Morphological and Molecular Responses of Lateolabrax maculatus Skeletal Muscle Cells to Different Temperatures

Temperature strongly modulates muscle development and growth in ectothermic teleosts; however, the underlying mechanisms remain largely unknown. In this study, primary cultures of skeletal muscle cells of Lateolabrax maculatus were conducted and reared at different temperatures (21, 25, and 28 °C) in both the proliferation and differentiation stages. CCK-8, EdU, wound scratch and nuclear fusion index assays revealed that the proliferation, myogenic differentiation, and migration processes of skeletal muscle cells were significantly accelerated as the temperature raises. Based on the GO, GSEA, and WGCNA, higher temperature (28 °C) induced genes involved in HSF1 activation, DNA replication, and ECM organization processes at the proliferation stage, as well as HSF1 activation, calcium activity regulation, myogenic differentiation, and myoblast fusion, and sarcomere assembly processes at the differentiation stage. In contrast, lower temperature (21 °C) increased the expression levels of genes associated with DNA damage, DNA repair and apoptosis processes at the proliferation stage, and cytokine signaling and neutrophil degranulation processes at the differentiation stage. Additionally, we screened several hub genes regulating myogenesis processes. Our results could facilitate the understanding of the regulatory mechanism of temperature on fish skeletal muscle growth and further contribute to utilizing rational management strategies and promoting organism growth and development.


Introduction
Fish skeletal muscle is an important source of high-quality protein for human food. The growth performance of skeletal muscle determines food production and economic benefit in the aquaculture industry [1]. In contrast to mammals, for which hypertrophy of existing muscle fibers is the primary mechanism of skeletal muscle growth during the postnatal stages, fish skeletal muscle mass grows by increasing both the number and size of muscle fibers, exhibiting an indeterminate growth pattern that combines hyperplasia and hypertrophy throughout its lifespan [2,3]. The generation of muscle and myogenesis involves a series of steps, including the activation, proliferation, and differentiation of muscle stem cells (MSCs, also known as satellite cells), the fusion of precursors to form syncytial myotubes, and the maturation of myotubes to generate contractile myofibers [4][5][6]. The process of muscle formation is highly regulated by many myogenic regulatory factors (MRFs) and signaling molecules [6][7][8][9].
The indeterminate growth observed in most fishes is influenced by environmental factors such as temperature [10], nutrition [11], salinity [12] and oxygen concentrations [13]. Among them, temperature strongly modulates muscle development in ectothermic teleosts [10]. For example, in gilthead sea bream, myogenesis progression and the expression levels of myofibril organization-related genes such as UNC45 and Hsp90α were modified by temperature during the early stages of the life cycle [14]. By using digital morphometry and immunolabeling to quantitatively examine Pax7, myogenin, and H3P, the proliferation and differentiation of muscle precursor cells (MPCs) were proven to be changed by different temperatures, which was an underlying reason for the lasting effects on muscle cellularity and body growth of European pearlfish (Rutilus frisii meidingeri) [10]. Recent studies in European sea bass (Dicentrarchus labrax) and Senegalese sole (Solea senegalensis) have indicated that temperature-dependent muscle growth is regulated by epigenetic factors and that higher temperature increases myogenin expression and improves muscle growth mediated by DNA methylation or miRNA [15,16]. However, the underlying mechanisms of temperature-regulated muscle growth in teleosts remain largely unknown.
The in vitro culture of muscle cells is an effective system for studying the process of myogenesis and investigating the effects of various regulatory factors on muscle formation and growth [17,18]. The murine myoblast cell line C2C12 has been used extensively as an examination model for mammals and has led to numerous important discoveries in the mechanisms of myogenesis [19,20]. As muscle cell lines are unavailable in most species, primary cultures of skeletal muscle cells have become a viable alternative for muscle biology research [21,22], which can facilitate direct investigation of the impact of precisely controlled conditions, such as changes in cultivation temperatures. For instance, studies on cultured myogenic stem cells (satellite cells) in pigs and turkeys demonstrated that temperature significantly changed gene expression patterns in proliferating and differentiating muscle cells and thus had a profound effect on skeletal muscle growth [23,24]. For teleosts, muscle cells have been isolated and cultured in vitro from zebrafish (Danio rerio) [25], rainbow trout (Oncorhynchus mykiss) [26], and black rockfish (Sebastes schlegelii) [22]. Nevertheless, as mentioned above, myogenesis in fish has several unique characteristics compared to that in endothermal mammals and birds, and the regulation of fish skeletal muscle growth and development, especially the underlying mechanisms controlling the response to different temperatures, is still poorly understood and needs to be further investigated.
The spotted sea bass (Lateolabrax maculatus) is one of the most important marine-culture fish species in China, with annual production exceeding 190,000 tons in recent years (China Fishery Statistical Yearbook, 2021). Due to its high nutritional value and pleasant taste, spotted sea bass have become increasingly popular and require an improvement in growth as well as muscle mass to meet market demand. Therefore, understanding the principle of myogenesis in spotted sea bass is fundamentally important to explore valuable possibilities for application to fish culture production. Additionally, the rearing range of water temperature of L. maculatus is 16-33 • C, and the phenomenon of temperature-dependent growth has been reported in this species [27][28][29], but knowledge of the regulatory mechanism of skeletal muscle growth and development is still lacking. In this study, we employed primary cultures of L. maculatus skeletal muscle cells, in combination with comparative transcriptome analysis, to investigate the effects of different temperatures on in vitro skeletal muscle development. The aims of our study were (1) to establish a method for primary cultures of L. maculatus skeletal muscle cells in vitro; (2) to evaluate the effects of temperature on the proliferation, differentiation, and migration of cultured L. maculatus skeletal muscle cells; (3) to identify the key genes and pathways responsible for the proliferation and differentiation of skeletal muscle cells; and (4) to understand the potential molecular mechanism accounting for the phenotypic differences in skeletal muscle cells at the proliferation and differentiation stages in response to different temperatures. The results from this study will provide the basis for revealing how temperature affects myogenesis in L. maculatus and potentially be used to develop effective management strategies to be beneficial to skeletal muscle growth.

