Chronic macrophage activation derails muscle repair by disrupting mannose-receptor-linked plasticity revealed by endogenous irg1/acod1 tracking

Spencer, Caroline G.; Hamilton, Matthew; Bedsole, Ethan; Wei, Yingshan N.; Rojas, Alison M.; Burciu, Andrew; Zhu, John; Sabin, Keith Z.; Shiau, Celia E.

doi:10.1038/s41467-025-68204-3

Download PDF

Article
Open access
Published: 07 January 2026

Chronic macrophage activation derails muscle repair by disrupting mannose-receptor-linked plasticity revealed by endogenous irg1/acod1 tracking

Nature Communications volume 17, Article number: 1466 (2026) Cite this article

5167 Accesses
1 Citations
22 Altmetric
Metrics details

Subjects

Abstract

Macrophages are central drivers of chronic inflammation, yet how a sustained inflammatory state alters their function remains unclear. Using GFP knock-in zebrafish targeting irg1/acod1 that marks macrophage activation, we track the dynamic transitions of macrophage states during acute muscle injury under homeostatic and chronically inflamed conditions, induced by genetic mutation of nlrc3l. In the chronic inflammation model, muscle repair is impaired and expression of the mannose receptor mrc1b/cd206 is severely downregulated in a myd88-dependent manner. Two reparative macrophage subtypes, defined by their cellular behavior and single-cell transcriptomics profile, clustering and muscle-encasing, are lost. A chronic infection model recapitulates these defects, underscoring the link to macrophage mrc1b repression. Depleting either mrc1b or macrophages impairs muscle repair. Reinstating normal macrophage states by restoring macrophage nlrc3l expression or ablating myd88-mediated inflammatory pathways rescues muscle repair in nlrc3l mutants. Contrary to conventional discrete states, we identify hybrid M1/M2 macrophage states post-injury. While transient during normal injury response, a pro-inflammatory hybrid state persists during chronic activation, which restricts macrophage heterogeneity, represses mrc1b, and inhibits intracellular cathepsin K accumulation, a hallmark of reparative subtypes. Thus, our study provides mechanistic insight into the dynamics of macrophage activation during muscle injury and repair, and how these processes are modulated under chronic inflammation.

Identification of macrophages in normal and injured mouse tissues using reporter lines and antibodies

Article Open access 16 March 2022

Enteric glial cells favor accumulation of anti-inflammatory macrophages during the resolution of muscularis inflammation

Article Open access 07 September 2022

Coordinated chemokine expression defines macrophage subsets across tissues

Article 02 May 2024

Introduction

Immune dysregulation that causes persistent inflammatory cell infiltration can lead to chronic tissue damage, including secondary injury beyond the original insult^1,2,3,4,5. Macrophages play a central role in this process, as inappropriate or prolonged activation drives chronic inflammation^1,3,4,6. However, the impact of altered macrophage states on injury responses and tissue repair, particularly in the complex intact living organism, remains incompletely understood. Furthermore, the mechanisms by which macrophages balance the need to promote acute inflammation for tissue repair with preventing chronic inflammation and associated collateral damage remain elusive.

Skeletal muscle has been a prominent model for studying immune regulation of tissue injury and repair^3,6,7. This model highlights the critical interactions among macrophages (the primary inflammatory cells), injured muscle cells, satellite cells (muscle-specific stem cells), and other cell types, all of which contribute to tissue recovery following injury^1,3,6,8. Recent research underscores the essential and evolutionarily conserved role of acute inflammation in tissue repair, with evidence spanning from zebrafish to humans^7,9,10,11. Although prior studies have provided insights into the role of macrophages in responding to and facilitating acute muscle injury repair^1,3,6,8, including via single-cell transcriptomics and trajectory analysis to infer temporal changes and cellular interactions^12,13, direct real-time in vivo imaging of macrophage behaviors and interactions throughout the injury response remains lacking. Time-lapse imaging, combined with a rigorous longitudinal microscopy approach, enabled us to track macrophages across distinct phases of muscle injury and repair within the same subjects. Such analyses are essential for fully and precisely understanding the progression from inflammation to resolution at both the cellular and organismal levels in an intact living system. Macrophages are believed to transition from a pro-inflammatory state to a pro-repair state at different phases of injury response^1,8. However, there remains debate about whether these macrophage states are distinct and non-overlapping, or if they represent different subpopulations^3,14,15. Understanding these transitions is critical, as macrophage dysregulation is implicated in various diseases, including diabetes, cancer, and neurodegenerative disorders, where chronic activation of macrophages or microglia contributes to pathology^1,5,15.

Metabolic reprogramming is a hallmark of macrophage activation and involves the upregulation of the mitochondrial enzyme aconitate decarboxylase (acod1), also known as immune-responsive gene 1 (irg1), in response to diverse immune triggers, including microbial infections and endogenous damage-associated molecular patterns (DAMPs)^16,17,18,19. The transcriptional induction of irg1 in macrophage activation has been well-characterized in both zebrafish and mammals^{17,18,20,21,22}. To investigate macrophage state transitions during injury in real time within a localized and controlled environment, we generated a GFP knock-in allele at the zebrafish irg1/acod1 locus. This genetically modified allele allows tracing and isolation of macrophages at different activation states based on differential irg1 expression. Previous studies in zebrafish have shown that irg1 induction is macrophage-specific following bacterial infection, as determined through co-localization with cell-specific markers and macrophage ablation approaches^17,22. While an earlier study created a transposon-generated fluorescent reporter for irg1²², we anticipated that the precise GFP knock-in allele would offer a more quantitative readout of endogenous irg1 expression, ensuring reliable and consistent quantification across individuals and clutches while avoiding non-specific expression. Our direct comparison of this knock-in model with a conventional transposon-mediated irg1 reporter for analysis of activated macrophages supported these expectations.

To explore the consequences of chronic macrophage activation in vivo, we employed the zebrafish GFP knock-in allele in a mutant background st73 lacking the NOD-like receptor nlrc3l, a mutation known to cause chronic inflammation^23,24. These mutants display constitutive macrophage activation in the absence of overt immune challenges²⁴, allowing consistent tracking of chronic macrophage activation without the need to experimentally induce immune responses, thereby minimizing variability. Through dynamic imaging and longitudinal analysis, we evaluated the chronic macrophage activation and demonstrated that it led to the depletion of two key functional macrophage subtypes required for muscle injury clearance and repair in vivo. A parallel model of chronic macrophage activation by persistent E. coli infection mirrored these phenotypes. Furthermore, impaired tissue repair mimics the effects of macrophage depletion, and is linked to a significant reduction of the mannose receptor mrc1b (also known as CD206). Single-cell analysis revealed a profound shift in the transcriptional landscape of the chronically activated macrophages toward a dominating inflammatory state (sustained expression of inflammatory proteinase mmp9, tnfa, cxcl11.1 among other inflammatory genes) while still activating reparative pathway genes. This resulted in reduced cellular heterogeneity and loss of reparative subtypes required for muscle repair. Thus, chronic macrophage activation impairs muscle injury repair not by enhancing or expanding inflammatory functions, but by disrupting reparative programs—particularly through mrc1b downregulation.

Taken together, this study focuses on the dynamic transitions and shifts in macrophage cell states by visualizing these processes in real time during an injury response. By forcing all macrophages toward chronic pro-inflammatory states using the nlrc3l mutants, we test how adaptable and reversible macrophage states are upon an injury perturbation within the intact living organism. Notably, chronic activation drives myd88-dependent repression of the mannose receptor mrc1b/cd206 in macrophages in both the genetic and persistent infection model of chronic inflammation. These insights reveal a previously unrecognized mechanism linking chronic inflammation to a severe loss of mannose receptor as a cause for reduced macrophage plasticity and impaired tissue repair.

Results

Generation of GFP knock-in in zebrafish irg1/acod1 to monitor macrophage activation

We employed a 48-bp short DNA homology-mediated end joining CRISPR-Cas9 strategy, termed GeneWeld^25,26, to facilitate the integration of a P2A-GFP reporter cassette into exon 2 of the zebrafish irg1/acod1 locus (referred to as irg1 herein), proximal to the coding sequence initiation (Fig. 1). Following the selection of irg1 gRNA #2 based on its superior efficacy in inducing CRISPR-induced double-stranded DNA breaks, the genomic region encompassing the gRNA was sequenced from multiple wild-type zebrafish to identify the upstream and downstream 48 bp homology sequences for cloning into the gene targeting GeneWeld pGTag vector (Fig. 1a). Wild-type zebrafish harboring exact matches to the short homology sequences flanking the irg1 gRNA target site were chosen as parents for single-cell zebrafish embryo injections for CRISPR/Cas9 targeting (Fig. 1a). Subsequent analysis revealed a germline transmission rate of less than 1% for precise GFP knock-in, validated by site-specific PCR assays with two correctly targeted founder fish (#7 and #31) in the F1 generation (Fig. 1b, c). The precise in-frame integration of P2A-GFP at the zebrafish irg1 locus was confirmed via Sanger sequencing in both F1 and F2 generations (Fig. 1c). Notably, zebrafish carrying the GFP knock-in allele of irg1, designated as “irg1-KI:GFP,” exhibited segregation consistent with Mendelian inheritance patterns expected for a single-site integration.

Fig. 1: Construction of zebrafish irg1/acod1 GFP knock-in for tracking the dynamics of macrophage activation in vivo. — **Fig. 1: Construction of zebrafish *irg1/acod1* GFP knock-in for tracking the dynamics of macrophage activation in vivo.**

Furthermore, to test the function of the new knock-in reporter, we devised a dual labeling system utilizing irg1-KI:GFP to label activated macrophages coupled with mpeg1:BFP, which labels all macrophages with BFP (Fig. 1d). Upon LPS activation, macrophages in the brain, also known as microglia, robustly expressed GFP, with varying levels at the single-cell level, indicative of the capacity of the knock-in reporter to capture heterogeneity in activation levels compared to the uniformly low basal GFP expression observed in uninjected wild-type controls (Fig. 1d).

Since the GFP knock-in (KI) model effectively replaces the functional irg1 locus, creating a knock-in knockout condition, we independently created an irg1 knockout mutation bcz13 using CRISPR/Cas9 that causes a premature stop codon in irg1 (Supplementary Fig. 1). This mutation was studied to verify that heterozygous individuals did not exhibit a phenotype and were functionally equivalent to wild-type (Supplementary Fig. 1). This validated the use of the irg1-KI:GFP as a reliable reporter for irg1 expression without confounding effects from a knock-in knockout situation. Using a systemic challenge assay involving brain injection of zebrafish embryos²⁷, we analyzed gene expression changes post-E. coli infection to compare different irg1 genotypes for five genes involved in inflammatory signaling (il1b, tnfa, irg1) and immune cell increase (mfap4, mpx) (Supplementary Fig. 1). No distinction was observed between wild-type and heterozygous siblings, while homozygous irg1^bcz13 mutants showed significant increases in several genes, suggesting intact irg1 function in heterozygous fish akin to wild type (Supplementary Fig. 1). We, therefore, used heterozygous GFP knock-in transgenic zebrafish, carrying only one copy of the transgene, to track normal endogenous irg1 expression in all experiments.

Previous studies using whole-mount RNA in situ hybridization (WMISH) in zebrafish demonstrated a clear bimodal distribution of irg1 gene expression in macrophages, with irg1 specifically detected in macrophages at high levels following infection or other immune challenges and normally no expression at baseline^17,24. In contrast, an irg1 fluorescent reporter using GFP, generated via Tol2 transposon-mediated transgenesis, showed expression in wild-type macrophages even at baseline, with a significant increase post-challenge in a macrophage-specific manner²². Considering the variability in number and location of GFP insertions by Tol2-mediated integration, we anticipated that the precise single-site GFP knock-in at the endogenous irg1 locus would offer a more accurate reflection of irg1 transcriptional levels. To directly address any differences between the GFP knock-in and Tol2-mediated GFP transgenesis, we generated a similar Tol2-mediated irg1 reporter line, utilizing approximately 4.8 kb of regulatory sequence (Supplementary Fig. 2). Comparing the Tol2 and KI GFP reporters, we found that the KI reporter offered lower baseline GFP expression, a larger dynamic GFP range, and no off-target expression, while there was no difference in the reference macrophage BFP reporter used alongside the Tol2 or KI construct (Supplementary Fig. 3). These results also suggest fluorescent reporters are more sensitive in revealing very low levels of irg1 that would be below detection by WMISH.

Tracking macrophage activation in auto-inflammatory nlrc3l mutants reveals a MyD88 dependence

To systematically and continuously observe macrophage state changes in a living system without the need to provoke an immune response that is inherently variable from individual to individual, we utilized the irg1-KI:GFP reporter in nlrc3l^st73 mutants. These mutants exhibit auto-inflammatory macrophages at baseline in the absence of any overt immune challenges^23,24, but were previously distinguishable only by marker analysis in fixed samples, not in living individuals. Introducing a single copy of irg1-KI:GFP into nlrc3l^st73 heterozygous fish and crossing them with non-transgenic nlrc3l^st73 heterozygous individuals showed a Mendelian inheritance pattern, with about 50% of progeny inheriting the GFP knock-in reporter as expected for a single site GFP insertion. The irg1-KI:GFP reporter enabled rapid sorting of live nlrc3l homozygous mutants under a fluorescent stereoscope based on a strong GFP induction that is indicative of activated macrophages not observed in sibling and wild-type counterparts (Fig. 2a–d). These mutant macrophages also tend to aggregate on the yolk, thereby offering an ideal location to assay differential activation (Figs. 2a and 3a, b). A complete correspondence (100%) between the irg1-KI:GFP phenotype and nlrc3l^st73 genotype was confirmed by PCR-restriction-based genotyping after sorting.

