The gilthead sea bream (Sparus aurata, Linnaeus 1758) is one of the most important species in the Mediterranean aquaculture of the last decades as it represents a source of energy and proteins of high quality and thus, it has a high commercial value (1). However, there are still areas of their physiology that are not well described; therefore, more research is needed to optimize their growth and development in terms of farming.

In contrast to other vertebrate groups, some fish species exhibit indeterminate growth, which means that they can grow in length and weight beyond the maturation stage for as long as they live (2). Most of this growth is due to an increase in somatic tissue, specifically skeletal muscle, and implies new muscle cells recruitment to increase the number of muscle fibers (hyperplasia), as well as the increase in size of pre-existing fibers (hypertrophy). These phenomena are known to be an important determinant of the characteristics of flesh texture and, from an aquaculture perspective, muscle quality is of main interest (3, 4).

The growth hormone (Gh) and insulin-like growth factors (Igfs) axis is the most important endocrine system controlling many physiological processes of the body (5). The Gh is the central regulatory molecule of somatic growth in vertebrates, including teleost fish, and although its main target is the liver it can exert significant direct effects on other tissues such as skeletal muscle. However, it is well known that most of the effects of Gh on growth are indirect, through the Igfs (6). The large tissue distribution of Igfs-producing cells including their receptors, together with the extensive distribution of Gh receptors (Ghrs), make this mediation possible. Many studies, including several in fish, have demonstrated systemic effects as well as the existence of local paracrine/autocrine actions of Gh and Igfs (7, 8). In muscle, Igfs directly stimulate a variety of signals including proliferation and differentiation of satellite cells (SCs), hypertrophy of myoblasts, and inhibition of cell atrophy. In gilthead sea bream, it has been shown that Igf-2 alone appears to be an important stimulator of proliferation, while Igf-1 seems to have a synergetic activity with Gh, playing different roles in the control of myogenesis (9). Moreover, in the same species, three splice variants have been described (igf-1a, igf-1b and igf-1c) with differential responses depending on the physiological or stimulatory condition (10). Igfs exert these actions through the binding and activation of their cellular receptors and in turn multiple intracellular signal transduction cascades, including the phosphatidylinositol 3-kinase (Pi3k)-Akt-Tor cascade, which controls an important gene expression program (11).

The process of muscle generation, known as myogenesis, consists in that SCs, the muscle stem cells that can be recognized for the expression of Pax3/7 (12), proliferate and differentiate into myoblasts that fuse together (i.e., hyperplasia) or with pre-existing fibers (i.e., hypertrophy) to form multinucleated myofibers. Myogenesis is regulated by the myogenic regulatory factors (Mrfs), a set of four transcription factors that are expressed sequentially being Myf5 and MyoD the first to be activated controlling SCs activation and proliferation, and later Myogenin and Mrf4 that coordinate myoblasts differentiation and fusion (13). In adulthood, those processes can also occur in response to challenging conditions that threaten muscle homeostasis. Thanks to the ability of muscle cells to modify both, the amount and types of proteins being synthesized within them, defined as muscle plasticity, it is possible to cope with any stimulus that disrupts the normal condition of the organism (14). In this framework, muscle regulation implies a protein turnover to ensure the correct function of the myofibers, i.e., a balance between protein synthesis and degradation. The main catabolic systems that control muscle proteolysis have been also described in gilthead sea bream and are the ubiquitin-proteasome complex (UbP); calpains, a family of Ca²⁺-dependent proteases formed by catalytic and regulatory small subunits; cathepsins, some of them related to the autophagy-lysosome system; and caspases, the apoptosis protease system (15, 16).

