatime L, et al. Molecular actors of inflammation and their signaling pathways: mechanistic insights from zebrafish. Biology. (2023) 12:153. doi: 10.3390/biology12020153
27. Nguyen-Chi M, Laplace-Builhe B, Travnickova J, Luz-Crawford P, Tejedor G, Phan QT, et al. Identification of polarized macrophage subsets in zebrafish. Elife. (2015) 4:e07288. doi: 10.7554/eLife.07288
28. Hammond FR, Lewis A, Speirs ZC, Anderson HE, Sipka T, Williams LG, et al. An arginase 2 promoter transgenic line illuminates immune cell polarisation in zebrafish. Dis Model Mech. (2023) 16:dmm049966. doi: 10.1242/dmm.049966
29. Begon-Pescia C, Boireau S, Boyer-Clavel M, Lutfalla G, and Nguyen-Chi M. Preparing sequencing grade RNAs from a small number of FACS-sorted larvae macrophages isolated from enzyme free dissociated zebrafish larvae. MethodsX. (2022) 9:101651. doi: 10.1016/j.mex.2022.101651
30. Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, and Schilling TF. Stages of embryonic development of the zebrafish. Dev Dyn. (1995) 203:253–310. doi: 10.1002/aja.1002030302
31. Phan QT, Sipka T, Gonzalez C, Levraud JP, Lutfalla G, and Nguyen-Chi M. Neutrophils use superoxide to control bacterial infection at a distance. PloS Pathog. (2018) 14:e1007157. doi: 10.1371/journal.ppat.1007157
32. Kim D, Pertea G, Trapnell C, Pimentel H, Kelley R, and Salzberg SL. TopHat2: Accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biol. (2013) 14:R36. doi: 10.1186/gb-2013-14-4-r36
33. Gentleman RC, Carey VJ, Bates DM, Bolstad B, Dettling M, Dudoit S, et al. Bioconductor: open software development for computational biology and bioinformatics. Genome Biol. (2004) 5. doi: 10.1186/gb-2004-5-10-r80
34. Huber W, Carey VJ, Gentleman R, Anders S, Carlson M, Carvalho BS, et al. Orchestrating high-throughput genomic analysis with Bioconductor. Nat Methods. (2015) 12:115–21. doi: 10.1038/nmeth.3252
35. Benjamini Y and Hochberg Y. Controlling the false discovery rate: A practical and powerful approach to multiple testing. J R Stat Soc Ser B Stat Methodol. (1995) 57:289–300. doi: 10.1111/j.2517-6161.1995.tb02031.x
36. Ge SX, Jung D, Jung D, and Yao R. ShinyGO: A graphical gene-set enrichment tool for animals and plants. Bioinformatics. (2020) 36:2628–9. doi: 10.1093/bioinformatics/btz931
37. Tang D, Chen M, Huang X, Zhang G, Zeng L, Zhang G, et al. SRplot: A free online platform for data visualization and graphing. PloS One. (2023) 18:e0294236. doi: 10.1371/journal.pone.0294236
38. Laplace-Builhé B, Barthelaix A, Assou S, Bohaud C, Pratlong M, Severac D, et al. NRG1/ErbB signalling controls the dialogue between macrophages and neural crest-derived cells during zebrafish fin regeneration. Nat Commun. (2021) 12:6336. doi: 10.1038/s41467-021-26422-5
39. Leiba J, Sipka T, Begon-Pescia C, Bernardello M, Tairi S, Bossi L, et al. Dynamics of macrophage polarization support Salmonella persistence in a whole living organism. Elife. (2024) 13:e89828. doi: 10.7554/eLife.89828
40. Mori R, Wauman J, Icardi L, van der Heyden J, De Cauwer L, Peelman F, et al. TYK2-induced phosphorylation of Y640 suppresses STAT3 transcriptional activity. Sci Rep 2017–71. (2017) 7:15919. doi: 10.1038/s41598-017-15912-6
