Author response: Apoptosis recognition receptors regulate skin tissue repair in mice
Year: 2023
Type: peer-review
Abstract: Full text Figures and data Side by side Abstract Editor's evaluation eLife digest Introduction Results Discussion Methods Data availability References Decision letter Author response Article and author information Abstract Apoptosis and clearance of apoptotic cells via efferocytosis are evolutionarily conserved processes that drive tissue repair. However, the mechanisms by which recognition and clearance of apoptotic cells regulate repair are not fully understood. Here, we use single-cell RNA sequencing to provide a map of the cellular dynamics during early inflammation in mouse skin wounds. We find that apoptotic pathways and efferocytosis receptors are elevated in fibroblasts and immune cells, including resident Lyve1+ macrophages, during inflammation. Interestingly, human diabetic foot wounds upregulate mRNAs for efferocytosis pathway genes and display altered efferocytosis signaling via the receptor Axl and its ligand Gas6. During early inflammation in mouse wounds, we detect upregulation of Axl in dendritic cells and fibroblasts via TLR3-independent mechanisms. Inhibition studies in vivo in mice reveal that Axl signaling is required for wound repair but is dispensable for efferocytosis. By contrast, inhibition of another efferocytosis receptor, Timd4, in mouse wounds decreases efferocytosis and abrogates wound repair. These data highlight the distinct mechanisms by which apoptotic cell detection coordinates tissue repair and provides potential therapeutic targets for chronic wounds in diabetic patients. Editor's evaluation The manuscript reports important new information on how dead cells are cleared from the wound site in order to promote effective repair of the damaged tissue. Interestingly, the authors find that the components of this clearance pathway are abnormally high in diabetics who have difficulty healing wounds and their results suggest that tampering them down may be a therapy to restore normal wound healing. https://doi.org/10.7554/eLife.86269.sa0 Decision letter Reviews on Sciety eLife's review process eLife digest Our skin is constantly exposed to potential damage from the outside world, and it is vital that any injuries are repaired quickly and effectively. Diabetes and many other health conditions can hamper wound healing, resulting in chronic wounds that are both painful and at risk of becoming infected, which can lead to serious illness and death of patients. After an injury to the skin, the wound becomes inflamed as immune cells rush to the site of injury to fight off infection and clear the wound of dead cells and debris. Some of these dead cells will have died by a highly controlled process known as apoptosis. These so-called apoptotic cells display signals on their surface that nearby healthy cells recognize. This triggers the healthy cells to eat the apoptotic cells to remove them from the wound. Previous studies have linked changes in cell death and the removal of dead cells to chronic wounds in patients with diabetes, but it remains unclear how removing dead cells from the wound affects healing. Justynski et al. used a genetic technique called single-cell RNA sequencing to study the patterns of gene activity in mouse skin cells shortly after a wound. The experiments found that, as the area around the wound started to become inflamed, the wounded cells produced signals of apoptosis that in turn triggered nearby healthy cells to remove them. Other signals relating to the removal of dead cells were also widespread in the mouse wounds and treating the wounds with drugs that inhibit these signals resulted in multiple defects in the healing process. Further experiments used the same approach to study samples of tissue taken from foot wounds in human patients with or without diabetes. This revealed that several genes involved in the removal of dead cells were more highly expressed in the wounds of diabetic patients than in the wounds of other individuals. These findings indicate that for wounds to heal properly it is crucial for the body to detect and clear apoptotic cells from the wound site. Further studies building on this work may help to explain why some diabetic patients suffer from chronic wounds and help to develop more effective treatments for them. Introduction Proper tissue function and homeostasis require efficient and effective repair of injury. Repair of mammalian tissues requires highly dynamic changes in cellular heterogeneity and communication to correctly heal tissue, usually resulting in a scar rather than true tissue