The innate immune regulator MyD88 dampens fibrosis during zebrafish heart regeneration

Goumenaki, Pinelopi; Günther, Stefan; Kikhi, Khrievono; Looso, Mario; Marín-Juez, Rubén; Stainier, Didier Y. R.

doi:10.1038/s44161-024-00538-5

Download PDF

Article
Open access
Published: 13 September 2024

The innate immune regulator MyD88 dampens fibrosis during zebrafish heart regeneration

Nature Cardiovascular Research volume 3, pages 1158–1176 (2024)Cite this article

572 Accesses
1 Citations
20 Altmetric
Metrics details

Subjects

Abstract

The innate immune response is triggered rapidly after injury and its spatiotemporal dynamics are critical for regeneration; however, many questions remain about its exact role. Here we show that MyD88, a key component of the innate immune response, controls not only the inflammatory but also the fibrotic response during zebrafish cardiac regeneration. We find in cryoinjured myd88^−/− ventricles a significant reduction in neutrophil and macrophage numbers and the expansion of a collagen-rich endocardial population. Further analyses reveal compromised PI3K/AKT pathway activation in the myd88^−/− endocardium and increased myofibroblasts and scarring. Notably, endothelial-specific overexpression of myd88 reverses these neutrophil, fibrotic and scarring phenotypes. Mechanistically, we identify the endocardial-derived chemokine gene cxcl18b as a target of the MyD88 signaling pathway, and using loss-of-function and gain-of-function tools, we show that it controls neutrophil recruitment. Altogether, these findings shed light on the pivotal role of MyD88 in modulating inflammation and fibrosis during tissue regeneration.

Macrophages directly contribute collagen to scar formation during zebrafish heart regeneration and mouse heart repair

Article Open access 30 January 2020

Cardiac injury modulates critical components of prostaglandin E₂ signaling during zebrafish heart regeneration

Article Open access 20 February 2020

Dual genetic tracing reveals a unique fibroblast subpopulation modulating cardiac fibrosis

Article 23 March 2023

Main

Myocardial infarction is the most common cause of cardiac injury in humans, often leading to an irreversible loss of heart muscle tissue^1,2. Unlike the adult mammalian heart, the zebrafish heart can regenerate lost tissue upon different types of injuries, making it a powerful animal model to study heart regeneration^3,4,5,6.

Immediately after cardiac injury, tissue damage triggers the activation of a sterile inflammatory response, with immune cells infiltrating the injured area, clearing dead cells and debris, modifying the extracellular matrix (ECM) and initiating signaling cascades^7,8. As shown previously, this response is programmed to be resolved timely and is essential for triggering the regenerative response in both zebrafish^9,10,11 and neonatal mice¹². On the other hand, in the non-regenerative adult mammalian heart, inflammation persists longer and has been linked with tissue damage after myocardial infarction^13,14. Thus, it is essential that the inflammatory response is tightly controlled to allow for successful regeneration¹⁴.

Cardiac inflammation after injury involves the release of damage-associated molecular pattern (DAMP) molecules from dying and stressed cells, which bind to Toll-like receptors (TLRs)^15,16,17. Importantly, most TLRs, as well as interleukin-1 receptors (IL-1Rs), signal through the same adaptor molecule, myeloid differentiation factor 88 (MyD88)^18,19. While the MyD88 signaling pathway is involved in the activation of the immune response, a comprehensive understanding of its role in cardiac repair and regeneration remains elusive, with studies reporting conflicting results^14,20,21. As excessive and persistent inflammation has been linked with adverse regeneration outcomes, studies have often focused on blocking MyD88 or TLR function to restrict inflammation. Limiting inflammation in MyD88 or TLR loss-of-function (LOF) mouse models improved their regenerative potential^22,23,24. Conversely, TLR4-MyD88 activation in mesenchymal stem cells conferred cardioprotective effects in cardiac ischemia–reperfusion injury models in mouse²⁵, TLR4-MyD88 activation in cardiomyocytes (CMs) reduced their apoptosis in vitro²⁶, and MyD88 cardioprotective effects were highlighted in a rat aortic banding model²⁷. The critical role of MyD88 signaling in preserving cardiac function and limiting progression to heart failure was also shown in a dominant negative MyD88 transgenic mouse model²⁸. Depletion of MyD88 in T cells results in increased inflammation and fibrosis after transverse aortic constriction in mouse²⁹. While most of these studies were carried out in non-regenerative mammalian models, understanding the role of MyD88 in a regenerative system should provide insights into the pathogenesis of cardiovascular diseases and the development of new treatment approaches. In addition, the precise roles of the MyD88 signaling axis beyond cardiovascular inflammation and its contribution to other responses necessary for regeneration have so far been largely overlooked. Therefore, we directed our study toward unraveling the cell-specific functions of MyD88 signaling during cardiac regeneration in zebrafish.

In this study, we use newly generated genetic tools and transcriptomic profiling to investigate the role of MyD88 signaling during zebrafish cardiac regeneration at several stages and across different cell types; we find that MyD88 not only has a pivotal role in the immune response, but also a previously unidentified function in limiting endothelial-mediated fibrosis. Specifically, we show that in addition to its role in neutrophil and macrophage recruitment, MyD88 is important to limit a fibrotic response by regulating the transcriptome of endocardial and mesenchymal cells and by limiting myofibroblast numbers, fibrin levels and scar sizes. Furthermore, we identify a critical role for MyD88 signaling in regulating the phosphoinositide 3 kinase (PI3K)/AKT pathway in the injured endocardium. Notably, using an endothelial-specific mutant rescue strategy, we reveal the essential role of MyD88 in endothelial cells to promote neutrophil recruitment and limit the fibrotic response. Mechanistically, we identify the endocardial chemokine gene cxcl18b as a transcriptional target of the MyD88 signaling pathway and showed that it controls neutrophil recruitment.

Results

Reduced inflammatory cell numbers in injured myd88 ^−/− ventricles

Genes encoding MyD88 and MyD88 pathway-related signaling components are expressed in a wide range of tissues and cell populations even in the absence of tissue damage and infection^15,30,31. To begin to understand how MyD88 signaling affects different cell populations during cardiac regeneration in zebrafish, we dissociated cryoinjured ventricles from myd88 LOF mutants (myd88^−/−)³² and wild-type (WT) (myd88^+/+) siblings, sorted live cells and performed single-cell RNA sequencing (scRNA-seq) analysis (Fig. 1a and Supplementary Fig. 1a). We focused our analysis at the early time point of 24 hours post cryoinjury (hpci), when myd88 expression peaks in the cryoinjured zebrafish heart (Supplementary Fig. 1b)¹⁰. In addition, as the receptors of the MyD88 signaling pathway can rapidly recognize cardiac tissue damage through binding their ligands, 24 hpci is a good time point because it is within the early acute inflammatory phase³³. We performed quality control analysis, applied filtering cutoffs and obtained a dataset consisting of 11,735 cells (8,181 and 3,554 myd88^+/+ and myd88^−/− cells, respectively). To assess cell type diversity, we took into consideration known marker genes and identified five major cell cluster populations, including endocardial cells, coronary endothelial cells (cECs), myeloid cells, epicardial and epicardium-derived cells (EPDCs) and mesenchymal cells (Fig. 1b,c)^34,35,36. We did not detect a CM population, presumably because their large size was incompatible with our sample preparation pipeline as reported previously³⁶. As myeloid cells activate the MyD88 signaling pathway^10,22,37,38, we decided to perform unbiased subclustering of the myeloid population to investigate their diversity (Fig. 1d). This analysis revealed five distinct myeloid clusters, of which four expressed macrophage marker genes (for example, mpeg1.1, c1qa) at higher levels and one expressed neutrophil marker genes (for example, mpx, lyz) at higher levels (Fig. 1d and Extended Data Fig. 1a,b). We identified additional genes highly expressed in this presumed neutrophil cluster and indeed observed their enrichment in neutrophils in published datasets (Extended Data Fig. 1c)^34,35,36. Comparison of the myeloid populations between myd88^+/+ and myd88^−/− ventricles revealed pronounced differences, such as a strong reduction of the neutrophil and the marco macrophage clusters in myd88^−/− ventricles (Fig. 1d). We also observed that the expression of inflammatory genes was increased in these two clusters, suggesting that they contain pro-inflammatory cells (Fig. 1e). myd88^−/− ventricles also exhibited expansion of a macrophage subpopulation that is transcriptionally unique (enriched genes: gpx1a, mrc1b, timp4.2, lxn, lta4h, csf3r, cfh) and is not present in myd88^+/+ ventricles (Extended Data Fig. 1d,e).

**Fig. 1: Reduced numbers of pro-inflammatory cells in cryoinjured *myd88*^−/− ventricles.**

Neutrophils are known pro-inflammatory mediators and immediate responders to tissue damage^39,40,41. They can shape the regenerative outcome because they influence important regeneration events, including clearance of dead cells and debris from the injured area, macrophage polarization, effective revascularization, epicardial activation and myocardial regeneration^{42,43,44,45,46,47}. The MyD88 signaling axis, once activated, facilitates neutrophil recruitment^22,23,37,48. To further investigate the neutrophil phenotype observed in cryoinjured myd88^−/− ventricles (Fig. 1d), we used the TgBAC(mpx:GFP) neutrophil reporter line to determine neutrophil numbers after injury. While at 6 hpci we did not observe obvious differences between myd88^−/− and myd88^+/+ sibling ventricles (Extended Data Fig. 1f,g), at 24 hpci neutrophil numbers were significantly reduced in myd88^−/− ventricles and their numbers remained reduced at least until 96 hpci (Fig. 1f,g). We also counted the number of neutrophils at 6 days post sham (dps) surgery and 7 days post cryoinjury (dpci), but did not observe any significant differences at these time points (Extended Data Fig. 1f,g). Together these data indicate that neutrophil numbers were reduced at 24 and 96 hpci in myd88^−/− ventricles.

As with neutrophils, the MyD88 signaling axis contributes to macrophage recruitment^10,37,38,49. Thus, we used the Tg(mpeg1:EGFP) macrophage reporter line to determine whether macrophage numbers were also affected in cryoinjured myd88^−/− ventricles. Both scRNA-seq analysis and immunostaining for mpeg1:EGFP⁺ cells revealed that total macrophage numbers were not affected in myd88^−/− ventricles at 24 hpci (Fig. 1h,i and Extended Data Fig. 1h). While we did not observe a significant reduction in total macrophage numbers at 24 hpci, our transcriptomic data showed that some macrophage populations might be affected, with the marco macrophage population being severely reduced and the hp-1 macrophage population clearly increased in myd88^−/− ventricles (Fig. 1d and Extended Data Fig. 1d,e). As the macrophage response initiates and develops later than the neutrophil response after injury, we set out to analyze cryoinjured ventricles at a later time point, when a macrophage phenotype might be more obvious. To this end, we selected 96 hpci, a time point when macrophage numbers increases significantly in WT zebrafish^10,34,50,51, and indeed found a reduction in mpeg1:EGFP⁺ cells in myd88^−/− ventricles compared with myd88^+/+ siblings (Fig. 1h,i).

Taken together, these transcriptomic analyses and immunostaining data reveal that during zebrafish cardiac regeneration, MyD88 is important for neutrophil and macrophage recruitment and for enhancing the inflammatory state of immune cells.

MyD88 signaling attenuates fibrosis in injured ventricles

Besides analyzing myeloid cells, the most extensively studied activators of the MyD88 signaling pathway^10,22,37,38, we wanted to determine which other cell types were important for MyD88 function during cardiac regeneration. As reported previously, endothelial cells express myd88 and components of the MyD88 signaling axis, such as the tlr genes^52,53. To investigate the state and role of endocardial cells in the absence of MyD88 function during regeneration, we performed unbiased subclustering analysis of the endocardial population in our scRNA-seq dataset. This analysis revealed the presence of four distinct endocardial clusters, represented in different proportions in cryoinjured myd88^−/− and myd88^+/+ ventricles (Fig. 2a). Specifically, the most abundant endocardial cell cluster found in myd88^+/+ ventricles (irx5a endocardial cells) was significantly reduced in myd88^−/− ventricles. Interestingly, the most abundant endocardial cluster in myd88^−/− ventricles displayed high expression levels of genes encoding collagens (col1a2, col1a1a, col5a1, col1a1b, col5a2a, col6a2, col6a1) and genes associated with fibrosis (postna, gstm.3, sparc, acta2)^34,54. Hence, we annotated this cluster as ‘collagen-rich endocardial cells’ (Fig. 2a,b). Pseudotime trajectory analysis indicated that cells transition from the wound endocardial cluster serpine1 (ref. ⁵⁵) to the other endocardial clusters (Extended Data Fig. 2a). To investigate the relationship between endocardial clusters, we also performed velocity analysis. The resulting data indicate that endocardial cells in myd88^−/− ventricles tend to progress toward the collagen-rich endocardial state (Extended Data Fig. 2b). As shown previously in mouse^56,57,58 and zebrafish^35,59, the endocardium contributes to the activated fibroblast cell population and to α-smooth muscle actin (αSMA)⁺ myofibroblasts in injured hearts; thus, increased expression of collagen and fibrotic genes in the cryoinjured myd88^−/− endocardium (Fig. 2a,b) could reflect an increase in these endocardium-derived cells. αSMA⁺ myofibroblasts are the key contributors to ECM fibrotic remodeling and scar formation in the injured heart^56,58,60. Moreover, in the context of MyD88 signaling, MyD88 depletion in T cells results in increased cardiac fibroblast to myofibroblast transformation in vitro, as indicated by elevated αSMA and collagen type 1 expression in fibroblasts²⁹. To investigate the role of MyD88 in myofibroblast differentiation during zebrafish cardiac regeneration, we immunostained for αSMA expression on sections of cryoinjured ventricles carrying the ET(krt4:EGFP) endocardial enhancer trap line (Fig. 2c). At 96 hpci, we found an increased abundance of αSMA⁺ cells in myd88^−/− compared with myd88^+/+ siblings (Fig. 2c,d), which remained elevated until 7 dpci (Extended Data Fig. 2c,d). To differentiate between αSMA⁺ cells of different origins, we quantified intraventricularly localized αSMA⁺ cells, presumably of endocardial and fibroblast origin, and excluded superficially localized αSMA⁺ cells, presumably of epicardial, fibroblast and perivascular origin^35,61. In line with the fibrotic transcriptomic profile of the myd88^−/− endocardium (Fig. 2a,b), at 96 hpci we found an increased number of intraventricular αSMA⁺ cells in myd88^−/− zebrafish compared with myd88^+/+ siblings (Fig. 2d). Superficially localized αSMA⁺ cells were also elevated in cryoinjured myd88^−/− ventricles (Extended Data Fig. 2e), but not to the same extent as the intraventricularly localized αSMA⁺ cells. Altogether, these data indicate that MyD88 suppresses myofibroblast differentiation during zebrafish cardiac regeneration.