41. Rhee SG and Bae SH. The antioxidant function of sestrins is mediated by promotion of autophagic degradation of Keap1 and Nrf2 activation and by inhibition of mTORC1. Free Radic Biol Med. (2015) 88:205–11. doi: 10.1016/j.freeradbiomed.2015.06.007
42. Chen H, Li G, Liu Y, Ji S, Li Y, Xiang J, et al. Pleiotropic roles of CXCR4 in wound repair and regeneration. Front Immunol. (2021) 12:668758. doi: 10.3389/fimmu.2021.668758
43. Zhong X, Li X, Liu F, Tan H, and Shang D. Omentin inhibits TNF-α-induced expression of adhesion molecules in endothelial cells via ERK/NF-κB pathway. Biochem Biophys Res Commun. (2012) 425:401–6. doi: 10.1016/j.bbrc.2012.07.110
44. Lepenies B, Martinez-Pomares L, Madan T, and Chiffoleau E. C-type lectin-like receptors as emerging orchestrators of sterile inflammation represent potential therapeutic targets. Front Immunol. (2018) 9:227. doi: 10.3389/fimmu.2018.00227
45. Liu C, Liu X, He Z, Zhang J, Tan X, Yang W, et al. Proenkephalin-A secreted by renal proximal tubules functions as a brake in kidney regeneration. Nat Commun. (2023) 14:7167. doi: 10.1038/s41467-023-42929-5
46. Ferat-Osorio E, Sánchez-Anaya A, Gutiérrez-Mendoza M, Boscó-Gárate I, Wong-Baeza I, Pastelin-Palacios R, et al. Heat shock protein 70 down-regulates the production of toll-like receptor-induced pro-inflammatory cytokines by a heat shock factor-1/constitutive heat shock element-binding factor-dependent mechanism. J Inflamm (Lond). (2014) 11:19. doi: 10.1186/1476-9255-11-19
47. Nagashima M, Fujikawa C, Mawatari K, Mori Y, and Kato S. HSP70, the earliest-induced gene in the zebrafish retina during optic nerve regeneration: Its role in cell survival. Neurochem Int. (2011) 58:888–95. doi: 10.1016/j.neuint.2011.02.017
48. Christin Häger S and Nylandsted J. Annexins: players of single cell wound healing and regeneration. Commun Integr Biol. (2019) 12:162–5. doi: 10.1080/19420889.2019.1676139
49. Manigat LC, Granade ME, Taori S, Miller CA, Vass LR, Zhong XP, et al. Loss of diacylglycerol kinase α Enhances macrophage responsiveness. Front Immunol. (2021) 12:722469. doi: 10.3389/fimmu.2021.722469
50. Frances A and Cordelier P. The emerging role of cytidine deaminase in human diseases: A new opportunity for therapy? Mol Ther. (2020) 28:357–66. doi: 10.1016/j.ymthe.2019.11.026
51. Fu A, Alvarez-Perez JC, Avizonis D, Kin T, Ficarro SB, Choi DW, et al. Glucose-dependent partitioning of arginine to the urea cycle protects β-cells from inflammation. Nat Metab. (2020) 2:432–46. doi: 10.1038/s42255-020-0199-4
52. Lin M, Ji X, Lv Y, Cui D, and Xie J. The roles of TRAF3 in immune responses. Dis Markers. (2023) 2023:7787803. doi: 10.1155/2023/7787803
53. Sommer F, Torraca V, Kamel SM, Lombardi A, and Meijer AH. Frontline Science: Antagonism between regular and atypical Cxcr3 receptors regulates macrophage migration during infection and injury in zebrafish. J Leukoc Biol. (2020) 107:185–203. doi: 10.1002/JLB.2HI0119-006R
54. Oehlers SHB, Flores MV, Hall CJ, O’Toole R, Swift S, Crosier KE, et al. Expression of zebrafish cxcl8 (interleukin-8) and its receptors during development and in response to immune stimulation. Dev Comp Immunol. (2010) 34:352–9. doi: 10.1016/j.dci.2009.11.007
55. Torraca V, Otto NA, Tavakoli-Tameh A, and Meijer AH. The inflammatory chemokine Cxcl18b exerts neutrophil-specific chemotaxis via the promiscuous chemokine receptor Cxcr2 in zebrafish. Dev Comp Immunol. (2017) 67:57–65. doi: 10.1016/j.dci.2016.10.014
