Fibroblast growth factor 21/extracellular signal-regulated kinase 1/2 pathway as a key link mediates the effects of low-protein diet on skeletal muscle fiber type transformation in pigs

Abstract

Low-protein (LP) diets play a pivotal role in inducing skeletal muscle remodeling toward an oxidative phenotype in pigs, the underlying targets and mechanisms remain uncertain. This study aimed to investigate whether fibroblast growth factor 21 (FGF21), an endocrine signal associated with low protein intake, mediates the effect of LP diets on myofiber type transition in pigs. Twenty-four healthy castrated weaned pigs (Landrace × Yorkshire) with an initial body weight (BW) of 7.25 ± 0.12 kg were assigned and fed one of two dietary treatments: a normal-protein (NP group) diet containing 19.78% crude protein (CP) or an LP diet containing 16.91% CP (LP group), over a 28-day trial. The LP diet was supplemented with crystalline amino acids to achieve an equal content of limiting amino acids. Additionally, half of the pigs in each dietary group were randomly administered an extracellular signal-regulated kinase 1/2 (ERK1/2) inhibitor (0.5 mg/kg BW, i.p.) weekly, while the remaining pigs received an equivalent dose of the vehicle (dimethyl sulfoxide). After blood sampling, all pigs were euthanized to collect liver and longissimus dorsi (LD) muscle samples. In vitro, we also explored the FGF21-induced myofiber conversion in porcine skeletal muscle satellite cells (PSCs), which were transfected with overexpression plasmid vector and ERK1/2 inhibitor. Results demonstrated that FGF21 was robustly increased by LP diets (P < 0.01), evidenced by elevated mRNA and protein expression levels of FGF21 in liver and LD muscle, as well as increased serum FGF21 concentration. Furthermore, pigs fed the LP diet exhibited a higher proportion of oxidative myofibers (P < 0.001), increased expression level of slow myosin heavy chain (MyHC) (P < 0.05) and myoglobin (P = 0.080), and enhanced succinate dehydrogenase activity (P < 0.01) in LD muscle. These effects were largely attenuated by ERK1/2 inhibitor administration. The activation of ERK1/2 and mTOR complex 1 (mTORC1) in the LD muscle of pigs fed LP diets was also suppressed by ERK1/2 inhibitor treatment (P < 0.05). In vitro, FGF21 overexpression increased slow MyHC protein expression, ERK1/2 phosphorylation and mTORC1 activity in PSCs (P < 0.05). Furthermore, inhibition of the ERK1/2 signaling significantly eliminated the FGF21-induced enhancement of oxidative myofiber expression and mTORC1 activity in PSCs (P < 0.05). Collectively, our findings suggest that the FGF21-ERK1/2 pathway serves as a key link mediator of LP diet-induced myofiber specification in pigs, and provide new insights into the understanding of the metabolic benefits of LP diets in pigs.

1. Introduction

Skeletal muscle, as the largest organ system in the body, has a significant ability to support physical movement, metabolic processes and overall health (Argilés et al., 2016). Generally, mammalian skeletal muscle fibers are grossly categorized as type Ⅰ, ⅡA, ⅡX and ⅡB based on their different myosin heavy chain (MyHC) isoform expressions, which differ in terms of structural properties, contractile patterns and metabolic functions (Schiaffino and Reggiani, 2011). Type Ⅰ myofibers (slow-twitch oxidative) have higher levels of mitochondria and oxidative enzymes than fast-twitch glycolytic type ⅡB myofibers. Type ⅡA myofibers (fast-twitch oxidative) are intermediate between Type Ⅰ and ⅡB myofibers, which are primarily referred to oxidative metabolism. Type ⅡX myofibers are more similar to glycolytic myofibers in contractile patterns and metabolic function (Pette and Staron, 2000; Schiaffino and Reggiani, 2011). The proportions of different myofibers have a high degree of plasticity, and they can transform each other in the event of internal or external stimuli such as pathology and nutrition. It should be noted that myofiber type specification is a vital factor directly related to muscle metabolic capacity and quality. For example, a higher proportion of glycolytic myofibers in skeletal muscle is positively associated with diabetes/obesity or associated comorbidities (Duan et al., 2017; Fisher et al., 2017). Conversely, individuals undergoing endurance exercise training may significantly increase oxidative myofibers with few glycolytic myofibers in skeletal muscle (Egan and Zierath, 2013). In livestock, the transformation of glycolytic myofibers to the oxidative myofibers plays a pivotal role in producing high-quality meat, especially in the early stages of development (Joo et al., 2013; Zhao et al., 2023).