**Fig. 2: Fibrotic phenotype in cryoinjured *myd88*^−/− ventricles.**

Activation of MyD88 signaling in mesenchymal stem cells reportedly conferred cardioprotective effects via Stat3 in a mouse model of cardiac ischemia–reperfusion injury²⁵. As myofibroblasts, a mesenchymal cell type, are affected in myd88^−/− ventricles, we decided to take a closer look at the mesenchymal population in our scRNA-seq dataset. Unbiased subclustering analysis indicated the presence of three mesenchymal subclusters. Strikingly, strong differences between the myd88^−/− and myd88^+/+ samples resulted in one cluster (hapln1a mesenchymal cells) being of almost completely myd88^−/− and another (cst14a.1 mesenchymal cells) being of almost completely myd88^+/+ identity (Extended Data Fig. 2f). Specifically, the myd88^−/− hapln1a mesenchymal cell cluster exhibited increased expression of endothelial-to-mesenchymal transition and fibrotic-related genes, such as sox9a and acta2 (Extended Data Fig. 2g,h)⁵⁹.

As our data pointed to an endocardial-related fibrotic phenotype in cryoinjured myd88^−/− ventricles, we next analyzed additional endocardial processes that could be affected in the absence of myd88 function. Endocardial hyperinvasion of the injured area has been linked with increased endothelial-to-mesenchymal transition and fibrotic remodeling⁵⁹. However, we did not observe any obvious differences between myd88^−/− and myd88^+/+ siblings with respect to the krt4:EGFP⁺ and Cdh5⁺ endocardial cell area covering the injury at 96 hpci (Extended Data Fig. 3a,b) or 7 dpci (Extended Data Fig. 3c,d), respectively. Additionally, the Aldh1a2 activation pattern was similar in myd88^−/− and myd88^+/+ sibling ventricles at 96 hpci (Extended Data Fig. 3a)⁵⁵, indicating that only endothelial-to-mesenchymal transition processes were affected in cryoinjured myd88^−/− ventricles.

Blocking MyD88 signaling in uninjured murine hearts leads to increased fibrosis²⁸. To investigate the regenerative potential and scar resolution abilities of cryoinjured myd88^−/− ventricles, we performed acid fuchsin orange G (AFOG) staining. As reported previously, scar tissue in 14 dpci WT hearts consists of an extensive collagen network and a ring-like peripheral fibrin structure^5,33,62. In contrast, the scar tissue in myd88^−/− ventricles at 14 dpci displayed thicker fibrin-rich areas in comparison with their myd88^+/+ siblings (Fig. 2e), and these increased fibrin levels persisted until at least until 30 dpci (Fig. 2e–g). However, the collagen proportions within the scars remained similar between myd88^−/− and myd88^+/+ siblings (Extended Data Fig. 4a). Additionally, at 30 dpci, there were more myd88^−/− ventricles with bigger scar areas compared with their myd88^+/+ siblings (Extended Data Fig. 4b). We also examined scars at 90 dpci to analyze fibrotic tissue persistence, but did not observe significant differences at this time point (Extended Data Fig. 4c,d).

In summary, the fibrotic-like transcriptomic profile (enrichment in collagen and fibrotic genes) observed in both endocardial and mesenchymal cells, along with the increased number of total and intraventricular αSMA⁺ cells, increased fibrin levels and larger scars in cryoinjured myd88^−/− hearts, collectively indicate that MyD88 signaling has a role in limiting fibrosis during zebrafish cardiac regeneration.

Endocardial MyD88 signaling activates the PI3K/AKT pathway

To gain mechanistic insights into how MyD88 signaling regulates endocardial-specific processes during regeneration, we sorted endocardial cells from myd88^−/− and myd88^+/+ siblings using the ET(krt4:EGFP) line at 96 hpci and conducted bulk RNA-seq analysis (Fig. 3a and Supplementary Fig. 2). In line with the previously observed fibrotic phenotype (Fig. 2), gene set enrichment analysis (GSEA) revealed an increase in the regulation of the cellular response to transforming growth factor-β stimulus in the cryoinjured myd88^−/− endocardium (Extended Data Fig. 5a). Additionally, GSEA (Extended Data Fig. 5a) and closer examination of the most differentially expressed genes (Fig. 3b) revealed that immune-associated processes, such as TLR, nucleotide-binding oligomerization domain (NOD)-like receptor, C-type lectin receptor and cytosolic DNA-sensing signaling pathways, as well as the mitogen-activated protein kinase (MAPK) signaling pathway, were impaired in the myd88^−/− endocardium. As the MAPK signaling pathway has been linked with the activated endocardium in both the cardiac resection^63,64 and cryoinjury⁵¹ models, we immunostained for phosphoERK (pERK), an indicator of activated MAPK signaling, in ET(krt4:EGFP) ventricles and verified its presence in the endocardium at 96 hpci. However, we did not observe statistically significant differences in the pERK⁺ cell area covering the injury between myd88^−/− and myd88^+/+ siblings (Extended Data Fig. 6a,b), indicating that defective MAPK signaling was not responsible for the endocardial phenotype observed in myd88^−/− ventricles. This observation is in line with a previous study where lipopolysaccharide treated Myd88^−/− mice displayed activation of the MAPK signaling pathway⁶⁵.

**Fig. 3: The PI3K/AKT signaling pathway is suppressed in the *myd88*^−/− endocardium.**

GSEA and examination of the most differentially expressed genes (DEGs) (Fig. 3b, Supplementary Table 1 and Extended Data Fig. 5a,b) further indicated that activation of the PI3K/AKT pathway is impaired in the myd88^−/− endocardium. Additionally, GSEA terms related to upstream (ErBB, IGF, VEGF, FGF and NGF pathways) and downstream components (mTOR pathway) of PI3K/AKT signaling were substantially downregulated in the cryoinjured myd88^−/− endocardium (Extended Data Fig. 5a,b)^{64,66,67,68,69}. Interestingly, MyD88 signaling can also regulate PI3K/AKT pathway activation^28,70,71. Hence, we costained for phospho-Akt (pAkt), which indicates activation of PI3K/AKT signaling when present in the nucleus, together with the nuclear endothelial cell marker Fli1 in sections of ET(krt4:EGFP) ventricles and found a tendency for reduced activation of the PI3K/AKT pathway in myd88^−/− endocardial cells at 96 hpci (Fig. 3c,d). Given that PI3K/AKT activation affects cell proliferation^68,72, we then assessed endocardial proliferation at 96 hpci, when it reaches high levels in WT^55,68. Notably, we found significantly reduced proliferation in myd88^−/− endocardial cells at 96 hpci (Fig. 3e,f).

In summary, these findings underscore the critical role of MyD88 in regulating signaling cascades in the injured endocardium. Particularly, we identified impairments in the immune response, in the activation of the PI3K/AKT pathway and in endocardial cell proliferation in cryoinjured myd88^−/− ventricles.

MyD88 signaling promotes revascularization and CM repopulation

Revascularization is one of the earliest processes observed in the injured area and it is vital for effective cardiac tissue regeneration^73,74. Following cryoinjury of the zebrafish ventricle, cECs undergo proliferation and initiate the sprouting of new vessels, both superficially and intraventricularly. While intraventricular revascularization is orchestrated by endocardial vascular endothelial growth factor A signaling, superficial revascularization is partially regulated by Apelin signaling⁷⁴. We found that both endocardial VEGF and Apelin signaling pathways were impaired in cryoinjured myd88^−/− ventricles (Extended Data Fig. 5a). Additionally, neutrophils, which were significantly reduced in myd88^−/− ventricles (Fig. 1d,f,g), express VEGF and promote revascularization^42,44,46. Hence, we used the Tg(-0.8flt1:RFP) reporter line, which labels cECs and assessed their proliferation at 96 hpci, when cEC proliferation peaks in cryoinjured WT ventricles⁷⁴. We found that cEC proliferation was significantly reduced at 96 hpci in myd88^−/− hearts when compared with myd88^+/+ siblings (Extended Data Fig. 7a,b). While cEC proliferation recovered by 7 dpci (Extended Data Fig. 7c,d), coronary vessel coverage in the injured area remained impaired at 7 dpci (Extended Data Fig. 7e,f) in myd88^−/−. Together, these data reveal that revascularization of the injured area was affected in cryoinjured myd88^−/− ventricles.

During cardiac regeneration in zebrafish, CMs undergo dedifferentiation and proliferation to repopulate the injured tissue^75,76,77 and this process is heavily influenced by the microenvironment surrounding the CMs⁷⁸. Given the altered inflammatory and fibrotic environment in cryoinjured myd88^−/− ventricles, along with the impaired numbers and functions of various cell types that precede CM appearance in the injured area, we decided to assess CM behavior. We observed a significant reduction in CM proliferation in the injury border zone at 96 hpci and 7 dpci in myd88^−/− cryoinjured ventricles (Fig. 4a–d). We also examined CM proliferation at 14 dpci and dedifferentiation at 96 hpci⁷⁹ but did not find any significant differences (Extended Data Fig. 8a–d). Furthermore, we quantified CM protrusive activity toward the injury at 72 hpci and 7 dpci and found that while the number of protrusions was the same between myd88^−/− and myd88^+/+ siblings, CM protrusions were shorter in myd88^−/− ventricles at 7 dpci (Fig. 4e–g). These results highlight the essential role of MyD88 in facilitating efficient CM proliferation and protrusion toward the injured tissue during cardiac regeneration.

**Fig. 4: Reduced CM proliferation and reduced length of CM protrusions toward the injured tissue in cryoinjured *myd88*^−/− ventricles.**

Endothelial myd88 overexpression reverses myd88 ^−/− phenotypes

To investigate whether the loss of MyD88 in endothelial cells was the contributing factor for the fibrotic phenotype observed in cryoinjured myd88^−/− ventricles, we specifically overexpressed (OE) myd88 in endothelial cells. For this purpose, we generated a Tg(fli1a:myd88,EGFP) line that expresses myd88 and EGFP under the control of a bidirectional fli1a promoter. As expected, myd88 expression was increased in transgenic zebrafish when compared with non-transgenic sibling larvae and adult cryoinjured ventricles (Extended Data Fig. 9a), thereby validating the overexpression ability of the tool. In addition, the transgenic EGFP signal colocalized with the endocardial Aldh1a2 signal in cryoinjured ventricles, validating the endothelial-specific expression profile of the Tg(fli1a:myd88,EGFP) line (Extended Data Fig. 9b). We analyzed myd88^−/− and myd88^+/+ siblings carrying the overexpression transgene and found that the previously elevated αSMA⁺ cells (total and intraventricular) (Fig. 2c,d) not only reverted to control levels, but also significantly decreased in cryoinjured myd88^−/− ventricles at 96 hpci (Fig. 5a,b). To assess whether endothelial-specific myd88 overexpression could improve the regeneration potential of cryoinjured myd88^−/− ventricles, we analyzed scars at 30 dpci. Remarkably, Tg(fli1a:myd88,EGFP); myd88^−/− and myd88^+/+ sibling ventricles were indistinguishable in terms of fibrin abundance and scar area size (Fig. 5c–f). Altogether, these data further indicate that endothelial MyD88 function is critical to limit fibrosis in the regenerating zebrafish heart.

**Fig. 5: *myd88* overexpression in endothelial cells rescues the fibrotic, scarring and neutrophil phenotypes in cryoinjured *myd88*^−/− ventricles.**

Endothelial cells activate immune-regulatory programs, secrete cytokines and chemokines, and facilitate early immune cell recruitment in response to cardiac injury^14,52,55,80. They also express MyD88 and components of the MyD88 signaling axis, including TLRs^52,53. To determine whether myd88 overexpression in endothelial cells could also restore the low neutrophil levels observed in cryoinjured myd88^−/− ventricles (Fig. 1d,f,g), we quantified the number of Mpx⁺ cells and found that neutrophil levels were indeed restored at 96 hpci (Fig. 5g,h), indicating that MyD88 function in endothelial cells is sufficient to rescue the neutrophil phenotype. Overall, these endothelial-specific myd88 overexpression experiments support the model that endothelial-specific MyD88 activation promotes neutrophil recruitment in the cryoinjured heart.

MyD88-Cxcl18b signaling promotes neutrophil recruitment after cardiac injury

To gain a deeper understanding of the targets of the MyD88 signaling pathway during zebrafish cardiac regeneration, we conducted bulk RNA-seq analysis on myd88^−/− and myd88^+/+ untouched (UT) ventricles and injured tissues at 1 and 24 hpci (Fig. 6a). As anticipated, we observed decreased expression of several immune-related genes in cryoinjured myd88^−/− tissues, including cxcl18b (Fig. 6b). Through cross-examination of our transcriptomic datasets with publicly available datasets, we found that cxcl18b was significantly upregulated (P_adj < 0.05 and FDR > 1) at early time points after cardiac injury in WT zebrafish^10,81,82. cxcl18b is also upregulated in zebrafish after infection and fin amputation, where it regulates neutrophil recruitment^32,83,84,85. Importantly, a previous study reported that cxcl18b upregulation is downstream of MyD88 signaling in infection models³²; however, the specific role of Cxcl18b during cardiac regeneration remains to be elucidated. Regarding its expression pattern, cxcl18b is expressed in caudal hematopoietic and endothelial cells in infected zebrafish larvae⁸³ and in endocardial cells in regenerating zebrafish hearts³⁴. We used a Tg(cxcl18b:EGFP) reporter line⁸³ to look more closely at cxcl18b expression and observed clear EGFP expression in endocardial and epicardial cells at 24 hpci (Extended Data Fig. 10a).