56. Jablonski KA, Amici SA, Webb LM, Ruiz-Rosado JDD, Popovich PG, Partida-Sanchez S, et al. Novel markers to delineate murine M1 and M2 macrophages. PloS One. (2015) 10:e0145342. doi: 10.1371/journal.pone.0145342
57. Orecchioni M, Ghosheh Y, Pramod AB, and Ley K. Corrigendum: Macrophage Polarization: Different Gene Signatures in M1(LPS+) vs. Classically and M2(LPS–) vs. Alternatively Activated Macrophages. Front Immunol. (2020) 11:234. doi: 10.3389/fimmu.2019.01084
58. Cudak N, López-Delgado AC, Keil S, and Knopf F. Fibroblast growth factor pathway component expression in the regenerating zebrafish fin. Gene Expr Patterns. (2023) 48:119307. doi: 10.1016/j.gep.2023.119307
59. Floss T, Arnold HH, and Braun T. A role for FGF-6 in skeletal muscle regeneration. Genes Dev. (1997) 11:2040–51. doi: 10.1101/gad.11.16.2040
60. Cai Q, Wu G, Zhu M, Ge H, Xue C, Zhang Q, et al. FGF6 enhances muscle regeneration after nerve injury by relying on ERK1/2 mechanism. Life Sci. (2020) 248:117465. doi: 10.1016/j.lfs.2020.117465
61. Loynes CA, Lee JA, Robertson AL, Steel MJG, Ellett F, Feng Y, et al. PGE2 production at sites of tissue injury promotes an anti-inflammatory neutrophil phenotype and determines the outcome of inflammation resolution in vivo. Sci Adv. (2018) 4:eaar8320. doi: 10.1126/sciadv.aar8320
62. LeBert DC, Squirrell JM, Rindy J, Broadbridge E, Lui Y, Zakrzewska A, et al. Matrix metalloproteinase 9 modulates collagen matrices and wound repair. Dev. (2015) 142:2136–46. doi: 10.1242/dev.121160
63. Li L, Zhang J, and Akimenko MA. Inhibition of mmp13a during zebrafish fin regeneration disrupts fin growth, osteoblasts differentiation, and Laminin organization. Dev Dyn. (2020) 249:187–98. doi: 10.1002/dvdy.112
64. Iyer RP, Jung M, and Lindsey ML. MMP-9 signaling in the left ventricle following myocardial infarction. Am J Physiol - Heart Circulatory Physiol. (2016) 311:H190–8. doi: 10.1152/ajpheart.00243.2016
65. Lindsey ML. Assigning matrix metalloproteinase roles in ischaemic cardiac remodelling. Nat Rev Cardiol. (2018) 15:471–9. doi: 10.1038/s41569-018-0022-z
66. Wculek SK, Dunphy G, Heras-Murillo I, Mastrangelo A, and Sancho D. Metabolism of tissue macrophages in homeostasis and pathology. Cell Mol Immunol. (2022) 19:384–408. doi: 10.1038/s41423-021-00791-9
67. Almeida L, Dhillon-LaBrooy A, and Sparwasser T. The evolutionary tug-of-war of macrophage metabolism during bacterial infection. Trends Endocrinol Metab. (2024) 35:235–48. doi: 10.1016/j.tem.2023.11.002
68. Varga T, Mounier R, Horvath A, Cuvellier S, Dumont F, Poliska S, et al. Highly dynamic transcriptional signature of distinct macrophage subsets during sterile inflammation, resolution, and tissue repair. J Immunol. (2016) 196:4771–82. doi: 10.4049/jimmunol.1502490
69. Mouton AJ, DeLeon-Pennell KY, Rivera Gonzalez OJ, Flynn ER, Freeman TC, Saucerman JJ, et al. Mapping macrophage polarization over the myocardial infarction time continuum. Basic Res Cardiol. (2018) 113:26. doi: 10.1007/s00395-018-0686-x
70. Xie Y, Tolmeijer S, Oskam JM, Tonkens T, Meijer AH, and Schaaf MJM. Glucocorticoids inhibit macrophage differentiation towards a pro-inflammatory phenotype upon wounding without affecting their migration. DMM Dis Model Mech. (2019) 12:dmm037887. doi: 10.1242/dmm.037887