One of the phenomena that can alter normal muscle functions and body homeostasis are plasma membrane disruptions induced by muscle contraction as well as severe damage caused by traumatic injuries or muscle diseases, leading all these processes to fiber rupture. Nevertheless, skeletal muscle presents an intrinsic mechanism capable of regenerating the entire damaged contractile system. The fundamental role in muscle regeneration is played by the SCs resident in the muscle, which are activated in response to various previous stimuli that may come from the tissue itself or from other systems of the organism (17). When necrosis of the affected myofibers occurs because of the trauma, first an inflammatory response is initiated with the release of different cytokines, as the interleukins, from neutrophils, among other immune cells (18, 19). These induce macrophages infiltration and activate the proteolytic systems to remove all the dead material, followed by an activation of SCs to generate new fibers or to fuse with undamaged fibers. Finally, newly formed fibers maturate and remodel the regenerated muscle to recover its normal functions (20).

In addition, there are other factors that can also have critical roles during myogenesis. Proliferating SCs are characterized by the presence of proliferating cell nuclear antigen (Pcna), which is classically known to stimulate the entry of SCs to the cell cycle (21), and of hepatocyte growth factor receptor (cMet) that turns on also cell proliferation in skeletal muscle (22, 23). On the other hand, two novel muscle-specific fusogens, Myomaker and Myomixer, have been described and could play a fundamental role in the stage in which myoblast fusion takes place. In fact, the lack of some of these two muscle-specific membrane proteins results in an incomplete myogenesis since cells cannot fuse and therefore multinucleated myofibers are not formed (24). Other factors with an important role are caveolae, which are invaginations of the plasma membrane that function as messenger centers in tissues, including skeletal muscle, controlling signal transduction. The transmembrane proteins that enable the formation of these structures are members of the caveolin family (25). Caveolin-1 (Cav-1) was the first member described and was identified as being expressed in quiescent SCs but not in mature myofibers and is downregulated when cells begin to proliferate (26). The other two members of the family are Caveolin-2 (Cav-2) and Caveolin-3 (Cav-3). Characterization of the former revealed that colocalizes with Cav-1 and is co-expressed in the same circumstances. Otherwise, Cav-3, unlike Cav-1, is involved in myoblast differentiation and myotube fusion, so that a deficiency of Cav-3 results in cellular immaturity (27). Finally, another family of proteins essential for the maintenance of skeletal muscle homeostasis in adults are the Wnts, although information about the role of these molecules in fish is very limited. Canonical Wnt signaling regulates the differentiation of myoblasts, while the non-canonical pathway controls the self-renewal of SCs and the growth of muscle fibers (28, 29).

In some teleost fishes, like the gilthead sea bream, and as opposed to mammals, the skeletal system is defined as acellular since the skeleton is generated and maintained by chondrocytes, osteoblasts, and osteoclasts but it lacks osteocytes within the calcified extracellular matrix (ECM) (30, 31). Bone lineage determination and differentiation of osteoblasts from mesenchymal stem cells (MSCs) implies various key regulatory actors including specific transcription factors, mineral availability, and environmental conditions (32). The transcription factor expression pattern that controls osteogenesis was elucidated in gilthead sea bream few years ago and it was demonstrated that there is conservation of transcripts compared to mammals (33). Indeed, Runx2 is crucial for the commitment of the osteoprogenitor cells from MSCs and for the differentiation of osteoblasts; besides it coordinates the expression of other genes, such as osteopontin (op), osteonectin (on) and osteocalcin (ocn), among others, which are key factors regulating the ECM mineralization (34). In addition, it is essential to point out that bone is among the only tissues that has a cell type whose main action is to destroy the own tissue, the osteoclast (35). This function is known as bone remodeling and it is crucial to guarantee mineral homeostasis, having regulatory implications also in other tissues such as the skeletal muscle (36).

Bones represent the attachment location for skeletal muscles and therefore this tissue is also important to cope with threatening conditions. Hence, the musculoskeletal system is a complex interconnected structure (37). Together, the musculoskeletal system enables locomotion and has an essential metabolic role, and the interaction of muscle and bone tissues goes beyond the mechanical and involves a great deal of biochemical communication (38, 39). Both tissues are a relevant source of signaling molecules that can act in an autocrine, paracrine, or endocrine way. It is for these reasons that the functions and processes of each individual tissue are regulated by the other one (40, 41).

