1 Introduction

Myostatin (MSTN) has been well-known as a negative regulator of muscle growth and development. Its mutation produces a “double-muscle” phenotype, which shows its inestimable commercial value in improving meat production of livestock and poultry, and providing high-quality protein for humans (Fan et al., 2022; Chen et al., 2021b). Due to its role in promoting muscle atrophy and cachexia, MSTN has been recognized as a promising therapeutic target to offset the loss of muscle mass (Lee, 2021; Baig et al., 2022; Wijaya et al., 2022).

MSTN is highly conserved in mammals, and mutations in the MSTN gene, either artificially or naturally, will result in increased skeletal muscle weight and produce a “double-muscle” phenotype, which has been reported in many species, including cattle, sheep, and pigs, rabbits, and humans (Grisolia et al., 2009; Dilger et al., 2010; Kambadur et al., 1997). However, mutations at different loci of the MSTN often produce variety of different phenotypes, and its molecular mechanism of skeletal muscle growth and development remains controversial (Hanset and Michaux, 1985; Grobet et al., 1997; Wegner et al., 2000; Kambadur et al., 1997; Marchitelli et al., 2003). More than 77 natural mutation sites of MSTN have been reported in various sheep breeds, most of these mutations were found to be located in the non-coding regions, and did not affect MSTN activity (Kijas et al., 2007; Sjakste et al., 2011; Han et al., 2013; Dehnavi et al., 2012). In addition to introns, it is still possible that mutations in regulatory regions and exons may not affect the sheep phenotypes (Pothuraju et al., 2015; Kijas et al., 2007; Boman and Vage, 2009; Boman et al., 2009).

Fibroblast growth factor 5 (FGF5) belongs to the fibroblast growth factor (FGF) family and is a secretory signaling protein. FGF5 played an inhibitory effect on mouse hair growth (Hebert et al., 1994), and its natural mutation can lead to a significant increase in hair growth in angora mice (Sundberg et al., 1997). Subsequent studies have also successively confirmed the inhibitory effect of FGF5 on mammalian hair growth and is recognized to be a negative regulator of hair growth (Kehler et al., 2007; Dierks et al., 2013; Yoshizawa et al., 2015; Legrand et al., 2014; Higgins et al., 2014).

FOS-like 1 (FOSL1), also named Fos-related antigen 1 (FRA1), is a member of the Fos subfamily of activator protein 1 (AP-1) superfamily. The Fos family proteins are involved in cell proliferation, differentiation, and transformation. FOSL1 and even AP-1 family members are generally recognized to be related to various cancers. However, accumulating evidence indicate that AP-1 family proteins play a critical role in skeletal muscle cell proliferation, differentiation, and muscle development (Puntschart et al., 1998; Liu et al., 2010). FOSL1 has also been characterized as an important component of the response of skeletal muscle during aging and a lipid biosynthesis-related factor (Mathes et al., 2021; Wang et al., 2017). Furthermore, FOSL2 and c-Jun, another member of the AP-1 family, were found to inhibit myoblast differentiation (Alli et al., 2013; Bengal et al., 1992). In all, it is foreseeable that the role of FOSL1 in myogenesis will increasingly emerge.

In this study, to increase both meat and wool production, we first produced the MSTN and FGF5 dual-gene biallelic homozygous mutations sheep by the increased delivery ratio of Cas9 mRNA to sgRNA targeting MSTN and FGF5. The MSTN^Del273C mutation with FGF5 knockout sheep highlights a dominant “double-muscle” phenotype by decreasing the muscle fiber cross-sectional area and increasing the number of muscle fibers per unit area. Then, we used the MSTN and FGF5 dual-gene biallelic homozygous mutations sheep to unravel the molecular mechanism of the “double-muscle” phenotype and myofiber hyperplasia.

2 Materials and Methods

2.1 Tissue sample collection and preparation

Gluteus medius and longissimus dorsi were harvested from WT and MSTN^Del273C mutation with FGF5 knockout (MF^-/-) sheep, and three WT sheep and four MF^-/- F1 generation sheep (half-sib) were used for feeding and slaughter. All samples were immediately frozen in liquid nitrogen and then stored at −80°C until analysis. All sheep are raised by the national feeding standard NT/T815-2004. All procedures performed for this study were consistent with the National Research Council Guide for the Care and Use of Laboratory Animals. All experimental animal protocols in this study were approved and performed following the requirements of the Animal Care and Use Committee at China Agricultural University (AW02012202-1-3). All surgeries were performed under sodium pentobarbital anesthesia, and all efforts were made to minimize any suffering experienced by the animals used in this study.

2.2 Cell isolation, culture, and transfection

Sheep skeletal muscle satellite cells were isolated and cultured as previously described (Chen et al., 2021a). In brief, the muscle tissues of the hind limbs from 3-month-old sheep fetuses were cut into small pieces, digested with 0.2% collagenase type II (Gibco, Grand Island, NY) at 37℃ for 1 h, and then centrifuged at 1000 rpm for 10 min. The precipitates were continued digested with 0.25% trypsin (Gbico, Grand Island, NY) at 37℃ for 30 min, and digestion was terminated with serum containing medium. The cell suspension was successively filtered through 100, 200 and 400 mesh cell sieves. After this, the cells were centrifuged at 1000 rpm for 10 min, resuspended in growth medium (GM) containing DMEM/F12 (Gbico, Grand Island, NY) with 20% fetal bovine serum (FBS, Gibco) and 1% penicillin-streptomycin liquid (Gbico, Grand Island, NY), and cultured for 2-3 times with differential adhesion. To induce differentiation, the cells were cultured to 70% confluence in GM, and followed by an exchange to differentiation medium (DM) containing DMEM high glucose (Gbico, Grand Island, NY) with 2% horse serum (HS, Gibco) and 1% penicillin-streptomycin liquid to culture 24 h, 48 h, and 72 h. To produce viral solution for over-expression of the target gene, it was subcloned into the XbaI and BamHI sites of the lentiviral vector by seamless cloning. HEK 293T cells were co-transfected with the envelope plasmid pMD2.G, the packaging plasmid psPAX2 and the target plasmid at a mass ratio of 1:2:4. Then, the culture medium was collected at 48h and 72h after transfection, and the cell debris was removed by filtration. Then, the sheep skeletal muscle satellite cells were infected with packaged lentivirus when they were cultured to 60%-70% confluence in 96-well, 24-well or 6-well plates. Finally, cells were collected for analysis after infection at 24 h or 48 h. All the primer sequences of gene cloning were listed in Table S1.