**Fig. 6: *cxcl18b* is activated by MyD88 signaling and controls neutrophil recruitment.**

We next investigated cxcl18b expression using real-time quantitative polymerase chain reaction (RT–qPCR) and observed clear downregulation in cryoinjured myd88^−/− ventricles compared with myd88^+/+ siblings (Fig. 6c). We then developed gain-of-function and LOF genetic tools to investigate Cxcl18b function during heart regeneration. As Cxcl18b is a chemokine that attracts neutrophils, we used the gain-of-function tool to determine whether cxcl18b overexpression could restore the low neutrophil levels observed in cryoinjured myd88^−/− ventricles (Fig. 1d,f,g). For this purpose, we generated a Tg(hsp70l:LBL-cxcl18b-t2a-mCherry) line, which allows for conditional overexpression of cxcl18b under the control of the hsp70l promoter upon Cre-mediated recombination using the HOTcre system⁸⁶. To achieve global cxcl18b overexpression, we injected transgenic embryos with Cre mRNA and performed heat shocks (Extended Data Fig. 10b). Successful overexpression of cxcl18b and mCherry was observed using RT–qPCR (Extended Data Fig. 10c). We raised Cre mRNA-injected (cxcl18b overexpression) myd88^+/− and myd88^−/− zebrafish and uninjected (control) myd88^+/− and myd88^−/− Tg(hsp70l:LBL-cxcl18b-t2a-mCherry); TgBAC(mpx:GFP) siblings to adulthood. We then performed heat shocks and cryoinjury and quantified the number of mpx:GFP⁺ cells at 24 hpci (Fig. 6d). Cre mRNA-mediated recombination was validated by the detection of the recombination marker, mCherry (Extended Data Fig. 10d). While the number of mpx:GFP⁺ cells was reduced in control cryoinjured myd88^−/− ventricles, when we overexpressed cxcl18b, there was no significant difference in neutrophil numbers between myd88^−/− and myd88^+/− samples, indicating that cxcl18b overexpression is sufficient to at least partially rescue the low neutrophil numbers observed in cryoinjured myd88^−/− ventricles (Fig. 6e,f).

To understand how loss of Cxcl18b might affect neutrophil recruitment during cardiac regeneration, we generated a cxcl18b mutant allele using the CRISPR–Cas9 technology. As many chemokines have high sequence homology and to avoid transcriptional adaptation events^87,88, we specifically generated a full locus deletion allele (Extended Data Fig. 10e). To assess cxcl18b expression levels, we performed larval fin fold amputations, which trigger cxcl18b upregulation in WT cells^84,85, and collected larvae 6 hours post amputation (hpa) for RT–qPCR analysis. As anticipated, cxcl18b full locus deletion mutants completely lacked cxcl18b expression (Extended Data Fig. 10f). To examine whether lack of Cxcl18b could affect neutrophil recruitment in cryoinjured ventricles, we quantified neutrophil abundance at 24 and 96 hpci. Notably, while the Mpx⁺ cell count was unaffected at 24 hpci, there was a significant difference between cxcl18b^−/− and cxcl18b^+/+ siblings at 96 hpci (Fig. 6g,h).

In summary, these findings together indicate that Cxcl18b is a chemokine whose expression is upregulated upon cardiac cryoinjury by MyD88 and is at least partly responsible for neutrophil recruitment to the injured area.

Discussion

After cardiac injury, the zebrafish heart mounts a robust innate immune response, which is crucial for the regeneration process^9,10,11. The precise regulation of this early immune response is essential, underscoring the need for a detailed understanding of its regulators and underlying mechanisms¹⁴. Key activators of this response are the TLR and IL-1R signaling pathways, which signal predominantly through the adapter protein MyD88 (refs. ^16,17,18,19). While previous studies mostly focused on immune cells to investigate MyD88 function, in this study we examined other cell types of the heart and identified an unexplored function for MyD88: not only does it regulate the inflammatory response, but it also limits the endocardial fibrotic response to injury. Specifically, we observed a significant reduction in pro-inflammatory neutrophil and macrophage populations in cryoinjured myd88^−/− ventricles. More surprisingly, we also observed in myd88^−/− ventricles (1) the expansion of a collagen-rich endocardial population, (2) compromised endocardial PI3K/AKT pathway activation, (3) an increased number of myofibroblasts and (4) increased fibrin levels, as well as bigger scars. Our data further revealed that lack of MyD88 signaling impairs CM behavior in the injured area. By endothelial cell-specific overexpression of myd88, we showed that it has an essential role in the endothelial response to injury by regulating neutrophil recruitment to the injury site as well as fibrosis. Mechanistically, we identified the chemokine gene cxcl18b as a target of the MyD88 signaling pathway that controls neutrophil recruitment (Fig. 7). Overall, these data highlight the beneficial role of MyD88 activation in the regenerative response to cardiac injury in zebrafish and provide insights on how pathways activated very quickly after injury can shape the regenerative outcome.

**Fig. 7: Proposed model for the role of MyD88 in fibrosis and neutrophil recruitment following cardiac cryoinjury in zebrafish.**

While some studies suggested that MyD88 signaling promotes the regenerative process, others reported improved outcomes when MyD88 signaling was blocked^14,20,21. Our data clearly show that MyD88 is a beneficial molecule for the regenerative process in zebrafish. The discrepancy with studies reporting detrimental effects of MyD88 signaling on regeneration mostly lies in the model used. Specifically, while inflammation is essential in the early stages of regeneration^9,10,11, prolonged and excessive inflammation can lead to adverse outcomes^14,20. In non-regenerative mouse models, inflammation tends to be uncontained; thus, limiting inflammation by blocking MyD88 signaling can enhance the regeneration potential^{22,23,24,89,90,91}. Accordingly, blocking MyD88 function has been linked with protective effects in experimental settings (for example, endotoxin shock) after lipopolysaccharide administration^23,92,93. Once again, these data reflect special cases where inflammation is uncontrolled and excessive.

In addition, following myd88 overexpression in endothelial cells, we successfully reversed the reduced neutrophil number and the enhanced fibrosis phenotypes observed in cryoinjured myd88^−/− ventricles, indicating that MyD88 is required in endothelial cells for the regenerative response. However, myd88 overexpression in endothelial cells in cryoinjured myd88^+/+ ventricles did not lead to an increase in neutrophil numbers (Figs. 1g and 5h) or an improvement in the fibrotic response (Figs. 2c–g and 5a–f), indicating that MyD88 signaling is optimally tuned in WT conditions. Studies reported that endothelial-specific MyD88-deficient mice exhibit features of decreased inflammation⁹⁴, while stimulation of the MyD88 pathway in aortic endothelial cells in rabbits promoted inflammation⁹⁵. In light of our data, it will be interesting to investigate in injured mammalian hearts which cells activate MyD88 signaling and the function of the MyD88 pathway in endocardial cells.

Myeloid cells serve as key activators of the MyD88 signaling pathway in response to injury or infection^{22,37,38,39,40}. The myd88 LOF mutant used in this study has also been used in the context of larval zebrafish tail fin regeneration, where it was reported that MyD88 controls neutrophil and macrophage recruitment to the wound³⁷. In line with these findings, we observed reduced neutrophil and macrophage numbers in cryoinjured myd88^−/− ventricles. However, we do not anticipate major differences between myd88^−/− and WT zebrafish in the absence of an insult (injury or infection) that could trigger MyD88 pathway activation. As shown previously, early leukocyte hematopoiesis, migration, basal motility and phagocytosis are not affected by MyD88 deficiency^32,37. In addition, transcriptomic analysis from uninfected myd88^−/− and WT embryos did not reveal any differences in immune-related genes other than myd88, which was explained by the lower stability of the mutant transcript³². In our study, we also did not observe any significant differences in neutrophil count between sham-injured myd88^−/− and myd88^+/+ ventricles (Extended Data Fig. 1f,g). Altogether, these data indicate that the observed phenotypes in cryoinjured myd88^−/− ventricles are due to the mutant’s inability to activate the pathway in response to injury rather than to preexisting developmental defects.

We expanded our investigation beyond the conventional activators of the MyD88 signaling pathway, myeloid cells, to explore the impact of the loss of MyD88 function on other cell types. Our transcriptomic analysis pointed to endocardial cells, which in myd88^−/− ventricles display a fibrotic phenotype. Bulk RNA-seq analysis of endocardial cells revealed impairment in the activation of the MAPK and PI3K/AKT pathways in the myd88^−/− endocardium. We elected to focus on MyD88 function in the PI3K/AKT pathway because it was more strongly affected and has not been explored as extensively as the MAPK pathway in the endocardial response during cardiac regeneration in zebrafish^51,63,64. For these and other studies, we then overexpressed myd88 in endothelial cells in cryoinjured myd88^−/− ventricles, which led to the reversal of the reduced neutrophil count and the enhanced fibrosis. Taken together, these findings underscore the importance of the MyD88 signaling pathway in endocardial cells and thus support the hypothesis that at least some of the myd88^−/− phenotypes after cardiac injury could originate mostly in the endocardium. Unlike most myeloid cells, which need to be recruited to the injured tissue, endocardial cells are inherently present and abundant in the heart^36,96. Additionally, a previous study showed that MyD88 signaling in hematopoietic cells alone was not sufficient to trigger the inflammatory response in the injured mouse heart⁹⁷. Specifically, myocardial infarction in Myd88^−/− and WT mice reconstituted with WT bone marrow cells led to no differences in cytokine or chemokine levels⁹⁷. Furthermore, endothelial cells activate immune-regulatory programs, secrete cytokines and chemokines, and initiate immune cell recruitment^52,55,80. They express the tlr and myd88 genes^52,53 and at least a subgroup of endothelial cells, vascular endothelial cells, rely exclusively on MyD88 for TLR activation⁹⁸. Additionally, signaling cascades commonly activated by MyD88, such as nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) and PI3K/AKT, get activated in endothelial cells^69,99. These studies, together with our data, emphasize the pivotal role of early MyD88 signaling activation in endocardial cells and support the hypothesis that the fibrotic phenotype in cryoinjured myd88^−/− ventricles initiates in the endocardium.

During cardiac development and regeneration, the endocardium has a crucial role in signaling to CMs^55,100, while the ECM supports CM behavior^101,102,103. Given the altered inflammatory and fibrotic environment observed in cryoinjured myd88^−/− ventricles, along with the impaired endocardial response preceding CM appearance in the injured area, we investigated CM behavior. Our findings revealed decreased CM proliferation and reduced CM protrusion length in cryoinjured myd88^−/− ventricles. We hypothesize that the myd88^−/− CM environment is not sufficiently permissive for CM protrusions to extend as they do in myd88^+/+ siblings. However, we cannot exclude the possibility that CMs also activate MyD88 signaling, as activity of NF-κB, a MyD88 target, has also been reported in CMs¹⁰⁴. Unfortunately, our scRNA-seq analysis did not manage to capture a CM population; similar challenges with CM populations have been reported in other transcriptomic studies^36,105. In addition, it is probable that epicardial cells also activate MyD88 signaling during cardiac regeneration in zebrafish. The number of superficially localized αSMA⁺ cells, presumably of epicardial, fibroblast and perivascular origin, was also elevated in cryoinjured myd88^−/− ventricles (Extended Data Fig. 2e). To draw more definitive conclusions about the role of MyD88 in CMs, epicardial cells and other cardiac cell populations, additional cell type-specific studies will need to be performed.

Our data also reveal that the expression of Cxcl18b, a neutrophil chemoattractant produced in the endocardium, is induced after MyD88 activation. Notably, we observed that cxcl18b overexpression partially rescued the reduced neutrophil count in cryoinjured myd88^−/− ventricles (Fig. 6e,f). Neutrophil recruitment via Cxcl18b was previously reported not to be dependent on the dosage of Cxcl18b because receptor saturation might occur or neutrophils might not be able to get stimulated further⁸³. Our inability to fully rescue neutrophil numbers with cxcl18b overexpression might, at least in part, be due to these issues or the involvement of other chemoattractants. We also generated a cxcl18b full locus deletion allele and showed that homozygous mutants displayed significantly reduced neutrophil numbers at 96 hpci. As there was no clear difference in neutrophil count at 24 hpci between cryoinjured cxcl18b^−/− and cxcl18b^+/+ ventricles, we again speculate that Cxcl18b is one of several chemokines secreted after injury and required for neutrophil recruitment⁸³, and that its absence alone is not sufficient to trigger a difference at early time points.

In summary, our findings highlight the crucial role of MyD88 signaling in cardiac regeneration. We propose that MyD88 regulates both the inflammatory and fibrotic responses and that its early activation in the endocardium is vital for successful regeneration. Additionally, we identified the chemokine Cxcl18b as a target of MyD88 signaling that contributes to neutrophil recruitment. Gaining a deeper understanding of the early innate immune mechanisms activated in the injured heart is of vital importance because they can influence subsequent processes and thereby impact the regenerative outcome. In addition, a comprehensive understanding of the MyD88 pathway’s contribution to the regenerative response will determine whether targeting a molecule of the innate immune system could be considered as a treatment target to limit pathogenesis in cardiovascular diseases.