71. Bohaud C, de la Cruz J, Terraza C, Barthelaix A, Laplace-Builhé B, Jorgensen C, et al. Lactate metabolism coordinates macrophage response and regeneration in zebrafish. Theranostics. (2022) 12:3995–4009. doi: 10.7150/thno.65235
72. Eichner RD, Lim LK, and Hunt NH. Glycogen synthesis in resident and thioglycollate-elicited rat peritoneal exudate cells. Int J Biochem. (1983) 15:1485–8. doi: 10.1016/0020-711X(83)90084-8
73. Ma J, Wei K, Liu J, Tang K, Zhang H, Zhu L, et al. Glycogen metabolism regulates macrophage-mediated acute inflammatory responses. Nat Commun. (2020) 11:1769. doi: 10.1038/s41467-020-15636-8
74. Mathias JR, Perrin BJ, Liu T-X, Kanki J, Look AT, and Huttenlocher A. Resolution of inflammation by retrograde chemotaxis of neutrophils in transgenic zebrafish. J Leukoc Biol. (2006) 80:1281–8. doi: 10.1189/jlb.0506346
75. Hou Y, Khatri P, Rindy J, Schultz Z, Gao A, Chen Z, et al. Single-cell transcriptional landscape of temporal neutrophil response to burn wound in larval zebrafish. J Immunol. (2024) 213:469–80. doi: 10.4049/jimmunol.2400149
76. Walton EM, Cronan MR, Beerman RW, and Tobin DM. The macrophage-specific promoter mfap4 allows live, long-term analysis of macrophage behavior during mycobacterial infection in zebrafish. PloS One. (2015) 10:e0138949. doi: 10.1371/journal.pone.0138949
77. Kirchberger S, Shoeb MR, Lazic D, Wenninger-Weinzierl A, Fischer K, Shaw LE, et al. Comparative transcriptomics coupled to developmental grading via transgenic zebrafish reporter strains identifies conserved features in neutrophil maturation. Nat Commun. (2024) 15:1792. doi: 10.1038/s41467-024-45802-1
78. Lepidi S, Kenagy RD, Raines EW, Chiu ES, Chait A, Ross R, et al. MMP9 production by human monocyte-derived macrophages is decreased on polymerized type I collagen. J Vasc Surg. (2001) 34:1111–8. doi: 10.1067/mva.2001.119401
79. Tan TK, Zheng G, Hsu TT, Wang Y, Lee VWS, Tian X, et al. Macrophage matrix metalloproteinase-9 mediates epithelial-mesenchymal transition in vitro in murine renal tubular cells. Am J Pathol. (2010) 176:1256–70. doi: 10.2353/ajpath.2010.090188
80. Denans N, Tran NTT, Swall ME, Diaz DC, Blanck J, and Piotrowski T. An anti-inflammatory activation sequence governs macrophage transcriptional dynamics during tissue injury in zebrafish. Nat Commun. (2022) 13:5356. doi: 10.1038/s41467-022-33015-3
Keywords: inflammation, macrophages, polarization, transcriptomic profile, wound healing, zebrafish
Citation: Begon-Pescia C, Manchon L, Özbilgiç R, Boireau S, Louis A, Georges S and Nguyen-Chi M (2026) Temporal dynamics of macrophage transcriptional profiles during zebrafish wound healing. Front. Immunol. 16:1721612. doi: 10.3389/fimmu.2025.1721612
Received: 09 October 2025; Accepted: 23 December 2025; Revised: 22 December 2025;
Published: 19 January 2026.
Edited by:
Guirong Wang, Upstate Medical University, United StatesReviewed by:
Aleksandr Bobrovskikh, Russian Academy of Sciences (RAS), RussiaYiran Hou, University of Wisconsin-Madison, United States
Copyright © 2026 Begon-Pescia, Manchon, Özbilgiç, Boireau, Louis, Georges and Nguyen-Chi. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Mai Nguyen-Chi, bWFpLWV2YS5uZ3V5ZW4tY2hpQHVtb250cGVsbGllci5mcg==