Regenerating Skeletal Muscle Compensates for the Impaired Macrophage Functions Leading to Normal Muscle Repair in Retinol Saturase Null Mice

Skeletal muscle repair is initiated by local inflammation and involves the engulfment of dead cells (efferocytosis) by infiltrating macrophages at the injury site. Macrophages orchestrate the whole repair program, and efferocytosis is a key event not only for cell clearance but also for triggering the timed polarization of the inflammatory phenotype of macrophages into the healing one. While pro-inflammatory cytokines produced by the inflammatory macrophages induce satellite cell proliferation and differentiation into myoblasts, healing macrophages initiate the resolution of inflammation, angiogenesis, and extracellular matrix formation and drive myoblast fusion and myotube growth. Therefore, improper efferocytosis results in impaired muscle repair. Retinol saturase (RetSat) initiates the formation of various dihydroretinoids, a group of vitamin A derivatives that regulate transcription by activating retinoid receptors. Previous studies from our laboratory have shown that RetSat-null macrophages produce less milk fat globule-epidermal growth factor-factor-8 (MFG-E8), lack neuropeptide Y expression, and are characterized by impaired efferocytosis. Here, we investigated skeletal muscle repair in the tibialis anterior muscle of RetSat-null mice following cardiotoxin injury. Our data presented here demonstrate that, unexpectedly, several cell types participating in skeletal muscle regeneration compensate for the impaired macrophage functions, resulting in normal muscle repair in the RetSat-null mice.


Introduction
Skeletal muscle is frequently injured, but temporary damage is compensated by the remarkable regenerative capacity of this tissue [1]. The regeneration process involves several interrelated phases. In the first inflammation phase, tissue-resident macrophages (Mφs) sensing the damage initiate a local, sterile inflammation and recruit further immune cells to the injury site. Then, quiescent myogenic stem cells, called satellite cells (SCs), become activated and differentiate into myoblasts, which proliferate and fuse. Finally, in the tissue remodeling phase, the growth of muscle fibers, resolution of inflammation, revascularization, and innervation of new fibers take place [2]. University of Debrecen. To check whether the RetSat −/− mice are indeed full knock out, we determined the NEO cassette expression by qRT-PCR in several tissues and found it to be expressed (Supplementary Figure S1). All animal experiments were approved by the Animal Care and Use Committee of the University of Debrecen, with permission numbers 7/2016 and 7/2021/DEMÁB.

Cardiotoxin-Induced Muscle Injury Model
Mice were anesthetized with 2.5% isoflurane using a SomnoSuite device. The muscle damage was induced by injecting into the tibialis anterior (TA) muscle 50 µL of 12 µM CTX (Latoxan, Valence, France), dissolved in phosphate-buffered saline (PBS). This concentration of CTX induces severe muscle injury, facilitating detection of more significant alterations in the subsequent regeneration process in the absence of regeneration-related genes but still allows full regeneration as detected 3 months after the injury. Mice were sacrificed, and TA muscles were harvested at various time points following injury and processed for further experiments.

Hematoxylin/Eosin and Immunofluorescent Staining of the Regenerating Muscle
TA muscles from control mice or at the indicated days post-injury were dissected for histological assessment. The muscles were snap-frozen in liquid nitrogen-cooled isopentane and kept at −80 • C. Seven micrometer cryosections were cut at −20 • C using a 2800 Frigocut microtome (Leica, St Jouarre, France) and were kept at −20 • C until further analysis. Hematoxylin/eosin (H&E) staining was performed to assess the overall morphology and presence of necrotic fibers following injury. Images from the sections were taken using an AMG EVOS cl microscope (Thermo-Fisher Scientific, Waltham, MA, USA).
To calculate the cross-sectional areas (CSA) and collagen-stained areas, frozen muscle sections were incubated in 10 mM citric acid-sodium citrate buffer (pH 6.0) for 15 min then in blocking solution (50% FBS in PBS) for 1 h at room temperature. After blocking, samples were labeled with Dylight 488 conjugated anti-laminin B (PA5-22901, Invitrogen, Carlsbad, CA, USA) (1:100) and anti-collagen 1 antibody (SAB4500362, Sigma-Aldrich (Budapest, Hungary)) (1:100) at 4 • C overnight followed by Alexa Fluor 488 conjugated Goat anti-Rabbit IgG secondary antibody followed by washing three times with PBS. The nuclei were labeled with 4 µg/mL 4 ,6-diamidino-2-phenylindole (DAPI) (Invitrogen, Carlsbad, CA, USA), and the slides were mounted with glass coverslips. Images were taken on a FLoid Cell Imaging Station fluorescent microscope (Thermo-Fisher Scientific, Waltham, MA, USA) and analyzed using ImageJ v1.52 software (National Institutes of Health, Bethesda, MD, USA) with a muscle morphometry plugin. Areas with fibers containing centrallylocated nuclei were considered regenerating muscle parts. CSAs are reported in µm 2 , while collagen content is reported as the percentage of the total examined regenerating area.