Methods

Zebrafish husbandry and handling

Zebrafish larvae were raised under standard conditions. Adult fish were maintained in 3.5-l tanks at a stock density of ten fish per liter with the following parameters: water temperature, 27–27.5 °C; light–dark cycle, 14:10; pH, 7.0–7.5; conductivity, 750–800 µS cm⁻². Zebrafish were fed 3–5 times a day, depending on age, with granular and live food (Artemia salina). Health monitoring was performed twice a year. All procedures performed on animals conformed to the guidelines from Directive 2010/63/EU of the European Parliament on the protection of animals used for scientific purposes and were approved by the Animal Protection Committee (Tierschutzkommission) of the Regierungspräsidium Darmstadt (reference nos. B2/1218 and B2/1229).

Zebrafish lines

The following mutant and transgenic lines were used: myd88^hu3568 (ref. ³²), cxcl18b^bns683 (this study), TgBAC(mpx:GFP)ⁱ¹¹⁴ (ref. ¹⁰⁶), Tg(mpeg1:EGFP)^gl22 (ref. ¹⁰⁷), ET(krt4:EGFP)^sqet33-1A ref. ¹⁰⁸, abbreviated as ET(krt4:EGFP), Tg(-0.8flt1:RFP)^hu5333 (ref. ¹⁰⁹), Tg(fli1a:myd88,EGFP)^bns703 (this study), Tg(cxcl18b:EGFP)^ibl150 (ref. ⁸³) and Tg(hsp70l:loxP-TagBFP-loxP-cxcl18b-t2a-mCherry)^bns660 (this study), abbreviated as Tg(hsp70l:LBL-cxcl18b-t2a-mCherry).

Generation of zebrafish mutant and transgenic lines

Sequence analysis for the generation of new lines was performed using the ApE software (v.2.0.61). To generate the cxcl18b^bns683 full locus deletion allele, the CRISPR–Cas9 technology was used, as described previously^110,111,112. The single-guide RNAs (sgRNAs) GGAAAGTTACAAGGAAATGC and GGAAGTTTGGATGATTCTAA were designed with a CRISPR design tool (https://www.crisprscan.org/) targeting sequences upstream and downstream of the 5′ and 3′ untranslated regions, respectively. sgRNAs were transcribed using a MEGAshortscript T7 Kit (Thermo Fisher Scientific) and later purified with the RNA Clean & Concentrator Kit (Zymo Research). Then, 50 pg of each sgRNA were coinjected with 100 pg of cas9 mRNA into zebrafish one-cell-stage WT embryos. Injected embryos were raised to adulthood and later screened for founder identification. To detect the full locus deletion allele, we performed PCR using the forward 5′-GTACCCTGGTTAATGACTAATCCTAGTT-3′ and reverse 5′-AGCTGATCAGAACGCACAGTAACG-3′ primers, leading to a 310-bp product. To detect the WT allele, we performed PCR using the forward 5′- ATCCACAAAAACAGCAGGGC-3′ and reverse 5′- CATCTTCAGCGAGTCGGTGT-3′ primers, leading to a 739-bp product.

To generate the Tg(fli1a:myd88,EGFP)^bns703 line, in which a bidirectional fli1a promoter drives the expression of both myd88 and EGFP, we modified the fli1a:mCherry,EGFP construct provided by C. Helker. The myd88 coding sequence was cloned and subsequently used to replace the mCherry coding sequence. Cloning was performed using the In-Fusion HD Cloning Kit (Takara Bio). Then, 10 pg of the final construct were coinjected with 25 pg of Tol2 mRNA into zebrafish one-cell-stage WT embryos. Injected embryos positive for EGFP were raised to adulthood and later screened for founder identification.

To generate the Tg(hsp70l:loxP-TagBFP-loxP-cxcl18b-t2a-mCherry)^bns660 line, the hsp70l:loxP-TagBFP-loxP-il11ra-t2a-mCherry construct was modified⁵⁹. The cxcl18b coding sequence was cloned and subsequently used to replace the il11ra coding sequence. Cloning was performed using the In-Fusion HD Cloning Kit. Then, 10 pg of the final construct were coinjected with 25 pg of Tol2 mRNA into zebrafish one-cell-stage WT embryos. Injected embryos were heat-shocked and embryos positive for TagBFP were raised to adulthood and later screened for founder identification. To induce recombination, 12.5 pg of Cre mRNA were injected into Tg(hsp70l:loxP-TagBFP-loxP-cxcl18b-t2a-mCherry) one-cell-stage embryos. Cre mRNA-injected embryos were heat-shocked and embryos positive for mCherry were raised to adulthood.

Cardiac cryoinjury and heat shock treatments

Cardiac cryoinjury was performed in adult zebrafish hearts as described previously^3,4,5. Zebrafish were briefly anesthetized with tricaine and transferred on a wet sponge with their ventral side up. A small incision was made on the chest area exposing the heart. A cryoprobe precooled in liquid nitrogen was applied to the ventricular apex until thawing. Cryoinjured fish were transferred into fresh system water and left to recover. Heat shock treatments to induce the expression of the hsp70l-driven transgene Tg(hsp70l:loxP-TagBFP-loxP-cxcl18b-t2a-mCherry) were performed by incubating the adult fish or the embryos in preheated system or egg water (39 °C) for 1 h. Several heat shocks were performed as described in the corresponding experimental plans. It is important to note that heat shock alone, as a stress response, can promote an immune response, including the influx of inflammatory cells^113,114, potentially explaining the discrepancy in neutrophil count in cryoinjured ventricles observed with (Fig. 6f) and without (Fig. 1g) heat shock treatments.

Histological analysis and imaging

For the histological analysis, zebrafish hearts were fixed in 4% paraformaldehyde for 1 h at room temperature and then preserved overnight in 30% (w/v) sucrose solution prepared in 1× PBS at 4 °C. Hearts samples were embedded in O.C.T. (Tissue-Tek) and stored at −80 °C until further use. Eight and 50-μm-thick cryosections were collected on SuperFrost Plus slides (Thermo Fisher Scientific) using the Leica CM1950 cryostat and stored at −20 °C.

For AFOG staining, slides were thawed for 15 min at room temperature, rinsed twice with 0.1% Triton X-100 in 1× PBS to remove O.C.T., incubated in Bouin’s solution for 2 h at 60 °C and then stained according to the manufacturer’s instructions (AFOG staining kit, BioGnost) and without hematoxylin solution⁵⁹. Stained slides were imaged using a Nikon SMZ25 stereo microscope coupled with a Nikon Digital Sight DS-Ri1 camera.

For immunofluorescence staining, slides were thawed for 15 min at room temperature, rinsed twice with 0.1% Triton X-100 in 1× PBS to remove O.C.T. and permeabilized with 0.5% Triton X-100 in 1× PBS for 20 min (2 h for 50-μm-thick cryosections) at room temperature. Cryosections were then incubated in blocking buffer solution (1× PBS, 2% (v/v) goat serum, 0.2% Triton X-100 and 1% dimethyl sulfoxide) for 1 h at room temperature. Then, cryosections were incubated with primary antibodies in blocking buffer solution overnight at 4 °C, rinsed three times every 10 min with 0.1% Triton X-100 in 1× PBS and incubated with secondary antibodies in blocking buffer solution for 3 h at room temperature. Lastly, immunostained cryosections were rinsed three times every 10 min with 0.1% Triton X-100 in 1× PBS, incubated in DAPI (1:10,000 dilution, Sigma-Aldrich) for 5 min at room temperature and mounted with fluorescence mounting medium (cat. no. S3023, Agilent Dako) for imaging. MEF2, PCNA, pAkt and pERK immunostaining was performed as described previously⁷³. MEF2 and pERK immunostaining was supplemented with an additional step after O.C.T. removal, which consisted of antigen retrieval in 10 mM sodium citrate buffer with 0.05% (v/v) Tween-20 (all from Sigma-Aldrich), pH 6.0, for 7 min at 95 °C.

Primary antibodies used were: anti-GFP at 1:500 dilution (chicken, cat. no. GFP-1010, Aves Labs); anti-αSMA at 1:200 dilution (rabbit, cat. no. GTX124505, GeneTex); anti-Fli1 (clone EPR4646) at 1:100 dilution (rabbit, cat. no. ab133485, Abcam); anti-pAkt (clone 6F5) at 1:200 dilution (mouse, cat. no. 05-1003, Sigma-Aldrich); anti-PCNA (clone PC10) at 1:200 dilution (mouse, cat. no. sc-56, Santa Cruz Biotechnology); anti-MEF2 at 1:100 dilution (rabbit, cat. no. DZ01398, Boster Bio); anti-Mpx at 1:200 dilution (rabbit, cat. no. GTX128379, GeneTex); anti-Aldh1a2 (clone G-2) at 1:100 dilution (mouse, cat. no. sc-393204, Santa Cruz Biotechnology); anti-Aldh1a2 at 1:200 dilution (rabbit, cat. no. GTX124302, GeneTex); anti-zf-Cdh5 at 1:100 dilution (rabbit, cat. no. AS-55715, AnaSpec); anti-pERK (clone D13.14.4E) at 1:100 dilution (rabbit, cat. no. 4370S, Cell Signaling Technology); anti-RFP at 1:200 dilution (rabbit, cat. no. 600-401-379, Rockland Immunochemicals); N2.261 at 1:20 dilution (mouse, developed by H. M. Blau and obtained from the Developmental Studies Hybridoma Bank); and anti-DsRed at 1:200 dilution (recognizing mCherry, Living Colors, rabbit, cat. no. 632496, Takara Bio). Secondary antibodies used (all at 1:500 dilution) were: anti-chicken IgG (H+L) Alexa Fluor 488 (goat, cat. no. A-11039, Invitrogen); anti-mouse IgG (H+L) Alexa Fluor 488 (goat, cat. no. A-11029, Invitrogen); anti-mouse IgG (H+L) Alexa Fluor 568 (goat, cat. no. A-11004, Invitrogen); and anti-rabbit IgG (H+L) Alexa Fluor 647 (goat, cat. no. A-21244, Invitrogen). Phalloidin-Alexa Fluor 568 (cat. no. A12380, Thermo Fisher Scientific) was used at a 1:200 dilution. Imaging was performed using a ZEISS LSM 800 Observer inverted confocal microscope, a ZEISS Cell Observer Spinning Disk inverted confocal microscope, a ZEISS Axioscan 7 microscope, a Nikon Ni-E Eclipse widefield microscope equipped with a SlideExpress 2 slideloader (Märzhäuser), a SOLA Light Engine (Lumencor) and a DS-Qi2 Mono Digital Microscope Camera (Nikon). For the ZEISS and Nikon microscopes, the ZEN Blue Edition and NIS-AR software v.5.3 were used, respectively. Wholemount ventricle imaging was performed using a Nikon SMZ25 microscope coupled with a Nikon Digital Sight DS-Ri1 camera and the NIS-Elements v.4.30 software.

Quantification and statistical analysis

Quantification was done in two or three non-consecutive sections per ventricle and with the ZEN Blue Edition software. Quantification of the mpx:GFP⁺, mpeg1:EGFP⁺ and Mpx⁺ cell numbers was performed in the injured tissue and in peripheral border zone areas (100 μm). The mpx:GFP⁺ cell number analysis for the myd88^+/− control condition (Fig. 6e,f) was performed separately. Quantification of the total αSMA⁺ cell number was performed in the injured tissue, and of the intraventricular αSMA⁺ cell number was performed in the injured tissue excluding the superficial/peripheral αSMA⁺ cells (that is, quantification was based on cell localization and not on the presence of an additional cell marker). Quantification of intraventricularly localized αSMA⁺ cells did not include double-positive αSMA⁺ krt4:EGFP⁺ cells for the following reasons: (1) the endocardium has not fully extended into the injured tissue at 96 hpci (the time point of our analysis), as indicated by krt4:EGFP⁺ expression, and (2) even though αSMA⁺ cells can derive from the endocardium, it is not known how long they retain the expression of endocardial markers. The αSMA⁺ cell number analysis displayed in Fig. 2d is the result of two independent experiments, each of which contained samples of both genotypes from the same batch. For the endocardial pAkt activation and proliferation analysis, the percentage of pAkt⁺ and PCNA⁺ endocardial cells was calculated as a ratio of the total number of endocardial cells (krt4:EGFP⁺Fli1⁺) in a 50-μm wide area on the basal-most side of the injured tissue, as described previously⁵⁵. For the krt4:EGFP⁺, Cdh5⁺ and pERK⁺ cell area analysis, the fluorescent area within the injured tissue was measured and then divided by the total injured tissue area. For the CM dedifferentiation and proliferation analysis, the percentage of N2.261⁺ and PCNA⁺ CMs was calculated as a ratio of the total number of CMs (MEF2⁺) in peripheral border zone areas (100 μm). CM protrusions and protrusion lengths were measured in two non-consecutive 50-μm-thick cryosections with the largest injured area from each heart. The start of the protrusions was defined as the injury border and the end of the protrusions as the point where the F-actin signal disappeared in the injured tissue, as described previously⁷⁷. All protrusions measured were extending toward the injured tissue. For the cEC proliferation analysis, the percentage of PCNA⁺ cECs was calculated as a ratio of the total number of cECs (-0.8flt1:RFP⁺) in the injured tissue and in peripheral border zone areas (200 μm). For the coronary vessel coverage analysis, the percentage of the fluorescence intensity was calculated as a ratio of the background fluorescence in the injured tissue using ImageJ (v.1.53c) in wholemount images. For the scar area analyses, scar areas were selected based on the combined occurrence of collagen and fibrin within the ventricle. The average of the ratio of the 2–3 biggest scar areas to the total ventricular areas were calculated using ImageJ (v.1.53c). Percentages of scar areas (relative to ventricular areas) were then grouped based on their size. Fibrin and collagen measurements were performed as described previously⁶².