In this scenario, the main objective of this work is to perform an in vivo study to characterize skeletal muscle regeneration in gilthead sea bream. For this purpose, a mechanical injury will be produced in the muscle and changes in the transcription of key genes involved in the growth regulation of skeletal muscle, including the different systems and molecules mentioned above will be evaluated. In addition, several bone gene markers will be studied to determine the implication of this close-related tissue regulating myogenesis.

Juvenile gilthead sea breams were injured in the epaxial white muscle with a 14G needle. Then, muscle histological analysis plus assessment of changes in the transcription of key genes from muscle and bone tissues during a 30-day regeneration period was performed. The histological evaluation revealed at day 1 fibers in the process of necrosis, while 16 days after the injury no traces of damaged or necrotic fibers were observed in the gilthead sea bream. Our results agree with those of Rowlerson et al. (44), who found in the same species that muscle regeneration was weakly detected seven days post-injury, still with some necrotizing fibers, but at 21 days post-lesion many areas were regenerated, and the presence of new young fibers could be observed. In fact, in our study, at day 30 the myotomes showed a complete normal morphology with many new and small muscle fibers present.

The Gh-Igfs axis is the main regulator of muscle growth and contributes to the process of regeneration after tissue injury in vertebrates (20, 45) including fish (46). In this sense, the contribution of Igf-1 during tissue repair has been recognized (47) but the role of Igf-2 is not so well known although its contribution to myogenesis was suggested by demonstrating binding to the Igf-1 receptor, activating the Akt pathway and then MyoD (48, 49). Furthermore, Ge et al. (50) found in mice that igf-2 mRNA levels also increased drastically during early muscle regeneration. In this study, the peak of igf-2 at the beginning of muscle repair (i.e., day 1) is consistent with this early and pre-Igf-1 function at the onset of myogenesis. In fact, it has been previously observed in gilthead sea bream myocytes that Igf-2 contributes to muscle regulation mainly in the early stages of development and before Igf-1 (9, 51). Indeed, Igf-1 action takes place slightly later towards the initial stages of myoblast differentiation (52). In agreement with that, in the present study, total igf-1 expression decreased at day 1, recovering the expression levels at days 4, 8, 16 and 30 after injury, probably mainly due to the upregulation of the splice variant igf-1b at day 2 post-injury. Concerning receptors, ghr-1 expression followed a similar profile to that of igf-1 with a rapid decrease the first days after injury and a progressive and slow increase afterwards, returning to initial levels at day 16 when the animals recovered from the injury, a response that agrees with its anabolic role in this species (8, 53).

The rebuilding of skeletal muscle after an injury requires a fine regulation of proteinases expression. In that sense, calpains, cathepsins and the UbP are the major systems for muscle protein degradation, and they may contribute to the remodeling of skeletal muscle during regeneration (54). Previously, we have found important transcriptional responses of these proteolytic genes in gilthead sea bream subjected to sustained exercise, different diets, or the combination of both factors (15, 55, 56) or in response to 21 days of fasting (57). In this study, we found at least one gene expressed from each one of the three proteolytic systems at different times during regeneration. In fact, the expression of capn2 and ctsl was increased at day 1 and 2 respectively, while mafbx increased at day 8 post-injury and murf1 at day 30 compared to the early days. In a study done in rats that were injected with 0.75% bupivacaine, a drug that at this concentration produces a chemical injury, the enzymatic activity of Capn1 and Capn2 was maximum at day 5 post-injury, that of Cathepsin L and D at day 3, and the proteasome peaked at day 5 to progressively return to control levels by day 21 (54). This profile of proteolytic gene expression can be comparable to the expression profiles obtained in this study in gilthead sea bream and suggests that the sequence of activation of proteolytic genes during muscle regeneration appears to be quite well conserved through vertebrates. Specifically, calpains seemed to have a more relevant role in proliferation of gilthead sea bream cultured myocytes and the UbP system in cells differentiation (58). In addition, it has been demonstrated that mafbx interacts with myod regulating its expression; therefore, in this study mafbx might be inhibiting myod expression prematurely at day 8 in order to ensure the correct activity of Mrfs from day 16 onwards (59). In summary, some of these proteolytic genes could be necessary in our study species, to modulate myogenesis, as observed previously in mammals (60). Thus, these genes could be involved in myoblast proliferation and differentiation (61, 62) but are also required for myoblast fusion (63, 64).