Quantification of Necrotic Areas
Necrotic myofibers were defined as pink pale patchy fibers infiltrated by basophil single cells and quantified as described previously [11]. Briefly, 4 non-overlapping microscope view field areas were digitally captured from 6-8 H&E stained TA muscle sections at 200-fold magnification. The percentage of necrotic area/relative to the total regenerating area was calculated after manual outlining of the necrotic fibers in the sections.

Generation of Bone Marrow-Derived Macrophages (BMDMs) for NEO Cassette Expression Analysis
Bone marrow progenitors were obtained from the femur of 2 to 4-month-old RetSat +/+ and RetSat −/− mice by lavage with sterile physiological saline. Cells were differentiated for 5 days in DMEM medium supplemented with 10% conditioned medium derived from L929 cells, as a source for macrophage colony-stimulating factor (M-CSF); and 2 mM glutamine, 100 U/mL penicillin, and 100 mg/mL streptomycin at 37 • C in 5% CO 2 . Non-adherent cells were washed away every second day.   The APC non-stained cells mainly involved the Sca1 bright, CD140a + , integrin-α7− FAP cells, and the integrin-α7 + , Sca1−, CD140a− SC cells. The absolute cell count was based on the ratio of the cells of interest to the microbeads within the measured samples.

In Vitro Phagocytosis Assay by F4/80 + Cells
Phagocytosis experiments were carried out as described previously [30]. Briefly, target C2C12 cells were induced to undergo necrosis by heating the cells at 65 • C for 10 min. Some C2C12 cells were labeled with 1 µM CellTracker Deep Red Dye (ThermoFisher, Waltham, MA, USA), and some were not. Since our previous studies indicated that attenuated efferocytosis of RetSat-null macrophages is related to a decreased production of MFG-E8 during long-term efferocytosis [25], to induce MFG-E8 production, first the non-labeled cells were added to 5-carboxyfluorescein diacetate (CFDA) (6 µM)-stained F4/80 + cells isolated from the day 4 CTX-injured collagenase-treated TA muscles by magnetic beads (Miltenyi Biotec, Gladbach, Germany) at 5:1 ratio (dead cell/F4/80 + cell) plated in 8-well chamber slides (Gräfelfing, Germany) (3 × 10 5 /well). After 5 h co-culture, F4/80 + cells were washed and further incubated with labeled necrotic C2C12 cells for an additional 2 h, after which target cells were washed away extensively. In some cultures, Mφs were detached by trypsinization, and the percentage of engulfing cells was determined using a Becton Dickinson FACSCalibur flow cytometer (Becton, Dickinson and Company, Franklin Lakes, NJ, USA). Some other cultures were fixed in 1% paraformaldehyde. Representative fluorescent images were taken at a FLoid Cell Imaging Station.

Statistical Analysis
All the data presented represent the results of at least three independent experiments, and all data are presented as dots or mean or median ± SD or ± SEM. Statistical analysis was performed using two-tailed, unpaired Student's t-test and ANOVA with post-hoc Tukey HSD test. The equal variance of the sample groups was tested by an F-test. * denotes p < 0.05, ** denotes p < 0.01.