It is well established that decreasing dietary crude protein (CP) level in diets contributes to animal husbandry sustainability, including reducing pressure on protein ingredient supply, decreasing nitrogen excretion, and attenuating intestinal damage, while not impairing growth performance compared to traditional diets (Brink et al., 2022; Duarte et al., 2024; Wang et al., 2018). Importantly, the quantity and quality of dietary protein is strongly linked to skeletal muscle development and metabolic capacity. Muscle rich in oxidative myofibers possess more superior flavor quality than that rich in glycolytic myofibers, which is partly attributed to the contents of free amino acids (AAs) (Li et al., 2018; Yan et al., 2023). It is reported that a low-protein (LP) diet affects muscle growth process and remodeling in pigs (Li et al., 2018; Zhou et al., 2021).LP (15% CP) diets with supplemental crystalline AAs could positively affect meat quality, likely through increasing the skeletal muscle type Ⅰ and/or ⅡA fiber expression in growing and finishing pigs (Li et al., 2018). The LP (12% CP) diet containing a serine-to-glycine ratio 1:2, decreased fast-MyHC protein expression, but increased slow-MyHC expression in skeletal muscle of pigs when compared to pigs fed the normal protein (NP, 16% CP) diet (Zhou et al., 2021). However, the targets and specific mechanism by which LP diets promote the formation of oxidative myofibers in pigs is still not well understood.

Fibroblast growth factor 21 (FGF21), a pivotal modulator of cell metabolism, plays a vital role in many physiological processes including glucose homeostasis, injury protection and metabolic diseases as a hepatokine, adipokine, and myokine (Fisher and Maratos-Flier, 2016). Further, it has been reported that FGF21 could facilitate the transition of muscle fibers towards oxidative phenotype in mice, which is currently being considered as a potential candidate for remodeling the myofiber type (Liu et al., 2017). In addition, as a downstream effector of FGF21, activation of extracellular signal-regulated kinase 1/2 (ERK1/2) is involved in FGF21-mediated multiple biological actions including cardioprotective (Zhang et al., 2015) and hepatoprotective (Huang et al., 2023) effects. Moreover, ERK1/2 is required for skeletal myoblast terminal differentiation, which acts in promoting the transition of myofibers from fast to slow (Boyer et al., 2019; Li and Johnson, 2006). Thus, the activation of ERK1/2 by FGF21 in skeletal muscle may be involved in FGF21-induced myofiber type switching. More importantly, FGF21 represents a vital endocrine signal involved the control of metabolic homeostasis by dietary protein restriction (Laeger et al., 2014). However, whether the improvement of muscle oxidative metabolisms by LP diets was mediated through the activation of FGF21-ERK1/2 signaling pathway in pigs remains largely unclear.

Therefore, we hypothesized that LP diets induce oxidative fiber-type switching in skeletal muscle via activation of FGF21-ERK1/2 signaling pathway. To evaluate our hypothesis, we conducted a study to investigate the effect of LP diets on muscle aerobic oxidative capacity and myofiber type composition in pigs and the involved regulatory role of the FGF21-ERK1/2 pathway. Additionally, we further explored the potential mechanism behind these effects by examining porcine skeletal muscle satellite cells (PSCs).

2. Materials and methods

2.1. Animal ethics statement

All experiments were in accordance with the guidelines of the Animal Care and Use Committee of Jiangxi Agricultural University (permit No. JXAULL-2021–035).

2.2. Experimental design and diets

Twenty-four healthy castrated Landrace × Yorkshire pigs at 24 days old with an initial average body weight (BW) of 7.25 ± 0.12 kg were randomly assigned into two dietary treatments (12 pigs per treatment), consisting of an NP diet containing 19.78% CP or a LP diet containing 16.91% CP generated by adjusting the content of soybean meal and puffed corn. Diets were isoenergetic and the limiting AAs (lysine, methionine, threonine, tryptophan, valine, leucine and isoleucine) were supplemented to meet the National Research Council (2012)-recommended requirements for weaned pigs (Table 1). Nutrient levels in the experimental diets were analyzed by the methods of AOAC (2005) for CP (method 954.01), dry matter (method 930.15), crude fat (method 920.39), crude fiber (method 962.09), ash (method 942.05), calcium (method 927.02), and phosphorus (method 965.17). Briefly, nitrogen was analyzed using the Kjeldahl nitrogen method with a Kjeldahl apparatus (Kjeltec 2100, Foss, Sweden), and CP was calculated as nitrogen × 6.25. The dry matter was determined by oven-drying for 24 h at 103 °C. Crude fat was extracted with petroleum ether after acid hydrolysis to recover saponified fat (Soxtec System HT Tecator, Hillerød, Denmark; 1047 Hydrolyzing Unit and 1043 Extraction Unit). Crude fiber was analyzed by acid/alkaline hydrolysis of insoluble residues. Ash concentration was determined by incineration in an electric muffle furnace (SX2-10–13, Shanghai Shiyan Electric Furnace Co., Ltd., Shanghai, China) at 550 °C for 6 h. The calcium content analyzed using potassium permanganate titration, and total phosphorus content was analyzed colorimetrically by the molybdo-vanadate method, described by AOAC (2005). The organic matter, digestible energy and available phosphorus, and standardized ileal digestible (SID) amino acid content was calculated by dry matter minus crude ash, using the database of NRC (2012), and multiplying the SID coefficients obtained from NRC (2012) database by the AA content of feed ingredients, respectively. Pigs were housed in single cages (1.0 m × 1.5 m) in temperature (26 ± 2 °C) and humidity-controlled conditions (60% ± 5%), and had ad libitum access to feed and water. During the trial stage, one-half of the pigs in each dietary treatment were randomly allotted to receive ERK1/2 inhibitor (PD184352) dissolved in dimethyl sulfoxide (DMSO; 0.5 mg/kg body weight, i.p.) weekly, whereas the remaining pigs were allotted to receive an equivalent dose of the vehicle only (DMSO). The dosage of ERK1/2 inhibitor in pigs was based on the previous studies indicating that the ERK1/2 signaling pathway in newborn pigs could be inhibited effectively by administration of 0.1 to 1.0 mg/kg ERK inhibitor (Armstead et al., 2010, 2014; Kovács et al., 2018). Thus, we chose an average injection dose of 0.5 mg/kg for our study. The two levels of the two main effects of diets and injection were combined to constitute four treatment groups: (1) NP alone, (2) NP + inhibitor (NP + IN), (3) LP alone, (4) LP + inhibitor (LP + IN). The experiment lasted for four weeks.