regeneration. Cell death is a common event during tissue injury, and several studies from hydra to mice have shown the importance of apoptosis in the initiation of inflammation to drive reparative processes (Greenhalgh, 1998). Proper initiation and subsequent resolution of inflammation is essential for tissue repair and progression to the proliferation stage of healing, when fibroblasts, blood vessels, and other tissue-specific cells proliferate and migrate, forming new tissue to repair the wound. While several signaling factors have been shown to induce apoptosis in wounds (Guerin et al., 2021), the mechanisms by which apoptotic cells are recognized and regulate tissue repair are not well understood. The skin is an excellent model to define the mechanisms by which apoptotic cells regulate tissue repair. After injury, mammalian skin undergoes stages of repair beginning with inflammation, which removes debris and pathogens. As inflammation regresses, the proliferative phase leads to the coordination of epidermal keratinocytes, fibroblasts, endothelial, and immune cells to reseal the epidermal barrier and generate a reparative scar including new ECM production and revascularization (Eming et al., 2014). Apoptosis occurs after skin injury, and phagocytosis of apoptotic cells – or efferocytosis – by macrophages reduces inflammatory signaling and repair in several tissues (Bosurgi et al., 2017a; Peiseler and Kubes, 2019). Yet, it is unclear how apoptosis controls skin wound healing. Apoptotic cell death is characterized by cytomorphological alterations, DNA fragmentation, activation of caspases and other regulators, and finally membrane alterations including outer membrane exposure of phosphatidylserine (PtdSer), which allows the recognition of apoptotic cells by cellular receptors on phagocytes (Elmore, 2007). The most well-studied receptors that allow phagocytes to bind and phagocytose apoptotic cells include the TAM (Tyro3, Axl, and Mertk) tyrosine kinases and the TIM (T cell immunoglobulin and mucin domain) family of receptors (Lemke, 2019). While TIM receptors can directly bind PtdSer, TAM receptors require their ligands growth-arrest-specific 6 (Gas6) and protein S (Pros1) to bind to PtdSer (Lew et al., 2014; Elliott et al., 2017). To understand the cellular and molecular mechanisms by which apoptosis regulates skin wound healing, we performed single-cell RNA sequencing (scRNA-seq) on cells from murine wound beds 24 hr and 48 hr after injury. We found that transcriptional alterations in apoptotic pathways occur in this interval in fibroblasts, monocytes/macrophages, neutrophils, and dendritic cells. In addition, inhibition of two efferocytosis receptors, Axl and Timd4, abrogates proper wound repair. The results provide an atlas of cellular dynamics during the early stages of wound healing and reveal the essential role of the recognition and clearance of apoptotic cells in driving tissue repair after injury. Results Dramatic transcriptional heterogeneity during early skin inflammation after injury To assess cellular and molecular heterogeneity of the wound bed during the inflammatory phase, we performed scRNA-seq on cells isolated from 4 mm full-thickness biopsy punches on mouse back skin at 24 hr and 48 hr after injury (Shook et al., 2016). To ensure that we also captured the immediately adjacent tissue as well as cells that may have migrated into the wound site, we used a 6 mm biopsy punch to collect the tissue before isolating cells using enzymatic digestion (Figure 1A). By training a neural network to identify cell types based on expression of established marker genes (as in Wasko et al., 2022; Figure 1—figure supplement 1A–B), we classified four major cell types in the scRNA-seq data, including monocytes/macrophages (Mono/MO), neutrophils (Neut), dendritic cells (DC), and fibroblasts (FB) at both timepoints (Figure 1B–E). Figure 1 with 1 supplement see all Download asset Open asset Dynamic transcriptional heterogeneity and apoptosis are observed in murine wound beds 24 hr and 48 hr after injury. (A) Schematic of experimental design. (B) UMAP plot of single-cell RNA sequencing (scRNA-seq) data for cells from 24 hr wound beds in murine back skin. (C) UMAP plot of scRNA-seq data for cells from 48 hr wound beds in murine back skin. (D) UMAP plot of scRNA-seq data for cells from both 24 hr and 48 hr wound beds in murine back skin annotated by timepoint. (E) Top: UMAP plot of scRNA-seq data for cells from 24 hr and 48 hr wound beds annotated by cell identity. Bottom: Heatmap of differentially expressed marker genes in 24 hr and 48 hr wound beds. (F) Gene ontology terms enriched