All statistical analyses were performed in GraphPad Prism (v.9). Distribution of data in each group was assessed using the Shapiro–Wilk normality test. Data that were normally distributed were further analyzed with a two-tailed Student’s t-test. Data that were not normally distributed were further analyzed with a two-tailed Mann–Whitney U-test. The significance level was set to 0.05 for all tests. The exact P values are indicated in the figures. The error bars in the figures represent the mean ± s.d.

Tissue dissociation and cell sorting

For the scRNA-seq experiment, cardiac cells were isolated from a pool of four myd88^+/+ and four myd88^−/− ventricles, and for the endocardial bulk RNA-seq experiment from a pool of four ET(krt4:EGFP); myd88^+/+ and four ET(krt4:EGFP); myd88^−/− ventricles for each sample. Cell isolation was performed according to the manufacturer’s instructions (Pierce Primary Cardiomyocyte Isolation Kit, cat. no. 88281, Thermo Fisher Scientific) and with the following modifications: incubation was performed at 30 °C for 20 min, followed by resuspension in 1× Hanks’ Balanced Salt Solution (cat. no. 14175053, Gibco) with 0.25% BSA. The cell suspension was passed through a round bottom polystyrene test tube fitted with a 35-µm nylon mesh filter cap (Falcon, cat. no. 352235, Corning). DAPI (cat. no. D954, Sigma-Aldrich) was added before sorting. For the scRNA-seq experiment, resuspended cells were sorted using a BD FACSAria III Cell Sorter (BD Biosciences) equipped with a 100-µm nozzle and at an instrument pressure setting of 20 psi. Dead cells were excluded using DAPI excited by a 30-mW 405-nm laser paired with a 450/50-nm band-pass filter. For the endocardial bulk RNA-seq experiment, resuspended cells were sorted using an Invitrogen Bigfoot Spectral Cell Sorter (Thermo Fisher Scientific) equipped with a 100-µm nozzle tip and at an instrument pressure setting of 20 psi. Dead cells were excluded using DAPI excited by a 100-mW 355-nm laser paired with 455/14-nm band-pass filter. EGFP fluorescence was measured with a 100-mW 488-nm excitation paired with a 530/30-nm band-pass filter. Sorted EGFP⁺ cells were resuspended in ice-cold QIAzol Lysis Reagent (QIAGEN), flash-frozen in liquid nitrogen and kept at −80 °C until RNA extraction. Cytometric data were recorded using the FACSDiva software v.8.0.1 (BD Biosciences) and the Sasquatch software v.1.19.2 (Thermo Fisher Scientific). The .fcs files were analyzed using FlowJo v.10.8.1 (BD Life Sciences).

scRNA-seq analysis

For the scRNA-seq experiment, cells were counted with a Moxi cell counter and diluted according to the manufacturer’s protocol to obtain 10,000 single-cell data points per sample. Each sample was run separately on a lane in a Chromium controller with the Chromium Next GEM Single Cell 3′ Reagent Kits v.3.1 (10x Genomics). scRNA-seq library preparation was done using a standard protocol. Sequencing was done on NextSeq 2000 system and raw reads were aligned against the zebrafish genome (DanRer11) and counted using STARsolo¹¹⁵ followed by secondary analysis in annotated data format. Preprocessed counts were further analyzed using Scanpy¹¹⁶. Basic cell quality control was conducted by taking the number of detected genes and mitochondrial content into consideration. We removed 32 cells that did not express more than 300 genes or had a mitochondrial content greater than 10%. Furthermore, we filtered 7,823 genes if they were detected in fewer than 30 cells (<0.01%). The raw counts per cell were normalized to the median count over all cells and transformed into log space to stabilize variance. We initially reduced the dimensionality of the dataset using principal component analysis, retaining 50 principal components. Final data visualization was done using the scVelo^117,118 and CELLxGENE packages.

Bulk RNA-seq analysis

For bulk RNA-seq of endocardial cells (Fig. 3a), RNA was isolated from the fluorescence-activated cell-sorted endocardial cells using the miRNeasy Micro Kit (QIAGEN) combined with on-column DNase digestion (RNase-free DNase Set, QIAGEN) to avoid contamination by genomic DNA (gDNA). Subsequent RNA quality control analysis, complementary DNA (cDNA) preparation and sequencing were performed by Novogene. mRNA was purified from total RNA with poly-T oligo-attached magnetic beads, went through fragmentation and cDNA was synthesized. Trimmomatic¹¹⁹ was used to trim the reads. Reads longer than 15 nucleotides after trimming were kept and aligned to the Ensembl zebrafish genome v.danRer11 (Ensembl release 109) with STAR aligner (v.2.7.10a)¹¹⁵. Alignments were filtered with the Picard tool (v.3.0.0) to remove duplicates, multimapping, ribosomal or mitochondrial reads. Gene counts were generated using the featureCounts tool (v.2.0.4), taking all reads overlapping annotated exons into account and excluding those overlapping multiple genes¹²⁰. Finally, the raw count matrix was normalized and contrasts were created and analyzed using DESeq2 (v.1.36.0)¹²¹. Genes were classified as significantly differentially expressed at an average count greater than five, multiple testing P_adj < 0.05 and −0.585 < log₂ fold change > 0.585. Heatmaps with DEGs were obtained using the WIlsON¹²². To perform GSEA, we used the Python package gseapy¹²³. As gene sets, we used 162 custom Danio rerio mapped gene sets from reactome and 3,023 gene sets from Gene Ontology. Gene ranking was performed according to the DESeq-derived P_adj multiplied by the direction (+ or −) of the log₂ fold change. The GSEA analysis was run for gene sets with a min_size = 5, max_size = 1,000 and permutation_num = 1,000. From the resulting lists, representative sets were selected.

For bulk RNA-seq of untouched ventricles and injured tissues (Fig. 6a), RNA was isolated from three pooled ventricles or three pooled injured tissues per sample using the miRNeasy Micro Kit (QIAGEN) combined with on-column DNase digestion (RNase-free DNase Set, QIAGEN) to avoid contamination by gDNA. Two biological replicates were prepared for each condition. RNA and library preparation integrity were verified with LabChip Gx Touch 24 Analyzer (PerkinElmer). 400 ng of total RNA was used as input for the VAHTS Stranded mRNA-seq Library Preparation V6 according to the manufacturer’s protocol (Vazyme). Sequencing was performed on an NextSeq 500 instrument (Illumina) using v2 chemistry with 1 × 75-bp single end setup. Trimmomatic v.0.39 was used to trim reads after a quality drop below a mean of Q15 in a window of five nucleotides, keeping only filtered reads longer than 15 nucleotides¹¹⁹. Reads were aligned against the Ensembl zebrafish genome v.danRer11 (Ensembl release 99) with STAR v.2.7.3a¹¹⁵. Alignments were filtered with Picard v.2.21.7 to remove duplicates, multimapping, ribosomal or mitochondrial reads. Gene counts were established with featureCounts v.1.6.5 aggregating reads overlapping exons, excluding those overlapping multiple genes¹²⁰. The raw count matrix was normalized with DESeq2 v.1.26.0 (ref. ¹²¹). Contrasts were created with DESeq2 based on the raw count matrix. Genes were classified as significantly differentially expressed at average count greater than five, multiple testing P_adj < 0.05 and −0.585 < log₂ fold change > 0.585. Heatmaps with DEGs were obtained using WIlsON¹²².

RT–qPCR

For RT–qPCR, RNA was extracted from three pooled cryoinjured tissues (1 hpci) and ventricles (96 hpci) (Fig. 6c) or from single larvae (Extended Data Fig. 9a and Extended Data Fig. 10c,f) or from single cryoinjured ventricles (96 hpci) (Extended Data Fig. 9a) per biological replicate using the TRIzol Reagent (Thermo Fisher Scientific) using the phenol–chloroform extraction protocol. Total RNA was purified with the RNA Clean & Concentrator Extraction Kit (Zymo Research) according to the manufacturer’s instructions. At least 250 ng of total RNA per sample was reverse-transcribed with the Maxima First Strand cDNA Synthesis Kit (Thermo Fisher Scientific) according to the manufacturer’s instructions. All reactions were performed with three technical replicates using the SYBR Green PCR Master Mix (Thermo Fisher Scientific) on the CFX Connect Real-Time System (CFX Manager 3.1, Bio-Rad Laboratories) and with the following program: preamplification at 95 °C for 7 min followed by 39 cycles of amplification at 95 °C for 5 s and 60 °C for 20 s, using a melting curve from 60 to 92 °C with an increment of 1.0 °C every 5 s. Gene mRNA levels were normalized against the rpl13a mRNA levels and fold changes were calculated using the \(2-^{\Delta\Delta_{\mathrm{Ct}}}\) method. The RT–qPCR primer sequences are listed in Supplementary Table 2. The average Ct values of the RT–qPCR are listed in Supplementary Table 3.

Reporting summary

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

Data availability

The scRNA-seq, RNA-seq of endocardial cells and RNA-seq of untouched ventricles and injured tissues data reported in this study have been deposited in the Gene Expression Omnibus under accession nos. GSE262247, GSE262351 and GSE262169, respectively.