Fig. 2: irg1-KI:GFP knock-in reporter enables isolation and characterization of macrophage subpopulations based on activation states. — **Fig. 2: *irg1-KI:GFP* knock-in reporter enables isolation and characterization of macrophage subpopulations based on activation states.**

Fig. 3: Screening yolkball macrophages distinguishes macrophage states at scale using irg1-KI:GFP in zebrafish, enabling epistasis analysis that reveals a genetic interaction between nlrc3l and myd88. — Fig. 3: Screening yolkball macrophages distinguishes macrophage states at scale using *irg1-KI:GFP* in zebrafish, enabling epistasis analysis that reveals a genetic interaction between *nlrc3l* and *myd88.*

To profile macrophage states in nlrc3l mutants, we used FACS to sort macrophages based on GFP levels (high, low, negative) in irg1-KI:GFP mutants and their heterozygous and wild-type siblings (Fig. 2b, c). Flow cytometry analysis revealed an irg1-high population in mutants surpassing LPS-activated control macrophages in irg1 expression level (Fig. 2d). Bulk RNA-sequencing of sorted populations identified distinct gene sets upregulated in mutant macrophages, including immune activation, cell stress, and inflammatory response pathways (Fig. 2e–h). Sorted macrophages confirmed their identity with elevated expression of macrophage-specific markers (c1qb, c1qc, csf1ra, havcr1, irf8, lgals3bpb, marco, mfap4, mpeg1.1)^28,29, although a few markers showed inconsistent upregulation in mutant macrophages (Fig. 2g). Functional annotation highlighted increased expression of vesicle- and transport-related genes in irg1-high mutant macrophages, consistent with abnormal vesicle-filled phenotype observed in mutant macrophages with high irg1 expression (Fig. 2i), while irg1-low mutant macrophages showed enrichment in ECM and cell adhesion pathways (Fig. 2f–h). Although irg1-low mutant macrophages showed similarly low irg1 levels as control sibling macrophages, they modestly expressed pro-repair M2-like genes (arg2, irf3, ptger2a)^28,29, and significantly downregulated the mannose receptor mrc1b/cd206, a M2-associated marker^30,31. A substantial downregulation was also seen in irg1-high mutant macrophages (Fig. 2g). Additionally, irg1-low mutant cells displayed an enrichment for neutrophil-specific markers (mpx, lyz)²⁹(Fig. 2g). To directly confirm these cellular changes in vivo, we imaged double transgenic zebrafish expressing both neutrophil and irg1-KI:GFP reporters in vivo, and observed weak irg1 expression in mutant neutrophils in stark contrast to the absence of irg1 in control sibling neutrophils (Fig. 2i), suggesting that some irg1-low mutant cells are likely neutrophils exhibiting altered states. Furthermore, in vivo imaging revealed an unusually strong induction of inflammatory cytokine tnfa expression at baseline in almost all macrophages using the tnfa:GFP transcriptional reporter³², which is predominately macrophage cell-autonomous as it can be reversed by restoring macrophage nlrc3l expression (Supplementary Fig. 4). Weak tnfa in select few neutrophils was also observed in nlrc3l mutants, while no tnfa was detected in control sibling counterparts (Fig. 2i and Supplementary Fig. 4). These findings highlight substantially altered macrophage and neutrophil states in nlrc3l mutants, and similarly low irg1 immune cell expression in mutants and siblings can render significantly different cellular states.

Additionally, unsupervised hierarchical clustering³³ of cell-specific expression levels for a target gene set, followed by protein network analysis³⁴, identified significantly elevated TNF-alpha and non-canonical NF-kB pathways in nlrc3l mutant immune cells (Supplementary Fig. 5). This gene set comprised 55 genes known to be enriched in macrophages, neutrophils, or immune responses (Supplementary Fig. 5). Analysis was performed separately on sorted macrophages (high irg1-KI:GFP+ cells) and sorted neutrophils (lyz:GFP+ cells) from nlrc3l mutants compared to control sibling counterparts (Supplementary Fig. 5). These results revealed notable upregulation of inflammation-associated genes in nlrc3l mutant macrophages, particularly TNF-alpha and components of the non-canonical NF-kB pathway (nfkb2, traf2b, nik/map3k14a, relb) (Supplementary Fig. 5). Interestingly, mutant neutrophils exhibited expression of genes we found typically restricted to zebrafish macrophages (mfap4, tnfa, tlr22, tnfaip8l2b, tnfrs1b/TNFR2, traf6), suggesting a phenotypic shift resembling activated macrophages (Supplementary Fig. 5).

Given significant upregulation of TNF-alpha and NF-kB pathway genes in mutant macrophages that may account for their chronic inflammatory state, we aimed to investigate the role of the central mediator MyD88 by analyzing nlrc3l/myd88 double mutants (Fig. 3a, b, d). MyD88 is crucial for TNF-alpha production and NF-kB pathway activation via Toll-like receptors (TLRs) and interleukin-1 receptors (IL-1Rs), and for non-canonical NF-kB signaling through TNF receptor-associated factors (TRAFs)^35,36,37. In nlrc3l mutants, restoring macrophage nlrc3l expression, deleting myd88, or both, reversed the inappropriate macrophage activation, as evidenced by irg1 and tnfa downregulation, with the combination fully restoring a normal macrophage state on the yolk ball (Fig. 3d, e) and in the trunk (Supplementary Figs. 4, 6). The increase in irg1 expression in activated macrophages corresponded with the induction of macrophage tnfa, while restoring a normal macrophage state in nlrc3l mutants reduced irg1 to baseline levels (Fig. 3d, e) and eliminated tnfa expression (Supplementary Figs. 4, 6), validating irg1 and tnfa as consistent markers of macrophage activation in zebrafish, in agreement with previous reports that examined each marker individually^17,38,39. Furthermore, because several Nod-like receptors are known to interact with the adaptor protein called apoptosis-associated speck like protein containing a CARD (also known as ASC) to form cytosolic multiprotein complexes called inflammasomes, responsible for activation of inflammatory signaling, we explored whether the inappropriate macrophage activation in nlrc3l mutants was ASC-dependent. We generated nlrc3l/asc double mutants that showed no change from the nlrc3l mutant activation phenotype or impaired muscle repair after injury, thereby suggesting nlrc3l functioned independently of asc in this context (Fig. 3c and Supplementary Fig. 7a–d). Given that MyD88 mediates TLR signaling responsible for bacterial recognition and response³⁵, we evaluated macrophage activation in nlrc3l mutants under germ-free versus conventionally raised conditions (Supplementary Fig. 7e–g). Our findings showed that while commensal microbes contributed, nlrc3l mutant macrophages themselves retained substantial intrinsic activation (Supplementary Fig. 7e–g). In agreement with this, full rescue of inappropriate macrophage activation in nlrc3l mutants required both macrophage rescue and myd88 deletion, as neither intervention alone produced complete rescue consistently (Supplementary Fig. 6). This indicates contributions from both cell-intrinsic defects and likely non-cell-autonomous signals, supported by the need to delete myd88, a gene mediating TLR signaling. Taken together, these findings highlight the intricate nature of the inappropriate macrophage activation in nlrc3l mutants, where MyD88-dependent signaling, microbial cues, and internal cellular dysregulation collectively contribute to their chronic inflammatory activation state, in addition to the altered state of neutrophils. These results implicate both cell-autonomous and non-cell-autonomous mechanisms in the inappropriate immune activation.

Chronic macrophage activation impairs response to acute skeletal muscle injury

We focused on tracking macrophage states during tissue injury due to the controlled and localized perturbation injury provides, which progresses through distinct phases from inflammation to resolution, with macrophages playing a central role^6,11,40. This is particularly evident in skeletal muscle injury where macrophages are thought to transition from a pro-inflammatory to a pro-repair state in both mammals and zebrafish^1,6,41,42. Despite this, the precise cellular mechanisms by which macrophages coordinate an effective muscle injury response and repair in vivo, especially their dynamic transformations into diverse subtypes, have remained elusive through static or single-cell transcriptome analysis in previous studies lacking in vivo dynamic imaging. To address this gap, we developed an acute skeletal muscle injury model in zebrafish involving a major tail amputation (Fig. 4a). We found this model to trigger robust immune responses, where both irg1 (Fig. 4b and Supplementary Fig. 8) and tnfa (Supplementary Fig. 9) are highly induced in macrophages in wild-type zebrafish post-injury, and significant skeletal muscle repair (Fig. 5), distinct from previous zebrafish tail injury studies which focus on less disruptive tail fin truncation and regeneration without extensive muscle injury^11,40,43,44.

Fig. 4: In vivo longitudinal and time-lapse imaging tracks macrophage activation during acute injury response, revealing severe deficiencies in auto-inflammatory nlrc3l mutant. — **Fig. 4: In vivo longitudinal and time-lapse imaging tracks macrophage activation during acute injury response, revealing severe deficiencies in auto-inflammatory *nlrc3l* mutant.**

Fig. 5: Chronic inflammatory activation of macrophages impairs acute muscle injury repair, reversible by restoring wild-type macrophages or eliminating myd88. — Fig. 5: Chronic inflammatory activation of macrophages impairs acute muscle injury repair, reversible by restoring wild-type macrophages or eliminating *myd88.*

We used in vivo longitudinal and timelapse imaging to monitor individual zebrafish embryos, employing a dual labeling system that marks all macrophages and tracks their activation status (Fig. 4b and Supplementary Fig. 10). This approach revealed a consistent macrophage response to injury in wild-type zebrafish (Fig. 4b, c). Initially, macrophages are recruited to the injury site, peaking during the inflammatory phase between 24–30 h post-amputation (hpa) (Fig. 4b, c). Concurrently, there is robust expression of irg1 (Fig. 4b, c and Supplementary Fig. 8) and significant tnfa induction (Supplementary Fig. 9) in a subset of macrophages at the cut site. Timelapse imaging captured dynamic irg1 induction in macrophages from 5 to 14 h post-injury, both in those migrating to and already present at the cut site (Supplementary Fig. 10 and Supplementary Movie 1). Macrophage recruitment primarily occurred via migration outside circulation: dorsally along the dorsal longitudinal anastomotic vessel (DLAV), ventrally through the caudal vein plexus (CVP), or through the vasculature-free midline (Supplementary Fig. 10 and Supplementary Movie 1). Most macrophages at the cut site within 24 hpa exhibited intermediate to high GFP expression, indicating robust activation with elevated irg1 expression (Fig. 4b). Subsequently, macrophages migrated away from the injury site, redistributing throughout the body during immune deactivation, coinciding with tissue repair and regeneration from 48 hpa onwards (Fig. 4b and Supplementary Fig. 8). Timelapse imaging showed macrophages and neutrophils engaging in reverse migration away from the injury site, and dynamically surveying the repairing tissue, accompanied by a significant macrophage irg1 downregulation (Supplementary Movie 2). By 96 hpa, complete tissue repair was evident with the tail muscle fully healed, though regeneration to original length did not occur (Fig. 4b).

In stark contrast, chronically activated mutant macrophages maintained consistently high levels of irg1 and tnfa expression both at homeostasis and post-injury (Fig. 4c and Supplementary Figs. 9, 11, 12), yet were deficient from the initiation of the inflammatory response to the acute muscle injury (Fig. 4d–g). Although mutants had comparable macrophage numbers to siblings at baseline, they showed no significant increase post-injury, unlike controls which about tripled in number (Fig. 4d). Mutants also had elevated neutrophil counts at baseline similar to post-injury levels seen in controls (Fig. 4d). Moreover, these mutants did not form the characteristic large and dense macrophage cluster at the cut site observed at ~ 24 hpa; instead, they exhibited a void in this region where removal of injured and dying cells, and regeneration of myocytes were expected (Fig. 4e, Supplementary Figs. 11, 12, 13). These mutant macrophages interacted as smaller groups of cells that persisted even through the expected resolution phase at 72 hpa and 96 hpa (Supplementary Fig. 12), as depicted in the graphical plots comparing wild-type and mutant macrophage responses (Fig. 4c). Furthermore, while tnfa expression was restricted to macrophages at the cut site in control animals ( ~ 31% of macrophages post-injury and nearly none at baseline), nearly all macrophages in the mutants, including those in the caudal hematopoietic tissue (CHT), showed uniformly high tnfa expression ( ~ 90% at baseline and >95% post-injury) (Supplementary Fig. 9). Single cell irg1 profiling reveals distinct differences in the patterns of immune cell activation between mutant and control macrophages and neutrophils (Supplementary Fig. 11). Mutant macrophages at baseline and post-injury consistently exhibit high irg1 expression levels comparable to or generally higher than activated control macrophages post-injury (Fig. 4d and Supplementary Fig. 11). Additionally, a small fraction of neutrophils in mutants show an atypical induction of irg1 and tnfa expressions at baseline compared with control neutrophils which express neither gene at baseline (Fig. 2i and Supplementary Figs. 4, 11). After injury, irg1 expression is detectable in a small number of neutrophils—on average 7.6% in siblings and 10.7% in nlrc3l mutants at 20 hpa (and 4.3% at baseline in mutants, none in siblings) (Supplementary Fig. 11c). These data show that most macrophages are bright irg1-KI:GFP+ after injury and in baseline mutants, while neutrophils rarely cross the detectable threshold at 30 A.U. (Supplementary Fig. 11b, c). Thus, despite low-level expression in a few neutrophils, irg1-KI:GFP remains a reliable reporter for tracking macrophage-specific responses. Overall, when detectable in injured controls or in mutants, whether at baseline or post-injury, neutrophil irg1 or tnfa expression is significantly weaker and limited to a select number of neutrophils, in contrast to the broad and strong induction of these genes in activated macrophages (Fig. 4d and Supplementary Figs. 9, 11). This data is also consistent with the previously discussed bulk RNA-seq and imaging results in Fig. 2g, i where irg1-low cells from nlrc3l mutants include that of the neutrophil cell type besides macrophages.