The Effects of Temperature on the Proliferation, Differentiation and Migration of Skeletal Muscle Cells
In vitro EdU and CCK-8 assays were employed to evaluate the effect of temperature on cell proliferation. The EdU assay showed that the ratio of EdU+ cells was positively correlated with temperature; in comparison with that at 21 • C, the percentage of EdU+ cells cultured at 25 • C and 28 • C increased 2.43-fold and 3.5-fold, respectively ( Figure 1A,B). For the CCK-8 assay, the absorbance was determined at a wavelength of 450 nm to assess cell proliferation. As shown in Figure 1C, the OD values at 450 nm increased with increasing temperature (28 • C > 25 • C > 21 • C). These results indicated that the proliferative rate of skeletal muscle cells was increased as the temperature rises.   The effect of temperature on differentiation was revealed by the NFI assay. As shown in Figure 1D,E, although no significant difference was observed between cells cultured at 25 • C and 28 • C, the nuclear fusion index (NFI) values at 25 • C and 28 • C were remarkably higher than those at 21 • C. Notably, a number of multinucleated myotubes (with 3 nuclei per cell) were observed only in cells cultured at 28 • C ( Figure 1D). These results indicated that the differentiation progress of skeletal muscle cells was induced by high temperature.
In addition, an in vitro wound scratch assay was performed to evaluate the effect of temperature on cell migration. The wound closure rates of cells cultured at 21 • C, 25 • C, and 28 • C were calculated at different times postwounding. The results showed that the wound closure rates increased remarkably with increasing temperature, although the difference between cells cultured at 25 • C and 28 • C was not significant at 24 h postwounding ( Figure 1F,G). Overall, the above results indicated that higher temperatures significantly accelerated the proliferation, differentiation, and migration of skeletal muscle cells.