2.3 Total RNA isolation and real-time quantitative PCR (RT-qPCR)

The total RNA of tissues and cells was isolated using TRIzol reagent (Sangon Biotech, Shanghai, China) following the manufacturer’s protocol. In short, after tissues or cells were lysed, chloroform was added to separate the organic and inorganic phases, followed by precipitation with isopropanol and ethanol in turn, and finally, the RNA was dissolved in DEPC water. Then, the first strand cDNA was prepared using PrimeScript II 1st Strand cDNA Synthesis Kit (Takara, Beijing, China). qPCR was performed using 2× SYBR Green qPCR Mix (Low ROX) (Aidlab Biotechnologies, Beijing, Chian) in a Stratagene Mx3000P (Agilent Technologies, SUA). With GAPDH mRNA as endogenous control, the relative expression level of genes was calculated by the 2^-ΔΔCt method. All primers used were listed in Table S2.

2.4 Western blot

Tissue or cell samples were lysed in RIPA buffer (Solarbio, Beijing, China) supplemented with protease and phosphatase inhibitor cocktail (Beyotime, Beijing, China) for total protein extraction. Then, equal amounts of tissue or cell lysate were resolved by 10% SDS-PAGE and transferred onto PVDF membranes (Millipore, USA). The membranes were blocked with 5% BSA for 1h, incubated with primary antibody at 4℃ overnight, then incubated with secondary antibody for 1h before detection. The fold change of protein was normalized to GAPDH for quantitative analysis by ImageJ software. The antibodies information was listed in Table S3.

2.5 5-Ethynyl-2′-deoxyuridine (EdU) assay

At 24 h after transfection, sheep skeletal muscle satellite cells were incubated at 37℃ for 2 h in 96-well plates with 50 µM EdU (RiboBio, Guangzhou, China). Then, fixed the cells in 4% paraformaldehyde for 30 min and neutralized using 2 mg/mL glycine solution. The Apollo^® staining solution which contains EdU was added and incubated at room temperature for 30 min in the dark to label the DNA in the synthesis stage, the nuclear was then counterstained with DAPI. The number of EdU positive cells was counted from the images of five random fields obtained with an inverted fluorescence microscope at a magnification of 100×. EdU labeling index was expressed as the number of EdU-positive cell nuclei/total cell nuclei.

2.6 Cell counting kit-8 (CCK-8) and cell cycle detection

Skeletal muscle satellite cells were seeded in 96-well plates and cultured for appropriate time according to different experimental treatments. Then, 10 μL CCK-8 solution was added to each well and incubated at 37°C in a 5% CO₂ incubator for 2 h, and then the absorbance at 450 nm was measured with a microplate reader.

The cultured skeletal muscle satellite cells were digested with trypsin, centrifuged at 1000 g for 5 min to collect the cell pellet, washed once with ice-cold PBS, and then 1 mL of ice-cold 70% ethanol was added to fix the cells overnight at 4°C. The next day, the cells were washed with ice-cold PBS again, and the cells were incubated with 0.5 mL PI staining solution at 37°C for 30 min and collected by flow cytometry at low speed.

2.7 Immunofluorescence staining

Sheep skeletal muscle cells were fixed in 4% paraformaldehyde for 30 min, permeabilized in 0.1% Triton X-100 for 20 min and blocked with 5% normal goat serum for 30 min at room temperature, and then incubated with primary antibody at 4℃ overnight. Next, the fluorescent secondary antibody was added and incubated at 37℃ for 1 h in the dark, and the nuclear was then counterstained with DAPI. The immunofluorescence images from five random fields were captured with an inverted fluorescence microscope.

2.8 Chromatin Immunoprecipitation (ChIP)

The cells were fixed with 1% formaldehyde for 10 min at room temperature, then neutralized with 1× glycine solution for 5 min, and washed twice with ice-cold PBS. Then, the cells were collected with a cell scraper and resuspended in 0.5 mL cell lysis buffer (10 mM HEPES, 0.5% NP-40, 1.5 mM MgCl2, 10 mM KCl, pH 7.9) containing protease inhibitor cocktail (Beyotime, Beijing, China) and incubated on ice for 15 min to release the cytoplasm. Next, cell pellets were collected by centrifugation at 800 g for 5 min at 4°C and resuspended in 0.5 mL nuclear lysis buffer (50 mM Tris, 10 mM EDTA, 0.3% SDS, pH 8.0) containing protease inhibitor cocktail. After the DNA was fragmented by ultrasonication, the suspension was centrifuged at 4°C 12000g for 10 min, and the supernatant was collected. And 50 μL supernatant of each sample was diluted with 450 μL of ChIP dilution buffer (0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris-HCl pH 8.0, 167 mM NaCl), then 5 μg primary antibody was added and incubated overnight at 4°C with rotation. The next day, 20 μL protein A/G magnetic beads were added to each sample and incubated at 4°C with rotation for 2h. Then, the magnetic beads were respectively washed once with low-salt wash buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl pH 8.0, 150 mM NaCl), high-salt wash buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 500mM NaCl, 20 mM Tris-HCl pH 8.0), LiCl wash buffer (0.25 M LiCl, 1% NP-40, 1% sodium deoxycholate, 1 mM EDTA, 10 mM Tris-HCl pH 8.0), and TE buffer (10 mM Tris-HCl pH 8.0, 1 mM EDTA) at 4°C, and the supernatant was discarded. Next, 100 μL ChIP elution buffer (1% SDS, 100 mM NaHCO₃) containing proteinase K was added to each sample, then incubated at 62°C overnight, and the DNA was finally purified by a purification column.