In line with this, as a part of the characterization of the recently identified fusogens Myomaker and Myomixer in gilthead sea bream, aiming to estimate their functions, gene expression of these molecules in addition to the Mrfs were studied under different myogenesis conditions including this model of muscle regeneration (65). All these molecules showed low expression values until day 8, and then started to increase to reach a peak at day 16 post-wound to subsequently slowly decrease to recover initial levels by day 30. Recently, Manneken et al. (66) reviewed the process of muscle regeneration in zebrafish (Danio rerio), focusing on the differences to amniotes myogenesis during both, development, and repair. It is noticeable the similarity of the transcriptional profile of the different Mrfs analyzed in that study with those in gilthead sea bream. Moreover, Rowlerson et al. (44) showed that at 7 days post-lesion a weak reaction to desmin antibody appeared only in bigger fibers but not in presumed newly formed fibers; and at 21 days the regeneration was well advanced in many muscle areas. Forcina et al. (67) and Musarò (14) also reviewed the mechanisms regulating muscle regeneration in mammals, indicating that the MRFs participate in the process between 4 and 7 days post-injury, reaching the maturation and functional recovery of the muscle between days 10 and 15. The profile is again very similar to that observed in our previous study in gilthead sea bream (65) although the faster myogenic regeneration process in mammals could be explained by their higher body temperature compared to fish. In fact, in rainbow trout (Oncorhynchus mykiss), a species that grows at a lower water temperature than gilthead sea bream, the complete muscle regeneration in a similar experiment took longer (i.e., day 30) and myomaker and myomixer indeed peaked later, between days 16 and 30 after injury (68).

Other regulatory factors of muscle growth like pax7, cmet, caveolins, wnt5b, ogn, vegfa or myostatins followed the same tendency, with non-significant small fluctuations up to day 8 but significant increases in expression at 16 and/or 30 days post-lesion. Indeed, the parallelism between the expression profiles of these genes and the Mrfs makes sense due to its recognized role during mammalian myogenesis. For instance, very little is known about the role of caveolins in fish, but the increase observed in cav1 and cav3 suggests their participation in fish myogenesis activating SCs as described in mice and humans (26, 69); following pax7 and cmet, factors that are known to stimulate cell cycle activation and proliferation of SCs (21, 22). Moreover, the progressive increase in the expression of wnt5b was indicative of its potential role at the end of muscle regeneration during myoblast differentiation and fusion although a larger contribution from this or other members of the Wnt family could be expected in the early stages (28). However, again, the information in fish is very scarce and probably their role is not exactly the same as in mammals (29). The expression of ogn1 and ogn2 in the muscle increased from day 8. In a previous study, Costa et al. (70) determined the expression of ogn in primary cultures of gilthead sea bream myocytes and observed a significant increase in their expression at day 8 of culture development, at the stage of differentiation and fusion of myocytes to form myotubes. In the present study, it is possible to consider that ogn transcripts in muscle are also participating in the fusion of the newly formed fibers. Regarding the expression of vegfa, the peak observed at 16 days post-injury is consistent with the main period of new muscle fibers formation plus maturation, and therefore, increased angiogenesis and myogenic factors highest expression; supported by the role previously demonstrated for this factor in muscle remodeling in gilthead sea bream (56). Finally, in the case of mstn1 and 2, their gene expression was increased by day 30, which fits with their known role in gilthead sea bream as negative regulators of myogenesis, therefore helping to the completion of the process in this regeneration scenario (71). Thus, the period between days 8 and 16 after the injury seems to be in this species the main period of activation of myogenic factors to induce muscle regeneration. This is a little later that the timing described in mammals but faster than that observed in rainbow trout (67, 68, 72, 73).