Table 1.

Composition and nutrients level of the experimental diets (as-fed basis, %).

Item	NP	LP	Nutrients level	NP	LP
Ingredients			Analyzed composition
Expanded corn	59.00	66.65	Crude protein	19.78	16.91
Dehulled soybean meal	19.00	10.80	Crude fat	5.39	5.21
Low protein whey powder	6.00	6.00	Crude fiber	2.53	2.11
Fish meal	4.00	4.00	Organic matter	85.04	85.17
Soybean protein concentrate	6.00	6.00	Calcium	0.71	0.70
Soybean oil	2.60	2.30	Total phosphorus	0.61	0.59
L-Lysine HCl, 98.5%	0.30	0.50	Calculated composition
DL-Metionine, 98.5%	0.10	0.16	DE, MJ/kg	15.40	15.28
L-Threonine, 97.5%	0.10	0.22	Available phosphorus	0.38	0.37
L-Tryptophan, 98.0%	0.02	0.06	SID Lysine	1.35	1.35
L-Valine, 98.5%	0.00	0.14	SID Methionine	0.42	0.44
Leucine, 98.5%	0.00	0.10	SID Threonine	0.81	0.81
Isoleucine, 98.5%	0.00	0.12	SID Tryptophan	0.22	0.22
Dicalcium phosphate	0.78	0.85	SID Valine	0.86	0.86
Limestone	0.70	0.70	SID Leucine	1.58	1.50
Choline chloride	0.10	0.10	SID Isoleucine	0.80	0.79
NaCl	0.30	0.30
Premix1	1.00	1.00
Total	100.00	100.00

NP = normal protein; LP = low-protein; DE = digestible energy; SID = standardized ileal digestible.

¹Supplied per kilogram of diet: vitamin A, 2200 IU; vitamin D₃, 220 IU; vitamin E, 16 IU; vitamin K₃, 0.50 mg; vitamin B₁, 1.00 mg; vitamin B₆, 7.00 mg; vitamin B₁₂, 0.0175 mg; riboflavin, 3.50 mg; biotin, 0.05 mg; folic acid, 0.3 mg; niacin, 30.00 mg; pantothenic acid, 10.00 mg; Zn, 100.00 mg; Mn, 4.00 mg; Fe, 100.00 mg; Cu, 5.0 mg; I, 0.14 mg; Se, 0.30 mg.

2.3. Sample collection and preparation

At the beginning and the end of the experiment, pigs were weighed individually and feed intake was recorded every week. The feed to gain ratio was calculated based on the feed intake and body weight gain. At the experiment end, blood from the pigs was collected via the anterior vena cava, and centrifuged at 3000 × g at 4 °C for 15 min to collect serum. After blood sampling, all pigs were euthanized with an intravenous injection of pentobarbital sodium according to a previous study (Zou et al., 2021). Then, the liver and longissimus dorsi (LD) muscle between the 10th and 11th rib were promptly collected, frozen in liquid nitrogen and stored at −80 °C until subsequent biochemical analysis. The LD muscle samples adjacent to the 10th rib were cut into a rectangle (about 1.5 cm × 0.6 cm × 0.6 cm) following the muscle fiber orientation, placed in a perforated tube, rapidly frozen by liquid nitrogen, and transferred to −80 °C for storage until subsequent histochemical analysis.

2.4. Histological analysis

Frozen LD muscle samples were sectioned (6 μm) with a freezing microtome (CM1850, Leica, Wetzlar, Germany) at −20 °C. The staining process of myofibers was performed as described previously (Cai et al., 2023). The myofibers were categorized as type Ⅰ, ⅡA and ⅡB based on their myosin adenosine triphosphatase (mATPase) activities. At least 3 visual fields of each sample section were randomly chosen and photographed using a Nikon microscope. We counted the total fiber number and each myofiber type number by Image-Pro Plus 6.0 (Media Cybernetics, MD, USA) and calculated the percentages of different types of myofibers.

2.5. Muscle metabolic enzyme and lactic acid content assay

Homogenates of LD muscle samples were prepared as described previously (Zou et al., 2016). Briefly, approximately 0.1 g LD muscle were weighed and ground into a homogenate in ice-cold 0.9% NaCl solution (1:9, wt/vol) using grinding beads. Then, the homogenates were centrifuged at 4 °C, 3500×g for 10 min, to obtain the supernatant. The protein concentration was determined using Pierce BCA Protein Assay kit (Thermo Scientific, Waltham, MA, USA). The activities of creatine kinase (CK), succinate dehydrogenase (SDH), lactate dehydrogenase (LDH) and lactic acid (LA) content in muscle homogenates were determined by using commercial kits (Nanjing Jiancheng Institute of Bioengineering, Nanjing, Jiangsu, China).