Loss of RetSat Does Not Alter the Regeneration Program in the Tibialis Anterior Muscle Following Cardiotoxin Injury
To study the possible role of RetSat in muscle development, we compared the muscle weights and the myofiber CSAs of control and CTX-treated TA muscles from RetSat +/+ and RetSat −/− mice. There was no significant difference between the body and TA muscle weights of RetSat +/+ and RetSat −/− mice ( Figure 1A,B). The mean and median CSA of the control TA muscles were also similar between the two strains ( Figure 1C), indicating that embryonic TA muscle development is normal in the RetSat-deficient mice. mean and median myofiber CSAs at days 10 and 22 post-injury. The weights and CSAs of the regenerating muscles were also similar between the two strains ( Figures 1D,E); neither did we find a difference in the number of newly formed fibers with 2 or more central nuclei or in the mean number of central nuclei per fiber, which are indicators of myoblast fusion during the muscle regeneration, between RetSat +/+ and RetSat -/-mice determined at days 10 and 22 postinjury ( Figure 1F). The fiber size distribution of the muscles before and after injury was also similar between RetSat +/+ and -/-mice ( Figure 1G). muscles with their representative immunofluorescence pictures of laminin (green) and DAPI (blue) nuclear staining. ImageJ software was used to examine 500 or more myofibers in each sample. Scale bars, 100 μm. The data are presented as a mean ± SEM (n = 6). Asterisks indicate statistical significance (* p < 0.05); ns, not significant. (G) Myofiber size distribution in the control and regenerating RetSat +/+ and RetSat −/− TA muscles with their representative immunofluorescence pictures of laminin (green) and DAPI (blue) nuclear staining. ImageJ software was used to examine 500 or more myofibers in each sample. Scale bars, 100 µm. The data are presented as a mean ± SEM (n = 6). Asterisks indicate statistical significance (* p < 0.05); ns, not significant.
To investigate the effect of RetSat ablation on skeletal muscle regeneration, we induced muscle damage by injecting CTX into the TA muscles and measured their weights and the mean and median myofiber CSAs at days 10 and 22 post-injury. The weights and CSAs of the regenerating muscles were also similar between the two strains ( Figure 1D,E); neither did we find a difference in the number of newly formed fibers with 2 or more central nuclei or in the mean number of central nuclei per fiber, which are indicators of myoblast fusion during the muscle regeneration, between RetSat +/+ and RetSat -/mice determined at days 10 and 22 post-injury ( Figure 1F). The fiber size distribution of the muscles before and after injury was also similar between RetSat +/+ and -/mice ( Figure 1G).
Histological examination revealed no obvious morphological difference between the control muscles of RetSat +/+ and RetSat −/− mice. On day 4, the regenerating muscles of both the wild-type and RetSat −/− mice displayed local necrosis and abundant leukocyte infiltration ( Figure 2). By day 10, most of the necrotic tissue was cleared from the muscles, and by day 22 post-injury, the gross histological architecture of the muscles of both RetSat +/+ and RetSat −/− mice had been largely restored, and necrotic fibers were no longer visible. Histological examination revealed no obvious morphological difference between the control muscles of RetSat +/+ and RetSat −/− mice. On day 4, the regenerating muscles of both the wild-type and RetSat −/− mice displayed local necrosis and abundant leukocyte infiltration ( Figure 2). By day 10, most of the necrotic tissue was cleared from the muscles, and by day 22 post-injury, the gross histological architecture of the muscles of both RetSat +/+ and RetSat −/− mice had been largely restored, and necrotic fibers were no longer visible.  Previously, we detected lower phagocytic capacity of the Mφs in RetSat −/− mice [25] and observed a similar finding when we used necrotic myoblasts as target cells in the in vitro phagocytosis assay performed by muscle-derived Mφs ( Figure 3A,B); therefore, we determined the sizes of the necrotic areas in control and regenerating TA muscles. As shown in Figure 3C, we detected similar necrotic area sizes between the two strains of mice, indicating that the in vivo clearance of dead fibers is not affected by the loss of RetSat.
observed a similar finding when we used necrotic myoblasts as target cells in the in vitro phagocytosis assay performed by muscle-derived Mϕs ( Figure 3A,B); therefore, we determined the sizes of the necrotic areas in control and regenerating TA muscles. As shown in Figure 3C, we detected similar necrotic area sizes between the two strains of mice, indicating that the in vivo clearance of dead fibers is not affected by the loss of RetSat.
During muscle repair, the deposition of extracellular matrix proteins is transiently increased, which is required for the regulation of SC, and for myoblast proliferation and differentiation [31]. Therefore, we decided to determine the amount of collagen 1 in the control and regenerating TA muscles. In both mouse strains, there was a temporal increase in collagen 1 deposition at day 10 post-injury, which decreased by day 22, as compared to their own nonregenerating muscles; however, there was no significant difference between the two strains ( Figure 3D,E). To assess the possible impact of RetSat ablation on gene expression and SC cell proliferation and differentiation in the control and regenerating TA muscles, we determined the number of SC cells ( Figure 4A) and the expression of myogenic genes, such as the Pax7, MyoD, and myogenin transcription factors involved in myoblast proliferation and During muscle repair, the deposition of extracellular matrix proteins is transiently increased, which is required for the regulation of SC, and for myoblast proliferation and differentiation [31]. Therefore, we decided to determine the amount of collagen 1 in the control and regenerating TA muscles. In both mouse strains, there was a temporal increase in collagen 1 deposition at day 10 post-injury, which decreased by day 22, as compared to their own non-regenerating muscles; however, there was no significant difference between the two strains ( Figure 3D,E).
To assess the possible impact of RetSat ablation on gene expression and SC cell proliferation and differentiation in the control and regenerating TA muscles, we determined the number of SC cells ( Figure 4A) and the expression of myogenic genes, such as the Pax7, MyoD, and myogenin transcription factors involved in myoblast proliferation and differentiation and that of the myosin heavy chain (MYHC)1 differentiation marker. During muscle regeneration, the mRNA expression of Pax7, MyoD, and myogenin transiently increased, Cells 2022, 11, 1333 9 of 15 while that of RetSat and MYHC1 transiently decreased in the TA muscles; however, with the exception of RetSat, there was no significant difference in their expression between the two mouse strains ( Figure 4B). Cells 2022, 11, x FOR PEER REVIEW 9 of 16 differentiation and that of the myosin heavy chain (MYHC)1 differentiation marker. During muscle regeneration, the mRNA expression of Pax7, MyoD, and myogenin transiently increased, while that of RetSat and MYHC1 transiently decreased in the TA muscles; however, with the exception of RetSat, there was no significant difference in their expression between the two mouse strains ( Figure 4B). Altogether, these data imply that loss of RetSat does not affect the number of SCs in the skeletal muscle, nor does the loss of RetSat impact skeletal muscle development or regeneration.