In the healing process of a tissue, acute inflammation and immune cells play a critical role in almost all phases of regeneration. Damaged muscle fibers undergoing necrosis induce the infiltration of many types of immune cells (74, 75). Therefore, the infiltrating cells help clearing dead cells from the injured area as well as secrete different types of cytokines to recruit more immune cells in addition to producing cellular responses to regulate muscle cell activation, proliferation, and differentiation (19). In our study, it has been possible to observe a first wave of il-6 and il-15 increased expression in the early days after injury. In a mouse model of cardiotoxin-induced muscle injury, Zhang et al. (76) showed how high levels of il-6 were detected from the first day after injury (24 h) and demonstrated that this cytokine is crucial for macrophage infiltration, and that the lack of it in a knockout mouse model inhibited myoblast proliferation. Moreover, some studies support the fact that Il-6 also induces proteolysis of muscle cells by activating and modulating proteolytic systems (77, 78). On the other hand, il-15 is highly expressed in mammalian muscle, and it is believed that, in addition to its inflammatory function in a pro-inflammatory context, it may be a myokine as it promotes myogenesis in response to exercise as well as facilitates muscle regeneration (79, 80). The results of our study could be showing such mentioned functions, and in the regeneration in gilthead sea bream both Il-6 and Il-15 could be acting in a pro-inflammatory phase in addition to favoring myogenesis, although we cannot distinguish if these cytokines are specifically expressed in muscle cells, immune cells, or both. Furthermore, Tnfα has been reported to play an important role in muscle regeneration in mammals (81, 82). This cytokine can attract muscle SCs to the damaged site to promote their proliferation and differentiation (83). The increased expression of tnfα in gilthead sea bream muscle 16 days after injury could be indicative of this function since its upregulation occurred at the same time as the entire myogenic program was initiated. In fact, the releasing of a whole continuum of pro-inflammatory cytokines into the injured muscle stimulates the infiltration of monocytes that go on to exhibit an inflammatory profile to carry out phagocytosis and then, switch to anti-inflammatory macrophages to stimulate myogenesis (84–86). The colony stimulating factor 1 (Csf1) controls the production, differentiation, and function of macrophages and its receptor (Csf1R) mediates the biological effects (87). In our study, the strong increase detected in the expression of csf1r from day 4 on could indicate the arrival to the muscle of macrophages after the wave of pro-inflammatory interleukins. Moreover, the expression of this receptor remains elevated up to day 30, which could mean the maintenance of a population of anti-inflammatory and myogenesis-supporting macrophages until the end of the regenerative process.