2.6. Isolation and culture of PSCs

Porcine skeletal muscle satellite cells were isolated from the LD muscle of 1-day-old Large White male piglets as previously described with some modifications (Lv et al., 2022). Briefly, LD muscle was minced and digested with 0.1% type Ⅱ collagenase (17101–015, Gibco, Grand Island, NY, USA) for 90 min at 37 °C, following digestion of 0.25% trypsin for 10 min. Digestion was stopped by DMEM/F12 (Gibco) medium containing 20% fetal bovine serum (FBS; Gibco). Cells were cultured in growth medium (DMEM/F12 containing 20% FBS and 1% penicillin-streptomycin) on Matrigel-coated cell culture plates at 37 °C and 5% CO₂. The fresh medium was replaced every two days. When the cells were grown to about 90% confluence, the medium was changed to the myogenic differentiation medium (DMEM/F12 containing 2% horse serum and 1% penicillin-streptomycin).

2.7. Cell transfection and treatments

The porcine pcDNA3.1-FGF21 plasmid (pcDNA3.1 is a mammalian expression vector) was constructed by Shanghai GenePharma Co. Ltd., China. For gene overexpression, we transfected PSCs at 70%-80 % confluence with 4 μg pcDNA3.1-FGF21 plasmid or empty pcDNA3.1 vector using 9 μL Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) in each well of a 6-well plate for 24 h as described by the manufacturer's instruction. Then, the PSCs were induced to myogenic differentiation for 6 days as described above, and the medium and cells were harvested for the following analysis. To investigate the mechanism, the FGF21-overexpression PSCs were also treated with ERK1/2 inhibitor (2 μmol/L, Selleck Chemicals, IN), and then induced to myogenic differentiation for 6 days. For RNA interference, small interfering RNAs (siRNAs) target FGF21 were synthesized by GenePharma. The siRNA sequences are shown in Table S1. When the PSCs grew to 50%-60% confluence, 160 μmol/L siRNA-FGF21 or its negative control (NC) were transfected into cells by 9 μL Lipofectamine 2000 (Invitrogen) in each well of a 6-well culture plate.

2.8. Cell immunofluorescence staining

The PSCs were seeded in 12-well culture plate and transfected and treated as described above. For isolated and differentiated PSCs fixed in 4% paraformaldehyde, the immunofluorescence staining was conducted as described previously with a fluorescence microscope (IX73-DP80, Olympus, Tokyo, Japan) (Dey et al., 2014). The primary immunofluorescence staining antibodies (paired box gene 7 [Pax7], MyHC, fast MyHC and slow MyHC) were purchased from Abcam. DAPI staining was used to visualize cell nuclei.

2.9. FGF21 protein quantification

Enzyme-linked immunosorbent was performed to determine FGF21 protein content in serum and cell medium by the assay kits (BioVendor R&D, Modrice, Czech Republic).

2.10. Quantitative real-time PCR (qRT-PCR) analysis

LD muscle samples were used to extract total RNA by TransZol Up Plus RNA Kit (TransGen Biotech, Beijing, China). The RNA concentration and purity was determined via a nucleic-acid/protein analyzer (Beckman Coulter DU800, Fullerton, CA, USA). Then, reverse transcription was conducted using the EasyScript One-Step gDNA Removal and cDNA Synthesis SuperMix Kit (TransGen Biotech). Table S2 lists the detailed primers sequences. The PCR reaction was processed on a CFX96 RT-PCR Detection System (Bio-Rad Laboratories, Hercules, CA, USA) (Zou et al., 2021). Target gene transcript levels were calculated using the 2^−ΔΔCt method with β-actin as an internal control (Livak and Schmittgen, 2001).

2.11. Immunoblotting analysis

Immunoblotting analysis was conducted as reported previously (Li et al., 2024b). Briefly, equal weight protein from LD samples were separated via 10% SDS-PAGE and then transferred to the PVDF membrane using a wet trans-blotting system. After blocking the membranes in 5% fat-free milk, the proteins were identified with primary antibodies. The primary antibodies used were as follows: antibodies against FGF21 (cat. no. ab171941), slow-MyHC (cat. no. ab11083) and fast-MyHC (cat. no. ab91506) were purchased from Abcam (Cambridge, MA, USA) and were diluted 1:1000; antibodies against ERK1/2 (cat. no. 9102), phospho-ERK1/2 (p-ERK1/2) (cat. no. 9101), mammalian target of rapamycin (mTOR) (cat. no. 2972), phospho-mTOR (p-mTOR) (cat. no. 2971), ribosomal protein S6 (S6) (cat. no. 2217) and phospho-S6 (cat. no. 4858) were obtained from Cell Signaling Technology (Danvers, MA, USA) and were diluted 1:1000; antibodies against β-actin (cat. no. 81115-1-RR), Myoglobin (cat. no. 16048-1-AP), and peroxisome proliferator-activated receptor-γ coactivator-1α (PGC-1α) (cat. no. 66369-1-lg) were obtained from Proteintech Group (Wuhan, China) and were diluted 1:1000; antibodies against Raptor (cat. no. sc-81537) and phospho-Raptor (p-Raptor) (cat. no. 12778–2) were purchased from Santa Cruz Biotechnology (Dallas, TX, USA) and Signalway Antibody (Greenbelt, MD, USA), respectively, and were diluted 1:1000. The correspondence HRP-linked secondary antibodies (cat. no. SA00001–1 and SA00001–2, Proteintech Group) were diluted 1:5000. Bands were visualized and quantified with Bio-Rad ChemiDoc imaging systems (Bio-Rad).