Decreased Recruitment of Mϕs and Neutrophils after Injury in the Absence of RetSat
Following injury, muscle repair is initiated by the migration of inflammatory cells to the injury site. To determine the composition of leukocytes in the early phase of muscle regeneration, we performed a flow cytometric analysis of magnetically separated CD45 + cells from collagenase digested muscles. In accordance with previous observations, we observed early infiltration of neutrophils at day 2 post-injury, followed by an increasing number of Mϕs at days 3 and 4 in wild-type mice. However, in the absence of RetSat, a significantly decreased number of CD45 + cells infiltrated the injured muscle ( Figure 5A). In line with this observation, at day 2 post-injury, a significantly decreased gene expression level of monocyte chemoattractant protein-1 (MCP-1) was detected ( Figure 5B), whereas the neutrophil/Mϕ ratios did not change ( Figure 5C). Altogether, these data imply that loss of RetSat does not affect the number of SCs in the skeletal muscle, nor does the loss of RetSat impact skeletal muscle development or regeneration.

Decreased Recruitment of Mφs and Neutrophils after Injury in the Absence of RetSat
Following injury, muscle repair is initiated by the migration of inflammatory cells to the injury site. To determine the composition of leukocytes in the early phase of muscle regeneration, we performed a flow cytometric analysis of magnetically separated CD45 + cells from collagenase digested muscles. In accordance with previous observations, we observed early infiltration of neutrophils at day 2 post-injury, followed by an increasing number of Mφs at days 3 and 4 in wild-type mice. However, in the absence of RetSat, a significantly decreased number of CD45 + cells infiltrated the injured muscle ( Figure 5A). In line with this observation, at day 2 post-injury, a significantly decreased gene expression level of monocyte chemoattractant protein-1 (MCP-1) was detected ( Figure 5B), whereas the neutrophil/Mφ ratios did not change ( Figure 5C).