2.9 RNA-seq

The rRNA was removed from each total RNA sample of the gluteus medius to construct a strand-specific transcriptome sequencing library, and the Illumina Novaseq 6000 sequencing platform was used to perform high-throughput sequencing with a paired-end read length of 150 bp. Raw data were transformed into clean reads by removing reads containing adapter, ploy-N and low-quality reads from raw data. At the same time, Q20, Q30, GC-content, and sequence duplication levels of the clean data were calculated. The genome index was constructed using Hisat2 software and the clean reads were mapped to the sheep reference genome (Oar Rambouillet v1.0), the featureCounts software was used for expression quantification, and DESeq2 software was used for differential expression analysis based on P-value < 0.05 and | log2 Fold Change | > 1.

2.10 Statistical analysis

All results are presented as the mean ± SEM. Statistical analyses of differences between groups were performed using a two-tailed Student’s t-test or chi-square test and P < 0.05 was considered statistically significant. *P < 0.05, **P < 0.01 and ***P<0.001.

3 Results

3.1 Elevated molar ratio of Cas9/sgRNA can efficiently generate biallelic homozygous mutant sheep

The sgRNAs for targeting were designed in the third exon of the MSTN and FGF5 genes, respectively, to generate mutant MSTN gene for quality trait improvement and mutant FGF5 gene for yield trait improvement sheep by CRISPR/Cas9 gene editing technology (Figure 1A, B). Both MSTN and FGF5 PCR products could be cleaved by T7E1 and the fragment sizes were also as expected, and grayscale analysis showed that the editing efficiency was 14.6% and 11.4%, respectively (Figure 1C), which indicates that the designed sgRNAs can achieve more efficient gene targeting.

Figure 1

The microinjection was performed according to the injection molar ratio of Cas9 mRNA:sgRNAs (1:2, 1:10, and 1:15), respectively. The number of embryos injected, recipients of nuclear transfer, pregnancy, and alive lambs per group were listed in Table S4. The subsequent gene mutation detection showed that a total of 3 lambs were mutated in the MSTN and FGF5 genes at a Cas9 mRNA:sgRNAs injection molar ratio of 1:2, with a gene editing mutation rate of 14.3% (4/28). However, all 3 lambs were chimeric, that is, there were both mutant and wild-type after editing, and the biallelic mutation rate was 0% (Table S4, Figure 1D). When the injection molar ratio of Cas9 mRNA:sgRNAs was 1:10, two lambs were mutated in MSTN and FGF5 genes, and the mutation rate of gene editing was 18.2 % (4/22). And these two lambs were all double-gene biallelic homozygous mutant lambs, that is, the alleles on the homologous chromosomes were mutated after the target gene was edited (Table S4, Figure 1D). While the injection molar ratio of Cas9 mRNA: sgRNAs was continuously increased to 1:15, one lamb had a mutation, which was a biallelic mutation of MSTN gene, and the gene editing mutation rate was 7.14% (1/14) (Table S4, Figure 1D). All the mutation forms of bases and amino acids in MSTN and FGF5 genes are shown in Figure 1E and Figure 1F, respectively. These results indicate that increasing the delivery molar ratio of Cas9 mRNA to sgRNA from 1:2 to 1:10 can greatly improve the efficiency of biallelic mutation in sheep.

3.2 The MSTN^Del273C mutation with FGF5 knockout sheep highlights a dominant “double-muscle” phenotype and muscle fiber hyperplasia

Among gene-edited sheep, a sheep with biallelic deletion of MSTN and biallelic mutation of FGF5 aroused our great interest. Specifically, gene editing caused a deletion of 3-base pairs of AGC in the third exon of MSTN (Figure 1E, F), resulting in the deletion of cysteine at amino acid position 273 (MSTN^Del273C) (Figure 1 E, F), which is highlighted by the “double-muscle” phenotype (Figure 2A, 2B). At the same time, a biallelic mutation in FGF5 caused the knockout of FGF5 gene and increased the density and length of hairs (Zhang et al., 2020). Given the prominent “double-muscle” phenotype of MF^-/- sheep, we examined its histological morphology of the gluteus medius and longissimus dorsi, and found that the fiber cell number per unit area of muscle tissue in MF^-/- sheep was significant (P<0.01) higher than that in WT sheep (Figure 2C, D). Further, we also found that the percentage of smaller myofiber in the gluteus medius and longissimus dorsi was clear higher in MF^-/- sheep than that in WT sheep, whereas the percentage of larger muscle fiber area was lower (Figure 2C, E, F). To determine whether this phenomenon was affected by FGF5 gene mutation, we also separately examined the muscle fiber morphology in FGF5 knockout sheep alone and found that FGF5 knockout alone had no significant (P>0.05) effect on muscle fiber size (Figure S1). All these results demonstrated that the dual-gene biallelic mutation MF^-/- sheep had well-developed hip muscles with smaller muscle fibers, and this phenotype was dominated by MSTN gene.