Comparative Transcriptome Analysis between Cells in the Proliferation and Differentiation Stages
Principal component analysis (PCA) of the gene expression was conducted to evaluate the association between samples. Results showed that in the PCA plot, the replicated samples within the same group were located close to each other, and samples of different groups were positioned spatially distinct ( Figure S1), indicating the reliability of our experiment. Gene expression levels were initially compared between cell samples at the two developmental stages (proliferation vs. differentiation), which were cultured at the same temperature. Therefore, three pairwise comparison groups were generated, including PL vs. DL, PM vs. DM, and PH vs. DH groups. As shown in Figure 2A, there were 1411 upregulated and 3009 downregulated differentially expressed genes (DEGs) in the PH groups compared with the DH group, 2509 upregulated and 3761 downregulated DEGs between the PM vs. DM groups, and 1213 upregulated and 1544 downregulated DEGs between the PL vs. DL groups. Specifically, a total of 1309 overlapping DEGs among the three comparison groups, including 540 upregulated and 749 downregulated genes in the proliferation groups compared with the differentiation groups, were identified ( Figure 2B). GO enrichment analysis was conducted for those 1309 common DEGs to provide insights into their potential functions. As shown in Figure 2C, upregulated DEGs in the proliferation stage compared with the differentiation stage were mainly enriched in DNA replication-related GO terms, such as single-stranded DNA binding, DNA secondary structure binding in molecular function (MF), replication folk, mitotic spindle, condensed chromosome kinetochore, and condensed chromosome with respect to the cellular component (CC) and regulation of chromosome segregation, mitotic nuclear division, mitotic sister chromatid segregation, sister chromatid segregation, nuclear chromosome segregation, and chromosome segregation with respect to biological processes (BP). In contrast, downregulated DEGs in the proliferation stage compared with the differentiation stage were significantly enriched in myogenic differentiation-related GO terms, including actin binding, actin filament binding, muscle alpha-actin binding in MF, myofilament, striated muscle thin filament, sarcomere, contractile fiber, Z disc, myofibril, I band, and myofilament with respect to CC and actin-myosin filament sliding, muscle filament sliding, muscle fiber development, striated muscle cell development, muscle cell development, muscle tissue morphogenesis, muscle organ development, striated muscle tissue development, and muscle tissue development with respect to BP ( Figure 2D).
The gene set enrichment analysis (GSEA) was performed to extract further biological insights from the whole gene expression datasets instead of limiting by the arbitrary threshold for the identified DEGs. Overall, DNA replication-related gene sets, such as DNA strand elongation, chromosome maintenance, and activation of the prereplicative complex, were significantly positively correlated with proliferation groups (gene sets that were upregulated in PH, PM, and PL) ( Figure 2E). Meanwhile, gene sets including striated muscle contraction, HSF1-dependent transactivation and activated NOTCH1 transmits signal to the nucleus were positively correlated with the differentiation group (gene sets that were downregulated in PH, PM, and PL) ( Figure 2F).

Identification of Significant Modules and Genes Associated with Temperature by WGCNA
Weighted gene co-expression network analysis (WGCNA) was performed to identify modules and genes highly correlated with temperature treatment. After filtering out the low-expressed genes (FPKM < 1) in 18 transcriptome sequencing libraries, 12,191 genes were retained for WGCNA. As a result, a total of 11 coexpression modules were generated, which are labeled using different colors in Figure 3A and are described on the y-axis of Figure 3B. Module-trait correlation calculation results indicated that four modules (green, blue, brown, and red modules) were significantly associated with temperature at proliferation or differentiation stages (PH, PL, DH, and DL groups); therefore, we selected these four modules for further analysis. To investigate the relationships of coexpressed genes and to identify key regulatory genes (hub genes), PPI networks were constructed by Cytoscape using the top 50 genes ranked by connectivity within each selected module. In detail, the green module was significantly correlated with the PH group (correlation coefficient = 0.82, p-value < 0.05). In total, 447 genes were included in this module, and kdelr2b, leap2, krtcap2, svep1, s100a14, colgalt1, znf706, mob1a, tmem200a, and bag2 were identified as the hub genes ( Figure 3C). For the blue module, which correlated with the PL group (correlation coefficient = 0.86, p-value < 0.05), a total of 1376 genes were included, and the frmd4a, anln, ncapg, sesn2, bard1, aldh1a1, inppl1, arhgap23, etf1, and tdh genes were identified as hub genes ( Figure 3D). For the brown module, which was closely related to the DH group (correlation coefficient = 0.65, p-value < 0.05), a total of 1168 genes were included in this module, and afmid, stip1, xirp1, tspan12, mib2, tekt4, glg1, stac3, fam117a, and thrap3 were recognized as hub genes ( Figure 3E). For the red module, which was closely associated with the DL group (correlation coefficient = 0.89, p-value < 0.05), 320 genes in total were included, and plin4, sell1, gcdh, pgd, arl6ip1, mtrf1, hadh, pla2g4c, klhl4l, and ypel3 were identified as hub genes ( Figure 3F).