2.12. Statistical analysis

The Shapiro–Wilk test indicated that all data sets followed a normal distribution. The data were analyzed using IBM SPSS 22.0 (SPSS, Inc., New York, NY, USA). The General Linear Model procedure via a 2 × 2 factorial treatment arrangement with the dietary treatment and ERK1/2 inhibitor being the main factors was used for multiple-group comparisons. The model was as follows:

$$Y_{ijk}=\mu +A_i+B_j+{\left(AB\right)}_{ij}+e_{ijk}\text{,}$$

where Y_ijk is the dependent variable, μ is the mean, A_i is fixed treatment effect (i = NP or LP for pigs, and i = pcDNA3.1 empty vector or pcDNA3.1-FGF21 plasmid for PSCs), B_j is ERK1/2 inhibitor effect (j = yes or no), AB_ij is the interaction effect between fixed treatment and ERK1/2 inhibitor, e_ijk is the random error. The Duncan method was used to compare the mean values among treatments. The two groups of measurements were analysed using an Unpaired Student's t-test. Results are showed as means and standard errors of the mean (SEM). P ≤ 0.05 was considered as significant difference and 0.05 < P ≤ 0.10 as a tendency.

3. Results

3.1. Myofiber-type composition in pigs was altered by LP diet and ERK1/2 inhibitor administration

As presented in Table 2, there was no significant difference in growth performance among the treatments (P > 0.05), suggesting the LP diet and ERK1/2 inhibitor administration did not have negative effects on pigs.

Table 2.

Effects of low-protein diet and ERK1/2 inhibitor administration on the growth performance of weaned piglets.¹

Item	NP2	NP + IN3	LP4	LP + IN5	SEM	P-value
						Protein	Inhibitor	Protein × Inhibitor
Initial BW, kg	7.33	7.19	7.19	7.28	0.125	0.914	0.920	0.651
Final BW, kg	16.02	16.22	16.18	15.62	0.276	0.698	0.748	0.508
ADFI, g	438	436	459	424	10.6	0.820	0.419	0.466
ADG, g	310	332	324	304	6.6	0.608	0.958	0.133
F/G	1.42	1.32	1.41	1.39	0.033	0.585	0.383	0.557

ERK1/2 = extracellular signal-regulated kinase 1/2; SEM = standard error of the mean; BW = body weight; ADFI = average daily food intake; ADG = average daily gain; F/G = feed to gain ratio.

¹n = 6 replicates per treatment.

²NP: pigs fed a normal-protein diet were administered the DMSO (control, i.p.) on a weekly basis.

³NP + IN: pigs fed a normal-protein diet were administered the ERK1/2 inhibitor (0.5 mg/kg body weight, i.p.) on a weekly basis.

⁴LP: pigs fed a low-protein diet were administered the DMSO (control, i.p.) on a weekly basis.

⁵LP + IN: pigs fed a low-protein diet were administered the ERK1/2 inhibitor (0.5 mg/kg body weight, i.p.) on a weekly basis.

The analysis of skeletal muscle characteristics further showed that the LP diet markedly increased the percentage of myofiber type Ⅰ (P < 0.001), and reduced the percentage of myofiber type Ⅱb (P = 0.010) in LD muscles of pigs, regardless of ERK1/2 inhibitor treatment. The increased percentage of myofiber type Ⅰ in LP-fed pigs was largely prevented by ERK1/2 inhibitor treatment (P < 0.001). Moreover, diets and ERK1/2 inhibitor had a significant interaction effect on the proportion of myofiber type Ⅰ (P = 0.022) (Table 3 and Fig. 1A). No difference was observed in muscle fiber density (P > 0.05) (Table 3). Next, we examined the effect of the LP diet and ERK1/2 inhibitor on the expression of myofiber-specific genes MyHC-7, MyHC-2, MyHC-1, and MyHC-4, which correspond to MyHC Ⅰ, MyHC ⅡA, MyHC ⅡX, and MyHC ⅡB, respectively. The myogenesis gene MyoG (P = 0.005) and slow oxidative myofiber-specific gene MyHC-7 (P = 0.002) were upregulated, while the fast glycolytic myofiber-specific gene MyHC-4 (P = 0.009) was downregulated in LD muscles of LP-fed group. ERK1/2 inhibitor treatment of pigs fed LP diets decreased the MyoG (P = 0.009) and MyHC-7 (P = 0.005) mRNA levels (Fig. 1B and C). In agreement, the LP-fed pigs had lower fast-MyHC protein expression (P = 0.001) and higher slow-MyHC (P = 0.016) and Myoglobin (P = 0.051) protein expression compared with the NP or NP + IN pigs, which were recovered by ERK1/2 inhibitor treatment (P < 0.05; Fig. 1D–G). These results suggest that ERK1/2 pathway may be involved in the control of slow oxidative myofiber-related marker expression activation involving dietary protein levels.