in each cell type compared to the full dataset. (G) Heatmap of differentially expressed apoptosis-related genes from 24 hr and 48 hr wound beds. (H) Immunostaining for cleaved caspase 3 (Act. Casp3) (green) in wound bed 24 hr after injury. Arrow indicates cleaved caspase 3+ cells. (I) Immunostaining for CD11c (green) and cleaved caspase 3 (red) in wound bed 24 hr after injury. Arrows indicate double-positive CD11c+ cleaved caspase 3+ cells. (J) Immunostaining for GFP (green) and cleaved caspase 3 (red) in PdgfraCreER;mTmG wound bed 24 hr after injury. Arrow indicates double-positive GFP+ cleaved caspase 3+ cells. * indicates scab. Scale bars = 500 µm. In E and G, expression indicates scaled log-normalized mRNA counts. Surprisingly, the 24 hr and 48 hr samples clustered separately with minimal overlap, suggesting that dramatic changes occurred in the first 2 days of inflammation (Figure 1B–D). The absolute number of macrophages and neutrophils increased >2-fold over this interval (Figure 1—figure supplement 1C). While approximately 70% of each of the three immune cell populations was collected at 48 hr, the majority of fibroblasts were found in the 24 hr population (Figure 1—figure supplement 1D). We also found that genes upregulated by each cell type were markedly different between timepoints (Figure 1—figure supplement 1E), indicating major changes to expression patterns in the same cell type over time. To explore the changes in the major cell types in early wound beds (fibroblasts, neutrophils, DCs, and monocytes/macrophages), we analyzed the genes that were significantly upregulated in each group relative to the mRNAs expressed by the full dataset to determine gene ontology (GO) terms that were enriched for each cell type (Figure 1F). Monocytes/macrophages, neutrophils, and DCs upregulated mRNAs involved in their specific function in inflammation for cytokine production, chemotaxis, and antigen presentation, respectively. Similarly, fibroblasts uniquely upregulated genes involved in extracellular matrix (ECM) organization. Interestingly, DCs, fibroblasts, and neutrophils upregulated genes involved in apoptosis and cell death (Figure 1F and G). Neutrophils upregulated the largest number of mRNAs for apoptosis including Casp3, which is cleaved to activate apoptosis, and S100a8 and S100a9, which induce apoptosis of several cell types (Yui et al., 2003). Fibroblasts and DCs also upregulated apoptotic genes ranging from receptors (Fgfr1 and Bmpr1a), cytokines (Il12b), to mediators of apoptotic pathways (Bcl2l14 and Cth). Since apoptosis also involves post-transcriptional activation of several proteins, we sought to confirm that apoptosis occurred during early skin wound inflammation in vivo. Sections of mouse wounds at 24 hr (Figure 1H) and 48 hr (Figure 1—figure supplement 1F) were stained with antibodies (Abs) against cleaved (active) caspase 3 to detect activation of apoptosis. We found that significantly more cleaved caspase 3+ cells were present in 48 hr wounds (Figure 1—figure supplement 1F). We also detected cleaved caspase 3 in CD11c+ dendritic cells (Figure 1I) and Pdgfra+ fibroblasts (Figure 1J) in 24 hr wound beds. Overall, these data suggest that apoptosis occurs during inflammation after skin injury. Apoptosis recognition receptors, ligands, and downstream factors are expressed in the wound bed Given the upregulation of mRNAs associated with apoptosis and the relatively low level of apoptotic cells in early wound beds, we hypothesized that efficient efferocytosis clears apoptotic cells during inflammation. Initially, we inspected mRNA levels of efferocytosis receptors, their ligands, and downstream factors in scRNA-seq data from 24 hr and 48 hr wounds (Figure 2A). Interestingly, fibroblasts upregulated several genes encoding efferocytosis receptors (Tyro3, Axl, Itgav, and Slc7a11), as well as genes encoding for ligands Gas6, Pros1, C3, C4b, and Mfge8. DCs, macrophages, and neutrophils also upregulated several receptors that mediate detection of apoptotic cells. Macrophages were enriched for the downstream activators Arg1 and Retnla, whereas other cell types upregulated Socs1 and Socs3, which are downstream of the TAM receptors (Rothlin et al., 2007). Figure 2 with 1 supplement see all Download asset Open asset Apoptosis detection genes are highly expressed in the wound bed. (A) Heatmap of differentially expressed efferocytosis pathway genes in 24 hr and 48 hr wound beds. Expression indicates scaled log-normalized mRNA counts. (B) Feature plots showing expression of Gas6, Lyve1, Timd4, and Retnla with Lyve1+ region highlighted. (C) Immunostaining for GFP (green) and