Functional Enrichment Analysis for Significant Modules Affected by Temperature
To further investigate the biological functions related to the four significant modules, GO enrichment analysis was performed for the genes within each module. The results showed that genes in the green module (related to the PH group) were mainly enriched in extracellular matrix (ECM) organization-related GO terms, including extracellular matrix structural constituent and collagen binding in MF, collagen trimer in CC and extracellular structure organization, extracellular matrix organization, connective tissue development, and collagen fibril organization in BP. In addition, GO terms including endoplasmic reticulum lumen, integral component of endoplasmic reticulum membrane, calcium-mediated signaling using intracellular calcium source, and regulation of cardiac muscle contraction by calcium ion signaling were also significantly enriched in the PH group. Genes in the blue module (related to the PL group) were mainly enriched in DNA damage response-and repair-related GO terms, including damaged DNA binding and nucleoside binding in MF, mitotic spindle, spindle, and condensed chromosome with respect to CC, as well as double-strand break repair, nucleotide-excision repair, DNA repair, signal transduction in response to DNA damage and negative regulation of the mitotic cell cycle with respect to BP. In addition, genes in the brown module (related to the DH group) were mainly enriched in heat stress response, calcium homeostasis, myogenic differentiation, and sarcomere assembly-related GO terms, such as titin binding and structural constituent of muscle in MF, M band, myosin II complex, muscle myosin complex, striated muscle in thin filament, myofilament, sarcomere, contractile fiber, myofibril, Z disc, and I band with respect to CC, and muscle contraction, skeletal muscle cell differentiation, cellular response to heat, myotube differentiation, skeletal muscle tissue development, actomyosin structure organization, sarcomere organization, muscle cell development, myofibril assembly, and striated muscle cell development with respect to BP. Meanwhile, genes in the red module (related to the DL group) were mainly enriched in immune response-related GO terms, such as primary lysosome and azurophil granule with respect to CC, and myeloid cell activation involved in immune response, neutrophilmediated immunity, neutrophil activation, neutrophil activation involved in immune response, leukocyte degranulation, and neutrophil degranulation with respect to BP ( Figure 4A).  In addition, GSEA was performed to identify gene sets with a statistically significant difference between the PH and PL groups, as well as the DH and DL groups. The results showed that the most significantly enriched gene sets were positively correlated with the PH group, including HSF1 activation, HSF1-dependent transactivation, and cholesterol biosynthesis, while DNA double-strand break repair, apoptosis, and gene silencing by RNA were positively correlated with the PL group. For the DH group, the significantly enriched gene sets included myogenesis, striated muscle contraction, and HSF1 activation. The TNFR2 noncanonical NF-kB pathway, signaling by the B-cell receptor (BCR), and signaling by interleukins were positively correlated with the DL group ( Figure 4B).

Putative Functional Roles of Genes in Skeletal Muscle Cells Affected by Temperature
Based on the above functional enrichment analysis results, in combination with manual literature searches about the functions of DEGs, we summarized and drew a putative schematic diagram presenting the main functional categories affected by temperature at proliferation and differentiation stages ( Figure 5A) and highlighted genes involved in skeletal myogenesis regulated by temperature ( Figure 5B).
High temperature (28 • C) promoted "HSF1 activation", "DNA replication" and "ECM organization"-related genes and pathways at the proliferation stage, which might be responsible for temperature sensation and promote the cell proliferation process ( Figure 6A). In addition to "HSF1 activation", it is worth noting that high temperature (28 • C) induced a large number of DEGs engaged in "Calcium activity", "Myogenic differentiation and myoblast fusion", and "Sarcomere assembly" at the differentiation stage, which may account for the significantly accelerated developmental process of skeletal muscle cells. In contrast, the most abundant functional classes affected by low temperature (21 • C) were "DNA damage and repair" and "Apoptosis" at the proliferation stage, as well as immune response-related pathways such as "Cytokine signaling" and "Neutrophil degranulation" at the differentiation stage ( Figure 5A).

Validation of RNA-Seq Data via qPCR
To validate the RNA-seq results, several DEGs (myomaker, fgf7, musk, ki67, myocilin, myoferlin, myogenin, myod1, and myf5) were selected for the qPCR analysis. Results showed that there was a positive correlation coefficient (R 2 ≥ 0.85) between the RNA-seq and qPCR data, indicating that bioinformatic analysis results were reliable ( Figure 6B). The key genes involved in skeletal myogenesis induced by high temperature (28 • C) at the differentiation stage. High temperature upregulated expressions of "HSF1 activation", "Calcium activity", "Myogenic differentiation and myoblast fusion", and "Sarcomere assembly" related genes, respectively, which may account for the significantly accelerated developmental process of skeletal muscle cells. The DEGs identified in this study were labeled with red color. SR: sarcoplasmic reticulum. The full names of genes are listed in Table S2.