References

Braunwald, E. The war against heart failure: the Lancet lecture. Lancet 385, 812–824 (2015).
Article PubMed Google Scholar
Talman, V. & Ruskoaho, H. Cardiac fibrosis in myocardial infarction—from repair and remodeling to regeneration. Cell Tissue Res. 365, 563–581 (2016).
Article CAS PubMed PubMed Central Google Scholar
González-Rosa, J. M., Martín, V., Peralta, M., Torres, M. & Mercader, N. Extensive scar formation and regression during heart regeneration after cryoinjury in zebrafish. Development 138, 1663–1674 (2011).
Article PubMed Google Scholar
Schnabel, K., Wu, C.-C., Kurth, T. & Weidinger, G. Regeneration of cryoinjury induced necrotic heart lesions in zebrafish is associated with epicardial activation and cardiomyocyte proliferation. PLoS ONE 6, e18503 (2011).
Article CAS PubMed PubMed Central Google Scholar
Chablais, F., Veit, J., Rainer, G. & Jaźwińska, A. The zebrafish heart regenerates after cryoinjury-induced myocardial infarction. BMC Dev. Biol. 11, 21 (2011).
Article PubMed PubMed Central Google Scholar
Wang, J. et al. The regenerative capacity of zebrafish reverses cardiac failure caused by genetic cardiomyocyte depletion. Development 138, 3421–3430 (2011).
Article CAS PubMed PubMed Central Google Scholar
Frangogiannis, N. G. The inflammatory response in myocardial injury, repair, and remodelling. Nat. Rev. Cardiol. 11, 255–265 (2014).
Article CAS PubMed PubMed Central Google Scholar
Anzai, A., Ko, S. & Fukuda, K. Immune and inflammatory networks in myocardial infarction: current research and its potential implications for the clinic. Int. J. Mol. Sci. 23, 5214 (2022).
Article CAS PubMed PubMed Central Google Scholar
Huang, W.-C. et al. Treatment of glucocorticoids inhibited early immune responses and impaired cardiac repair in adult zebrafish. PLoS ONE 8, e66613 (2013).
Article CAS PubMed PubMed Central Google Scholar
Lai, S.-L. et al. Reciprocal analyses in zebrafish and medaka reveal that harnessing the immune response promotes cardiac regeneration. eLife 6, e25605 (2017).
Article PubMed PubMed Central Google Scholar
de Preux Charles, A.-S., Bise, T., Baier, F., Marro, J. & Jaźwińska, A. Distinct effects of inflammation on preconditioning and regeneration of the adult zebrafish heart. Open Biol. 6, 160102 (2016).
Article PubMed PubMed Central Google Scholar
Han, C. et al. Acute inflammation stimulates a regenerative response in the neonatal mouse heart. Cell Res. 25, 1137–1151 (2015).
Article CAS PubMed PubMed Central Google Scholar
Farache Trajano, L. & Smart, N. Immunomodulation for optimal cardiac regeneration: insights from comparative analyses. npj Regen. Med. 6, 8 (2021).
Article PubMed PubMed Central Google Scholar
Lai, S.-L., Marin-Juez, R. & Stainier, D. Y. R. Immune responses in cardiac repair and regeneration: a comparative point of view. Cell. Mol. Life Sci. 76, 1365–1380 (2019).
Article CAS PubMed Google Scholar
Timmers, L. et al. The innate immune response in reperfused myocardium. Cardiovasc. Res. 94, 276–283 (2012).
Article CAS PubMed Google Scholar
Yang, D., Han, Z. & Oppenheim, J. J. Alarmins and immunity. Immunol. Rev. 280, 41–56 (2017).
Article CAS PubMed PubMed Central Google Scholar
Prabhu, S. D. & Frangogiannis, N. G. The biological basis for cardiac repair after myocardial infarction: from inflammation to fibrosis. Circ. Res. 119, 91–112 (2016).
Article CAS PubMed PubMed Central Google Scholar
Cohen, P. The TLR and IL-1 signalling network at a glance. J. Cell Sci. 127, 2383–2390 (2014).
CAS PubMed PubMed Central Google Scholar
Muzio, M., Ni, J., Feng, P. & Dixit, V. M. IRAK (Pelle) family member IRAK-2 and MyD88 as proximal mediators of IL-1 signaling. Science 278, 1612–1615 (1997).
Article CAS PubMed Google Scholar
Bayer, A. L. & Alcaide, P. MyD88: at the heart of inflammatory signaling and cardiovascular disease. J. Mol. Cell. Cardiol. 161, 75–85 (2021).
Article CAS PubMed PubMed Central Google Scholar
Frangogiannis, N. G. The immune system and cardiac repair. Pharmacol. Res. 58, 88–111 (2008).
Article CAS PubMed PubMed Central Google Scholar
Feng, Y. et al. Innate immune adaptor MyD88 mediates neutrophil recruitment and myocardial injury after ischemia-reperfusion in mice. Am. J. Physiol. Heart Circ. Physiol. 295, H1311–H1318 (2008).
Article CAS PubMed PubMed Central Google Scholar
Singh, M. V. et al. MyD88 mediated inflammatory signaling leads to CaMKII oxidation, cardiac hypertrophy and death after myocardial infarction. J. Mol. Cell. Cardiol. 52, 1135–1144 (2012).
Article CAS PubMed PubMed Central Google Scholar
Oyama, J. et al. Reduced myocardial ischemia-reperfusion injury in Toll-like receptor 4-deficient mice. Circulation 109, 784–789 (2004).
Article CAS PubMed Google Scholar
Chu, X. et al. Lipopolysaccharides improve mesenchymal stem cell-mediated cardioprotection by MyD88 and stat3 signaling in a mouse model of cardiac ischemia/reperfusion injury. Stem Cells Dev. 28, 620–631 (2019).
Article CAS PubMed PubMed Central Google Scholar
Zhu, X. et al. MyD88 and NOS2 are essential for Toll-like receptor 4-mediated survival effect in cardiomyocytes. Am. J. Physiol. Heart Circ. Physiol. 291, H1900–H1909 (2006).
Article CAS PubMed Google Scholar
Ha, T. et al. Blockade of MyD88 attenuates cardiac hypertrophy and decreases cardiac myocyte apoptosis in pressure overload-induced cardiac hypertrophy in vivo. Am. J. Physiol. Heart Circ. Physiol. 290, H985–H994 (2006).
Article CAS PubMed Google Scholar
Chen, W. Q. et al. Overexpressing dominant negative MyD88 induces cardiac dysfunction in transgenic mice. Chinese Sci. Bull. 55, 3569–3575 (2010).
Article CAS Google Scholar
Bayer, A. L. et al. T-cell MyD88 is a novel regulator of cardiac fibrosis through modulation of T-cell activation. Circ. Res. 133, 412–429 (2023).
Article CAS PubMed PubMed Central Google Scholar
Hardiman, G. et al. Genetic structure and chromosomal mapping of MyD88. Genomics 45, 332–339 (1997).
Article CAS PubMed Google Scholar
Hall, C. et al. Transgenic zebrafish reporter lines reveal conserved Toll-like receptor signaling potential in embryonic myeloid leukocytes and adult immune cell lineages. J. Leukoc. Biol. 85, 751–765 (2009).
Article CAS PubMed Google Scholar
van der Vaart, M., van Soest, J. J., Spaink, H. P. & Meijer, A. H. Functional analysis of a zebrafish myd88 mutant identifies key transcriptional components of the innate immune system. Dis. Model. Mech. 6, 841–854 (2013).
CAS PubMed PubMed Central Google Scholar
Chablais, F. & Jazwinska, A. The regenerative capacity of the zebrafish heart is dependent on TGFβ signaling. Development 139, 1921–1930 (2012).
Article CAS PubMed Google Scholar
Ma, H. et al. Functional coordination of non-myocytes plays a key role in adult zebrafish heart regeneration. EMBO Rep. 22, e52901 (2021).
Article CAS PubMed PubMed Central Google Scholar
Hu, B. et al. Origin and function of activated fibroblast states during zebrafish heart regeneration. Nat. Genet. 54, 1227–1237 (2022).
Article CAS PubMed PubMed Central Google Scholar
Rolland, L. et al. The regenerative response of cardiac interstitial cells. J. Mol. Cell. Biol. 14, mjac059 (2023).
Article PubMed Google Scholar
Hu, W. et al. A novel function of TLR2 and MyD88 in the regulation of leukocyte cell migration behavior during wounding in zebrafish larvae. Front. Cell Dev. Biol. 9, 624571 (2021).
Article PubMed PubMed Central Google Scholar
Roh-Johnson, M. et al. Macrophage-dependent cytoplasmic transfer during melanoma invasion in vivo. Dev. Cell 43, 549–562 (2017).
Article CAS PubMed PubMed Central Google Scholar
Nathan, C. Neutrophils and immunity: challenges and opportunities. Nat. Rev. Immunol. 6, 173–182 (2006).
Article CAS 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
Ma, Y., Yabluchanskiy, A. & Lindsey, M. L. Neutrophil roles in left ventricular remodeling following myocardial infarction. Fibrogenesis Tissue Repair 6, 11 (2013).
Article CAS PubMed PubMed Central Google Scholar
Peterson, E. A., Sun, J., Chen, X. & Wang, J. Neutrophils facilitate the epicardial regenerative response after zebrafish heart injury. Dev. Biol. 508, 93–106 (2024).
Article CAS PubMed Google Scholar
Horckmans, M. et al. Neutrophils orchestrate post-myocardial infarction healing by polarizing macrophages towards a reparative phenotype. Eur. Heart J. 38, 187–197 (2017).
CAS PubMed Google Scholar
Taichman, N. S., Young, S., Cruchley, A. T., Taylor, P. & Paleolog, E. Human neutrophils secrete vascular endothelial growth factor. J. Leukoc. Biol. 62, 397–400 (1997).
Article CAS PubMed Google Scholar
Ma, Y. Role of neutrophils in cardiac injury and repair following myocardial infarction. Cells 10, 1676 (2021).
Article CAS PubMed PubMed Central Google Scholar
Gong, Y. & Koh, D.-R. Neutrophils promote inflammatory angiogenesis via release of preformed VEGF in an in vivo corneal model. Cell Tissue Res. 339, 437–448 (2010).
Article PubMed Google Scholar
Xu, S. et al. Prolonged neutrophil retention in the wound impairs zebrafish heart regeneration after cryoinjury. Fish Shellfish Immunol. 94, 447–454 (2019).
Article CAS PubMed Google Scholar
Castoldi, A. et al. TLR2, TLR4 and the MYD88 signaling pathway are crucial for neutrophil migration in acute kidney injury induced by sepsis. PLoS ONE 7, e37584 (2012).
Article CAS PubMed PubMed Central Google Scholar
Latty, S. L. et al. Activation of Toll-like receptors nucleates assembly of the MyDDosome signaling hub. eLife 7, e31377 (2018).
Article PubMed PubMed Central Google Scholar
Bevan, L. et al. Specific macrophage populations promote both cardiac scar deposition and subsequent resolution in adult zebrafish. Cardiovasc. Res. 116, 1357–1371 (2020).
Article CAS PubMed Google Scholar
Cardeira-da-Silva, J. et al. Antigen presentation plays positive roles in the regenerative response to cardiac injury in zebrafish. Nat. Commun. 15, 3637 (2024).
Article CAS PubMed PubMed Central Google Scholar
Salvador, B. et al. Modulation of endothelial function by Toll like receptors. Pharmacol. Res. 108, 46–56 (2016).
Article CAS PubMed Google Scholar
Opitz, B., Eitel, J., Meixenberger, K. & Suttorp, N. Role of Toll-like receptors, NOD-like receptors and RIG-I-like receptors in endothelial cells and systemic infections. Thromb. Haemost. 102, 1103–1109 (2009).
Article CAS PubMed Google Scholar
Grivas, D., González-Rajal, A. & de la Pompa, J. L. Midkine-a regulates the formation of a fibrotic scar during zebrafish heart regeneration. Front. Cell Dev. Biol. 9, 669439 (2021).
Article PubMed PubMed Central Google Scholar
Münch, J., Grivas, D., González-Rajal, A., Torregrosa-Carrión, R. & de la Pompa, J. L. Notch signalling restricts inflammation and serpine1 expression in the dynamic endocardium of the regenerating zebrafish heart. Development 144, 1425–1440 (2017).
Article PubMed Google Scholar
Aisagbonhi, O. et al. Experimental myocardial infarction triggers canonical Wnt signaling and endothelial-to-mesenchymal transition. Dis. Model. Mech. 4, 469–483 (2011).
Article CAS PubMed PubMed Central Google Scholar
Alonso-Herranz, L. et al. Macrophages promote endothelial-to-mesenchymal transition via MT1-MMP/TGFβ1 after myocardial infarction. eLife 9, e57920 (2020).
Article CAS PubMed PubMed Central Google Scholar
Davis, J. & Molkentin, J. D. Myofibroblasts: trust your heart and let fate decide. J. Mol. Cell. Cardiol. 70, 9–18 (2014).
Article CAS PubMed Google Scholar
Allanki, S. et al. Interleukin-11 signaling promotes cellular reprogramming and limits fibrotic scarring during tissue regeneration. Sci. Adv. 7, eabg6497 (2021).
Article CAS PubMed PubMed Central Google Scholar
Fu, X. et al. Specialized fibroblast differentiated states underlie scar formation in the infarcted mouse heart. J. Clin. Invest. 128, 2127–2143 (2018).
Article PubMed PubMed Central Google Scholar
Crisan, M., Corselli, M., Chen, W. C. & Péault, B. Perivascular cells for regenerative medicine. J. Cell. Mol. Med. 16, 2851–2860 (2012).
Article CAS PubMed PubMed Central Google Scholar
Koth, J. et al. Runx1 promotes scar deposition and inhibits myocardial proliferation and survival during zebrafish heart regeneration. Development 147, dev186569 (2020).
Article CAS PubMed PubMed Central Google Scholar
Liu, P. & Zhong, T. P. MAPK/ERK signalling is required for zebrafish cardiac regeneration. Biotechnol. Lett. 39, 1069–1077 (2017).
Article CAS PubMed Google Scholar
Tahara, N. et al. The FGF-AKT pathway is necessary for cardiomyocyte survival for heart regeneration in zebrafish. Dev. Biol. 472, 30–37 (2021).
Article CAS PubMed PubMed Central Google Scholar
Kawai, T., Adachi, O., Ogawa, T., Takeda, K. & Akira, S. Unresponsiveness of MyD88-deficient mice to endotoxin. Immunity 11, 115–122 (1999).
Article CAS PubMed Google Scholar
Wen, X., Jiao, L. & Tan, H. MAPK/ERK pathway as a central regulator in vertebrate organ regeneration. Int. J. Mol. Sci. 23, 1464 (2022).
Article CAS PubMed PubMed Central Google Scholar
Fukazawa, R. et al. Neuregulin-1 protects ventricular myocytes from anthracycline-induced apoptosis via erbB4-dependent activation of PI3-kinase/Akt. J. Mol. Cell. Cardiol. 35, 1473–1479 (2003).
Article CAS PubMed Google Scholar
Ghafouri-Fard, S. et al. Interplay between PI3K/AKT pathway and heart disorders. Mol. Biol. Rep. 49, 9767–9781 (2022).
Article CAS PubMed PubMed Central Google Scholar
Shiojima, I. & Walsh, K. Role of Akt signaling in vascular homeostasis and angiogenesis. Circ. Res. 90, 1243–1250 (2002).
Article CAS PubMed Google Scholar
Kwon, O. S. et al. MyD88 regulates physical inactivity-induced skeletal muscle inflammation, ceramide biosynthesis signaling, and glucose intolerance. Am. J. Physiol. Endocrinol. Metab. 309, E11–E21 (2015).
Article CAS PubMed PubMed Central Google Scholar
Rhee, S. H., Kim, H., Moyer, M. P. & Pothoulakis, C. Role of MyD88 in phosphatidylinositol 3-kinase activation by flagellin/Toll-like receptor 5 engagement in colonic epithelial cells. J. Biol. Chem. 281, 18560–18568 (2006).
Article CAS PubMed Google Scholar
He, Y. et al. Targeting PI3K/Akt signal transduction for cancer therapy. Signal Transduct. Target. Ther. 6, 425 (2021).
Article CAS PubMed PubMed Central Google Scholar
Marín-Juez, R. et al. Fast revascularization of the injured area is essential to support zebrafish heart regeneration. Proc. Natl Acad. Sci. USA 113, 11237–11242 (2016).
Article PubMed PubMed Central Google Scholar
Marin-Juez, R. et al. Coronary revascularization during heart regeneration is regulated by epicardial and endocardial cues and forms a scaffold for cardiomyocyte repopulation. Dev. Cell 51, 503–515 (2019).
Article CAS PubMed PubMed Central Google Scholar
Kikuchi, K. et al. Primary contribution to zebrafish heart regeneration by gata4⁺ cardiomyocytes. Nature 464, 601–605 (2010).
Article CAS PubMed PubMed Central Google Scholar
Jopling, C. et al. Zebrafish heart regeneration occurs by cardiomyocyte dedifferentiation and proliferation. Nature 464, 606–609 (2010).
Article CAS PubMed PubMed Central Google Scholar
Beisaw, A. et al. AP-1 contributes to chromatin accessibility to promote sarcomere disassembly and cardiomyocyte protrusion during zebrafish heart regeneration. Circ. Res. 126, 1760–1778 (2020).
Article CAS PubMed PubMed Central Google Scholar
Moretti, L., Stalfort, J., Barker, T. H. & Abebayehu, D. The interplay of fibroblasts, the extracellular matrix, and inflammation in scar formation. J. Biol. Chem. 298, 101530 (2022).
Article CAS PubMed Google Scholar
Sallin, P., de Preux Charles, A. S., Duruz, V., Pfefferli, C. & Jaźwińska, A. A dual epimorphic and compensatory mode of heart regeneration in zebrafish. Dev. Biol. 399, 27–40 (2015).
Article CAS PubMed Google Scholar
Long, H. et al. Endothelial cells adopt a pro-reparative immune responsive signature during cardiac injury. Life Sci. Alliance 7, e202201870 (2023).
Article PubMed PubMed Central Google Scholar
Bednarek, D. et al. Telomerase is essential for zebrafish heart regeneration. Cell Rep. 12, 1691–1703 (2015).
Article CAS PubMed PubMed Central Google Scholar
King, B. L. et al. RegenDbase: a comparative database of noncoding RNA regulation of tissue regeneration circuits across multiple taxa. npj Regen. Med. 3, 10 (2018).
Article PubMed PubMed Central Google Scholar
Torraca, V., Otto, N. A., Tavakoli-Tameh, A. & Meijer, A. H. The inflammatory chemokine Cxcl18b exerts neutrophil-specific chemotaxis via the promiscuous chemokine receptor Cxcr2 in zebrafish. Dev. Comp. Immunol. 67, 57–65 (2017).
Article CAS PubMed Google Scholar
Xie, Y. et al. Glucocorticoids inhibit macrophage differentiation towards a pro-inflammatory phenotype upon wounding without affecting their migration. Dis. Model. Mech. 12, dmm037887 (2019).
Article CAS PubMed PubMed Central Google Scholar
Chatzopoulou, A. et al. Glucocorticoid-induced attenuation of the inflammatory response in zebrafish. Endocrinology 157, 2772–2784 (2016).
Article CAS PubMed Google Scholar
Hesselson, D., Anderson, R. M., Beinat, M. & Stainier, D. Y. Distinct populations of quiescent and proliferative pancreatic β-cells identified by HOTcre mediated labeling. Proc. Natl Acad. Sci. USA 106, 14896–14901 (2009).
Article CAS PubMed PubMed Central Google Scholar
Rossi, A. et al. Genetic compensation induced by deleterious mutations but not gene knockdowns. Nature 524, 230–233 (2015).
Article CAS PubMed Google Scholar
El-Brolosy, M. A. et al. Genetic compensation triggered by mutant mRNA degradation. Nature 568, 193–197 (2019).
Article CAS PubMed PubMed Central Google Scholar
Arslan, F. et al. Myocardial ischemia/reperfusion injury is mediated by leukocytic Toll-like receptor-2 and reduced by systemic administration of a novel anti-Toll-like receptor-2 antibody. Circulation 121, 80–90 (2010).
Article CAS PubMed Google Scholar
Timmers, L. et al. Toll-like receptor 4 mediates maladaptive left ventricular remodeling and impairs cardiac function after myocardial infarction. Circ. Res. 102, 257–264 (2008).
Article CAS PubMed Google Scholar
Luo, W. et al. Blockage of MyD88 in cardiomyocytes alleviates cardiac inflammation and cardiomyopathy in experimental diabetic mice. Biochem. Pharmacol. 206, 115292 (2022).
Article CAS PubMed Google Scholar
Hsu, A. Y. et al. Development and characterization of an endotoxemia model in zebra fish. Front. Immunol. 9, 607 (2018).
Article PubMed PubMed Central Google Scholar
Feng, Y., Zou, L., Chen, C., Li, D. & Chao, W. Role of cardiac- and myeloid-MyD88 signaling in endotoxin shock: a study with tissue-specific deletion models. Anesthesiology 121, 1258–1269 (2014).
Article CAS PubMed Google Scholar
Yu, M. et al. MyD88-dependent interplay between myeloid and endothelial cells in the initiation and progression of obesity-associated inflammatory diseases. J. Exp. Med. 211, 887–907 (2014).
Article CAS PubMed PubMed Central Google Scholar
Liu, S. et al. Blood flow patterns regulate PCSK9 secretion via MyD88-mediated pro-inflammatory cytokines. Cardiovasc. Res. 116, 1721–1732 (2020).
Article CAS PubMed Google Scholar
Patra, C. et al. The zebrafish ventricle: a hub of cardiac endothelial cells for in vitro cell behavior studies. Sci. Rep. 7, 2687 (2017).
Article PubMed PubMed Central Google Scholar
Feng, Y., Zou, L., Si, R., Nagasaka, Y. & Chao, W. Bone marrow MyD88 signaling modulates neutrophil function and ischemic myocardial injury. Am. J. Physiol. Cell Physiol. 299, C760–C769 (2010).
Article CAS PubMed PubMed Central Google Scholar
Harari, O. A., Alcaide, P., Ahl, D., Luscinskas, F. W. & Liao, J. K. Absence of TRAM restricts Toll-like receptor 4 signaling in vascular endothelial cells to the MyD88 pathway. Circ. Res. 98, 1134–1140 (2006).
Article CAS PubMed PubMed Central Google Scholar
Wullaert, A., Bonnet, M. C. & Pasparakis, M. NF-κB in the regulation of epithelial homeostasis and inflammation. Cell Res. 21, 146–158 (2011).
Article CAS PubMed Google Scholar
Qu, X., Harmelink, C. & Baldwin, H. S. Endocardial-myocardial interactions during early cardiac differentiation and trabeculation. Front. Cardiovasc. Med. 9, 857581 (2022).
Article CAS PubMed PubMed Central Google Scholar
Wu, C. C., Jeratsch, S., Graumann, J. & Stainier, D. Y. R. Modulation of mammalian cardiomyocyte cytokinesis by the extracellular matrix. Circ. Res. 127, 896–907 (2020).
Article CAS PubMed Google Scholar
Bassat, E. et al. The extracellular matrix protein agrin promotes heart regeneration in mice. Nature 547, 179–184 (2017).
Article CAS PubMed PubMed Central Google Scholar
Wang, J., Karra, R., Dickson, A. L. & Poss, K. D. Fibronectin is deposited by injury-activated epicardial cells and is necessary for zebrafish heart regeneration. Dev. Biol. 382, 427–435 (2013).
Article CAS PubMed Google Scholar
Karra, R., Knecht, A. K., Kikuchi, K. & Poss, K. D. Myocardial NF-κB activation is essential for zebrafish heart regeneration. Proc. Natl Acad. Sci. USA 112, 13255–13260 (2015).
Article CAS PubMed PubMed Central Google Scholar
Miranda, A. M. A. et al. Single-cell transcriptomics for the assessment of cardiac disease. Nat. Rev. Cardiol. 20, 289–308 (2023).
Article CAS PubMed Google Scholar
Renshaw, S. A. et al. A transgenic zebrafish model of neutrophilic inflammation. Blood 108, 3976–3978 (2006).
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 PubMed Central Google Scholar
Poon, K.-L., Liebling, M., Kondrychyn, I., Garcia-Lecea, M. & Korzh, V. Zebrafish cardiac enhancer trap lines: new tools for in vivo studies of cardiovascular development and disease. Dev. Dyn. 239, 914–926 (2010).
Article PubMed Google Scholar
Bussmann, J. et al. Arteries provide essential guidance cues for lymphatic endothelial cells in the zebrafish trunk. Development 137, 2653–2657 (2010).
Article CAS PubMed Google Scholar
Gagnon, J. A. et al. Efficient mutagenesis by Cas9 protein-mediated oligonucleotide insertion and large-scale assessment of single-guide RNAs. PLoS ONE 9, e98186 (2014).
Article PubMed PubMed Central Google Scholar
Varshney, G. K. et al. High-throughput gene targeting and phenotyping in zebrafish using CRISPR/Cas9. Genome Res. 25, 1030–1042 (2015).
Article CAS PubMed PubMed Central Google Scholar
El-Sammak, H. et al. A Vegfc-Emilin2a-Cxcl8a signaling axis required for zebrafish cardiac regeneration. Circ. Res. 130, 1014–1029 (2022).
Article CAS PubMed PubMed Central Google Scholar
Lee, C. T., Zhong, L., Mace, T. A. & Repasky, E. A. Elevation in body temperature to fever range enhances and prolongs subsequent responsiveness of macrophages to endotoxin challenge. PLoS ONE 7, e30077 (2012).
Article CAS PubMed PubMed Central Google Scholar
Lam, P.-Y., Harvie, E. A. & Huttenlocher, A. Heat shock modulates neutrophil motility in zebrafish. PLoS ONE 8, e84436 (2013).
Article PubMed PubMed Central Google Scholar
Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013).
Article CAS PubMed Google Scholar
Wolf, F. A., Angerer, P. & Theis, F. J. SCANPY: large-scale single-cell gene expression data analysis. Genome Biol. 19, 15 (2018).
Article PubMed PubMed Central Google Scholar
Bergen, V., Lange, M., Peidli, S., Wolf, F. A. & Theis, F. J. Generalizing RNA velocity to transient cell states through dynamical modeling. Nat. Biotechnol. 38, 1408–1414 (2020).
Article CAS PubMed Google Scholar
La Manno, G. et al. RNA velocity of single cells. Nature 560, 494–498 (2018).
Article CAS PubMed PubMed Central Google Scholar
Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120 (2014).
Article CAS PubMed PubMed Central Google Scholar
Liao, Y., Smyth, G. K. & Shi, W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics 30, 923–930 (2014).
Article CAS PubMed Google Scholar
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014).
Article PubMed PubMed Central Google Scholar
Schultheis, H. et al. WIlsON: Web-based Interactive Omics VisualizatioN. Bioinformatics 35, 1055–1057 (2019).
Article CAS PubMed Google Scholar
Fang, Z., Liu, X. & Peltz, G. GSEApy: a comprehensive package for performing gene set enrichment analysis in Python. Bioinformatics 39, btac757 (2023).
Article CAS PubMed Google Scholar