Figure 2

To further determine the heritability of the above phenotypes, we crossed MF^-/- sheep with homozygous WT sheep to produce the offspring generation of MF^-/- sheep, which were identified that the gene editing could be stably inherited to the offspring, and they were all mutant heterozygotes of MSTN and FGF5 monoalleles (MF^+/−) (Figure S2). Subsequently, 3 WT sheep and 4 MF^+/− sheep were slaughtered. Compared to WT sheep, the proportion of rib meat was significantly reduced in MF^+/− sheep, and the proportion of hind leg meat was significantly increased, the ratio of hind leg meat increased by 21.2% (Table 1). Further, the muscle weight of different parts in WT and MF^+/− sheep has no significant difference (Table S5). However, the proportion of gluteus medius in the carcass of MF^+/− sheep was significantly (P<0.01) increased, with a 26.3% increase compared to WT sheep (Figure 2G). In addition, there were no significant (P>0.05) differences in pH, color, drip loss, cooking loss, shearing force, and amino acid content of the longissimus dorsi between WT and MF^+/− sheep (Table S6-8). Compared with WT sheep, the cross-sectional area of gluteus medius muscle fibers in MF^+/− sheep was also smaller (Figure 2H-I), and the number of muscle fiber cells per unit area was significantly increased (P<0.001) (Figure 2H, J); the percentage of smaller muscle fiber area in MF^+/− sheep was significantly increased (P<0.05), while the percentage of larger muscle fiber area was significantly decreased (P<0.05) (Figure 2H, K), these results was consistent with that in MF^-/- sheep. All these results demonstrated that the dual-gene biallelic editing of the MSTN and FGF5 could be stably inherited by the offspring, and the later generation characterized excellent traits of high-yield meat.

Table 1

3.3 The MSTN^Del273C mutation with FGF5 knockout promotes skeletal muscle satellite cells proliferation and inhibits myogenic differentiation

Given that the MSTN^Del273C mutation with FGF5 knockout produced the phenotype of “double-muscle” and reduced muscle fiber cross-sectional area, we detected the in-situ expression of MSTN protein. The results showed that there was no significant differential expression of MSTN protein in both gluteus medius and longissimus dorsi of MF^-/- sheep compared with WT sheep (Figure S3A). Also, there was no significant difference in MSTN mRNA (Figure S3B) and protein expression (Figure S3C-D) in gluteus medius of WT and MF^+/− sheep. Further immunofluorescence also revealed no significant difference in MSTN protein expression in both myoblasts and myotubes (Figure S3E). These results suggested that the MSTN^Del273C mutation with FGF5 knockout does not affect the normal expression of MSTN.

The proliferation and differentiation of skeletal muscle satellite cells is a key step in muscle formation and development. The CCK-8 and EdU cell proliferation experiments showed that the proliferative rate of MF^+/− cells were highly significantly (P<0.01) elevated (Figure 3A) with a significant (P<0.05) increase in the rate of EdU-positive cells (Figure 3B, C) compared to WT cells. In addition, cell cycle detection showed a significant (P<0.01) reduce in the proportion of G1 phase and a significant increase (P<0.05) in the proportion of S phase in MF^+/− cells (Figure 3D, E). Meanwhile, the mRNA expression levels of the cell cycle marker genes CyclinB1, CDK4, Cyclin A1, Cyclin E1, and CDK2 were significantly increased (P<0.05) (Figure 3F). These results suggest that the MSTN^Del273C mutation with FGF5 knockout may promote cell proliferation by accelerating the cell cycle from G0/G1 phase to S phase.

Figure 3

Although the mRNA levels of MyoD1 and MyoG were significantly increased after induced differentiation 2 days in MF^+/− cells (Figure 3G), the mRNA level of MyHC (Figure 3G) and the protein levels of MyoD1, MyoG, and MyHC (Figure 3H, I) were dramatically decreased (P<0.05), suggesting that the MSTN^Del273C mutation with FGF5 inhibit myogenic differentiation. Meanwhile, the immunofluorescence staining of MyoG and MyHC in myotubes showed that myotube fusion index (Figure 3J, K), number of myotubes (Figure 3J, L), and number of nuclei per myotube (Figure 3J, M) were all highly significantly (P<0.01) reduced after inducing differentiation for 2 days of MF^+/− cells compared to WT cells, as was the myotube diameter at the maximum measured (Figure 3J, N). To continue follow-up tracing of the progression of myogenic differentiation, we also performed MyoG and MyHC immunofluorescence staining after inducing differentiation for 4 days (DM4) and 6 days (DM6), and the results showed that the differentiation capacity and fusion ability of MF^+/− cells were consistently significantly lower than WT cells during the ongoing differentiation process, as was the diameter of fused myotubes (Figure S4A-J). The reduced expression of myogenic differentiation markers further confirmed that the MSTN^Del273C mutation with FGF5 knockout consistently inhibits myogenic differentiation of skeletal muscle satellite cells (Figure S4K-N). Taken together, our results elucidated that the MSTN^Del273C mutation with FGF5 knockout inhibits myogenic differentiation of skeletal muscle satellite cells and induces a smaller myotube diameter of myotubes after induced differentiation, which may explain why the cross-sectional area of muscle fibers is decreased in MF^-/- and MF^+/− sheep.

3.4 FOSL1 is a key gatekeeper of MSTN^Del273C mutation with FGF5 knockout-mediated muscle phenotype

To elucidate the potential mechanism of the MSTN^Del273C mutation with FGF5 knockout result in smaller muscle fiber cross-sectional area and myotube diameter, the RNA-seq was performed in gluteus medius. A total of 25,958 genes were identified by RNA-seq, and principal component analysis (PCA) showed relatively well repeatability within sample groups and well discrimination between WT and MF^+/− samples (Figure 4A). With P-value < 0.05 and | log2 Fold Change | > 1 as the screening criterion, 295 differentially expressed genes (DEGs) were screened, including 79 up-regulated DEGs and 216 down-regulated DEGs (Figure 4B). The cluster analysis showed that the samples were clustered into WT groups and MF^+/− groups, with obvious differences between groups (Figure 4C). Pearson correlation analysis also showed high similarity within sample groups (Figure 4D).