Lyve1 (red) in Lyz2CreER;mTmG wound bed 24 hr after injury. Arrows indicate Lyve1+ cells. (D) Immunostaining for Gas6 (green) and Lyve1 (red) in wild-type (WT) wound bed 24 hr after injury. Arrows indicate double-positive Gas6+ Lyve1+ cells. (E) Immunostaining for Timd4 (green) in WT wound bed 24 hr after injury. Arrows indicate Timd4+ cells. (F) Immunostaining for CD11c (green) and Axl (red) in wound bed 24 hr after injury. Arrows indicate double-positive CD11c+ Axl+ cells. (G) Immunostaining for GFP (green) and Axl (red) in PdgfraCreER;mTmG wound bed 24 hr after injury. Arrows indicate double-positive GFP+ Axl+ cells. (H) Schematic of experimental design. (I) Immunostaining for CFSE-stained apoptotic neutrophils in wound beds 1 and 5 days after injury. (J) Example of CFSE-stained whole neutrophil in the wound bed. Scale bar = 100 µm. (K) Example of active efferocytosis of CFSE-stained whole neutrophil in the wound bed. Scale bar = 100 µm. (L) Quantification of instances of efferocytosis observed per mm2 in the wound bed. (M) Quantification of percentage of all stained cells that are undergoing efferocytosis in the wound bed. In L and M, n=6, error bars indicate mean ± SEM, unpaired t-test, *p<0.05, **p<0.01. * indicates scab. In C–I, scale bars = 500 µm. We noted that Axl's ligand, Gas6, and several other genes involved in efferocytosis were expressed predominantly in the fibroblast cluster, but were also lowly expressed in the monocyte/macrophage cluster (Figure 2A). Examining the UMAP plot to determine the heterogeneity of efferocytosis gene expression within individual cell types, we observed that Gas6 was highly expressed by a specific subset of monocyte/macrophage cells. These cells also overexpressed the resident macrophage marker Lyve1, the apoptosis receptor Timd4, and Retnla, a downstream factor of efferocytosis (Figure 2B), indicating that they may play a role in apoptosis detection and response in wound healing. Lyve1 has been identified as a marker for resident macrophages, which are distinct from the majority of wound macrophages that differentiate from bone marrow-derived monocytes and are recruited to the wound after injury (Lim et al., 2018; Wang et al., 2020). To confirm the presence of Lyve1+ resident cells in vivo, we used a Lyz2CreER;mTmG mouse model, in which myeloid cells in the bone marrow can be induced to express GFP prior to injury, such that any GFP+ cells observed in the wound bed are interpreted as newly recruited to the site of injury, while GFP- cells are interpreted to be resident to the skin. We observed that Lyve1 was expressed at the protein level both in wound beds (Figure 2C) and adjacent to the wound (Figure 2—figure supplement 1A) with immunofluorescence staining, and confirmed that these cells were resident rather than recruited to the wound bed after injury, since they did not express GFP. Further, we confirmed that Lyve1+ cells co-expressed Gas6 protein (Figure 2D) and that Timd4 protein (Figure 2E) was expressed in wound beds. The TAM receptor Axl was uniquely expressed in the single-cell dataset by both dendritic cells and fibroblasts (Figure 2A). Using immunofluorescence staining, we confirmed Axl expression in 24 hr wound beds in both dendritic cells (CD11c) (Figure 2F) and fibroblasts (Pdgfra-GFP) (Figure 2G), which displayed elongated GFP+ processes with central localization of Axl. Thus, while we cannot rule out Axl expression by other cell types in skin wounds, these data are consistent with our single-cell results, confirming that Axl is expressed by fibroblasts and dendritic cells during the inflammatory response of skin repair. We also analyzed mRNA expression of apoptosis-related genes by qPCR to determine their expression in wound beds compared to naive skin (Figure 2—figure supplement 1B). All efferocytosis receptor mRNAs studied, including the TAM receptors, were significantly upregulated in the wound bed compared to naive skin, while the ligand Gas6 was significantly downregulated. The downstream factor of efferocytosis Retnla was also significantly downregulated in the wound bed, while Socs1 and Socs3 were significantly upregulated. To analyze efferocytosis activity after injury, we intradermally injected labeled apoptotic neutrophils from bone marrow into wound beds 1, 3, and 5 days after injury. Wound beds were collected 1 hr after injections, sectioned, and imaged to quantify labeled cells (Figure 2H). We detected and quantified labeled intact neutrophils and efferocytosed neutrophils in wounds at all three timepoints (Figure 2H–M). While wounds at 5 days had significantly more cells undergoing efferocytosis (Figure 