Table 3.

Effects of low-protein diet and ERK1/2 inhibitor administration on myofiber composition and metabolic enzyme activities in skeletal muscle of weaned piglets.¹

Item	NP2	NP + IN3	LP4	LP + IN5	SEM	P-value
						Protein	Inhibitor	Protein × Inhibitor
Myofiber density, mm2	3281	3603	3787	3969	118.4	0.069	0.280	0.762
Myofiber composition, %
Type I	9.40c	7.84c	17.76a	12.41b	0.866	<0.001	<0.001	0.022
Type Ⅱa	36.54	33.99	39.73	37.65	1.456	0.265	0.448	0.938
Type Ⅱb	53.43a	58.16a	42.51b	49.93ab	1.962	0.010	0.084	0.692
SDH, U/mg protein	5.77b	5.54b	7.44a	6.47ab	0.233	0.003	0.127	0.335
LDH, U/g protein	430.16a	435.35a	303.73b	344.64ab	20.531	0.007	0.531	0.627
CK, U/mg protein	3.89	3.41	3.67	3.96	0.129	0.542	0.717	0.156
LA, mmol/g protein	1.48a	1.53a	1.08b	1.34ab	0.057	0.005	0.114	0.250

ERK1/2 = extracellular signal-regulated kinase 1/2; SEM = standard error of the mean; SDH = succinate dehydrogenase; LDH = lactate dehydrogenase; CK = creatine kinase; LA = lactic acid.

Within a row, means with different letters are significantly different (P < 0.05).

¹n = 6 replicates per treatment.

²NP: pigs fed a normal-protein diet were administered the DMSO (control, i.p.) on a weekly basis.

³NP + IN: pigs fed a normal-protein diet were administered the ERK1/2 inhibitor (0.5 mg/kg body weight, i.p.) on a weekly basis.

⁴LP: pigs fed a low-protein diet were administered the DMSO (control, i.p.) on a weekly basis.

⁵LP + IN: pigs fed a low-protein diet were administered the ERK1/2 inhibitor (0.5 mg/kg body weight, i.p.) on a weekly basis.

Fig. 1.

Effects of low-protein diet and ERK1/2 inhibitor administration on muscle fiber-type composition in LD muscle of pigs. (A) Representative ATPase staining in LD muscle sections. The darkest staining represents type Ⅰ fibers, the lightest staining represents type Ⅱb fibers and the middle staining is type Ⅱa fibers. (B) Relative mRNA expression of muscle growth-related genes. (C) Relative mRNA expression of myofibers-specific genes MyHC-7, MyHC-2, MyHC-1 and MyHC-4. (D-G) Representative images of immunoblotting (D) and relative protein contents of fast-MyHC (E), slow-MyHC (F) and myoglobin (G). n = 6 replicates per treatment. Mean values with different letters are significantly different (P < 0.05). ERK1/2 = extracellular signal-regulated kinase 1/2; LD = longissimus dorsi; Pax7 = paired box gene 7; MyoD = myogenic differentiation factor; MyoG = myogenin; Myf5 = myogenic factor 5; MyHC = myosin heavy chain. Treatment groups: NP means pigs fed a normal-protein diet were administered the DMSO (control, i.p.) on a weekly basis; NP + IN means pigs fed a normal-protein diet were administered the ERK1/2 inhibitor (0.5 mg/kg body weight, i.p.) on a weekly basis; LP means pigs fed a low-protein diet were administered the DMSO (control, i.p.) on a weekly basis; LP + IN means pigs fed a low-protein diet were administered the ERK1/2 inhibitor (0.5 mg/kg body weight, i.p.) on a weekly basis.

3.2. Regulation of muscle oxidative capacity by LP diet and ERK1/2 inhibitor administration in pigs

The increased slow oxidative myofibers is usually associated with an improvement in muscle oxidative capacity (Li et al., 2024a). As shown in Table 3, the pigs fed the LP diets exhibited higher SDH activity (P = 0.003), a hallmark of oxidative metabolism, and lower LDH activity (P = 0.007), a hallmark of glycolytic metabolism, as well as the LA content (P = 0.005) in LD muscles compared to pigs fed the NP diets (P < 0.01), regardless of ERK1/2 inhibitor injection. Moreover, administration of ERK1/2 inhibitor tended to decrease the SDH activity (P = 0.083), and increase the LA content (P = 0.058) in pigs fed the LP diets. There was no difference in muscle CK activity among the treatments. These results indicate that the LP diet and ERK1/2 signaling have major roles in regulating muscle oxidative capacity.

3.3. FGF21 is robustly increased by the LP diet in pigs

Taking into account FGF21 is identified as a key endocrine signal of low protein intake that is linked to the myogenic differentiation program (Laeger et al., 2014; Liu et al., 2017), we further explored the effect of the LP diet on the production and circulation of FGF21 in pigs. Pigs fed the LP diets had higher hepatic FGF21 gene expression (P = 0.004) and serum FGF21 levels (P < 0.001) than those in pigs fed the NP diets (Fig. 2A and B). Moreover, the hepatic FGF21 protein levels were also increased in the LP-fed pigs (P < 0.001) (Fig. 2C and D). A similar pattern of FGF21 expression was noticed in LD muscle of LP-fed pigs (Fig. 2A–C and E).