Figure 4

GO enrichment analysis showed that 295 DEGs were significantly enriched in signal transduction, positive regulation of cell proliferation, positive regulation of cell migration, skeletal muscle cell differentiation, muscle contraction, cardiac muscle contraction, and positive regulation of JNK cascade and so on in biological process (BP); in cellular component (CC), these DEGs mainly belonged to cell membrane, cytoplasm, perinuclear region of cytoplasm, and I band; and molecular function (MF) mainly included protein binding, nucleotide binding, ATP binding, and protein kinase activity (Figure 4E), indicating that these DEGs are significantly closely related to cell proliferation, myogenic differentiation, and muscle development. In addition, KEGG enrichment analysis showed that 295 DEGs were significantly enriched in cytokine-cytokine receptor interaction, MAPK signaling pathway, IL-17 signaling pathway, and TNF signaling pathway (Figure 4F), implying that DEGs were significantly associated with cell proliferation, differentiation, apoptosis, and inflammation. Furthermore, based on GO and KEGG enrichment analysis, 18 genes with a large difference were selected from 295 DEGs for validation at mRNA level in gluteus medius. The results showed that GLIS1 was significantly (P<0.05) up-regulated, and MYL10, FOSL1, PDPN, PLA1A, and ANKRD2 were significantly (P<0.05) down-regulated (Figure 4G).

In addition, we also verified the above six DEGs in the longissimus dorsi, in which FOSL1, PDPN and ANKRD2 were significantly (P<0.05) reduced in MF^+/− sheep (Figure S5A). Given that MSTN^Del273C mutation with FGF5 knockout promotes skeletal muscle satellite cell proliferation and inhibits differentiation, we dynamically monitored the expression of FOSL1, PDPN, and ANKRD2 during myogenic differentiation. The results indicated that both PDPN and ANKRD2 were significantly (P<0.05) up-regulated during myogenic differentiation (Figure S5B-C). More strikingly, FOSL1 mRNA level was strongly (P<0.01) decreased after induced differentiation (Figure 5A), and its expression diminished continuously with the differentiation progress. Furthermore, compared with WT cells, the mRNA expression levels of FOSL1 in MF^+/− cells at GM and DM2 were significantly (P<0.05) elevated (Figure 5B), suggesting that FOSL1 may play a crucial role in the proliferation and myogenic differentiation of skeletal muscle satellite cells.

Figure 5

As aforementioned above, the AP-1 transcription factor family member c-Fos inhibits myogenesis and MyoD1 expression by directly binding to the MyoD1 promoter region. Given that FOSL1 is a member of the AP-1 family, we, therefore, speculated that FOSL1 might have similar functions to c-Fos. Subsequent protein-protein interaction (PPI) analysis of FOSL1, c-Fos and MyoD1 further suggested that there was a potential interaction between FOSL1 and MyoD1 (Figure 5C). In addition, we also found that the mRNA expression level of c-Fos was highly significantly (P<0.01) reduced in MF^+/− myoblasts compared with WT cells, whereas the expression level of MyoD1 mRNA was dramatically (P<0.01) increased (Figure 5D). Given the physical interaction between c-Fos and MyoD1, the JASPAR database (https://jaspar.genereg.net/) was used to predict the binding of FOSL1 to MyoD1 promoter region, and found that two bZIP recognition sites in the MyoD1 promoter region had the most significant binding potential to FOSL1, of which motif 1 located in the -1839 to -1830 region and motif 2 was located in the -1273 to -1264 region (Figure 5E-G). Subsequently, ChIP-qPCR confirmed that FOSL1 directly binds to these two bZIP recognition sites in the MyoD1 promoter region (Figure 5H-I), indicating that FOSL1 plays an important role in the transcriptional regulation of MyoD1.

3.5 The MSTN^Del273C mutation with FGF5 knockout contribute to muscle phenotype via MEK-ERK-FOSL1 axis

As previously mentioned, FOSL1 may be a key gatekeeper of MSTN^Del273C mutation with FGF5 knockout-mediated muscle phenotype. Given that, we also investigated the protein levels of FOSL1 and c-Fos in GM and MD2, respectively. The results showed that the protein level of FOSL1 was significantly (P<0.05) reduced and c-Fos protein levels were significantly (P<0.05) decreased in MF^+/− cells at GM compared with WT cells (Figure 6A-B), whereas FOSL1 protein levels were significantly (P<0.05) diminished and c-Fos protein levels were highly significantly (P<0.01) elevated in MF^+/− cells after induced differentiation (Figure 6F-G), which further demonstrated the key role of FOSL1 on myogenesis. As demonstrated previously, enrichment analysis significantly enriched the MAPK signaling pathway. Compared with WT cells, the p-FOSL1 protein level of MF^+/− cells was strongly increased at GM (P<0.01) (Figure 6A-B), and there was no significant (P>0.05) difference between MEK1/2 and p-MEK1/2 protein levels (Figure 6A, C), but its downstream ERK1/2 protein level was extremely significantly (P<0.01) decreased, and accompanied by a significant (P<0.05) increase in p-ERK1/2 protein levels (Figure 6A, D). In addition, there was no significant (P>0.05) difference in p38 MAPK protein level, but p-p38 MAPK protein level was significantly (P<0.05) enhanced (Figure 6A, E). After induced differentiation, there was no significant (P>0.05) difference in p-FOSL1 protein level in MF^+/− cells compared with WT cells (Figure 6F-G), although both MEK1/2 and ERK1/2 protein levels were dramatically (P<0.01) inhibited, there was no significant (P>0.05) difference in their phosphorylated protein levels (Figure 6F, H-I). In addition, with a strong (P<0.01) decrease in the p38 MAPK protein level, the p-p38 MAPK protein level dramatically (P<0.01) increased (Figure 6F, J). These results suggested that MSTN^Del273C mutation with FGF5 knockout may regulate the expression and activity of FOSL1 via the MEK1/2-ERK1/2-FOSL1 axis to affect the proliferation and myogenic differentiation of skeletal muscle satellite cells.