2L), the efferocytosis rate was generally constant, with a slight but nonsignificant increase at 3 days (Figure 2J). Taken together, these data indicate that the machinery for multiple efferocytosis pathways are active in the inflammatory and proliferative stages of wound healing. Cell death signaling in human diabetic and non-diabetic wounds We next set out to explore whether these apoptosis and efferocytosis-related pathways were also relevant in pathological states associated with dysregulated wound healing, such as diabetes. Previous studies have indicated that apoptosis is increased in diabetic wounds, including an elevation in apoptotic lymphocytes (Arya et al., 2014). Additional studies have shown that hypoxic environments (such as those found in diabetic wounds) increase macrophage efferocytosis (Wang et al., 2023), though other studies have indicated that macrophage efferocytosis is defective in diabetic wounds (Khanna et al., 2010). To analyze the expression of transcriptional changes in apoptosis and efferocytosis signaling in foot wounds, we analyzed genes in these categories in four non-diabetic and five diabetic human patients (to be published elsewhere). We found several alterations in the expression of mRNAs associated with apoptotic signaling pathways between non-diabetic and diabetic foot ulcers (Figure 3—figure supplement 1A). Specifically, monocytes/macrophages increased expression of several apoptosis genes (including BCL2, FGFR3, FGFR1, CTH, CASP3, IL19, NME5, and S100A8/9) in diabetic foot ulcers compared to monocytes/macrophages in non-diabetic wounds. When we compared the expression of genes associated with the efferocytosis pathway between non-diabetic and diabetic foot ulcers, we found a striking increase in overall expression of efferocytosis pathway genes in diabetic wounds compared to non-diabetic wounds (Figure 3A). In particular, the expression of AXL increased in all cell types in diabetic wounds, while expression of its ligand GAS6 increased in all cell types with the exception of basal keratinocytes and mast cells. Figure 3 with 1 supplement see all Download asset Open asset Human diabetic wounds have increased efferocytosis signaling expression compared to non-diabetic wounds. (A) Heatmaps of differentially expressed genes related to efferocytosis in non-diabetic and diabetic wound beds. Expression indicates scaled log-normalized mRNA counts. (B) CellChat circle plot diagrams showing GAS6-AXL communication in non-diabetic and diabetic wound beds. (C) Left: Immunostaining for CD68 (green) and Gas6 (red) in non-diabetic foot wound. Right: Immunostaining for CD68 (green) and Gas6 (red) in diabetic foot ulcer. Scale bars = 50 µm. (D) Quantification of Gas6 corrected total fluorescence. Error bars indicate mean ± SEM. n=2 for Non-diabetic, and 6 for diabetic foot wound; ns, nonsignificant. To further investigate the AXL/GAS6 signaling pathway, we analyzed these data with CellChat, which quantitatively infers intercellular signaling networks from scRNA-seq data (Jin et al., 2021). This analysis revealed altered signaling between AXL and its ligand GAS6 between the two groups. In non-diabetic patient wounds, GAS6 expression by several cell types, including monocytes/macrophages and fibroblasts, stimulated signaling via AXL receptors on fibroblasts and pericytes. However, in diabetic patient wounds, GAS6 and AXL expression was more robust in several cell types. Notably, monocytes/macrophages increased GAS6 stimulation and newly expressed AXL to receive pathway signaling in diabetic wounds (Figure 3B). To validate these findings, we stained sections of non-diabetic foot wounds and diabetic foot ulcers with Abs against Gas6 and CD68, a macrophage marker. Indeed, we found that Gas6 staining was more prevalent in diabetic skin sections, and that it colocalized with the macrophage marker (Figure 3C). Further, when quantified via corrected total fluorescence (CTF), more Gas6 staining was observed in diabetic samples, though this was not significant. This supported the CellChat results, indicating that macrophages express Gas6 more highly in diabetic wounds. Taken together, these data indicate that diabetic wounds may activate and modulate GAS6-AXL signaling, and suggest a potential avenue for future stimulation is but not for Axl upregulation in skin we sought to the molecular mechanisms that induce Axl mRNA expression after injury. work that Axl expression was induced by receptor 3 stimulation (Rothlin et al., and that is essential for skin wound repair et al., Thus, we the role of signaling in Axl expression in the skin. scRNA-seq of early wounds revealed