Fig. 2.

FGF21 was rapidly and robustly induced in liver and LD muscle of pigs fed low-protein diet. (A) Relative mRNA expression of FGF21 in liver and LD muscle. (B) Serum FGF21 content. (C-E) Representative images of immunoblotting (C) and relative protein contents of FGF21 in liver (D) and LD muscle (E). n = 6 replicates per treatment. Mean values with different letters are significantly different (P < 0.05). FGF21 = fibroblast growth factor 21; LD = longissimus dorsi. Treatment groups: NP means pigs fed a normal-protein diet were administered the DMSO (control, i.p.) on a weekly basis; NP + IN means pigs fed a normal-protein diet were administered the ERK1/2 inhibitor (0.5 mg/kg body weight, i.p.) on a weekly basis; LP means pigs fed a low-protein diet were administered the DMSO (control, i.p.) on a weekly basis; LP + IN means pigs fed a low-protein diet were administered the ERK1/2 inhibitor (0.5 mg/kg body weight, i.p.) on a weekly basis.

3.4. The LP diet activates ERK1/2 signaling pathway in skeletal muscle of pigs

The ERK1/2 signaling pathway was activated by FGF21 in myoblasts (Ma et al., 2023), thus the ERK1/2 signaling-related molecules were further determined. As expected, ERK1/2 inhibitor injection significantly decreased the ERK1/2 phosphorylation level in LD muscle (P < 0.001) regardless of dietary protein levels. Additionally, there was a significant increase in muscle ERK1/2 phosphorylation level in LP-fed pigs compared with NP groups (P = 0.021) (Fig. 3A and B). The phosphorylation levels of mTOR, and its associated protein Raptor, in LD muscle were significantly greater in LP-fed pigs when compared with the NP-fed pigs (P < 0.05), which were recovered following ERK1/2 inhibitor administration (Fig. 3A–C and D). Similar effects were observed for S6 phosphorylation, a downstream target of mTOR (Fig. 3A and E). Moreover, the increased PGC-1α protein level in LP-fed pigs, a pivotal regulator of oxidative muscle fiber phenotype, tended to be decreased by administration of ERK1/2 inhibitor (P = 0.073) (Fig. 3A and F). These results indicate the mediatory role of FGF21-ERK1/2 signaling pathway in LP diet-induced promotion of the oxidative myofiber expression in pigs.

Fig. 3.

The low-protein diet activates ERK1/2-mTORC1 pathway in LD muscle of pigs. (A) Representative immunoblotting images of p-ERK1/2, ERK1/2, p-mTOR, mTOR, p-Raptor, Raptor, p-S6, S6 and PGC-1α. (B–F) Relative protein expression of the above proteins. n = 6 replicates per treatment. Mean values with different letters are significantly different (P < 0.05). ERK1/2 = extracellular signal-regulated kinase 1/2; mTOR = mammalian target of rapamycin; LD = longissimus dorsi; S6 = ribosomal protein S6; PGC-1α = peroxisome proliferator-activated receptor-γ coactivator-1α. Treatment groups: NP means pigs fed a normal-protein diet were administered the DMSO (control, i.p.) on a weekly basis; NP + IN means pigs fed a normal-protein diet were administered the ERK1/2 inhibitor (0.5 mg/kg body weight, i.p.) on a weekly basis; LP means pigs fed a low-protein diet were administered the DMSO (control, i.p.) on a weekly basis; LP + IN means pigs fed a low-protein diet were administered the ERK1/2 inhibitor (0.5 mg/kg body weight, i.p.) on a weekly basis.

3.5. FGF21 promoted the formation of oxidative myofiber in PSCs

Porcine skeletal muscle satellite cells were isolated to gain more insight into the role of FGF21 in controlling muscle fiber-type specification in vitro. For the isolated and differentiated cells, Pax7 (skeletal muscle satellite cell marker) and MyHC (myogenic differentiation marker) immunofluorescence staining showed Pax7⁺ cells and MyHC⁺ cells accounted for 98% of total cells, suggesting that the cells were skeletal muscle satellite cells (Fig. S1). Next, we investigated the FGF21 gene expression pattern during myogenesis. There was a remarkable increase in FGF21 mRNA levels during PSCs myogenic differentiation, which indicates an extraordinary role of FGF21 in regulating the myogenic differentiation (Fig. 4A). Porcine skeletal muscle satellite cells were transiently transfected with the pcDNA3.1-FGF21 plasmid vector to overexpress the FGF21 gene. As expected, the expression of FGF21 at both mRNA (P < 0.001) (Fig. 4B) and protein levels (P < 0.001) in cells (Fig. 4C and D) and FGF21 protein concentration (P < 0.001) in culture medium (Fig. 4E) were sharply increased in response to the overexpression. Overexpression of FGF21 increased the MyHC-7 (P = 0.008) and MyHC-2 (P = 0.014) mRNA expression, whereas decreased MyHC-4 mRNA level (P < 0.030) in differentiated PSCs (Fig. 4F). Further immunofluorescence staining indicated that overexpression of FGF21 significantly decreased fast MyHC-positive cells, and increased slow MyHC-positive cells (Fig. 4G and H). Consistently, we acquired the same results in protein expression analysis (Fig. 4I–K). Furthermore, overexpression of FGF21 significantly increased the phosphorylation levels of ERK1/2 (P = 0.027), Raptor (P = 0.033) and S6 (P = 0.044), and tended to increase the phosphorylation level of mTOR (P = 0.090) in differentiated PSCs (Fig. 5A–E). FGF21 overexpression also increased the PGC-1α protein level (Fig. 5F) (P = 0.001). These results demonstrate that the ERK1/2 pathway may be involved in the FGF21-induced promotion of the transition of myofiber from glycolytic to an oxidative phenotype.