Referring to genes expressed in bone, the small leucine-rich proteoglycans ogn1 and ogn2 have been characterized in gilthead sea bream during the in vitro development of different cell types including bone and muscle (70). Hence, their significantly increased expression in the bone tissue close to the muscle wound at days 4 and 8 post-injury, respectively, in the current study, agrees with their reported upregulation at the onset of osteoblasts as well as myocytes differentiation. Moreover, Tanaka et al. (88) also demonstrated the regulatory crosstalk role between bone and muscle of ogn, but references in fish are very poor. In this study, the earlier expression of ogn in bone and later in muscle could indicate a stimulus from bone to enhance the recovery of the injured muscle. Similarly, ocn expression modulation during regeneration with increased levels at day 4 versus day 8, support its role as a promising osteokine participating in the regulation of the muscle response after an injury in fish, as already demonstrated in mammals (89). Indeed, Ocn has been proven to be able to increase importantly muscle activity in elderly mice individuals (90) and to directly promote protein synthesis in myotubes, explaining why this hormone is responsible for muscle maintenance during aging (91). Thus, the increase of ocn at day 4 post-injury before the upregulation of Mrfs would support its stimulatory role in fish muscle too. The bmp2 and ctsk genes showed an early activation of expression at days 1 and 2 post-injury, respectively, although bmp2 expression decreased at day 8 recovering basal levels at the end of the experiment. This early upregulation of bmp2 could be related to the highest expression found for this molecule at the beginning of embryonic development in gilthead sea bream (92), but also to situations in which bone (and muscle) growth is stimulated, such as for example in Atlantic salmon (Salmo salar) reared at elevated water temperatures (93). Moreover, the peak of expression of ctsk at day 2 resembled the increase in expression observed after 20 days of fasting (57) suggesting that ctsk and therefore, bone resorption, may have a role in muscle remodeling after stressful conditions. The key osteogenic factor runx2 after a fall at day 4, peaked at day 8, keeping similar levels until the end of the experiment, perhaps suggesting an induction of osteogenesis at this time to support the subsequent muscle mass development. Furthermore, the inter-tissue inhibitory role of Myostatin has been already well characterized in mammals (89, 94). In our model species, a previous study of fasting and refeeding proved how myostatin could act coordinating musculoskeletal growth and suggested its potential role as osteokine in fish (57). In this regeneration study, a regulatory role for myostatin could also be hypothesized, since bone decreased mstn2 expression by day 4 in response to the muscle injury possibly to allow proper muscle remodeling, thus giving more insights of the importance of myostatin between the bone and muscle crosstalk in fish. Finally, the ECM marker on showed a progressive increase during regeneration with the maximum expression reached at day 30 coinciding with the highest expression observed for on at the end of in vitro and in vivo osteogenesis in gilthead sea bream according to its key role regulating the later stages of tissue development (33, 92). In summary, these results open a promising new line of research to study the possible crosstalk between bone and muscle in fish to regulate skeletal muscle development and growth.

In gilthead sea bream, the generation of a wound and the presence of damaged and dead muscle fibers firstly activated pro-inflammatory genes as il-6 and il-15 that contributed cleaning the injured tissue and stimulating the muscle regeneration process ( Figure 7 ). Then, proteolytic factors, first capn2 and ctsl, and later on the UbP system member mafbx allowed an effective degradation of the damaged tissue.

At the very early stage of 1-day post-injury, it was noticeable the increase of igf-2 and the decreases of ghr1 and total igf-1 expression. The return to basal levels at day 8 of igf-2 was parallel to the increase of igf-1 and simultaneous to the induction of many other genes’ expression involved in the regulation of myogenesis like pax7, cmet, cav1, cav3, and wnt5b, indicating the beginning of muscle remodeling. In fact, vegfa also showed a clear peak of expression at 16 days post-lesion, which parallels the Mrfs and the fusogens myomaker and myomixer patterns of expression previously described (65), thus suggesting that progression of myogenesis and angiogenesis is taking place resulting in new fibers formation.

Furthermore, it could be hypothesized that some bone-derived genes such as bmp2, ogn1, ocn and mstn2 could be participating by sending information to the muscle mainly at the early stages of the regenerative process to control harmonic musculoskeletal growth, thus potentially presenting a role as osteokines in fish.

Overall, this study shows how different mechanisms are orchestrated to recover an injured muscle and allows knowing for the first time in gilthead sea bream the timely contribution of different molecules along the regeneration process of damaged skeletal muscle fibers.

The original contributions presented in the study are included in the article/ Supplementary Materials . Further inquiries can be directed to the corresponding author.

The study was conducted according to the guidelines of the European Union Council directive (EU 2010/63) and approved by the Ethics Committee of the University of Barcelona (protocol code CEEA 37/20).

JG conceptualized the study. AO-T, MP-A, VJ-P, IG-P, AS-M and JG performed the sampling. AO-T, MP-A, VJ-P, FM and IG-P performed the laboratory analyses. AO-T, MP-A, VJ-P and JG analyzed and interpreted the data. IN, EC, JB and JG acquired funding. AO-T, MP-A, IN, EC, JF-B, DG, JB and JG drafted and critically reviewed the manuscript. All authors read and approved the final paper. The authors have declared no conflict of interests. All authors contributed to the article and approved the submitted version.