Discussion
Changes in temperature have already attracted particular attention to the aquaculture industry because of their remarkable impact on muscle, which ultimately influences growth performance and meat quality [30,31]. Understanding the regulatory mechanism of temperature on skeletal muscle could facilitate the development of a rational management strategy to promote growth and optimize aquaculture production. However, the knowledge to date of the temperature-regulatory systems involved in controlling the skeletal muscle growth of fish is still limited. In this study, we utilized a cell culture method to investigate the effects of temperature on the skeletal muscle growth of spotted sea bass. At the cellular level, morphological observation provided direct evidence to characterize temperature function. When incubated at different temperatures (21 • C, 25 • C, and 28 • C), skeletal muscle cells exhibited a significant increase in cell number, wound healing rate, and cell fusion in response to rising temperatures, which indicated accelerated proliferation, differentiation, and migration rates at higher temperatures. This phenomenon was consistent with our previous in vivo study of juvenile spotted sea bass, which showed that the growth performance indicated by the parameters of both specific growth rate (SGR) and weight gain rate (WGR) was significantly higher when reared at 28 • C in comparison with 20 • C after 30 days [32]. The impacts of temperature on myogenesis have also been studied using a primary cultured muscle cell model in chickens, which reported the significant influence of thermal stress during myogenic proliferation and differentiation stages [33,34].
We then performed comparative transcriptome analysis to investigate the underlying molecular mechanism regulating the responses of skeletal muscle cells to different temperatures. As we summarized in Figure 5A, within the range of living temperature conditions, higher temperatures could improve the myogenic proliferation process mainly by activating the HSF1 pathway, altering ECM organization, and accelerating DNA replication. As the most well-known temperature-sensitive transcription factor, HSF1 and its downstream HSPs have been demonstrated to be essential regulators of cell proliferation in several cell types [35,36]. For example, it was proven that the signaling pathway activated by HSF1-Hsp27 facilitates Hek293 cell proliferation [37]. In pregnant rats, HSF1-HSP90 was localized in myometrial cells and suggested to play important roles in myometrial proliferation [38]. Moreover, cell proliferation mediated by Ras-induced ROS was recognized to be associated with cell type-specific alterations of the HSF1/SESN3/p21Cip1/WAF1 pathways [39]. HSF1 enhanced HeLa cell proliferation, migration, and invasion by enhancing MTDH-VEGF-C expression in vitro [40]. Although little is known about the roles of the HSF1 pathway in muscle cell proliferation, evidence has shown that the heat stress-associated increase in skeletal muscle mass in mice may be induced by HSF1 and/or the HSF1-mediated stress response that activates muscle satellite cells and the Akt/p70S6K signaling pathway [41].
The ECM is known to modulate cellular morphology and act as a conduit between extracellular stimuli and cells by regulating proliferation, migration, differentiation, and survival [42]. Skeletal muscle contains three layers of ECM, which include the endomysium, perimysium, and epimysium. These three layers of connective tissue structure provide both structural support and biochemical cues that direct muscle formation [43]. Interactions between skeletal muscle myoblasts and their ECM have been extensively documented. Liu et al. demonstrated by analyzing the protein interaction signals between cells that ECM could act on skeletal muscle progenitor cell proliferation [44]. Stern et al. confirmed that myoblasts cultured on ECM extracted from adult rat leg muscles have enhanced proliferation ability [45]. The results also demonstrated that ECM could have a direct effect on leiomyoma smooth muscle cells (LSMCs) and participate in their proliferation through the interplay between the collagen matrix and the PDGF-stimulated MAPK pathway [46].