Myoblasts Compensate for Attenuated MFG-E8 Levels of Macrophages in RetSat-Null Mice
To investigate the impact of RetSat ablation on the polarization of Mφs and on their gene expressions, CD45 + cells from collagenase-digested regenerating muscles were magnetically separated at days 2, 3, and 4 post-injury and stained for the cell surface marker proteins F4/80, Ly6C, CD206, and MHCII. In addition, the expression of their genes was determined by quantitative PCR. Since to our surprise, based on the relative disappearance of necrotic areas ( Figure 3C), loss of RetSat in vivo did not seem to affect efferocytosis during skeletal muscle regeneration, we first checked whether muscle-derived CD45 + cells from the RetSat-null mice altered MFG-E8 levels. As seen in Figure 5D, expression of MFG-E8 mRNA within the CD45 + cells gradually increased until day 4 following cardiotoxininduced injury in both mouse strains. In accordance with our previous observations [25], the muscle-derived CD45 + cells in the RetSat-null mice also expressed significantly less MFG-E8. This finding is in accordance with the results of the in vitro phagocytosis assay, which demonstrated a decreased long-term efferocytosis capacity for the muscle-derived RetSat null macrophages ( Figure 3A,B).  However, in the regenerating muscle, we found no alterations in the levels of MFG-E8 mRNA in the RetSat-null mice as compared to their wild-type littermates ( Figure 5D), indicating that very likely myoblasts (the only cell type besides Mφs that is known to produce MFG-E8 in regenerating skeletal muscle) fully compensate for the attenuated MFG-E8 production of Mφs in the RetSat-null mice. In fact, using the data derived from the NPY mRNA expression in the CD45 + cells and in the total muscle ( Figure 5E), we could estimate that in the wild-type regenerating muscles, less than 1% of the MFG-E8 is derived from CD45 + cells. Since MFG-E8 is secreted into the tissue environment, myoblast-derived MFG-E8 is expected to become available for phagocytosing cells as well. Thus, our data indicate that in the regenerating muscles, independent of RetSat expression, macrophage-derived MFG-E8 has no limiting effect on the phagocytic capacity of macrophages. This observation partly explains why the in vivo efferocytosis did not change in the regenerating muscles of RetSat-null mice.

Altered NPY Levels Both in Mφs and in the Skeletal Muscle of RetSat-Null Mice
Next, we assessed the levels of expression of NPY in muscle-derived CD45 + cells. NPY has anti-inflammatory functions [26,27], and was also shown to promote angiogenesis [32]. NPY mRNA levels in CD45 + cells from the muscles of wild-type mice increased until day 3 following CTX-induced injury, and then they started to decrease. However, in accordance with our previous findings [25], muscle-derived CD45 + cells from RetSat-null mice lacked significant expression of NPY ( Figure 5E).
Skeletal muscle does not express NPY mRNA or protein [33], but sympathetic neurons do, and they co-release the NPY with noradrenaline following stimulation [34]. Accordingly, we detected expression of NPY mRNA in control muscles, and in muscles from day 10 following CTX-induced injury in both wild-type and RetSat-null mice ( Figure 5D). In this context, it is worth noting that sympathetic neurons regulate immune functions via NPY [35] and were reported to facilitate muscle repair [36]; we detected a significantly enhanced amount of NPY mRNA in the muscles of control and day-22 post-injury RetSatnull mice. The finding was similar when we checked the NPY levels in other organs, except for the brain, where NPY mRNA expression was similar. These data indicate that the loss of RetSat affects the mRNA expression of NPY not only in Mφs but also in the sympathetic neurons innervating various tissues.
In addition, in the muscles of wild-type mice, expression of NPY mRNA increased until day 3 following injury, then gradually decreased, whereas no expression of NPY mRNA was detected during this time period in the muscles of RetSat-null mice. Since the CD45 + cells of RetSat-null mice do not express NPY, this transient increase in the wild-type muscle could be a result of infiltrating NPY-expressing CD45 + cells.