Download references

Acknowledgements

We thank D. Grabski, N. Fukuda, C. Kremser, C. Buettner, S. Howard and E. Atzaraki for essential technical support; R. Ramadass for help with microscopy; and S. Allanki, J. Cardeira-da-Silva, M. Balakrishnan, T.-L. Tseng, S. Gupta, P. Sagvekar and T. Molina-Villa for input during the development of this work and comments on the paper. We thank our animal house staff for excellent support. We also thank A. Meijer (Leiden University) for sharing the myd88^hu3568 mutant and the Tg(cxcl18b:EGFP)^ibl150 line. R.M.-J. is currently supported by the Canadian Institutes of Health Research (PJT-178037) and a FRQS Junior-1 award. This research was supported by funds from the Max Planck Society and awards from the European Research Council under the European Union’s research and innovation programs (AdG 694455-ZMOD and AdG 101021349-TAaGC) to D.Y.R.S.

Funding

Open access funding provided by Max Planck Society.

Author information

Authors and Affiliations

Department of Developmental Genetics, Max Planck Institute for Heart and Lung Research, Bad Nauheim, Germany
Pinelopi Goumenaki & Didier Y. R. Stainier
DZHK German Centre for Cardiovascular Research, Partner Site Rhine-Main, Bad Nauheim, Germany
Pinelopi Goumenaki, Stefan Günther, Mario Looso & Didier Y. R. Stainier
Cardio-Pulmonary Institute (CPI), Bad Nauheim, Germany
Pinelopi Goumenaki, Stefan Günther, Mario Looso & Didier Y. R. Stainier
Bioinformatics and Deep Sequencing Platform, Max Planck Institute for Heart and Lung Research, Bad Nauheim, Germany
Stefan Günther
Flow Cytometry Service Group, Max Planck Institute for Heart and Lung Research, Bad Nauheim, Germany
Khrievono Kikhi
Bioinformatics Core Unit (BCU), Max Planck Institute for Heart and Lung Research, Bad Nauheim, Germany
Mario Looso
Centre Hospitalier Universitaire Sainte-Justine Research Centre, Montreal, Quebec, Canada
Rubén Marín-Juez
Department of Pathology and Cell Biology, University of Montreal, Montreal, Quebec, Canada
Rubén Marín-Juez

Authors

Pinelopi Goumenaki
View author publications
You can also search for this author in PubMed Google Scholar
Stefan Günther
View author publications
You can also search for this author in PubMed Google Scholar
Khrievono Kikhi
View author publications
You can also search for this author in PubMed Google Scholar
Mario Looso
View author publications
You can also search for this author in PubMed Google Scholar
Rubén Marín-Juez
View author publications
You can also search for this author in PubMed Google Scholar
Didier Y. R. Stainier
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

P.G., R.M.-J. and D.Y.R.S. conceptualized the study. P.G., R.M.-J. and D.Y.R.S. devised the study methodology. P.G., S.G., K.K. and M.L. carried out the investigation. All authors carried out the formal analysis. P.G. wrote the original paper draft. All authors reviewed and edited the paper draft. R.M.-J. and D.Y.R.S. supervised the study. D.Y.R.S. administered the project and acquired the funding.

Corresponding author

Correspondence to Didier Y. R. Stainier.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Cardiovascular Research thanks Herman Spaink, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Additional information

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

Extended data

Extended Data Fig. 1 Neutrophils and macrophages are affected in cryoinjured myd88^−/− ventricles.

a, UMAP representation of the macrophage and neutrophil clusters from the scRNA-seq analysis. b, Violin plots showing expression of known macrophage and neutrophil gene markers. c, Violin plots showing expression of neutrophil-enriched gene markers in the macrophage and neutrophil clusters. Genes selected from published datasets^34,35,36. d, UMAP representation of the myeloid subclusters from the scRNA-seq analysis. Area (iii) corresponds to a macrophage subcluster that is increased in cryoinjured myd88^−/− ventricles. e, UMAP representation of this myd88^−/− macrophage subcluster (blue) and the other macrophages (orange), and violin plots showing enriched genes in the myd88^−/− macrophage subcluster (iii). f, Representative images of immunostaining for GFP (neutrophils, white) with DAPI (DNA marker, blue) counterstaining on sections of TgBAC(mpx:GFP); myd88^+/+ and TgBAC(mpx:GFP); myd88^−/− ventricles at 6 hps, 6 hpci and 7 dpci. g, mpx:GFP⁺ cell numbers in TgBAC(mpx:GFP); myd88^+/+ and TgBAC(mpx:GFP); myd88^−/− ventricles at 6 hps and injured tissues and border zone areas (100 μm) at 6, 24 and 96 hpci, as well as 7 dpci. The dots in the graph represent individual ventricles; data are shown as the mean ± s.d.; n = 4 myd88^+/+, n = 5 myd88^−/− for 6 hps, n = 5 myd88^+/+, n = 5 myd88^−/− for 6 hpci, n = 9 myd88^+/+, n = 7 myd88^−/− for 24 hpci, n = 7 myd88^+/+, n = 8 myd88^−/− for 96 hpci, and n = 3 myd88^+/+, n = 4 myd88^−/− for 7 dpci. Statistical tests: Student’s t-test for 6 hps, 6, 24 hpci and 7 dpci, and Mann-Whitney U-test for 96 hpci. h, The pie charts show the proportion of macrophages in the scRNA-seq dataset at 24 hpci. Yellow dashed lines delineate the injured area; yellow arrowheads point to mpx:GFP⁺ cells. Scale bars, 100 μm.