Most notably, during the inflammatory phase at ~ 24 hpa, timelapse imaging revealed distinct cellular behavioral differences among activated macrophages at the cut site in wild-type and control siblings post-injury (Fig. 4e–g and Supplementary Movies 3 and 4). Macrophages at the injury site that displayed dynamically irregular and migratory cell shapes with high mobility (increased speed and distance traveled, Fig. 4f) are categorized as subtype 1 “mobile”. Other macrophages in close proximity to the repairing tail fin with increased vesicular content and higher phagocytic index (cell-to-cytosol volume ratio), indicating large intracellular vesicles such as phagosomes or endosomes (Fig. 4f) are categorized as subtype 2 “cluster”. The final macrophage population at the injury site exhibited largely stationary behavior with an elongated and columnar morphology oriented along the muscle fibers that are aligned with the body anterior-posterior axis is referred to as subtype 3 “encasing” herein (Fig. 4e–g, Supplementary Fig. 13). Subtype 3 macrophages encircle myocytes and take on a muscle-cell-like shape, showing the largest cell size and greater overall sphericity, particularly in the lowest 10^th percentile of the population distribution, indicating less variability in cell shape than the other subtypes (Fig. 4f, g, Supplementary Fig. 13). In mutants, our dynamic tracking of the injury response indicated an apparent absence of subtypes 2 and 3 macrophages, which were consistently observed in all control animals (n = 4 for each genotype, Fig. 4e–g and Supplementary Movies 3–6). These observations underscore the behavioral and morphological cellular diversity that may define distinct macrophage functions post-injury, which are compromised in chronically activated mutants.

Chronic macrophage activation linked with severe downregulation of mannose receptor causes unresolved muscle injury, reversible by reinstating a normal state

To delve into the functionally altered macrophages, we explored whether injured skeletal muscle in nlrc3l mutants could effectively repair and regenerate. To visualize muscle tissues, we used muscle actin antibody or actin-binding phalloidin staining, enabling co-localization of macrophages with injured or repairing myocytes (Fig. 5 and Supplementary Fig. 13). Detailed examination of macrophage-myocyte interactions identified three functional macrophage subtypes corresponding to those previously discussed: 1) mobile surveying cells, 2) clustering cells that appear likely phagocytic for clearing injured and dying cells, and 3) elongated stationary cells encasing injured or repairing muscle cells (Supplementary Fig. 13 and Supplementary Movies 3 and 4). Immediately after tail amputation (within 1 hpa), muscle injury was comparable across all genotypes (Supplementary Fig. 13). However, by 24 hpa, mutant macrophages inadequately infiltrated the cut site, although recruitment of neutrophils appeared normal, lacking both clustering subtype 2 and encasing subtype 3 (Fig. 5a, Supplementary Fig. 13 and Supplementary Movie 4). In contrast, sibling controls displayed macrophages that formed clusters appearing likely phagocytic (subtype 2) in the most severely damaged region of the injury site, and encasement of seemingly intact muscle fibers by macrophages (subtype 3) in the adjacent area (Supplementary Fig. 13). By 48 hpa, control siblings had mostly cleared injured cells and debris, resulting in a clean-cut tail end with a well-defined tissue boundary, and only occasional, if any, non-intact muscle fibers remaining (Fig. 5c, d and Supplementary 13). In nlrc3l mutants, although more macrophages were recruited to the injury site, they exhibited limited engulfment of damaged muscle cells, leaving significant cellular damage and debris compared to controls at the injury site (Fig. 5c–f). By 96 hpa, control siblings showed fully repaired muscle with regrowth at the tail tip, forming a pointed end (Fig. 5d-f). In contrast, mutants retained unresolved tissue and failed to form the expected muscle tip at the tail end (Fig. 5c–f). This coincided with deficient activation of satellite cells^12,45 critical for regenerating myocytes during active muscle repair^1,46, as evidenced by a significant reduction in the emergence of Pax7+ satellite cells^46,47 at the injury site observed at both 24 hpa and 48 hpa in nlrc3l mutants (Fig. 5a). These findings underscore that the impaired initial inflammatory response by macrophages in these mutants significantly hindered the ability of macrophages to phagocytose and support injured muscle repair and regeneration.

The loss of essential macrophage functional subtypes 2 and 3, critical for clearance of cellular damage and muscle support, in the chronically activated mutant macrophages prompted us to investigate whether similar outcomes could result from macrophage ablation. To address this, we utilized irf8 mutants⁴⁸, known to lack macrophages in steady-state zebrafish embryos at stages concurrent with our injury study. Interestingly, although initially devoid of macrophages, irf8 mutants after injury showed the presence of macrophages (Supplementary Fig. 14), likely due to injury-triggered hematopoiesis that is irf8 independent. Despite this partial macrophage recovery, muscle injury in macrophage-depleted irf8 mutants exhibited persistent impairment in repair and regeneration from 48 to 96 h post-amputation, resembling the phenotype of chronically activated nlrc3l mutants (Supplementary Fig. 14). These findings underscore that both chronic activation and depletion of macrophages lead to a similar impaired muscle repair outcome, suggesting a comparable loss of macrophage function during acute injury. To directly assess whether the impaired muscle injury repair was due to altered mutant macrophages, we restored normal macrophages devoid of inflammatory signatures by rescuing macrophage wild-type nlrc3l expression or deleting myd88 signaling. Both approaches effectively reinstated a normal macrophage state, reducing or eliminating irg1 and tnfa inductions (Fig. 3, and Supplementary Figs. 4, 6), and significantly facilitated recovery of injured muscle by 96 hpa to a level indistinguishable to controls (Fig. 5c–f). By contrast, genetic deletion of an inflammasome adaptor protein asc failed to restore a normal macrophage state or muscle injury repair in nlrc3l mutants (Fig. 3c, and Supplementary Fig. 7b–d), supporting that their chronic activation is asc-independent.

To investigate mechanisms underlying functional deficiencies of chronically activated mutant macrophages post-injury, we performed bulk RNA-seq on FACS-sorted macrophages from uninjured and 24 hpa zebrafish embryos. We identified several injury-induced upregulated immune cell genes, including significantly elevated matrix metalloproteinase-9 (mmp9), known as an inflammatory proteinase upregulated in various diseased and inflamed tissues that can facilitate tissue remodeling, cell migration and inflammatory signaling through cleaving and activating cytokines and other secreted proteins^49,50,51, and relb, a non-canonical NF-kB pathway transcription factor⁵² (Supplementary Fig. 15). qPCR analysis of additional animals confirmed excessive mmp9 expression in nlrc3l mutants at baseline and post-injury, implicating dysregulated remodeling activities that may potentially hinder macrophage ability to respond and phagocytose injured cells (Supplementary Fig. 15). Additionally, relb was significantly increased in nlrc3l mutants at baseline and even higher post-injury as shown by the qPCR analysis, suggesting sustained NF-kB activation that may compromise macrophage function in injury response compared to controls (Supplementary Fig. 15).

Furthermore, mrc1b, a gene encoding the mannose receptor (also known as CD206) of the C-type lectin receptor family crucial for phagocytosis, was shown earlier by RNA-seq analysis as significantly downregulated in baseline mutant macrophages (Fig. 2g), but not its paralog mrc1a (Fig. 6a). To assess the normal post-injury expression of mrc1b, we performed qPCR analysis and HCR in situ hybridization, revealing strong expression in injury-responsive macrophages clustered at the cut site, corresponding to subtypes 2 and 3 in control siblings, and in the ventral region appearing in the caudal hematopoietic tissue (CHT)(Fig. 6b–e). qPCR and HCR validation confirmed downregulation of mrc1b in nlrc3l mutant macrophages at both baseline and post-injury (Fig. 6b–e). This aligns with the loss of functional subtypes 2 and 3, linked to phagocytosis and muscle enclosure, respectively (Supplementary Fig. 13).

Fig. 6: Macrophage subtype loss and impaired muscle repair in chronically activated nlrc3l mutants are linked to mrc1b downregulation, reversible by myd88 deletion. — **Fig. 6: Macrophage subtype loss and impaired muscle repair in chronically activated *nlrc3l* mutants are linked to *mrc1b* downregulation, reversible by *myd88* deletion.**

The excessive upregulation of mmp9 and relb, combined with a severe loss of mrc1b/cd206 in nlrc3l mutant macrophages at baseline and post-injury, elucidates persistent molecular changes that likely impede critical macrophage functions for facilitating tissue repair processes. In support of this, knockdown of mrc1b using transient CRISPR gRNA injections with Cas9 was validated using T7 assay and sequencing (Supplementary Fig. 16); this resulted in similar reduction in cluster and muscle-encasing subtypes (2 and 3)(Supplementary Fig. 16) and repair defect as seen in chronically activated nlrc3l mutants, suggesting an essential role for mrc1b in muscle repair (Fig. 6f). The muscle repair defect was also confirmed in the stable mrc1b^sa18640 mutant zebrafish (Fig. 6g and Supplementary Fig. 16). These mechanistic alterations may be consequences of the chronic inflammatory activation state of nlrc3l mutant macrophages that hinders their ability to perform essential tissue repair functions.

Furthermore, myd88 knockout in nlrc3l mutants using double nlrc3l/myd88 mutants led to recovery of mrc1b expression in macrophages post-injury (Fig. 6h, i), thereby restoring a more homeostatic phenotype. These results are consistent with myd88-dependent signaling contributing to dysregulated macrophage state and significant macrophage mrc1b downregulation in nlrc3l mutants. To further explore the generalizability to additional paradigms of chronic macrophage activation, we developed a model based on persistent systemic E. coli infection (Supplementary Fig. 17a–c). We showed that similarly to nlrc3l mutants, the chronic macrophage activation caused by infection impaired normal skeletal muscle repair (Supplementary Fig. 17d, e), and was associated with a mrc1b downregulation and reduction of both clustering and encasing macrophage subtypes (Supplementary Fig. 17f–k).

Chronic macrophage activation erodes cellular heterogeneity, driving a universal pro-inflammatory program while still expressing repair-promoting genes

To better define the molecular impact of chronic macrophage activation and distinguish macrophage subtypes during muscle injury repair, we performed single-cell RNA-seq (scRNA-seq) on ~ 10,000 irg1⁺ cells from zebrafish at baseline (uncut), inflammatory phase (24 hpa), and resolution phase (48 hpa), as well as from nlrc3l mutants at 24 hpa (Fig. 7a and Supplementary Fig. 18). Seurat clustering of quality-filtered cells (n = 9603 comprised of 1174 cells (uncut), 3557 cells (24 hpa), 1847 cells (48 hpa), and 3025 nlrc3l mutant cells (24 hpa)) revealed 16 clusters spanning macrophages and neutrophils, with the vast majority of cells expressing canonical macrophage markers (mpeg1.1, csf1ra, mpeg1:BFP)(Fig. 7b–f, Supplementary Fig. 18). Comparing cell distributions across conditions, baseline and resolution (uncut and 48 hpa) cells showed substantial overlap in UMAP space, indicating similar cell states (Fig. 7d). In contrast, cells from 24 hpa (control and nlrc3l mutant) were more divergent—both from other conditions and from each other—with nlrc3l mutant cells preferentially occupying distinct but confined clusters (notably 1, 4, and 11)(Fig. 7d and Supplementary Fig. 18, 19). Neutrophils were identified as clusters 4, 11, and 12, marked by mpx and lyz (Fig. 7f and Supplementary Fig. 20), confirming prior observations of irg1 expression in some neutrophils only after injury especially in nlrc3l mutants (Fig. 2i and Supplementary Fig. 11). The remaining clusters (10, 13–15) were sparse and uncharacterized. The skewed clustering pattern in mutant cells, compared to the broader distribution of control cells, suggests a loss of macrophage heterogeneity under chronic activation (Fig. 7d and Supplementary Fig. 18).

Fig. 7: Single-cell profiling of irg1+ cells reveals that chronic inflammatory conditioning amplified macrophage divergence post-injury, causing subtype loss, restricted repair programs, mrc1b downregulation, and altered neutrophil states in nlrc3l mutants. — Fig. 7: Single-cell profiling of *irg1*+ cells reveals that chronic inflammatory conditioning amplified macrophage divergence post-injury, causing subtype loss, restricted repair programs, *mrc1b* downregulation, and altered neutrophil states in *nlrc3l* mutants.