During skeletal myogenesis, myoblasts irreversibly exit the cell cycle and differentiate and then fuse into multinucleated myotubes, which further mature into functional myofibers [4,47]. In our study, after 48 h of cultivation in differentiation medium, the skeletal muscle cells cultured at higher temperatures exhibited increased numbers of multinucleated myotubes ( Figure 1D). Consistent with the phenotypic observation, the RNA-seq analysis results revealed that numerous DEGs engaged in myogenic differentiation and myoblast fusion leading to myotube formation showed upregulated expression levels at higher temperatures. The associated DEGs and their potential involvement in myogenesis are presented schematically in Figure 5B and discussed as follows: (1) HSF1 activation-mediated genes (heat shock proteins), (2) genes mediating calcium activity, (3) genes mediating myogenic differentiation and myoblast fusion, and (4) genes mediating sarcomere assembly.
In this study, hspa1a and hsc70 belonging to the Hsp70 family, hsp90aa1.1, and serpinh1b (hsp47) were all upregulated at 28 • C compared with those at 21 • C. Evidence has wellconfirmed that Hsp90α1 is required for organized myofibril assembly in skeletal muscles of zebrafish embryos through facilitating myosin folding and assembly into organized myofibril filaments [48]. In terms of muscular adaptation, Hsp70 has been demonstrated to play a pivotal role in preventing apoptosis, influencing energy metabolism, and facilitating cellular processes after heat treatment [49]. Although the function of hsp47 in skeletal myogenesis during adaptation to temperature changes has not been explored for ectothermic teleosts, as a collagen-specific molecular chaperone, the expression of Hsp47 in rat skeletal muscle may regulate collagen production with gravitational conditions [50]. Therefore, we speculated that higher temperature within the range of suitable living conditions may activate HSPs, which not only function as the heat sensor but also accelerate the skeletal myogenesis process via facilitating myofibril assembly and maintaining cellular homeostasis.
The process of myoblast fusion during skeletal myogenesis is calcium regulated, depending on both the continuous presence of cell surface calcium and an adequate intracellular Ca 2+ [51]. Ca 2+ stores are essential for eliciting a precise spatiotemporal pattern of Ca 2+ signal in developing muscle cells [52]. In our study, ryr1, serca1a, and camk2b showed significantly higher expression levels at 28 • C than at 21 • C. RyR-mediated Ca 2+ dynamics have been demonstrated as a universal requirement for skeletal myogenesis [53]. Ryr1 is responsible for releasing Ca 2+ from the SR to the myoplasm and serca1a functions to uptake Ca 2+ from the myoplasm to the SR which maintains calcium homeostasis [53][54][55][56][57][58]. After the intracellular Ca 2+ concentration reaches a certain threshold value, subsequently activating camk2b and forming the Ca 2+ /CaM complex, further inhibiting histone deacetylase II (HDACII) and increasing nuclear abundance of myogenin and myocyte enhancer factor 2 (MEF2) [59], which synergistically regulate myogenesis and induce the development of skeletal muscle [60,61].
Myogenic specification and differentiation are coordinated by a series of myogenic transcription factors. Two key transcription factor families including the myogenic basic helix-loop-helix (bHLH) proteins and myocyte enhancer factor 2 (MEF2) are essential for both specification and differentiation of muscle progenitors [62]. Previous studies have shown that MEF2 factors are essential for muscle differentiation, and MyoD and MEF2 family members function combinatorially to activate transcription and myogenesis [63]. Additionally, mef2c has been proved as a direct transcriptional target of the bHLH proteins during skeletal muscle development [63]. In our results, the expression of mef2c showed upregulation at 28 • C compared with those at 21 • C. Besides, muscle-specific genes such as the myomaker, fitm1, stac3, mlip, myoz1b, and xkr8, which are activated transcriptionally by myogenic transcription factors via binding to their conserved E-box (CANNTG) and then trigger the myoblast fusion process [64][65][66][67][68][69][70], all exhibited higher expression levels at 28 • C than those at 21 • C. The induced expressions of these key genes regulating myogenesis might be the key cause for the accelerated skeletal muscle myogenic differentiation and myoblast fusion process at higher temperature.
Overall, according to our transcriptomic analysis results, the significantly accelerated proliferation, differentiation, and migration rates of skeletal muscle cells in response to rising temperatures revealed by morphological observations (Figure 1) might result from the high temperature-induced expression levels of genes involved in "HSF1 activation", "DNA replication" and "ECM organization" at the proliferation stage, as well as genes associated with "HSF1 activation", "Calcium activity", "Myogenic differentiation and myoblast fusion", and "Sarcomere assembly" at the differentiation stage ( Figure 5).