Figure 6

To investigate the role of FOSL1 on the proliferation and myogenic differentiation of skeletal muscle satellite cells, we successfully constructed FOSL1 gain-of-function model (Figure 7A, G-H). Our results showed that overexpression of FOSL1 significantly (P<0.01) increased the cell proliferation rate (Figure 7B), and the mRNA expression levels of cell proliferation-related marker genes also significantly (P<0.05) increased (Figure 7C). Meanwhile, the EdU cell proliferation assay further demonstrated the promoting effect of FOSL1 overexpression on cell proliferation (Figure 7D-E). Similarly, we also investigated the expression of c-Fos and MyoD1, respectively. As anticipated, the overexpression of FOSL1 inhibited the MyoD1 mRNA (P<0.01) and protein (P<0.05) expression levels (Figure 7F, G-H), and significantly (P<0.05) suppressed mRNA expression level of c-Fos (Figure 7F), and we also observed a significant (P<0.01) increase in the expression level of p-FOSL1 protein (Figure 7G-H). These results are consistent with what we observed in MF^+/− cells at GM, suggesting a potential inhibitory effect of p-FOSL1 protein levels on MyoD1.

Figure 7

In addition, we also investigated the effect of FOSL1over-expression on cell differentiation. Although the mRNA expression level of MyoG was significantly increased (P<0.01) with an up-regulated tendency on MyoD1 mRNA, the expression level of MyHC mRNA was significantly (P<0.05) reduced after FOSL1 overexpression (Figure 8A). More importantly, the protein expression levels of MyoD1, MyoG and MyHC were all significantly decreased (P<0.05), proving that the overexpression of FOSL1 inhibited the myogenic differentiation of skeletal muscle satellite cells. Meanwhile, the p-FOSL1 protein expression level was also significantly (P<0.01) elevated with the increase of FOSL1 protein expression level (Figure 8B-C). Subsequently, immunofluorescence staining further confirmed the significant (P<0.05) inhibitory effect of FOSL1 overexpression on cell differentiation (Figure 8D-E). Also, the number of myotubes, the number of nuclei per myotube, and the myotube diameter all significantly decreased (P<0.05) (Figure 8D, F-H). These results further demonstrated that elevated p-FOSL1 protein level inhibits myogenic differentiation and produces smaller myotubes.

Figure 8

To further ascertain this insight, the tert-butylhydroquinone (TBHQ), which can strongly activate ERK1/2 and increase p-ERK1/2 protein expression level, was used to activate ERK1/2 and act as an indirect activator of FOSL1. As expected, the addition of 20 μM TBHQ significantly (P<0.01) inhibited the myogenic differentiation of skeletal muscle satellite cells (Figure 8I-J). And, the number of myotubes was significantly increased (P<0.05) (Figure 8I, K), while the number of nuclei per myotube was significantly (P<0.05) decreased and produced a smaller myotube diameter (P<0.05) (Figure 8I, L-M). Taken together, these results repeatedly confirm that FOSL1 as a key gatekeeper of MSTN^Del273C mutation with FGF5 knockout-mediated muscle phenotype.

In short, our results shed light that the MSTN^Del273C mutation with FGF5 knockout mediated the activation of FOSL1 via MEK-ERK-FOSL1 axis, further promotes skeletal muscle satellite cell proliferation, and inhibits myogenic differentiation by inhibiting the transcription of MyoD1, and resulting in smaller myotubes (Figure 9). Our results demonstrate the potential mechanism for smaller muscle fibers of MSTN^Del273C mutation with FGF5 knockout sheep.

Figure 9

4 Discussion

4.1 Optimized Cas9 mRNA and sgRNA delivery ratio improves the efficiency of dual-gene biallelic homozygous mutations

The strategy for producing gene knockout animals by CRISPR/Cas9 gene editing system is usually to introduce the Cas9 mRNA and the sgRNA of the target gene into their prokaryotic embryos by microinjection. However, this “one-step” method often results in a “mosaic” of gene-edited offspring, that is, due to the fertilized egg will divide into multiple blastomeres successively many times, the editing ability and editing mode of the CRISPR/Cas9 system for each blastomere may be different, thereby resulting in the occurrence of chimeric mutant individuals carrying both the wild-type and mutant alleles (Wan et al., 2015). Such chimeric mutants have now been reported in gene knockout mice (Wang et al., 2013), rats (Bao et al., 2015), monkeys (Niu et al., 2014), pigs (Hai et al., 2014), sheep (Hongbing HAN, 2014), goats (Wang et al., 2015), rabbits (Lv et al., 2016), and humans (Wang and Yang, 2019) prepared by a “one-step” method using the CRISPR/Cas9 system. For studies involved in genetic phenotypes, chimeric gene knockout animals require further cross-breeding to obtain animals with a complete knockout of the target gene. Once required to generate multiple gene knockout animals, this time-consuming and laborious operation will become extremely difficult. Although many studies have been devoted to eliminating this widespread chimeric mutation (Sato et al., 2015; Sung et al., 2014; Kotani et al., 2015; Chen et al., 2015; Zhou et al., 2014; Tu et al., 2017; Wang et al., 2015), however, these optimizations did not bring about a significant improvement in the production efficiency of biallelic knockout animals. Here, we increased the delivery ratio of Cas9 mRNA to sgRNA from 1:2 to 1:10, which improve the efficiency of the homozygous mutation of the biallelic gene. This unprecedented optimization method not only improved the overall gene knockout efficiency, but also the obtained gene-edited offspring were all dual-gene biallelic mutation.

4.2 Phenotypes produced by MSTN mutations are mutation site-dependent

As mentioned previously, although MSTN mutations have been found to produce a “double-muscle” phenotype in multiple species, the microscopic phenotypes are different, and this difference is closely related to the mutation site and species types. In mice, the number of skeletal muscle fibers with MSTN gene knockout significantly increased by 86% (McPherron et al., 1997). A missense mutant MSTN only increased the number of mouse muscle fibers, while dominant negative MSTN resulted in increased muscle fiber cross-sectional area in mice, but not the number of muscle fibers (Nishi et al., 2002; Zhu et al., 2000). In addition, the use of MSTN neutralizing antibody on adult rats also resulted in an increased muscle fiber cross-sectional area (Haidet et al., 2008). In cattle, natural MSTN mutant Belgian Blue cattle had an increased number of muscle fibers and reduced muscle fiber diameter (Wegner et al., 2000). The muscle fiber cross-sectional area of longissimus dorsi and gluteus medius in sheep was significantly increased after a 4bp deletion of the first exon of MSTN (Zhiliang et al., 2004). In pigs, both the MSTN gene-edited Meishan and Hubei pigs showed a phenotype with increased muscle fiber density (Qian et al., 2015; Xu et al., 2013). Here, we prepared MSTN^Del273C mutation with FGF5 knockout sheep with 3-base pairs of AGC in the third exon of MSTN, which caused the deletion of cysteine at amino acid position 273. Its macroscopic phenotype is similar to that of the MSTN-edited sheep with the first exon knocked out 4-base pairs. Both of them showed an abnormally developed “double-muscle” phenotype of hip muscle, but the microscopic phenotype was exactly the opposite.