Using prior knowledge from in vivo imaging of macrophages and select markers, we identified Seurat clusters corresponding to previously characterized macrophage subtypes: mobile (subtype 1), clustering (subtype 2), and muscle-encasing (subtype 3)(Figs. 4e–g and 7d and Supplementary Fig. 19). Subtype 1 cells, known to express high levels of irg1 and tnfa, were enriched in cluster 1—predominantly composed of nlrc3l mutant and 24 hpa control cells (Fig. 7d and Supplementary Fig. 18d, e). Subtype 2, which appears only in 24 and 48 hpa control macrophages with strong irg1 expression, was identified in cluster 3 (Fig. 7d and Supplementary Fig. 18d, e). Subtype 3, found exclusively in 24 hpa controls with muscle-encasing macrophages, was mapped to cluster 9, distinguished by high mrc1b expression and predominance of 24 hpa control cells (Fig. 7d and Supplementary Fig. 18d, e). To molecularly profile these subtypes, we performed pseudobulk analysis comparing 24 hpa nlrc3l mutant and control macrophages. Clusters of interest (macrophage clusters 0, 1, 3, 4, 5, 9; neutrophil clusters 4, 11, 12) revealed key transcriptional signatures (Fig. 7e–i). Cluster 3 (subtype 2) showed enrichment for repair and homeostatic genes, particularly those involved in lipid metabolism (fabp4a, apoeb, apoc1) and phagocytosis/ECM remodeling (mertka, npc2.1, npc2.2, ctsk, ctsl.1, ifi30a) with moderate levels of pro-inflammatory genes (including tnfa and nfkb2). Cluster 9 (subtype 3) was characterized by upregulation of TGF-beta signaling (tgfbr1a, tgfbi, tgfbrap1) and endocytic pathways (cd302)(Fig. 7e). Overall, chronically activated nlrc3l mutant cells upregulated pro-inflammatory genes and negative regulators of inflammation— including TNF-alpha, interleukins, cytokines, type II interferon and NF-kB components (tnfa, ifngr1, il6, tgfb1b, il10ra, il4, nfkb2, traf2b, nik/map3k14a), while downregulating TGF-beta pathway genes that were normally upregulated in 24 hpa control macrophages (Fig. 7g–i). This shift points to a dysregulated and uniformly pro-inflammatory state in chronically activated mutant macrophages, consistent with and extending insights from bulk RNA-seq results (Fig. 2g).

To isolate macrophage-specific changes, we further subclustered the dataset using cells in clusters 0-3 and 5-9 (Fig. 7b), excluding neutrophils (Fig. 7j–m, Supplementary Fig. 19). UMAP projections of macrophage-only data revealed that nlrc3l mutant macrophages were concentrated in specific subclusters (notably 1 and 5), while control macrophages at various timepoints were broadly distributed (Fig. 7j, k). Baseline (control uncut) and 48 hpa control macrophages again showed close overlap, reinforcing their shared transcriptional identity. Despite high levels of pro-inflammatory genes (acod1, tnfa, cxcl11.1, mmp9), chronically activated mutant macrophages still express many repair-associated genes (mrc1b, ctsk, apoeb, fabp4a, tgfbi, sb:cb81 also known as ZFP36L1-like) and alternative activation markers (arg2, il4), albeit at lower levels (Fig. 7l, m). Among these, mrc1b was the most strongly downregulated gene in mutant macrophages at 24 hpa (Fig. 7l and Supplementary Fig. 19g). Although thought to support phagocytosis, mrc1b was unexpectedly most highly expressed in homeostatic and resolving macrophages (Fig. 7l, Supplementary Fig. 19g), rather than during the peak phagocytic period in the initial inflammatory phase of the injury response. Its strong suppression in chronically activated macrophages suggests that macrophage activation may negatively regulate mrc1b transcription and that its functional role in macrophage biology is more complex and not fully understood.

Moreover, a substantial proportion of mutant macrophages (ranging from ~10% to over 50%) co-expressed classical M1 and M2 markers, a phenomenon we termed hybrid M1/M2 states (Fig. 8a, Supplementary Fig. 19). These hybrid cells were less than 10% of the control 24 hpa macrophages, indicating that while hybrid states occur under normal conditions, chronic activation significantly enhances this mixed inflammatory/repair phenotype (Fig. 8a, Supplementary Fig. 19). Although gene markers corresponding to subtypes 2 (based on cluster 3 markers, apoeb + /ctsk + ) and 3 (based on cluster 9 markers tgfbi + /sb:cb81 + ) were present in nlrc3l mutant macrophages, they were significantly underrepresented and localized to regions with high expression of pro-inflammatory activation genes like irg1 and tnfa (Figs. 7l, m and 8a). This further supports the conclusion that chronic macrophage activation skews the transcriptional landscape toward a uniform, hybrid inflammatory state while eroding the normal cellular diversity required for balanced muscle repair.

Fig. 8: Chronic activation of macrophages in nlrc3l mutants restricts macrophage heterogeneity, favoring hybrid M1/M2 states, and impairs subtype specification and cathepsin K accumulation. — **Fig. 8: Chronic activation of macrophages in *nlrc3l* mutants restricts macrophage heterogeneity, favoring hybrid M1/M2 states, and impairs subtype specification and cathepsin K accumulation.**

Although some irg1⁺ cells post-injury were identified as neutrophils (clusters 4,11, and 12)(Fig. 7f), especially in nlrc3l mutants, their EGFP transcript counts were tens to hundreds times lower than in macrophages (Supplementary Fig. 19h), consistent with low reporter levels seen by in vivo imaging (Fig. 2i and Supplementary Fig. 11). Subclustering of neutrophils from scRNA-seq showed that most of these cells were from 24 hpa nlrc3l mutants and exhibited altered states, expressing both M1 and M2 markers (e.g., acod1, tnfa, mmp9, arg2)(Supplementary Fig. 20), consistent with bulk RNA-seq results showing phenotypic shift resembling activated macrophages (Fig. 2g and Supplementary Fig. 15). Overall, irg1-KI:GFP reliably tracks macrophages at baseline, when neutrophils do not express the reporter, and during activation, when macrophages express the reporter at a much higher intensity and abundance, clearly distinguishing them from neutrophils under perturbation.

Our scRNA-seq analysis identified an endolysosomal protease cathepsin K (ctsk)⁵³ as a key marker highly upregulated in clustering subtype 2 macrophages during the initial inflammatory phase of a muscle injury response at 24 hpa (Fig. 7m). To validate this, we conducted in vivo experiments showing that cathepsin K (CTSK) protein was not only abundant in clustering macrophages, as predicted, but enriched in muscle-encasing macrophages, forming large dense cytosolic accumulations (Fig. 8b), suggesting intracellular collagen and protein degradation^53,54 as a critical function shared by these subtypes at 24 hpa. Strikingly, these densely packed CTSK⁺ domains were entirely absent in chronically activated nlrc3l mutant macrophages post-injury (Fig. 8b), indicating a profound loss of a hallmark protease. During injury response, CTSK⁺ cells co-expressed tnfa, consistent with a hybrid reparative-inflammatory state in wild-type and control macrophages as suggested by the scRNA-seq data (Fig. 8a and Supplementary Figs. 18, 19). While mutant macrophages co-expressed tnfa and ctsk at the transcript level, ctsk expression was markedly reduced transcriptionally, and CTSK protein was detectably low (Fig. 8b and Supplementary Figs. 18, 19).

Overall, single-cell transcriptomic analysis revealed that chronic macrophage activation due to nlrc3l deletion reduced cellular heterogeneity, reinforced a dominant pro-inflammatory transcriptional program, and promoted hybrid M1/M2 states co-expressing inflammatory and repair-associated markers. These hybrid states, less frequent in control macrophages ( < 10% at 24 hpa), were significantly enriched in mutants. At least three macrophage subpopulations were identified, including the pro-inflammatory subtype 1 marked by high irg1 and tnfa, and two pro-repair subtypes: subtype 2, characterized by cholesterol efflux (fabp4a, apoeb, apoc1) and phagocytosis-related genes (mertka, npc2.2, ctsk, ctsl.1, ifi30a), and subtype 3, marked by TGF-β signaling (tgfbr1a, tgfbr1b, tgfbi). Both reparative subtypes were significantly depleted in mutants. Thus, chronic macrophage activation impairs muscle injury repair not by enhancing or diversifying inflammatory functions, but by disrupting reparative programs. Furthermore, pro-inflammatory gene expression underlies the transcriptome of these macrophages, constraining their plasticity despite showing expression of pro-repair pathways, albeit at a reduced level. Together, these results provide new mechanistic insights into how chronic inflammation disrupts macrophage-mediated repair by downregulating mrc1b/cd206, blocking cathepsin-K-rich degradative compartments, and promoting functionally deficient hybrid states, and also establish cathepsin K as a validated marker of reparative macrophage identity in zebrafish.

Discussion

Acute inflammation is essential for repairing skeletal muscle damage from trauma, infections, or toxins^1,2; however, the duration and intensity of the inflammatory response after the injury significantly affect the outcome. Chronic inflammation, characterized by a prolonged response, is associated with impaired muscle regeneration in mammals that can lead to fibrosis, myofiber degeneration, and persistent inflammation^1,2,6. Although the immune system, particularly macrophages, largely determines the nature of the inflammatory response to injury, the mechanisms by which macrophages regulate that process remain unclear.

In this study, we show that a deleterious nlrc3l mutation in zebrafish leads to chronic inflammatory activation of macrophages throughout the injury time course, marked by persistently high expression of a metabolic marker of macrophage activation irg1 and the pro-inflammatory cytokine tnfa in addition to a significantly upregulated mmp9 expression, a known secreted inflammatory proteinase upregulated in various chronic inflammatory diseases^49,50,55. The significant upregulation of mmp9 in nlrc3l mutant macrophages at baseline and after major tail cut injury (Supplementary Figs. 15 and 19) may reflect its role in sustaining the inflammatory environment, as it has been proposed to activate and regulate levels of cytokines, adhesion molecules, and growth factors, possibly activating TNF-alpha and IL-1B proteins^49,50,51. This contrasts sharply with the significant reduction or elimination of immune activation marked by high irg1 expression after the acute inflammatory phase by 72 hpa in wild-type macrophages (Fig. 4b). Using an endogenous reporter of macrophage activation, based on a GFP knock-in allele at the irg1 locus, we demonstrate that chronically activated macrophages impair muscle repair following tail amputation. Restoring normal macrophage function, either by reintroducing wild-type nlrc3l or inhibiting myd88-dependent inflammatory signaling, rescues muscle repair defects. These findings highlight the importance of properly controlled acute inflammation during injury response to prevent collateral and long-term muscle damage. This requirement for a balanced inflammatory response appears to be evolutionarily conserved, as we show in this study that prolonged excessive inflammation hinders skeletal muscle injury repair in zebrafish, mirroring the findings in mammals from both in vitro and in vivo methods^42,56,57.

The impact of prolonged inflammatory activation on macrophage heterogeneity and function during muscle repair remains poorly defined. Previous studies in rodents and human cells have shown that pro-inflammatory M1 macrophages promote satellite cell activation and myoblast proliferation, but inhibit differentiation, while pro-repair or anti-inflammatory M2 macrophages facilitate myoblast differentiation^7,8. Consistent with macrophages having an essential role in muscle repair, we show that depletion of macrophages in zebrafish, irrespective of state, impairs muscle repair. Similarly, depletion of mammalian monocytes/macrophages, such as by Ccr2 deletion, irradiation, diphtheria toxin, clodronate, or pharmacological inhibitors of migration, also impede normal muscle repair^8,58,59,60, highlighting a conserved and essential role for macrophages in muscle injury repair. While pro-inflammatory macrophages have been shown to promote myoblast expansion in vitro^8,61, our data suggest that chronically activated macrophages inhibit satellite cell activation (Fig. 5a), thereby preventing enough myoblasts to form and facilitate an effective muscle regeneration. Chronic activation may cause a significant imbalance in macrophage states, skewing heavily toward a chronically pro-inflammatory profile, but yet not functionally equivalent to the classic pro-inflammatory state. Since inflammatory myopathies can involve persistent M1 macrophages and elevated pro-inflammatory cytokines^62,63, our nlrc3l mutant provides a useful paradigm for exploring relevant mechanisms and potential therapeutic strategies. Results from this study raise the possibility that non-functional macrophages may be a driver of these disorders.

Through dynamic in vivo imaging of normal macrophage behavior following acute muscle injury, we observed that during the peak inflammatory phase (around timepoint 24 hpa), most macrophages at the injury site expressed high levels of M1 markers such as irg1 and tnfa. Interestingly, a subset of these macrophages concurrently expressed mrc1b/cd206, a marker typically associated with M2 macrophages⁶⁴, and this marker was most enriched in macrophages located in the muscle region where muscle cells are phagocytosed or encased by macrophages to facilitate muscle regeneration (Fig. 6a–d). This suggests that injury-responding macrophages initially categorized as pro-inflammatory based on markers expression can simultaneously adopt pro-repair functions, such as subtype 2 (phagocytosis) and subtype 3 (muscle-encasing), as shown in Fig. 8 and Supplementary Fig. 19. We found that macrophages display substantial functional plasticity, showing a dynamic mix of M1- and M2-like behaviors early on during the initial inflammatory phase of the injury response. They express markers and perform functions characteristic of both pro-inflammatory (M1-like) and pro-repair (M2-like) states, suggesting a broader overlap of activation states and challenging the conventional view of distinct macrophage states (Figs. 7 and 8). Single-cell analysis of normal control and chronically activated nlrc3l mutant macrophages indicate a small percentage ( < 10%) of normal macrophages during initial inflammatory phase of the injury response co-express alternative activation or pro-repair genes (such as mrc1b and arg2) with pro-inflammatory M1 marker tnfa, while this is significantly increased to as much as 50% in chronically activated nlrc3l mutant macrophages (Fig. 8a and Supplementary Fig. 19). Notably, the two functional subtypes 2 and 3 were largely absent in auto-inflammatory nlrc3l mutants based on multiple modes of analyses (in vivo imaging (Fig. 4 and Supplementary Fig. 13), bulk and single-cell RNAseq (Figs. 2, 7, and Supplementary Figs. 18, 19), where macrophages appear locked in a persistent M1-like pro-inflammatory state, losing the ability to concurrently adopt M2-like pro-repair functions although expressing low levels of reparative program genes.