Ethics Statement
All experiments were conducted in accordance with the guidelines of the Animal Research and Ethics Committee of Ocean University of China (Permit Number: 2014201). In addition, our research did not involve endangered or protected species.

I.
Cell proliferation at different temperatures. The isolated cells were plated at a density of 15,000 cells per well in culture flasks and incubated in a 25 • C incubator with GM for 24 h. A subset of cells was divided into three parts of equal amounts and cultured at experimental temperatures of 21 • C, 25 • C, and 28 • C. The GM was changed every 24 h. After 72 h of proliferation, the cell medium was removed, and the cells were rinsed with cold PBS and collected for RNA extraction ( Figure S2A). II.
Cell differentiation at different temperatures. The other part of the isolated cells was divided into three equal parts and cultured with GM for 72 h of proliferation at 25 • C, reaching approximately 90% confluency. Then, the GM was replaced with differentiation medium (DM) (L-15, 2% horse serum (G-CLONE, Beijing, China), 1% 100 U mL −1 penicillin, and 100 µg mL −1 streptomycin) to induce differentiation, and the cells were cultured in incubators at 21 • C, 25 • C, and 28 • C. The DM was changed every 24 h. After 48 h of differentiation, the cell medium was removed, and the cells were rinsed with cold PBS and collected for RNA extraction ( Figure S2B).
In brief, for the EdU assay, the cells were divided into three equal parts and placed into three 6-well cell culture plates cultured in a 25 • C incubator overnight with GM. EdU (10 µM) was added to each well, and the cells were incubated for 6 h at 21 • C, 25 • C, and 28 • C, fixed with 4% PFA for 15 min, permeabilized in 0.25% Triton X-100 for 10 min, and reacted with Click-iT reaction buffer for 30 min. Subsequently, the images were visualized by an Echo Revolve microscope (ECHO, Chicago, IL, USA). The percentage was calculated as the ratio of the number of EdU-positive cells to the number of total nuclei stained with 4 ,6-diamidino-2-phenylindole (DAPI, 2 mg/mL) in the image. Three microscopic vision fields of each well were randomly selected to calculate the ratio of positive cells.
For the CCK-8 assay, cells were divided into three equal parts, placed into three 96-well cell culture plates (5000 cells/well), and cultured in a 25 • C incubator overnight with GM. Subsequently, the cells were transferred and cultured at 21 • C, 25 • C, and 28 • C for 24 h, and then 10 µL CCK-8 reagent was added to the cell plate for 4 h. The capacity of cell proliferation was assessed by detecting the optical density (OD) of each well at 450 nm wavelength using a microplate reader (Bio-Rad, Hercules, CA, USA).

Cell Differentiation Assay
The effect of temperature on cell differentiation was measured by nuclear fusion index (NFI) assay. Cells were divided into three equal parts, placed into three 6-well cell culture plates, and cultured with GM for 72 h of proliferation at 25 • C until reaching approximately 90% confluency. Then, the GM was replaced with DM. Cells were induced to differentiate at 21 • C, 25 • C, and 28 • C for 48 h, and then the cell nuclei were stained with DAPI.
Briefly, cells were washed with cold PBS, fixed in 4% paraformaldehyde in PBS solution for 20 min, permeabilized in 0.25% Triton X-100 for 15 min, stained with 2 mg/mL DAPI for 2 min, and washed in PBS three times for 5 min. Images were taken under a fluorescence microscope (ECHO, Chicago, America). NFI was calculated manually as a ratio of the number of nuclei inside myotubes to the number of total nuclei after myotube formation at 21 • C, 25 • C, and 28 • C, as previously described [93]. Three microscopic vision fields of each well were randomly selected to calculate the myotube NFI, and the experiment was repeated three times.

In Vitro Wound Scratch Assay
The wound scratch assay was employed to evaluate the migration rates of cells cultured at different temperatures. First, cells were divided into three equal parts and placed into three 6-well cell culture plates, supplemented with GM, and cultured at a 25 • C incubator until they reached full confluency. A sterile 200 µL pipette tip was used to create a uniform scratch wound on the monolayer of cells, and then the cell debris was removed by washing with PBS 3 times. The scratch cells were cultured in serum-free medium at 21 • C, 25 • C, and 28 • C. Images were captured at 0, 12, and 24 h with an Echo Revolve microscope (ECHO, Chicago, America), and the scratch closure rate was measured using ImageJ software (v2.1.0, Bethesda, MD, USA).