4.3 The phenotype of MSTN^Del273C mutation with FGF5 knockout sheep is only controlled by MSTN

Although we produced the MSTN^Del273C mutation with FGF5 knockout sheep, FGF5 is currently recognized to be an important regulator of hair growth and development and there is no direct evidence of a potential regulatory role of FGF5 on muscle development (Xu et al., 2020; Zhang et al., 2020; Higgins et al., 2014; Hebert et al., 1994). Furthermore, there is no evidence of the crosstalk between MSTN and FGF5. Importantly, we also did not find a significant effect on muscle fiber phenotype in muscle morphological analysis of FGF5 knockout alone sheep. These evidence support whether it is homozygous mutant F0 generation or heterozygous mutant F1 generation, the reduction of muscle fiber cross-sectional area of MSTN^Del273C mutation with FGF5 knockout sheep is only controlled by MSTN, whereas not FGF5. These results indicate that MSTN may control skeletal muscle weight from two relatively independent aspects regulating the number of muscle fibers during embryonic development and the muscle fiber size after birth.

4.4 The MSTN^Del273C mutation with FGF5 knockout activate FOSL1 and promote cell proliferation

The proliferation and differentiation of skeletal muscle satellite cells is a key step in myogenesis and muscle development, which is a highly coordinated multistep biological process driven by many myogenic regulatory factors such as paired box families (Pax3/7), myogenic regulatory factors (Myogenin, MyoD, Myf5, and MRF4/6), myocyte enhancer factor 2 (MEF2) family proteins, these factors collectively regulate the expression of muscle specific genes and to control skeletal muscle development (Braun and Gautel, 2011). Aside from myogenic regulatory factor, MSTN has been repeatedly demonstrated to be involved in the proliferation (Thomas et al., 2000b; Taylor et al., 2001; Huang et al., 2007; Ge et al., 2020) and differentiation (Langley et al., 2002; Gao et al., 2020) of skeletal muscle satellite cells. MSTN negatively regulates the G1/S phase transition of cell cycle by specifically up-regulating cyclin dependent kinase inhibitors p21WAF1/CIP1, and reducing the level and activity of Cyclin-dependent kinase 2 (CDK2) protein in myoblasts (Thomas et al., 2000a; Joulia et al., 2003; McCroskery et al., 2003), resulting in the arrest of myoblasts in G1 phase of cell cycle, so as to maintain the static state of satellite cells. In this study, the MSTN^Del273C mutation with FGF5 knockout promoted the transformation of G1/S phase by reducing the proportion of cell cycle G1 phase and increasing the proportion of S phase, resulting in the activation of skeletal muscle satellite cells and entry into the cell cycle, and further promoting cell proliferation. Furthermore, FOSL1 has been repeatedly proved to promote the proliferation of a variety of cells, especially tumor cells (Sobolev et al., 2022; Talotta et al., 2020). In the current study, the MSTN^Del273C mutation with FGF5 knockout led to the increase of p-FOSL1 level, and the over-expression of FOSL1 also promoted cell proliferation, suggesting that the activated FOSL1 is a key factor in the proliferation of skeletal muscle satellite cells. Altogether, our results support that the MSTN^Del273C mutation with FGF5 knockout promotes the proliferation of skeletal muscle satellite cells by activating FOSL1.

4.5 FOSL1 binds to the MyoD1 promoter and inhibits its transcription

In this study, AP-1 family member FOSL1 was significantly reduced in MF^+/− sheep, and its expression were drastically reduced during myogenic differentiation, which was consistent with the decrease of FOSL1 expression during C2C12 differentiation (Tobin et al., 2016). Therefore, FOSL1 was recognized as a potential gatekeeper. As mentioned previously, c-Fos, a member of the AP-1 transcription factor family, has been shown to inhibit myogenesis and MyoD1 expression by directly binding to the MyoD1 promoter region (Li et al., 1992). Moreover, FOSL1 heterodimerizes with other transcription factors, such as the members of the bZIP family, and these dimers are either disabling the transcriptional activator complex or saving the interacting proteins from degradation in proteasomes (Sobolev et al., 2022). Therefore, we speculate that FOSL1 may have similar functions to c-Fos. PPI analysis of FOSL1, c-Fos and MyoD1 suggested a potential interaction between FOSL1 and MyoD1. Subsequently, we confirmed that FOSL1 directly binds to two bZIP recognition sites in the MyoD1 promoter region. Meanwhile, the overexpression of FOSL1 confirmed the potential inhibitory effect of FOSL1 and p-FOSL1 on MyoD1. In addition, FOSL2, another AP-1 family member, can also inhibit myoblast differentiation (Alli et al., 2013), which may support the inhibitory effect of FOSL1 on myogenic differentiation. In a word, these results fully support our hypothesis that FOSL1 binds the MyoD1 promoter and inhibits its transcription.