An important question is how the functional macrophage subtypes 2 and 3, categorized as “clustering” and “muscle-encasing”, respectively, can be distinguished at the molecular level and by their specific roles during injury. Addressing this will be crucial for elucidating the mechanisms that enable macrophages to perform the diverse array of essential functions in muscle injury repair. Insights from our scRNA-seq analysis highlighted distinct pathways elevated in the two pro-repair subtypes, whereby cholesterol efflux (fabp4a, apoeb, apoc1) and phagocytosis-related genes (mertka, npc2.2, ctsk, ctsl.1, ifi30a) primarily marked the clustering subtype 2 macrophages, and response to TGF-β signaling (tgfbr1a, tgfbr1b, tgfbi) is most highly induced in muscle-encasing subtype 3 macrophages (Fig. 7m and Supplementary Fig. 19). In vivo analysis of the subtype 2 marker ctsk, encoding the protease cathepsin K (CTSK), revealed a marked difference between control and nlrc3l mutant macrophages. In controls, a subset of both clustering and muscle-encasing macrophages showed high intracellular accumulation of the CTSK protein during the inflammatory phase of injury (Fig. 8b), a feature absent in chronically activated nlrc3l mutants. This CTSK accumulation, consistent across all wild-type and control animals analyzed, marks normal reparative macrophage activation and suggests an active macrophage role in collagen and matrix degradation. ctsk deficiency is known to impair breakdown of cholesterol and extracellular matrix, leading to foam-like macrophages in atherosclerosis⁶⁵, which the vesicle-engorged nlrc3l mutant macrophages may phenotypically mimic given their shared ctsk loss. While ctsk is best known as a cysteine protease involved in collagen degradation during osteoclast-mediated bone resorption, it is also upregulated under pathological and inflammatory conditions^53,66, including in epithelioid cells and multinucleated giant cells of macrophage origin⁶⁷, tumor-associated M2 macrophages⁶⁸, and human atherosclerotic plaques^65,69. Here, by contrast, we demonstrate that upregulation and accumulation of cathepsin K is part of a normal process in macrophage activation during tissue injury response, one that fails to occur in chronically activated nlrc3l mutants. This loss of functional plasticity mirrors the impaired muscle repair seen with macrophage depletion.

Our zebrafish model of acute muscle injury, induced by major tail amputation at 2 days post-fertilization, enables the targeted study of recruited macrophages and their interactions prior to the establishment of muscle-resident macrophages, offering relevant insights into mammalian inflammatory monocyte mechanisms^1,6,70. Although neutrophils are generally first to be recruited, macrophages rapidly appear at the cut site by 5 hpa, becoming the primary source of tnfa expression and necrotic cell clearance (Fig. 4). This suggests zebrafish macrophages implement functions typically attributed to mammalian neutrophils and monocytes recruited to the injury site. In nlrc3l mutants, neutrophils exhibit unexpected changes, including increased cell numbers and expression of the transcriptional indicators of inflammation (irg1, tnfa) typically confined to macrophages in zebrafish^17,39,40, indicating possible compensatory mechanisms when macrophages are dysfunctional. Interestingly, a previous study has described muscle lesions shielded by macrophages to protect them from further damage due to inflammatory neutrophils⁷¹. This phenomenon may resemble the muscle-encasing macrophages (subtype 3) we observed in zebrafish, although at a different spatial scale. In zebrafish post-injury, individual macrophages are seen wrapping around single myocytes rather than a cluster of macrophages enveloping an area of injured myocytes. The role and significance of neutrophil-macrophage interplay during the various phases of muscle injury response and resolution remain areas of ongoing investigation.

The significant loss of the mannose receptor mrc1b (an ortholog of human Mrc1) in chronically activated macrophages of nlrc3l mutants aligns with the observed loss of phagocytic and engulfment capabilities in these cells post-injury, offering a possible molecular explanation for the unresolved muscle injury. Furthermore, our study shows that chronic activation disrupts macrophage subtype dynamics, leading to the loss of mrc1b-expressing reparative macrophage subtypes during injury response in both genetic and persistent infection models, highlighting a broad link between chronic macrophage activation and inappropriate Mrc1 downregulation, with important implications for understanding the mechanism impeding inflammatory macrophages from supporting tissue repair. Consistent with this, a prior scRNA-seq study showed that Toxoplasma gondii infection-induced chronic inflammation impeded muscle repair following cardiotoxin injury in mice, which coincided with a loss of Mrc1+ macrophages, a subset typically enriched post-injury and suggested to facilitate repair⁴². The scRNA-seq analysis also identified macrophages co-expressing pro-repair Mrc1 and pro-inflammatory markers early in injury⁴², but their functions remain unclear, and validation of the co-expression of markers was lacking in vivo. Together, these findings indicate a possible conserved mechanism in fish and mammals by which chronic inflammation downregulates the mannose Mrc1 receptor expression in macrophages and thereby causes impaired muscle repair.

Mrc1 (also known as CD206) is a transmembrane glycoprotein belonging to the C-type lectin family, primarily expressed by tissue-resident macrophages, dendritic cells, and some lymphatic and endothelial cells^64,72. It is also upregulated in tumor-associated macrophages and during inflammatory responses⁶⁴. Mrc1 functions as a scavenger receptor on macrophages, especially within M2-like pro-repair subsets, binding to endogenous molecules (including lysosomal hydrolases and collagens), and microbial components^64,72. In zebrafish, mrc1a and mrc1b are orthologs of human Mrc1, with mrc1a mainly expressed in lymphatic endothelial cells^73,74 and weakly in some microglia^75,76, but mrc1b is much less characterized. We show that mrc1b is highly expressed in macrophages clustered at the muscle injury site corresponding to the clustering (likely phagocytic) and muscle-encasing macrophage subtypes 2 and 3, and its knockdown impairs muscle repair similarly to macrophage depletion (in irf8 mutants) or chronic activation (in nlrc3l mutants), indicating its essential role in regeneration. Our findings suggest possible sub-functionalization between the two paralogs, with the role of mammalian Mrc1 in macrophages predominantly fulfilled by mrc1b in zebrafish. This interpretation is also supported by previous scRNA-seq data, which identified specifically mrc1b expression in a subset of tissue-resident macrophages in adult zebrafish²⁸.

While previous studies have associated Mrc1/CD206 positive macrophages with wound repair^77,78, our single-cell transcriptomic analysis from isolating macrophages at different timepoints of the injury response suggests that mrc1b dynamically marks macrophage functional states, and its unexpected high expression at baseline rather than during peak phagocytic activity of the injury response provides evidence that its in vivo function is still incompletely understood (Fig. 7l, Supplementary Fig. 19g). Therefore, our findings reveal a role of mrc1b in distinguishing macrophage subtypes during injury response and tissue repair in zebrafish, where it may mediate macrophage functions beyond clearance and repair of injured tissue, such as differentiation of reparative and inflammatory macrophage subtypes (schematized in Fig. 6j). While the exact molecular mechanism regulating mrc1b remains unclear, our findings provide a foundation for future studies into how chronic inflammatory activation alters mrc1b expression in zebrafish. Moreover, depletion of MyD88 signaling partially restores mrc1b levels in nlrc3l mutants post-injury, suggesting that target genes of the NF-kB pathway downstream of MyD88 may repress mrc1b transcription directly or indirectly in zebrafish.

Integrating a zebrafish GFP knock-in irg1 reporter for tracking activated macrophages with acute muscle injury analysis in the nlrc3l mutant background offers a unique genetic paradigm amenable to in vivo single-cell and whole-body imaging for interrogating how chronic inflammation affects tissue injury and repair. While existing reporters like tnfa:GFP³² label macrophage activation, they reflect distinct pathways. tnfa, encoding a pro-inflammatory cytokine, marks inflammatory signaling, while irg1, a mitochondrial enzyme, tracks metabolic reprogramming. Both show partial overlap during peak inflammation ( ~ 24 hpa; Fig. 7l–m, Supplementary Figs. 18 and 19), but irg1 expression persists into resolution, indicating broader, sustained activation. Consistent with this, our scRNA-seq data reveal that irg1 is more broadly expressed post-injury than tnfa (Fig. 7, Supplementary Figs. 18 and 19). irg1-KI:GFP, generated via targeted knock-in, ensures consistent, quantifiable expression across animals. In contrast, the expression of tnfa:GFP, inserted via transposon, varies with copy number and integration site. Moreover, tnfa:GFP is expressed in non-immune cells, including CNS and PNS neurons⁷⁹ (Supplementary Figs. 4, 9), complicating immune-specific analysis, such as by FACS and transcriptomics. irg1-KI:GFP, by contrast, is predominantly macrophage-specific, where its upregulation marks a shift in activation state without detectable expression in non-immune cells. Thus, irg1-KI:GFP offers a stable, specific, and quantitative in vivo tool for tracking immune activation and metabolic state unmatched by existing reporters.

We show that chronic inflammation diminishes macrophage functional plasticity, inhibiting their ability to clear damaged tissue and promote regeneration (Fig. 6j). These findings highlight the importance of maintaining macrophage plasticity for effective tissue repair that is associated with a normal level of Mrc1. While Mrc1 has been a potential therapeutic target for delivering small molecules and mannosylated cargo into macrophages to treat cancer, infectious diseases, and other conditions^64,80, this study implicates that reconstituting Mrc1 in macrophages may be equally useful for chronic inflammatory conditions associated with deficient or non-functional macrophages. Our study demonstrates that chronic activation restricts macrophage cell heterogeneity and disrupts subtype dynamics, leading to loss of mrc1b-expressing clustering and encasing macrophage subtypes during injury responses in both genetic and infection models of chronic inflammation. It also results in the loss of high-density intracellular cathepsin K accumulation typically seen in reparative macrophages, suggesting deficiencies in intracellular protein degradation. These findings reveal that chronic activation suppresses mrc1b expression in a myd88-dependent manner and inhibits cathepsin K from accumulating at a high concentration, offering new insights into how chronic inflammation reprograms macrophages to impair their ability to support tissue repair.

Methods

Zebrafish transgenic and mutant lines

Embryos from wild-type, mutant, and transgenic backgrounds were derived from: nlrc3^st73²³, irf8^st95⁴⁸, myd88^b1358⁸¹, irg1/acod1^bcz13 (this study), asc^{bcz82/nc303cs} (this study), mrc1b^sa18640 (this study), irg1-KI:GFP (knock-in, this study), irg1:GFP (tol2-based, this study), tnfa:GFP^pd1028 ³², “macro-rescue” mpeg1:nlrc3l²⁴, lyz:mCherry⁸², mpeg1:GFP⁶¹, and mpeg1:BFP⁸² and raised at 28.5°C to stage for analysis. Genotyping primers and assays are described in Supplementary Data 1. This study was carried out in accordance with the approval of UNC-Chapel Hill Institutional Animal Care and Use Committee (protocols 19-132 and 22-103).

Generating GFP knock-in at the irg1/acod1 locus using GeneWeld platform

GFP knock-in followed protocol as previously described^25,26. High quality genomic DNA was used to identify the 48-bp upstream and downstream homologous arm sequences for cloning using Type IIS restriction enzymes BfuAI and BspQI for ligation into the GeneWeld (GW) plasmid, pGTag vector^25,26. The following primers were used to verify the correct insertions: 5’_pgtag_seq: 5’-GCATGGATGTTTTCCCAGTC-3’ and 3’_pgtag_seq: 5’-ATGGCTCATAACACCCCTTG-3’. Single-cell injections into zebrafish were conducted using a mix containing the universal gRNA, irg1 gRNA-2, Cas9 mRNA, and the irg1-targeting GW vector, and also without the irg1-targeting GW plasmid as a negative control. A total of 15 injected F₀ adults were outcrossed to generate stable F₁ founders. Progeny from only 1 out of 15 F₀ fish gave a GFP validation by PCR. Assessment of 49 adult F₁s yielded 2 fish (#7 and 31) carrying GFP that was localized to the irg1 locus (yielding a 0.3% success rate) as verified by PCR and Sanger sequencing using multiple primer sets as well as by experimental evidence of a Mendelian segregation of the reporter expression. This new fish line is named “irg1-KI:GFP”.