RNA Extraction and Transcriptome Sequencing
The total RNA of cells at proliferating and differentiating stages which were cultured at three experimental temperatures (21 • C, 25 • C, and 28 • C) ( Figure S2) was extracted by TRIzol reagent (Vazyme, Nanjing, China) according to the manufacturer's instructions. The RNA concentration and quality were evaluated by Nanodrop 2000 (Nanodrop Technologies, Wilmington, DE, USA), and the RNA integrity was confirmed by the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). A total of 18 sequencing libraries were constructed using the NEBNext ® Ultra™ RNA Library Prep Kit for Illumina ® (NEB, Ipswich, MA, USA) following the manufacturer's instructions and index codes were added to each sample to attribute sequences. Then all the libraries were sequenced on an Illumina Nova-Seq 6000 platform, generating 150 bp paired-end reads. The RNA-seq data have been deposited in the NCBI SRA database under the Accession number: PRJNA859992.

Weighted Gene Co-Expression Network Analysis (WGCNA) and Hub Genes Identification
After filtering the genes with FPKM values lower than 1, the weighted gene coexpression networks were constructed by the WGCNA package (v1.47) in R [98]. FPKM values of genes were imported to constructed co-expression modules. Expression correlation coefficients of these imported genes were calculated, and the soft threshold β = 14 was selected to build gene networks using a scale-free topology model. Genes with highly similar expression patterns were clustered into branches, and modules were identified by the tree cut algorithm. To identify significant modules related to proliferation or differentiation stages under different temperatures, module eigengenes (MEs) were used to calculate the module-trait relationship. The intramodular connectivity, the gene trait significance absolute value (GS), and the gene module membership (MM) within each significant module were calculated to screen hub genes. Here, the top 10 connectivity genes with GS > 0.2 and MM > 0.8 were considered hub genes. The screened hub genes and interaction network for each related module were subsequently visualized by Cytoscape (v3.8.2) [99].

Validation of Gene Expression by Real-Time PCR
Real-time PCR was performed for validating the bioinformatic analysis results of RNAseq data. A part of the RNA samples obtained from the above cell culture experiments was reverse-transcripted to cDNA using the HiScript III 1st Strand cDNA Synthesis Kit (+gDNA wiper) (Vazyme, Nanjing, China). Real-time PCR was conducted by Applied Biosystems 7300 machines (Applied Biosystems, Foster, CA, USA). Eight genes including myomaker, fgf7, musk, ki67, myocilin, myoferlin, myogenin, myod1, and myf5 were chosen for validation, and α-tubulin was regarded as the reference gene. The specific primers of selected genes are listed in Table S3. The reaction system of 10 µL included 5 µL of ChamQ™ SYBR Color qPCR Master Mix (Vazyme, Nanjing, China), 1 µL of cDNA templates, 0.2 µL of each primer, and 3.6 µL of ddH 2 O. The qPCR procedure was performed as follows: 95 • C for 30 s and 40 cycles of 95 • C for 10 s, 60 • C for 30 s, and 72 • C for 30 s. All qPCR experiments were performed by three biological sample replicates with three technical replicates. Then the relative gene expression values were calculated by the 2 − Ct method.

Statistical Analysis
The differences among different groups were analyzed by one-way analysis of variance (ANOVA) followed by Duncan's multiple range test. The difference was considered statistically significant for p-value < 0.05. Data were shown as the mean ± standard deviation (SD). All statistical analysis was performed by IBM SPSS Statistics 21.0 software (IBM SPSS, Turkey). The results were shown by GraphPad Prism software (v8.4.2) (GraphPad Software, San Diego, CA, USA).

Conclusions
In our study, we characterized cultured muscle cells of L. maculatus for the first time and investigated the impacts of different temperatures (21 • C, 25 • C, and 28 • C) on myoblast proliferation and differentiation at both the morphologic and molecular levels. The results of CCK-8, EdU, wound scratch, and nuclear fusion index assays revealed that higher temperatures significantly accelerated the proliferation, differentiation, and migration processes of myogenic cells in vitro compared to lower temperatures. Gene expression profiles were revealed by RNA-seq analysis. According to GO, GSEA, and WGCNA analyses, higher temperature induced HSF1-activation, DNA-replication, and ECM-organization processes at the proliferation stage; conversely, lower temperature induced DNA-damage, DNA-repair, and apoptosis processes. At the differentiation phase, higher temperatures induced HSF1-activation, calcium activity regulation, myogenic differentiation and myoblast fusion, and sarcomere assembly processes; conversely, lower temperatures induced cytokine-signaling and neutrophil-degranulation processes. Overall, the cell morphological observation and RNA-seq data presented here provide a significant improvement to our current understanding of how temperature affects L. maculatus skeletal muscle development and growth.  Data Availability Statement: Publicly available datasets were analyzed in this study. This data can be found here: PRJNA859992.