4.6 The MSTN^Del273C mutation with FGF5 knockout contribute to muscle phenotype via MEK-ERK-FOSL1 axis

MSTN dimer first binds to ActRIIB, and then binds to ALK4/5 to form a complex. Smad2/3/4 enters the nucleus to regulate the expression of target genes, and different transcription factors bind to Smad2/3/4 complex, resulting in different functions of Smad signaling pathway (Chen et al., 2021b). The nonclassical pathway of MSTN involves PI3K/Akt/mTOR signaling pathway and MAPK signaling pathway, which mainly includes ERKs, JNKs and p38 MAPK (Huang et al., 2007; Gui et al., 2012). All of those pathways are involved in the signal transduction pathway of MSTN and mediate the transcription of MRFs (Myogenin, Myf5, MyoD), MuRF-1 and Atrogin-1, to regulate myogenic differentiation and skeletal muscle quality (Chen et al., 2021b).

MSTN induces muscle fiber hypertrophy prior to satellite cell activation (Wang and McPherron, 2012) and inhibits IGF-I-induced increase in myotube diameter through Akt signaling pathway (Morissette et al., 2009). Recently, a study on the downstream target gene Smad2 of MSTN showed that the knock-out of Smad2 expression in primary myoblasts did not affect the efficiency of myogenic differentiation, but produced smaller myotubes with reduced expression of the terminal differentiation marker myogenin. In turn, the overexpression of Smad2 stimulated the expression of myogenin and enhanced cell differentiation and fusion (Lamarche et al., 2021). In our study, the MSTN^Del273C mutation with FGF5 knockout resulted in the inhibition of myogenic differentiation of skeletal muscle satellite cells, and the number of myotubes and the myotube size were significantly reduced.

As previously described, the DEGs of gluteus medius RNA-seq were significantly enriched in the MAPK signaling pathway. A recent study on glioma showed that FOSL1 can be activated by the Ras-MEK1/2-ERK1/2 axis in MAPK signaling pathway (Marques et al., 2021). Similarly, the activated MEK1/2-ERK1/2 axis in aged skeletal muscle also activates FOSL1 and increases the abundance of FOSL1 and the trans-activation capacity of the Fos-Jun heterodimer (Mathes et al., 2021). In our study, the MSTN^Del273C mutation with FGF5 knockout regulates FOSL1 expression and activity through MEK1/2-ERK1/2-FOSL1 axis and activated FOSL1 further inhibits myogenic differentiation of skeletal muscle satellite cells, resulting in smaller myotube diameter. However, puzzlingly, despite the high expression of p-FOSL1 in MF^+/− myoblasts, it did not significantly inhibit the transcription of MyoD1, which may be related to a dramatic enhance in c-Fos, or there might be other signaling pathways regulating MyoD1 after MSTN^Del273C mutation with FGF5 knockout. Furthermore, it has been demonstrated that the inhibition of MEK1/2 using MEK1/2-specific inhibitor PD184352 can significantly down-regulate FOSL1 expression (Mathes et al., 2021). To further confirm our hypothesis, TBHQ was used to activate ERK1/2 and act as an indirect activator of FOSL1. Interestingly, activated FOSL1 markedly inhibited myogenic differentiation of skeletal muscle satellite cells, which also resulted in smaller myotubes, but significantly increased the number of myotubes. Taken together, these results shed light on the potential mechanisms by which MSTN^Del273C mutation with FGF5 knockout leads to increased myofiber numbers and decreased fiber cross-sectional area.

5 Conclusion

In this study, we found that increasing the delivery ratio of Cas9 mRNA to sgRNA can improve the efficiency of the homozygous mutation of the biallelic gene. Based on this, we generated a MSTN^Del273C mutation with FGF5 knockout sheep, a dual-gene biallelic homozygous mutant, which highlights a dominant “double-muscle” phenotype. Both F0 and F1 generation mutants highlight the excellent trait of high-yield meat and the more number of muscle fibers per unit area. Our results suggested the MSTN^Del273C mutation with FGF5 knockout mediated the activation of FOSL1 via MEK-ERK-FOSL1 axis, further promotes skeletal muscle satellite cell proliferation, and inhibits myogenic differentiation by inhibiting the transcription of MyoD1, and resulting in smaller myotubes. This supports the myofiber hyperplasia that more number of muscle fibers and smaller cross sectional area, caused by the MSTN^Del273C mutation with FGF5 knockout.

Data availability statement

The raw sequence data reported in this paper have been deposited in the Genome Sequence Archive (Genomics, Proteomics & Bioinformatics 2021) in National Genomics Data Center (Nucleic Acids Res 2022), China National Center for Bioinformation / Beijing Institute of Genomics, Chinese Academy of Sciences (GSA: CRA008539) that are publicly accessible at https://ngdc.cncb.ac.cn/gsa.

Ethics statement

All experiments were performed in accordance with relevant guidelines and adhere to the ARRIVE guidelines (https://arriveguidelines.org/) for the reporting of animal experiments. All sheep are raised in accordance with the national feeding standard NT/T815-2004. All procedures performed were consistent with the National Research Council Guide for the Care and Use of Laboratory Animals. All experimental animal protocols were approved and performed in accordance with the requirements of the Animal Care and Use Committee at China Agricultural University (AW02012202-1-3).

Competing financial interests

The authors declare that there are no competing financial interests.

Author contributions

MMC performed the majority of experiments, data analysis, and drafted the manuscript. YZ performed a part of experiments and revised the manuscript. XLX, SJW and ZML helped with data analysis. XSZ, JLZ and XFG were responsible for the management of the feeding plant, slaughtering, and collecting samples. YMY helped to process some biological information data. SYQ, GY, SQW, HXL and AWW helped to collect and organize original data. GSL led the prokaryotic injection and embryo transfer. YL prepared the gene editing sheep. KY and HBH participated in project management. KY, FHL and ZXL conceived the project, revised manuscript and final approval of manuscript. All authors read and approved the final manuscript.

Acknowledgements

This work was supported by National Key Research and Development Program-Key Projects (2021YFF1000704, 2021YFD1200900), Natural Science Foundation of China (32072722), and National Transgenic Creature Breeding Grand Project (2016zx08008-003).

Figure Legends

Figure S1

Figure S2

Figure S3

Figure S4

Figure S5