Cloning and creating transgenic TOL2-based irg1:GFP reporter

Approximately 5 kb of regulatory sequence upstream of the irg1/acod1 coding sequence for the start codon was cloned from the BAC DKEY-57A22, which contained the zebrafish chromosomal region encompassing the irg1 region in chromosome 9, using the irg1p −4794 F and irg1p + 18 R primers as listed in Supplementary Data 1. This represents a longer fragment than that used in the previously published TOL2-based irg1 reporter by Sander et al.²². The ~5 kb regulatory sequence of irg1 was cloned into a Gateway-compatible p5E vector backbone to be combined with a pME-GFP coding sequence to create the irg1:GFP plasmid. Tol2-mediated transgenesis was used to integrate this reporter cassette into the zebrafish genome to create the new irg1:GFP stable transgenic line, as shown in Supplementary Fig. 2.

Tail amputation

Zebrafish embryos at 2 dpf were anesthetized in 0.02% MS-222 (tricaine) supplemented with 0.003% PTU. Embryos were amputated at the halfway point between the end of the yolk extension and tail fin end ( ~ 0.5 mm measured from tip of tail fin equivalent to ~8-10 somites from end of yolk extension) in a Sylgard lined petri dish using a 5 mm straight microsurgical blade with a 15 degree angle (Sharpoint Microsurigical Knives, 72-1551). After the operation, embryos were immediately transferred to fresh water with PTU and monitored for recovery. Most experiments focus on the initial inflammatory phase of the injury response, so embryos were typically collected 18-24 h after amputation for the 24 hpa timepoint.

In vivo time-lapse and static confocal imaging and analysis

All time-lapse and static z-stack imaging were performed using a Nikon A1R+ hybrid galvano and resonant scanning confocal system equipped with an ultra-high speed A1-SHR scan head and controller. Images were obtained using an apochromat lambda 40x water immersion objective (NA 1.15) or a plan apochromat lambda 20x objective (NA 0.75). Z-steps at 1–3 µm were taken at 40x and 3–5 µm at 20x. Different stages of zebrafish were mounted on glass-bottom dishes using 1.5% low-melting agarose and submerged in fish water supplemented with 0.003% PTU to inhibit pigmentation. All image acquisition parameters were kept constant within an experiment (including laser power, gain, speed, and resolution). Image analysis was performed using ImageJ Fiji version 2.9.0 for injury analysis at the tissue or cell population level, or Imaris 10.1.1.software (Bitplane Oxford) for dynamic imaging or large-scale single cell quantifications (fluorescence, numbers). Traced ROIs were measured for area and fluorescence levels from original, unadjusted raw data. For tail cut images, “cut site” is defined as the region within the two intact somites from the edge of the injured tail. The largest ROI within one intact somite from the edge of the injured tail and injured tissue is defined as the “cut site cell cluster” for analysis. Imaris analysis was used to generate 3-dimensional surfaces of cells of interest using Machine Learning Pixel Classification with the Fiji/ImageJ Plugin Labkit, followed by manual curation to ensure accuracy. Quantification of fluorescence levels and other measurements were taken from surfaces created for the different cells using Imaris. For movie analyses, macrophage surfaces were generated using customized algorithms, where surfaces were tracked over time, and machine learning segmentation was used on the GFP channel for macrophages with repeated training. Slicer extended section was set at 3 μm, the size of each Z-slice, and any ROIs less than 10 voxels were filtered out. Tracking was conducted using the autoregressive motion algorithm with a max distance of 20 μm. Repeated manual editing was required to separate clumps and cells in close contact using the cut surface tool, or to connect pieces that belonged to one cell using the unify tool at every timepoint. Tracks of each cell ROI were then aligned which required manual checks and editing to disconnect or reconnect to assemble a correctly integrated track, which was repeated for each ROI at every timepoint.

Fluorescent stereoscope imaging and analysis

Leica M165 FC microscope (Leica Microsystems, Germany) equipped with a Leica DFC9000 GT camera was used to acquire fluorescent images of live whole embryos. Embryos were mounted in petri dishes with 1% low-melting agarose and covered with water supplemented with 0.003% PTU, and imaged individually with consistent laser power and magnification settings within each experiment. Yolkball analysis was performed using ImageJ Fiji. Any visible fluorescent cells within the boundary of the yolkball were captured as an ROI. ROIs were created automatically by using the threshold tool to generate a mask of fluorescent cells and selecting the "Analyze Particles" option to create the ROI set. A freehand selection tool for manual tracing was used to generate additional ROIs of any missed cells from the automated threshold setting. ROIs were then used to measure area and fluorescence levels from original unadjusted images.

Whole mount immunohistochemistry

Embryos were fixed in 4% paraformaldehyde (PFA) for 2 h at room temperature or overnight at 4°C, followed by a series of PBST washes (phosphate-buffered saline with 0.2% Tween20) and passage through 100% methanol at −20°C until proceeding with immunostaining. Pax7 antibody staining required a shorter fixation protocol using 2% PFA fix for 20 min at room temperature. Blocking solution was PBST with 5% normal goat serum, followed by overnight incubation with the primary antibody at 4°C, then washed before proceeding to an overnight incubation with the appropriate secondary antibody at 4°C. The following primary antibodies were used in blocking solution: rabbit anti-smooth muscle actin (GTX100034, GeneTex) at 1:500, chicken anti-GFP (ab13970, Abcam) at 1:500, mouse anti-Pax7 (AB_528428, DSHB) at 1:100, and rabbit anti-CTSK (cathepsin K) at 1:500 (E7U5N, Cell Signaling Technology) followed by incubation with the appropriate secondary antibodies used at 1:500 to 1:2000. Use of antibodies were validated based on pattern of staining in control normal samples that matched description of reagent provided by the manufacturer and previously published data. DAPI (Roche) may be used in the last wash at 5 μg/mL for an hour. See Supplementary Data 1 for details of reagents used.

Phalloidin staining for muscle actin

Embryos were fixed in 4% paraformaldehyde (PFA) for 2 h at room temperature or overnight at 4°C, followed by several 10-min PBST washes and incubation with Phalloidin-iFluor 647 (Abcam) at 1:500 dilution in PBST overnight at 4°C. After staining, embryos were washed in PBST and mounted in 1% low-melting agarose for imaging.

Whole mount hybridization chain reaction (HCR)

Embryos, either injured or normal, were fixed in freshly made 4% PFA overnight at 4°C and stored in 100% methanol. To proceed with HCR, embryos were rehydrated, permeabilized using proteinase K, and processed through a series of reagents as per protocols previously published and provided by the manufacturer (Molecular Instruments). Probe targets and amplifiers were designed and made by Molecular Instruments as listed in Supplementary Data 1.

Isolation of macrophages for bulk transcriptome analysis by FACS

Embryos from nlrc3^st73 heterozygous incrosses were sorted as siblings or mutants based on a high induction of irg1-KI:GFP expression at 3 dpf. Around 20-50 embryos were dissociated per sample on ice in FACSmax Cell Dissociation Solution (Amsbio) in a microcentrifuge tube using a motorized pellet pestle (Fisherbrand) until the sample is completely homogenized. The cell suspension was filtered through a sterile 35 μm nylon mesh (Falcon, 352235) and another passage through a sterile 20 μm mesh strainer (EASYstrainer, 542120) with an additional 500 μL FACSmax solution run through. The dissociated cell suspension was then analyzed and sorted directly into RNA lysis buffer (Ambion) on a BD FACSMelody Cell Sorter. General gating was applied using forward and side scatter to distinguish populations for single live cells that represented macrophages expressing high and low GFP expression from the irg1-KI:GFP transgene. See gating strategy in Supplementary Fig. 21.

RNA isolation and qPCR

RNA was isolated using the RNAqueous-Micro kit RNA Isolation Procedure (Ambion). Whole larvae were lysed in 100–300 uL RNA lysis buffer. cDNA was made from 150 or 200 ng of total RNA using oligo (dT) primer with SuperScript IV reverse transcriptase (Invitrogen). qPCR was performed on the QuantStudio 3 Real-Time PCR System (Applied Biosystems) using SYBR Green. The delta-delta CT method was used to determine the relative levels of mRNA expression between experimental samples and controls. ef1α was used as the reference gene for the determination of the relative expression of all target genes. Primer sequences for qPCR analysis are listed in Supplementary Data 1.

Bulk RNA-seq analysis

Paired-end RNA-Sequencing was performed on FACS-sorted macrophages derived from a pool of 20-50 embryos at 3 dpf either at baseline or 24 hpa. Three independent biological replicates for each condition were used for sequencing on an Illumina NovaSeq6000 S2 or NovaSeq6000 SP using 2×50 base pairs, each replicate represents a different cohort of embryos from which macrophages were isolated by FACS. The mutants were pre-sorted by spontaneous induction of the irg1-KI:GFP, and the genotypes of all RNA samples were also verified by PCR-restriction enzyme assay-based genotyping prior to library preparation using the SMARTr HT+ Nextera XT for low-input samples. 100% correspondence was found between the sorted phenotypes based on the presence or absence of irg1 induction and the expected nlrc3l mutant or sibling genotypes, respectively. RNA-seq analysis followed the previously used protocol²⁴, which consists of modules and packages including Trimmomatic for data trimming, MultiQC for quality control, and BBMap for sequence alignment with the zebrafish reference genome (GRCz11). The differential expression analysis was performed using DESeq2. Gene ontology analysis was performed on the differentially expressed genes with Metascape³⁴ and clustered heat map analysis with NG-CHM³³.

CRISPR-Cas9 targeted mutagenesis of irg1/acod1, asc and mrc1b

CRISPR gRNA designs, and synthesis of Cas9 mRNA and gRNAs for targeting zebrafish irg1/acod1 (NCBI accession: NM_001126456.2; Gene ID: 795305), asc (NCBI accession: NM_131495.3; Gene ID: 57923), and mrc1b (NCBI accession: NM_001423726.1; Gene ID: 559502) followed previously described methods⁸². Co-injection of Cas9 mRNA and guide RNAs (gRNAs) was conducted in 1-cell stage zebrafish embryos. gRNA target sequences and genotyping primers are provided in Supplementary Data 1. To ensure high mutagenesis rate and large deletion mutations, two or more gRNAs were simultaneously injected with Cas9 mRNA for each gene. Injected clutches of embryos were validated to contain CRISPR-mediated mutagenesis by a T7 endonuclease assay.

E. coli infection model combined with tail amputation

To induce systemic bacterial infection, wild-type transgenic zebrafish carrying irg1-KI:GFP; mpeg1:BFP at 2 dpf were anesthetized in 0.02% Tricaine (MS-222) for direct microinjection into the brain ventricle either with 1 nl of ~1000 CFU/nL of a common laboratory E. coli strain (MG1655 [CGSC6300]) or 1 nl of phosphate-buffered saline (PBS) as negative controls alongside uninjected animals for each experiment. Previous studies showed brain ventricle microinjection in the zebrafish embryo rapidly spreads systemically throughout the body⁸². Persistence of macrophage activation based on irg1 expression was assessed using the irg1-KI:GFP reporter with a single injection (4–6 h before amputation) or double injections (day-of and ~24 hpa), either method caused a sustained irg1 induction and activation of macrophages. All data reflect a single brain injection to induce chronic inflammatory macrophage activation prior to muscle injury that lasted past the resolution phase at 72 hpa (Supplementary Fig. 17). After validation of a bacterial response 4-6 h post-injection based on a strong induction of the irg1-KI:GFP expression, the embryos were amputated to induce skeletal muscle injury as described previously. Zebrafish following tail amputation were placed back into the 28.5 °C incubator to allow recovery until embryo collection at 20-24 hpa for downstream analysis.

scRNA-seq cell preparations and sequencing

Transgenic zebrafish carrying the GFP knock-in reporter irg1-KI:GFP generated in our lab were used to isolate macrophages at distinct phases of acute muscle injury response. Samples were collected at 24 hpa, representing the initial inflammatory phase, and at 48 hpa, marking the onset of tissue repair following debris clearance. Uninjured (“uncut”) animals served as baseline controls. Responses were compared between wild-type (or control sibling) zebrafish and mutants with a deleterious nlrc3l mutation that causes chronic macrophage activation. To induce acute muscle injury, large tail amputations were performed on 2 dpf zebrafish embryos. Activated macrophages upregulate irg1, displaying strong GFP fluorescence, while steady-state macrophages express minimal irg1, resulting in faint GFP signal. All irg1-KI:GFP+ cells expressing any detectable level of GFP were isolated using FACS on a BD FACSMelody across four conditions: control uncut, control 24 hpa, control 48 hpa, and nlrc3l mutant 24 hpa. Two biological replicates were collected per condition, resulting in a total of eight samples, each consisting of pooled cells from 90 embryos. To maintain cell viability and quality, embryos were dissociated and sorted in batches of 30, with three such batches pooled to constitute a single sample. Sample collection was interleaved to evenly distribute conditions across the processing timeline. Embryo dissociation was performed in a buffer containing ROCK inhibitor, protease, and DNase with gentle mechanical grinding, and a holding solution composed of HBSS (without phenol red), 1% BSA, and ROCK inhibitor. Cells were kept on ice until downstream processing using the 10x Genomics Chromium X Controller and the GEM-X Universal 3’ Gene Expression v4 4-plex kit (PN-1000779) following the manufacturer’s instructions. Cells were first verified for viability, concentration, and singleness using acridine orange and propidium iodide on the LUNA-FX7 Dual Fluorescence Cell Counter (Logos Biosystems) prior to 10x Genomics preparations. Single-cell libraries were prepared using the GEM-X OCM 3’ Chip, where samples were uniquely indexed and pooled in groups of four into two different GEMs during loading via on-chip multiplexing (OCM) technology. Reverse transcription was performed using a C1000 thermal cycler (Bio-Rad) to create cDNA libraries tagged with a cell barcode and unique molecular index (UMI). Dynabeads MyOne SILANE beads (Invitrogen) were used to purify broken GEMs prior to cDNA amplification, and final libraries with Illumina-compatible adapters were purified with SPRIselect magnetic beads (Beckam Coulter) and quantified using an Agilent Bioanalyzer High Sensitivity DNA chip (Agilent). Libraries were pooled in an equimolar ratio and sequenced on one lane of a NovaSeq X Plus 10B flow cell (Illumina) in paired-end format at Admera Health, LLC (Read 1: 28 cycles, i7 index: 10 cycles, i5 Index: 10 cycles, Read 2: 90 cycles).