that mRNAs are highly expressed in neutrophils and macrophages with a expressed in dendritic cells and fibroblasts (Figure Interestingly, is the in that it is expressed by both dendritic cells and fibroblasts, which also express high levels of Axl in the single-cell dataset (Figure 2A). To determine stimulation Axl expression in the skin, we injected the RNA or a in naive mouse back skin of wild-type (WT) or We collected the site and area after 2 hr and the skin samples for qPCR and (Figure We first analyzed cytokine mRNA a of signaling that inflammation (Rothlin et al., 2007). While several inflammatory cytokines were not induced in skin injected with (Figure supplement was upregulated in WT mice injected with but not in the skin of mice (Figure confirming the of for activation of in the skin et al., Axl protein (Figure and mRNA (Figure were induced in naive skin and Axl mRNA and protein expression were in skin of mice (Figure and Figure supplement 1B). Axl protein expression around the site was quantified via (Figure confirming these Taken together, these data indicate that signaling is to activate Axl expression in naive skin. Figure 4 with 1 supplement see all Download asset Open asset signaling is to upregulate Axl in naive skin, but is not required for Axl expression in the wound bed. (A) Heatmap of differentially expressed genes in 24 hr and 48 hr wound beds. Expression indicates scaled log-normalized mRNA counts. (B) Schematic of and (C) mRNA expression of genes relative to indicates mRNA Error bars indicate mean ± SEM, unpaired t-test, **p<0.01. ns, nonsignificant. (D) Immunostaining for Axl (red) in naive back skin injected with or Arrows indicate Axl+ cells. * indicates site. (E) mRNA expression of Axl relative to 2 hr after indicates mRNA Error bars indicate mean ± SEM, unpaired t-test, ns, nonsignificant. (F) Quantification of corrected total fluorescence for Axl in a 1 mm the site. Error bars indicate mean ± SEM, with multiple (G) Schematic of and (H) Immunostaining for Axl (red) in naive back skin injected with or Arrows indicate Axl+ cells. * indicates site. (I) Immunostaining for Axl (red) in wound beds 24 hr after injury in wild-type (WT) or Arrows indicate Axl+ cells. * indicates scab. Scale bars = 500 µm. Since was elevated by signaling in naive skin, we whether intradermally into naive mouse back skin was to induce Axl protein expression (Figure after Axl protein was detected by in skin injected with but not skin (Figure and to determine activation was for Axl expression in skin wounds, we analyzed Axl expression in wound beds of In to our results in naive skin, Axl protein expression was stimulated in wound beds to WT mice (Figure Thus, these data suggest that while is to drive Axl expression in naive skin, mechanisms drive Axl upregulation within skin wounds in its Axl is required for skin wound healing on our data showing upregulation of Axl mRNA and protein expression in wound beds, in the of we hypothesized that Axl may play a role in wound healing. To whether Axl signaling was required for wound we injected mice 3 hr prior to injury with a or function which to Axl's extracellular and infection et al., and has been shown to inhibit Axl activity in et al., Figure Since Axl activity Axl mRNA expression in a et al., we analyzed Axl mRNA expression in wounds of While we a nonsignificant of Axl mRNA at 1 (D) injury, Axl mRNA was significantly downregulated in skin wounds with on 5 compared to injected wounds (Figure indicating that the Axl signaling in skin wounds. Figure 5 with 1 supplement see all Download asset Open asset Axl inhibition results in defects to wound repair and changes to inflammation. (A) Schematic of experimental design. (B) Axl mRNA expression to Error bars indicate mean ± SEM, with multiple (C) used to analyze cells isolated from wounds 3 days after and injury. cells were identified as macrophages, neutrophils, or dendritic cells via (D) Top: Quantification of cells by cell type in or wound beds 3 days after injury. Error bars indicate mean ± SEM, with multiple ns, nonsignificant. Bottom: Quantification of Axl+ cells by cell type in or wound beds 3 days after injury. Error bars indicate mean ± SEM, with multiple **p<0.01. ns, nonsignificant. (E) Left: and staining of wound beds 5 days after and injury. Immunostaining for (green) in wound beds 5 days after and injury. Right: Immunostaining for (green) in wound beds 5 days after and injury. * indicates scab. (F) Top:
Source: eLife
Language: English
Topic: Wound Healing and Treatments
Subfield: Rehabilitation
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $2,500
OpenAlex ID: W4390066676