A Transient Delay in the M1/M2 Phenotypic Switch in Mφs of RetSat-Null Mice during the Muscle Regeneration Process
Since proper timing of the M1/M2 phenotypic change of Mφs is central in guiding muscle regeneration, we followed the process by determining the time-dependent changes in the expression of M1-and M2-specific cell surface markers of Mφs, and in the expression of genes of the CD45 + cells. Though the loss of RetSat did not affect the in vivo efferocytotic capacity of Mφs, known to result in an altered polarization of Mφs, at post-injury day 3, we detected a delayed generation of Ly6C − CD206 + Mφs from the Ly6C + RetSat −/− NPY null pro-inflammatory cells ( Figure 5F). However, this delay disappeared by day 4, perhaps due to the high myoblast-derived MFG-E8 level that also promotes the M1/M2 conversion, and the appearance of CD206 + macrophages [37]. At the same time, the formation of the MHCII high expressing cells was not affected.
3.6. mRNA Expression of RetSat, Various Cytokines, and Growth Factors in the CD45 + Macrophages Derived from the Regenerating TA Muscles of RetSat +/+ and RetSat −/− Mice In accordance with our previous publication [38], the expression of RetSat mRNA increased in the engulfing CD45 + cells ( Figure 6A). In line with the lack of immunosuppressive NPY production, at day 2 post-injury as compared to their wild-type counterparts, CD45 + cells from RetSat-null mice expressed an increased amount of pro-inflammatory IL-1β, a cytokine that strongly contributes to SC cell proliferation and differentiation [39]. However, we did not find alterations in the expression of other pro-inflammatory cytokines, such as TNF-α or IL-6, nor did we find a change in the appearance of the mRNA expression of the M2-like IL-10, PPARγ, or GDF3 genes. However, the mRNA expression of three enzymes known to determine NO release, arginase 1 (Arg1), nitric oxide synthase 3 (NOS3), and the inducible iNOS, did change in such a way that they promoted long-term NO production. Proper GDF3 production might be associated with normal myoblast differentiation in RetSat-null mice. Prolonged NO production, together with elevated IL-1µ levels, on the other hand, might promote SC cell activation in the presence of fewer Mφs [40], might contribute to the initiation of angiogenesis in the absence of NPY [41], and might enhance efferocytosis by promoting the appearance of PS on the surface of apoptotic cells [42]. Previous studies have demonstrated that, in addition to Mϕs, eosinophils are also essential for proper muscle repair by producing IL-4 during muscle regeneration [43]. IL-4 contributes to the proper M2-like polarization of Mϕs [44]; but most importantly, it triggers the proliferation of FAP cells [43]. FAP cells work side by side with Mϕs to properly clear the necrotic cells and contribute to myogenesis [43,45]. In addition to IL-4, macrophage-derived TGF-β1 is also involved in the generation of FAP cells, and the number of FAPs correlates with the TGF-β1 levels [46]. In this context, it is worth noting that we observed significantly elevated TGF-β1 and IL-4 mRNA expression levels in the RetSat- Previous studies have demonstrated that, in addition to Mφs, eosinophils are also essential for proper muscle repair by producing IL-4 during muscle regeneration [43]. IL-4 contributes to the proper M2-like polarization of Mφs [44]; but most importantly, it triggers the proliferation of FAP cells [43]. FAP cells work side by side with Mφs to properly clear the necrotic cells and contribute to myogenesis [43,45]. In addition to IL-4, macrophagederived TGF-β1 is also involved in the generation of FAP cells, and the number of FAPs correlates with the TGF-β1 levels [46]. In this context, it is worth noting that we observed significantly elevated TGF-β1 and IL-4 mRNA expression levels in the RetSat-null CD45 + cells at post-injury days 2 and 4, respectively, and as a result, we detected unaltered FAP cell numbers in the RetSat null muscle during regeneration ( Figure 6B).

Conclusions
Previous investigations have shown that strong intercellular crosstalk between various populations of cells in the regenerating muscle drives and balances the process of skeletal muscle repair [47,48]. Our data presented in this paper demonstrate that through this crosstalk, several compensatory mechanisms are induced in the regenerating muscles of RetSat-null mice to replace the impaired functions of the Mφs, which is attributed to their attenuated MFG-E8 production, and to their lack of NPY expression. Thus, high levels of MFG-E8 produced by the regenerating muscles generally obviate the need for MFG-E8 release by macrophages for proper efferocytosis, independently of the loss of RetSat. Higher levels of IL-1β and NO produced by CD45 + cells replace the need for NPY (which might appear at the site of repair from the sympathetic neurons as well) and promote sufficient SC cell proliferation, despite the presence of fewer neutrophils and macrophages at the regeneration sites of the RetSa-null muscles. Increased production of TGF-β and IL-4 by CD45 + leukocytes, on the other hand, maintains the proper FAP cell proliferation, contributing to the proper in vivo dead cell clearance. Our data cannot distinguish whether these alterations are the result of RetSat ablation or if they would also be induced under other circumstances when fewer neutrophils and macrophages infiltrate the damaged skeletal muscle area. Nevertheless, as a result of the adapted crosstalk, normal skeletal muscle repair was observed following CTX-induced injury in the RetSat-null mice.

Data Availability Statement:
The data presented in this study are available upon request from the corresponding author.