scRNA-seq analysis

Eight samples spanning four conditions were processed across two Cell Ranger Multi runs. Custom transgenes (EGFP and TagBFP) were appended to the zebrafish reference genome (GRCz11, from 10x Genomics). Gene expression matrices were generated using Cell Ranger Multi with sample demultiplexing guided by configuration CSVs specifying sample IDs, reference paths, and OCM barcodes. Downstream analysis was performed in R using Seurat (v5). Cells were filtered using miQC with a posterior cutoff of 0.80, mitochondrial content column set as “percent.mt”, and nFeature_RNA as the gene feature input. Cells marked as “discard” by miQC were removed from the dataset. SCTransform was used for normalization and variance stabilization (SCT assay), followed by dimensionality reduction via PCA using the first 30 principal components. UMAPs were computed from these PCs, and clustering was performed using shared nearest neighbor (SNN) graph-based clustering across multiple resolutions, and 0.4 was selected for downstream analyses based on biological interpretability. Cells with expression >0 were classified as positive for each gene. For statistical comparison between conditions, pairwise t-tests with FDR correction were applied to both gene-positive fractions and average expression values, stratified by cluster. Differential expression analysis between conditions was performed using a pseudobulk DESeq2 approach. For each cluster and condition comparison, raw RNA counts were aggregated by sample and filtered to retain genes with at least 10 counts in more than one sample. Only samples with ≥5 cells were included, and comparisons were restricted to condition pairs with at least two biological replicates. DESeq2 was run with fitType = “parametric”, and estimateSizeFactors() was used with type = “poscounts”. All scripts were executed on the UNC-CH’s Longleaf computing cluster, and outputs were organized into directories for reference creation, quality control, clustering, visualization, and differential expression.

Statistics and reproducibility

For pairwise comparisons, unpaired two-tailed t-tests were performed unless otherwise noted. F test was used to compare variances. For unequal variances, Welch’s correction was used on the two-tailed t-test. For multiple comparisons of 3 or more groups, one-way ANOVA test was applied followed by multiple pair-wise comparisons to determine the pair(s) showing significant differences using FDR-adjusted p-values. Experiments were repeated at least twice independently with similar results. All graphical plots and statistical tests were generated using GraphPad Prism 10 unless otherwise noted.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

Data generated in this study are provided in the publication, Supplementary Information and Source Data File (Supplementary Data 2). RNA sequencing data is freely available at the GEO repository as GSE283438 for bulk RNA-seq, and GSE301079 for scRNA-seq.

References

Wang, X. & Zhou, L. The multifaceted role of macrophages in homeostatic and injured skeletal muscle. Front Immunol. 14, 1274816 (2023).
Article CAS PubMed PubMed Central Google Scholar
Tu, H. & Li, Y. L. Inflammation balance in skeletal muscle damage and repair. Front Immunol. 14, 1133355 (2023).
Article CAS PubMed PubMed Central Google Scholar
Tidball, J. G. Inflammatory processes in muscle injury and repair. Am. J. Physiol. Regul. Integr. Comp. Physiol. 288, R345–R353 (2005).
Article CAS PubMed Google Scholar
Hamidzadeh, K., Christensen, S. M., Dalby, E., Chandrasekaran, P. & Mosser, D. M. Macrophages and the Recovery from Acute and Chronic Inflammation. Annu Rev. Physiol. 79, 567–592 (2017).
Article CAS PubMed Google Scholar
Furman, D. et al. Chronic inflammation in the etiology of disease across the life span. Nat. Med 25, 1822–1832 (2019).
Article CAS PubMed PubMed Central Google Scholar
Tidball, J. G. Regulation of muscle growth and regeneration by the immune system. Nat. Rev. Immunol. 17, 165–178 (2017).
Article CAS PubMed PubMed Central Google Scholar
Yang, W. & Hu, P. Skeletal muscle regeneration is modulated by inflammation. J. Orthop. Transl. 13, 25–32 (2018).
Google Scholar
Arnold, L. et al. Inflammatory monocytes recruited after skeletal muscle injury switch into antiinflammatory macrophages to support myogenesis. J. Exp. Med 204, 1057–1069 (2007).
Article CAS PubMed PubMed Central Google Scholar
Duchesne, E., Dufresne, S. S. & Dumont, N. A. Impact of inflammation and anti-inflammatory modalities on skeletal muscle healing: from fundamental research to the clinic. Phys. Ther. 97, 807–817 (2017).
Article PubMed Google Scholar
Kyritsis, N. et al. Acute inflammation initiates the regenerative response in the adult zebrafish brain. Science 338, 1353–1356 (2012).
Article ADS CAS PubMed Google Scholar
Iribarne, M. Inflammation induces zebrafish regeneration. Neural Regen. Res 16, 1693–1701 (2021).
Article CAS PubMed PubMed Central Google Scholar
De Micheli, A. J. et al. Single-cell analysis of the muscle stem cell hierarchy identifies heterotypic communication signals involved in skeletal muscle regeneration. Cell Rep. 30, 3583–3595.e3585 (2020).
Article PubMed PubMed Central Google Scholar
Oprescu, S. N., Yue, F., Qiu, J., Brito, L. F. & Kuang, S. Temporal Dynamics and Heterogeneity of Cell Populations during Skeletal Muscle Regeneration. iScience 23, 100993 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Varga, T. et al. Highly Dynamic Transcriptional Signature of Distinct Macrophage Subsets during Sterile Inflammation, Resolution, and Tissue Repair. J. Immunol. 196, 4771–4782 (2016).
Article CAS PubMed Google Scholar
Parisi, L. et al. Macrophage polarization in chronic inflammatory diseases: killers or builders? J. Immunol. Res. 2018, 8917804 (2018).
Article PubMed PubMed Central Google Scholar
Wu, R., Kang, R. & Tang, D. Mitochondrial ACOD1/IRG1 in infection and sterile inflammation. J. Intensive Med 2, 78–88 (2022).
Article CAS PubMed PubMed Central Google Scholar
Hall, C. J. et al. Immunoresponsive gene 1 augments bactericidal activity of macrophage-lineage cells by regulating beta-oxidation-dependent mitochondrial ROS production. Cell Metab. 18, 265–278 (2013).
Article CAS PubMed Google Scholar
Lee, C. G., Jenkins, N. A., Gilbert, D. J., Copeland, N. G. & O’Brien, W. E. Cloning and analysis of gene regulation of a novel LPS-inducible cDNA. Immunogenetics 41, 263–270 (1995).
Article CAS PubMed Google Scholar
Michelucci, A. et al. Immune-responsive gene 1 protein links metabolism to immunity by catalyzing itaconic acid production. Proc. Natl. Acad. Sci. USA 110, 7820–7825 (2013).
Article ADS CAS PubMed PubMed Central Google Scholar
Basler, T., Jeckstadt, S., Valentin-Weigand, P. & Goethe, R. Mycobacterium paratuberculosis, Mycobacterium smegmatis, and lipopolysaccharide induce different transcriptional and post-transcriptional regulation of the IRG1 gene in murine macrophages. J. Leukoc. Biol. 79, 628–638 (2006).
Article CAS PubMed Google Scholar
Wu, R., Liu, J., Tang, D. & Kang, R. The dual role of ACOD1 in inflammation. J. Immunol. 211, 518–526 (2023).
Article CAS PubMed Google Scholar
Sanderson, L. E. et al. An inducible transgene reports activation of macrophages in live zebrafish larvae. Dev. Comp. Immunol. 53, 63–69 (2015).
Article CAS PubMed Google Scholar
Shiau, C. E., Monk, K. R., Joo, W. & Talbot, W. S. An anti-inflammatory NOD-like receptor is required for microglia development. Cell Rep. 5, 1342–1352 (2013).
Article CAS PubMed PubMed Central Google Scholar
Kwon, V. et al. Peripheral NOD-like receptor deficient inflammatory macrophages trigger neutrophil infiltration into the brain disrupting daytime locomotion. Commun. Biol. 5, 464 (2022).
Article CAS PubMed PubMed Central Google Scholar
Welker, J. M. et al. GeneWeld: Efficient Targeted Integration Directed by Short Homology in Zebrafish. Bio Protoc. 11, e4100 (2021).
Article CAS PubMed PubMed Central Google Scholar
Wierson, W. A. et al. Efficient targeted integration directed by short homology in zebrafish and mammalian cells. Elife 9, e53968 (2020).
Article CAS PubMed PubMed Central Google Scholar
Rojas, A. M. & Shiau, C. E. Brain-localized and intravenous microinjections in the larval zebrafish to assess innate immune response. Bio Protoc. 11, e3978 (2021).
Article PubMed PubMed Central Google Scholar
Zhou, Q. et al. Cross-organ single-cell transcriptome profiling reveals macrophage and dendritic cell heterogeneity in zebrafish. Cell Rep. 42, 112793 (2023).
Article CAS PubMed Google Scholar
Rougeot, J. 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. 10, 832 (2019).
Article CAS PubMed PubMed Central Google Scholar
Nawaz, A. et al. CD206(+) M2-like macrophages regulate systemic glucose metabolism by inhibiting proliferation of adipocyte progenitors. Nat. Commun. 8, 286 (2017).
Article ADS PubMed PubMed Central Google Scholar
Roszer, T. Understanding the mysterious M2 macrophage through activation markers and effector mechanisms. Mediators Inflamm. 2015, 816460 (2015).
Article PubMed PubMed Central Google Scholar
Marjoram, L. et al. Epigenetic control of intestinal barrier function and inflammation in zebrafish. Proc. Natl. Acad. Sci. USA 112, 2770–2775 (2015).
Article ADS CAS PubMed PubMed Central Google Scholar
Broom, B. M. et al. A Galaxy Implementation of Next-Generation Clustered Heatmaps for Interactive Exploration of Molecular Profiling Data. Cancer Res 77, e23–e26 (2017).
Article ADS CAS PubMed PubMed Central Google Scholar
Zhou, Y. et al. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat. Commun. 10, 1523 (2019).
Article ADS PubMed PubMed Central Google Scholar
Janssens, S. & Beyaert, R. A universal role for MyD88 in TLR/IL-1R-mediated signaling. Trends Biochem Sci. 27, 474–482 (2002).
Article CAS PubMed Google Scholar
Shi, J. H. & Sun, S. C. Tumor necrosis factor receptor-associated factor regulation of nuclear factor kappab and mitogen-activated protein kinase pathways. Front Immunol. 9, 1849 (2018).
Article PubMed PubMed Central Google Scholar
Gohda, J., Matsumura, T. & Inoue, J. Cutting edge: TNFR-associated factor (TRAF) 6 is essential for MyD88-dependent pathway but not toll/IL-1 receptor domain-containing adaptor-inducing IFN-beta (TRIF)-dependent pathway in TLR signaling. J. Immunol. 173, 2913–2917 (2004).
Article CAS PubMed Google Scholar
Tsarouchas, T. M. et al. Dynamic control of proinflammatory cytokines Il-1beta and Tnf-alpha by macrophages in zebrafish spinal cord regeneration. Nat. Commun. 9, 4670 (2018).
Article ADS PubMed PubMed Central Google Scholar
Nguyen-Chi, M. et al. Identification of polarized macrophage subsets in zebrafish. Elife 4, e07288 (2015).
Article PubMed PubMed Central Google Scholar
Nguyen-Chi, M. et al. TNF signaling and macrophages govern fin regeneration in zebrafish larvae. Cell Death Dis. 8, e2979 (2017).
Article CAS PubMed PubMed Central Google Scholar
Kaliya-Perumal, A. K. & Ingham, P. W. Musculoskeletal regeneration: A zebrafish perspective. Biochimie 196, 171–181 (2022).
Article CAS PubMed Google Scholar
Jin, R. M., Warunek, J. & Wohlfert, E. A. Chronic infection stunts macrophage heterogeneity and disrupts immune-mediated myogenesis. JCI Insight 3, e121549 (2018).
Article PubMed PubMed Central Google Scholar
Sehring, I. M. & Weidinger, G. Recent advancements in understanding fin regeneration in zebrafish. Wiley Interdiscip. Rev. Dev. Biol. 9, e367 (2020).
Article PubMed Google Scholar
Li, L., Yan, B., Shi, Y. Q., Zhang, W. Q. & Wen, Z. L. Live imaging reveals differing roles of macrophages and neutrophils during zebrafish tail fin regeneration. J. Biol. Chem. 287, 25353–25360 (2012).
Article CAS PubMed PubMed Central Google Scholar
Fu, X., Wang, H. & Hu, P. Stem cell activation in skeletal muscle regeneration. Cell Mol. Life Sci. 72, 1663–1677 (2015).
Article CAS PubMed PubMed Central Google Scholar
Pipalia, T. G. et al. Cellular dynamics of regeneration reveals role of two distinct Pax7 stem cell populations in larval zebrafish muscle repair. Dis. Model Mech. 9, 671–684 (2016).
Article CAS PubMed PubMed Central Google Scholar
Seger, C. et al. Analysis of Pax7 expressing myogenic cells in zebrafish muscle development, injury, and models of disease. Dev. Dyn. 240, 2440–2451 (2011).
Article CAS PubMed Google Scholar
Shiau, C. E., Kaufman, Z., Meireles, A. M. & Talbot, W. S. Differential requirement for irf8 in formation of embryonic and adult macrophages in zebrafish. PLoS One 10, e0117513 (2015).
Article PubMed PubMed Central Google Scholar
Fingleton, B. Matrix metalloproteinases as regulators of inflammatory processes. Biochim Biophys. Acta Mol. Cell Res 1864, 2036–2042 (2017).
Article CAS PubMed Google Scholar
Yabluchanskiy, A., Ma, Y., Iyer, R. P., Hall, M. E. & Lindsey, M. L. Matrix metalloproteinase-9: Many shades of function in cardiovascular disease. Physiol. (Bethesda) 28, 391–403 (2013).
CAS Google Scholar
Silva, N. J. et al. Inflammation and matrix metalloproteinase 9 (Mmp-9) regulate photoreceptor regeneration in adult zebrafish. Glia 68, 1445–1465 (2020).
Article PubMed PubMed Central Google Scholar
Sun, S. C. The non-canonical NF-kappaB pathway in immunity and inflammation. Nat. Rev. Immunol. 17, 545–558 (2017).
Article CAS PubMed PubMed Central Google Scholar
Mijanovic, O. et al. Cathepsin K in pathological conditions and new therapeutic and diagnostic perspectives. Int J. Mol. Sci. 23, 13762 (2022).
Article CAS PubMed PubMed Central Google Scholar
Fabrik, I. et al. Lung macrophages utilize unique cathepsin K-dependent phagosomal machinery to degrade intracellular collagen. Life Sci. Alliance 6, e202201535 (2023).
Article CAS PubMed PubMed Central Google Scholar
Wang, Y. et al. The role of matrix metalloproteinase 9 in fibrosis diseases and its molecular mechanisms. Biomed. Pharmacother. 171, 116116 (2024).
Article CAS PubMed Google Scholar
Chen, S. E., Jin, B. & Li, Y. P. TNF-alpha regulates myogenesis and muscle regeneration by activating p38 MAPK. Am. J. Physiol. Cell Physiol. 292, C1660–C1671 (2007).
Article CAS PubMed Google Scholar
Howard, E. E., Pasiakos, S. M., Blesso, C. N., Fussell, M. A. & Rodriguez, N. R. Divergent Roles of Inflammation in Skeletal Muscle Recovery From Injury. Front Physiol. 11, 87 (2020).
Article PubMed PubMed Central Google Scholar
Summan, M. et al. Macrophages and skeletal muscle regeneration: a clodronate-containing liposome depletion study. Am. J. Physiol. Regul. Integr. Comp. Physiol. 290, R1488–R1495 (2006).
Article CAS PubMed Google Scholar
Robertson, T. A., Grounds, M. D. & Papadimitriou, J. M. Elucidation of aspects of murine skeletal muscle regeneration using local and whole body irradiation. J. Anat. 181, 265–276 (1992).
PubMed PubMed Central Google Scholar
Lu, H. et al. Macrophages recruited via CCR2 produce insulin-like growth factor-1 to repair acute skeletal muscle injury. FASEB J. 25, 358–369 (2011).
Article CAS PubMed Google Scholar
Ellett, F., Pase, L., Hayman, J. W., Andrianopoulos, A. & Lieschke, G. J. mpeg1 promoter transgenes direct macrophage-lineage expression in zebrafish. Blood 117, e49–e56 (2011).
Article CAS PubMed Google Scholar
Rinnenthal, J. L. et al. Inflammatory myopathy with abundant macrophages (IMAM): the immunology revisited. Neuromuscul. Disord. 24, 151–155 (2014).
Article PubMed Google Scholar
Dalakas, M. C. Inflammatory muscle diseases. N. Engl. J. Med 373, 393–394 (2015).
Article CAS PubMed Google Scholar
Azad, A. K., Rajaram, M. V. & Schlesinger, L. S. Exploitation of the macrophage mannose receptor (cd206) in infectious disease diagnostics and therapeutics. J. Cytol. Mol. Biol. 1, 1000003 (2014).
PubMed PubMed Central Google Scholar
Lutgens, E. et al. Disruption of the cathepsin K gene reduces atherosclerosis progression and induces plaque fibrosis but accelerates macrophage foam cell formation. Circulation 113, 98–107 (2006).
Article CAS PubMed Google Scholar
Dai, R. et al. Cathepsin K: The Action in and Beyond Bone. Front Cell Dev. Biol. 8, 433 (2020).
Article ADS PubMed PubMed Central Google Scholar
Buhling, F. et al. Cathepsin K-a marker of macrophage differentiation? J. Pathol. 195, 375–382 (2001).
Article CAS PubMed Google Scholar
Wu, N. et al. Cathepsin K regulates the tumor growth and metastasis by IL-17/CTSK/EMT axis and mediates M2 macrophage polarization in castration-resistant prostate cancer. Cell Death Dis. 13, 813 (2022).
Article CAS PubMed PubMed Central Google Scholar
Barascuk, N. et al. Human macrophage foam cells degrade atherosclerotic plaques through cathepsin K mediated processes. BMC Cardiovasc Disord. 10, 19 (2010).
Article PubMed PubMed Central Google Scholar
Julier, Z., Park, A. J., Briquez, P. S. & Martino, M. M. Promoting tissue regeneration by modulating the immune system. Acta Biomater. 53, 13–28 (2017).
Article CAS PubMed Google Scholar
Uderhardt, S., Martins, A. J., Tsang, J. S., Lammermann, T. & Germain, R. N. Resident macrophages cloak tissue microlesions to prevent neutrophil-driven inflammatory damage. Cell 177, 541–555.e517 (2019).
Article CAS PubMed PubMed Central Google Scholar
van der Zande, H. J. P., Nitsche, D., Schlautmann, L., Guigas, B. & Burgdorf, S. The mannose receptor: from endocytic receptor and biomarker to regulator of (meta)inflammation. Front Immunol. 12, 765034 (2021).
Article PubMed PubMed Central Google Scholar
Venero Galanternik, M. et al. A novel perivascular cell population in the zebrafish brain. Elife 6, e24369 (2017).
Article PubMed PubMed Central Google Scholar
Das, R. N. et al. Generation of specialized blood vessels via lymphatic transdifferentiation. Nature 606, 570–575 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Yu, T., Chen, J., Wang, Y. & Xu, J. The embryonic zebrafish brain is exclusively colonized by pu.1-dependent and lymphatic-independent population of microglia. Sci. Adv. 10, eado0519 (2024).
Article ADS CAS PubMed PubMed Central Google Scholar
Green, L. A., O’Dea, M. R., Hoover, C. A., DeSantis, D. F. & Smith, C. J. The embryonic zebrafish brain is seeded by a lymphatic-dependent population of mrc1(+) microglia precursors. Nat. Neurosci. 25, 849–864 (2022).
Article CAS PubMed PubMed Central Google Scholar
Elaldi, R. et al. CD206+CD14- Skin-resident Macrophages and DC-T Cell clusters are spatial features characterizing non-Relapsing Cutaneous Squamous Cell Carcinoma. Cancer Immunol. Res https://doi.org/10.1158/2326-6066.CIR-25-0182 (2025).
Article PubMed Google Scholar
Shook, B., Xiao, E., Kumamoto, Y., Iwasaki, A. & Horsley, V. CD301b+ Macrophages Are Essential for Effective Skin Wound Healing. J. Invest Dermatol 136, 1885–1891 (2016).
Article CAS PubMed PubMed Central Google Scholar
Smith, C. J. et al. TNFa/TNFR2 signaling is required for glial ensheathment at the dorsal root entry zone. PLoS Genet 13, e1006712 (2017).
Article PubMed PubMed Central Google Scholar
Scodeller, P. et al. Precision Targeting of Tumor Macrophages with a CD206 Binding Peptide. Sci. Rep. 7, 14655 (2017).
Article ADS PubMed PubMed Central Google Scholar
Burns, A. R. et al. Interhost dispersal alters microbiome assembly and can overwhelm host innate immunity in an experimental zebrafish model. Proc. Natl. Acad. Sci. USA 114, 11181–11186 (2017).
Article ADS CAS PubMed PubMed Central Google Scholar
Yang, L. et al. Drainage of inflammatory macromolecules from the brain to periphery targets the liver for macrophage infiltration. Elife 9, e58191 (2020).
Article CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