Source data

Extended Data Fig. 2 Fibrotic cell populations in cryoinjured myd88^−/− ventricles.

a, Trajectory analysis revealing a potential transition from the serpine1 endocardial cluster to the collagen-rich, irx5a, and frzb endocardial clusters. b, Velocity analysis done by comparing pre-mRNA and mRNA levels to infer the relations between the cells in the endocardial clusters. c, Representative images of immunostaining for αSMA (myofibroblasts, white) with DAPI (DNA marker, blue) counterstaining on sections of cryoinjured myd88^+/+ and myd88^−/− ventricles at 7 dpci. d, Total αSMA⁺ cell numbers in myd88^+/+ and myd88^−/− injured tissues at 7 dpci. The dots in the graph represent individual ventricles; data are shown as the mean ± s.d.; n = 4 myd88^+/+ and n = 5 myd88^−/−. Statistical test: Student’s t-test. e, Superficially localized αSMA⁺ cell numbers in myd88^+/+ and myd88^−/− injured tissues at 96 hpci. The dots in the graph represent individual ventricles; data are shown as the mean ± s.d.; n = 13 myd88^+/+ and n = 11 myd88^−/−. Statistical test: Mann-Whitney U-test. f, UMAP representation of the mesenchymal subclusters from the scRNA-seq analysis. g, UMAP representation of acta2 expression levels in the mesenchymal subclusters. h, Heatmap showing gene expression levels in the mesenchymal subclusters. Yellow dashed lines delineate the injured area; yellow arrowheads point to αSMA⁺ cells. Scale bars, 100 μm.

Source data

Extended Data Fig. 3 Endocardial area coverage is not affected in cryoinjured myd88^−/− ventricles.

a, Representative images of immunostaining for EGFP (endocardial cells, green) and Aldh1a2 (endocardial and epicardial cell activation marker, magenta) with DAPI (DNA marker, blue) counterstaining on sections of cryoinjured ET(krt4:EGFP); myd88^+/+ and ET(krt4:EGFP); myd88^−/− ventricles at 96 hpci. b, Ratio of krt4:EGFP⁺ cell area within the injured tissue in cryoinjured myd88^+/+ and myd88^−/− ventricles at 96 hpci. The dots in the graph represent individual ventricles; data are shown as the mean ± s.d.; n = 7 myd88^+/+ and n = 6 myd88^−/−. Statistical test: Student’s t-test. c, Representative images of immunostaining for Cdh5 (endothelial cell marker, white) on 50-μm-thick sections of cryoinjured myd88^+/+ and myd88^−/− ventricles at 7 dpci. d, Ratio of Cdh5⁺ cell area within the injured tissue in cryoinjured myd88^+/+ and myd88^−/− ventricles at 7 dpci. The dots in the graph represent individual ventricles; data are shown as the mean ± s.d.; n = 7 myd88^+/+ and n = 4 myd88^−/−. Statistical test: Mann-Whitney U-test. Yellow dashed lines delineate the injured area. Scale bars, 100 μm.

Source data

Extended Data Fig. 4 Bigger scars at 30 but not at 90 dpci in cryoinjured myd88^−/− ventricles.

a, Graph showing the percentage of the collagen/scar area at 14 and 30 dpci. The dots in the graph represent individual ventricles, and data are shown as the mean ± s.d.; n = 6 myd88^+/+ and n = 6 myd88^−/−. Statistical tests: Student’s t-test. b, Graph showing representation of groups (y axis) of different scar area sizes (different colors) at 30 dpci in cryoinjured myd88^+/+ and myd88^−/− ventricles. c, Representative images of AFOG staining on sections of cryoinjured myd88^+/+ and myd88^−/− ventricles at 90 dpci. d, Graph showing representation of groups (y axis) of different scar area sizes (different colors) at 90 dpci in cryoinjured myd88^+/+ and myd88^−/− ventricles. Black dashed lines delineate the scar area. Scale bars, 200 μm.

Source data

Extended Data Fig. 5 Enrichment analysis terms affected in the cryoinjured myd88^−/− endocardium at 96 hpci.

a, Selected GSEA plots for GO and KEGG pathway terms from transcriptomic analysis in the cryoinjured myd88^−/− versus myd88^+/+ endocardium at 96 hpci. Plots with red and blue lines indicate processes enriched in myd88 (mutant; myd88^−/−) and WT (myd88^+/+), respectively. b, Plot showing Reactome pathway terms enriched in the cryoinjured myd88^+/+ (WT) versus myd88^−/− (myd88 mutant) endocardium at 96 hpci.

Source data

Extended Data Fig. 6 pERK⁺ area coverage is not significantly affected in cryoinjured myd88^−/− ventricles.

a, Representative images of immunostaining for EGFP (endocardial cells, green) and pERK (activated MAPK signaling pathway, magenta) with DAPI (DNA marker, blue) counterstaining on sections of cryoinjured ET(krt4:EGFP); myd88^+/+ and ET(krt4:EGFP); myd88^−/− ventricles at 96 hpci. b, Ratio of pERK⁺ cell area within the injured tissue in cryoinjured myd88^+/+ and myd88^−/− ventricles at 96 hpci. The dots in the graph represent individual ventricles; data are shown as the mean ± s.d.; n = 7 myd88^+/+ and n = 5 myd88^−/−. Statistical test: Student’s t-test. Yellow dashed lines delineate the injured area. Scale bars, 100 μm.

Source data

Extended Data Fig. 7 Revascularization phenotype in cryoinjured myd88^−/− ventricles.

a, c, Representative images of immunostaining for RFP (coronary endothelial cells, green) and PCNA (proliferation marker, magenta) with DAPI (DNA marker, blue) counterstaining on sections of cryoinjured Tg(-0.8flt1:RFP); myd88^+/+ and Tg(-0.8flt1:RFP); myd88^−/− ventricles at 96 hpci (a) and 7 dpci (c). b, d, Percentage of proliferating coronary endothelial cells in the injured tissues and border zone areas (200 μm) at 96 hpci (b) and 7 dpci (d). The dots in the graphs represent individual ventricles; data are shown as the mean ± s.d.; n = 7 myd88^+/+, n = 6 myd88^−/− at 96 hpci (b), and n = 4 myd88^+/+, n = 5 myd88^−/− at 7 dpci (d). Statistical tests: Student’s t-test. e, Wholemount images of cryoinjured Tg(-0.8flt1:RFP); myd88^+/+ and Tg(-0.8flt1:RFP); myd88^−/− ventricles at 7 dpci. f, Percentage of coronary vessel coverage in the injured tissues of cryoinjured Tg(-0.8flt1:RFP); myd88^+/+ and Tg(-0.8flt1:RFP); myd88^−/− ventricles at 7 dpci. The dots in the graphs represent individual ventricles; data are shown as the mean ± s.d.; n = 4 myd88^+/+ and n = 6 myd88^−/−. Statistical test: Student’s t-test. Yellow dashed lines delineate the injured area; yellow arrowheads point to proliferating coronary endothelial cells. Scale bars, 100 μm (a,c), 200 μm (e).

Source data

Extended Data Fig. 8 CM dedifferentiation at 96 hpci and proliferation at 14 dpci are not significantly affected in cryoinjured myd88^−/− ventricles.

a, Representative images of immunostaining for MEF2 (CM nuclei, green) and N2.261 (dedifferentiation marker, magenta) on sections of cryoinjured myd88^+/+ and myd88^−/− ventricles at 96 hpci. b, Quantification of the number of dedifferentiating CMs in border zone areas (100 μm) at 96 hpci. The dots in the graphs represent individual ventricles; data are shown as the mean ± s.d.; n = 5 myd88^+/+ and n = 5 myd88^−/−. Statistical test: Student’s t-test. c, Representative images of immunostaining for MEF2 (CM nuclei, green) and PCNA (proliferation marker, magenta) on sections of cryoinjured myd88^+/+ and myd88^−/− ventricles at 14 dpci. d, Quantification of proliferating CMs in border zone areas (100 μm) at 14 dpci. The dots in the graphs represent individual ventricles; data are shown as the mean ± s.d.; n = 5 myd88^+/+ and n = 4 myd88^−/−. Statistical test: Student’s t-test. Yellow dashed lines delineate the injured area; yellow arrowheads point to dedifferentiating (a) and proliferating (c) CMs. Scale bars, 100 μm.

Source data

Extended Data Fig. 9 Validation of the Tg(fli1a:myd88,EGFP) line.

a, RT-qPCR analysis of myd88 mRNA levels in transgenic (Tg(fli1a:myd88,EGFP)) and non-transgenic 24 hpf larvae and 96 hpci ventricles. Data are shown as the mean ± s.d.; n = 3 (no transgene) and n = 3 (Tg(fli1a:myd88,EGFP)) 24 hpf larvae and n = 6 (no transgene) and n = 6 (Tg(fli1a:myd88,EGFP)) 96 hpci ventricles. Statistical tests: Student’s t-test. Ct values listed in Supplementary Table 3. b, Representative images of immunostaining for EGFP (endothelial cells, green) and Aldh1a2 (endocardial and epicardial cell activation marker, magenta) with DAPI (DNA marker, blue) counterstaining on a section of a cryoinjured Tg(fli1a:myd88,EGFP); myd88^+/+ ventricle at 96 hpci. Yellow dashed lines delineate the injured area. Scale bars, 100 μm.

Source data

Extended Data Fig. 10 Validation of cxcl18b tools.

a, Representative images of immunostaining for EGFP (cxcl18b expression, white) with DAPI (DNA marker, blue) counterstaining on a section of a cryoinjured Tg(cxcl18b:EGFP) ventricle at 24 hpci; n = 1. b, Experimental plan for Cre mRNA injected and uninjected (control) Tg(hsp70l:LBL-cxcl18b-t2a-mCherry) sibling larvae. c, RT-qPCR analysis of cxcl18b and mCherry mRNA levels in Cre mRNA-injected and uninjected (control) Tg(hsp70l:LBL-cxcl18b-t2a-mCherry) larvae. The dots in the graph represent individual larvae; data are shown as the mean ± s.d.; n = 4 (Control) and n = 4 (Cre mRNA injected). Statistical tests: Student’s t-test. Ct values listed in Supplementary Table 3. d, Representative images of immunostaining for mCherry (cxcl18b overexpressing cells, white) on sections of cryoinjured ventricles from uninjected (control) myd88^+/−, Cre mRNA injected (cxcl18b OE) myd88^+/−, uninjected (control) myd88^−/− and Cre mRNA injected (cxcl18b OE) myd88^−/− siblings at 24 hpci; all zebrafish carry the hsp70l:LBL-cxcl18b-t2a-mCherry transgene; n = 6 (myd88^+/− control), n = 5 (myd88^+/− cxcl18b OE), n = 6 (myd88^−/− control), and n = 5 (myd88^−/− cxcl18b OE). Experimental plan in Fig. 6d. e, Genomic sequence of the cxcl18b^bns683 full locus deletion allele. f, Experimental plan and RT-qPCR analysis of cxcl18b mRNA levels in cxcl18b^+/+ and cxcl18b^−/− larvae 6 hours post fin fold amputation (hpa). The dots in the graph represent individual larvae; data are shown as the mean ± s.d.; n = 4 cxcl18b^+/+ and n = 3 cxcl18b^−/−. Statistical test: Student’s t-test. Ct values listed in Supplementary Table 3. Yellow dashed lines delineate the injured area. Scale bars, 100 μm.

Source data

Supplementary information

Supplementary Information

Supplementary Figs. 1 and 2, Tables 1–3, Methods and References.

Reporting Summary

Source data

Source Data

Statistical source data for Figs. 1–7 and Extended Data Figs. 1–10.

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

Goumenaki, P., Günther, S., Kikhi, K. et al. The innate immune regulator MyD88 dampens fibrosis during zebrafish heart regeneration. Nat Cardiovasc Res 3, 1158–1176 (2024). https://doi.org/10.1038/s44161-024-00538-5

Download citation

Received: 07 April 2024
Accepted: 06 August 2024
Published: 13 September 2024
Issue Date: September 2024
DOI: https://doi.org/10.1038/s44161-024-00538-5

This article is cited by

Success in heart regeneration depends on endocardial innate immune signaling
- Kazu Kikuchi
Nature Cardiovascular Research (2024)

Subjects

Abstract

Similar content being viewed by others

Main

Results

Reduced inflammatory cell numbers in injured myd88 −/− ventricles

MyD88 signaling attenuates fibrosis in injured ventricles

Endocardial MyD88 signaling activates the PI3K/AKT pathway

MyD88 signaling promotes revascularization and CM repopulation

Endothelial myd88 overexpression reverses myd88 −/− phenotypes

MyD88-Cxcl18b signaling promotes neutrophil recruitment after cardiac injury

Discussion

Methods

Zebrafish husbandry and handling

Zebrafish lines

Generation of zebrafish mutant and transgenic lines

Cardiac cryoinjury and heat shock treatments

Histological analysis and imaging

Quantification and statistical analysis

Tissue dissociation and cell sorting

scRNA-seq analysis

Bulk RNA-seq analysis

RT–qPCR

Reporting summary

Data availability

References

Acknowledgements

Funding

Author information

Authors and Affiliations

Contributions

Corresponding author

Ethics declarations

Competing interests

Peer review

Peer review information

Additional information

Extended data

Supplementary information

Source data

Rights and permissions

About this article

Cite this article

Share this article

This article is cited by

Search

Quick links

Reduced inflammatory cell numbers in injured myd88 ^−/− ventricles

Endothelial myd88 overexpression reverses myd88 ^−/− phenotypes