We thank Maura McGrail and Jeff Essner for the generous sharing of GeneWeld plasmids and protocols, Michel Bagnat for the tnfa:GFP line, Brian Conlon and Anika Rueppell for the E. coli MG1655 support, and Peng Huang for reagent support. We are grateful to Mike Vernon and staff of UNC High-Throughput Sequencing Facility (supported by P30-CA016086 and P30-ES010126) for library preparations and RNA-sequencing, Gabrielle Cannon of UNC Advanced Analytics Core (UNC CGIBD, P30 DK034987) for scRNA-seq library preparations, researchers of UNC Bioinformatics and Analytics Research Collabrative (BARC): Tara Brennan, Matt Niederhuber, Ismael Gomez, and Austin Hepperla for processing and analzying scRNA-seq data, and Michelle Altemara and staff of UNC Zebrafish Aquaculture Core Facility for zebrafish housing and care. This work was supported by NIH NIGMS grant R35GM124719 to C.E.S.

Author information

These authors contributed equally: Caroline G. Spencer, Matthew Hamilton.

Authors and Affiliations

Department of Biology, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
Caroline G. Spencer, Matthew Hamilton, Ethan Bedsole, Yingshan N. Wei, Alison M. Rojas, Andrew Burciu, John Zhu, Keith Z. Sabin & Celia E. Shiau
Department of Microbiology and Immunology, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
Celia E. Shiau

Authors

Caroline G. Spencer
View author publications
Search author on:PubMed Google Scholar
Matthew Hamilton
View author publications
Search author on:PubMed Google Scholar
Ethan Bedsole
View author publications
Search author on:PubMed Google Scholar
Yingshan N. Wei
View author publications
Search author on:PubMed Google Scholar
Alison M. Rojas
View author publications
Search author on:PubMed Google Scholar
Andrew Burciu
View author publications
Search author on:PubMed Google Scholar
John Zhu
View author publications
Search author on:PubMed Google Scholar
Keith Z. Sabin
View author publications
Search author on:PubMed Google Scholar
Celia E. Shiau
View author publications
Search author on:PubMed Google Scholar

Contributions

C.G.S.: data analysis and visualization, experimentation (injury model, time course analysis, epistasis experiments, multiple mutant phenotyping, timelapse and static injury imaging, CRISPR), writing; M.H.: data analysis and visualization, experimentation (bulk and single-cell RNA-seq analysis, time course analysis, imaging, injury analysis, cloning, qPCR, HCR, scRNA-seq, IHC), writing; E.B.: data analysis and visualization, experimentation (KI versus TOL2 imaging, CRISPR, injury analysis, scRNA-seq, infection); Y.N.W.: data analysis and visualization, experimentation (subtype and muscle imaging, genetic experiments, infection); A.M.R.: experimentation (generation of GFP knock-in and TOL2 lines, germ-free experimentation); A.B.: experimentation (FACS sorting and protocol development, bulk RNA-seq preparations); JZ: experimentation (KI versus TOL2 analysis); K.S.: experimentation (initial IHC analysis of injury); C.E.S.: conceptualization and experimental designs, data analysis and visualization, supervision, funding acquisition, writing- original draft and all revisions. All authors reviewed and edited the manuscript.

Corresponding author

Correspondence to Celia E. Shiau.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks So-ichiro Fukada, Victoriano Mulero and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information (download PDF )

Description of Additional Supplementary Files (download PDF )

Supplementary Data 1 (download XLSX )

Supplementary Data 2 (download XLSX )

Supplementary Movie 1 (download MP4 )

Supplementary Movie 2 (download MP4 )

Supplementary Movie 3 (download MP4 )

Supplementary Movie 4 (download MP4 )

Supplementary Movie 5 (download MP4 )

Supplementary Movie 6 (download MP4 )

Reporting Summary (download PDF )

Transparent Peer Review file (download PDF )

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Spencer, C.G., Hamilton, M., Bedsole, E. et al. Chronic macrophage activation derails muscle repair by disrupting mannose-receptor-linked plasticity revealed by endogenous irg1/acod1 tracking. Nat Commun 17, 1466 (2026). https://doi.org/10.1038/s41467-025-68204-3

Download citation

Received: 19 November 2024
Accepted: 19 December 2025
Published: 07 January 2026
Version of record: 09 February 2026
DOI: https://doi.org/10.1038/s41467-025-68204-3