Fig. 4.

FGF21 is expressed in differentiated PSCs and promotes slow-oxidative myofibers program. (A) The relative mRNA expression of FGF21 during myogenic differentiation in PSCs. (B-E) Relative mRNA (B) and protein (C and D) levels of FGF21 in PSCs and FGF21 protein content in culture medium (E) after transiently transfecting PSCs with pcDNA3.1-FGF21 or empty pcDNA3.1 vector (control group). (F–K) The PSCs transiently transfected with pcDNA3.1-FGF21 plasmid or empty pcDNA3.1 vector were induced to myogenic differentiation for 6 days. The relative mRNA expression of MyHC-7, MyHC-2, MyHC-1 and MyHC-4 (F). Representative photographs of fast-MyHC (G) and slow-MyHC (H) immunofluorescence staining. Representative images of immunoblotting (I) and relative protein contents of fast-MyHC (J) and slow-MyHC (K). n = 6 replicates per treatment. FGF21 = fibroblast growth factor 21; PSCs = porcine skeletal muscle satellite cells; MyHC = myosin heavy chain.

Fig. 5.

FGF21 overexpression activates the ERK1/2-mTORC1 pathway in differentiated PSCs. The PSCs transiently transfected with pcDNA3.1-FGF21 plasmid or empty pcDNA3.1 vector were induced to myogenic differentiation for 6 days. (A) Representative immunoblotting images. (B-E) Ratio of phosphorylated ERK1/2 to total ERK1/2 (B), phosphorylated mTOR to total mTOR (C), phosphorylated Raptor to total Raptor (D) and phosphorylated S6 to total S6 (E). (F) PGC-1α relative protein expression. n = 6 replicates per treatment. ERK1/2 = extracellular signal-regulated kinase 1/2; mTOR = mammalian target of rapamycin; S6 = ribosomal protein S6; PGC-1α = peroxisome proliferator-activated receptor-γ coactivator-1α; PSCs = porcine skeletal muscle satellite cells.

3.6. FGF21 knockdown attenuated the expression of oxidative myofiber marker genes in PSCs

Next, we further investigated whether FGF21 knockdown could affect the fate determination of PSCs. The PSCs were transfected with three different anti-FGF21 RNAi molecules to analyze the FGF21 knockdown efficiency. The siRNA-450 (P = 0.041) and siRNA-520 (P = 0.009) oligonucleotides significantly reduced the FGF21 mRNA levels in PSCs (Fig. 6A). Thus, we selected siRNA-520 for the follow-up study. As expected, FGF21 protein levels in culture medium and PSCs were sharply decreased compared to the negative control group (NC) (P < 0.01) (Fig. 6B–D). Knockdown of FGF21 reduced the MyHC-7 mRNA level (P = 0.047) and increased MyHC-4 mRNA level (P = 0.016) in differentiated PSCs (Fig. 6E). Furthermore, immunofluorescence staining showed that the percentage of fast MyHC-positive cells were increased, while the percentage of slow MyHC-positive cells were decreased in the FGF21-knockdown group (Fig. 6F and G). Similar changes in fast MyHC and slow MyHC protein expression were also observed (Fig. 6H–J). In addition, knockdown of FGF21 significantly decreased the phosphorylation levels of ERK1/2 (P < 0.010) and mTOR (P = 0.009), and its associated protein Raptor (P < 0.034) and downstream target S6 (P < 0.001) (Fig. 7A–E). The PGC-1α protein level was also decreased in the FGF21-knockdown group (P = 0.031) (Fig. 7F). These results demonstrate that FGF21 knockdown suppressed the expression of slow oxidative myofiber markers.

Fig. 6.

FGF21 knockdown attenuates slow-oxidative myofiber-associated genes expression. (A) Three different siRNA targeting FGF21 (siRNA-209, siRNA-450 and siRNA-520) were transiently transfected into PSCs, and the efficiency was determined by qPCR. (B-D) FGF21 protein content in medium (B) and PSCs (C and D) after transiently transfecting PSCs with siRNA-FGF21 or a scrambled siRNA as a negative control (NC). (E-J) The PSCs transiently transfected with siRNA-FGF21 or negative control were induced to myogenic differentiation for 6 days. Relative mRNA expression of MyHC-7, MyHC-2, MyHC-1 and MyHC-4 (E). Representative photographs of fast-MyHC (F) and slow-MyHC (G) immunofluorescence staining. Representative images of immunoblotting (H) and relative protein contents of fast-MyHC (I) and slow-MyHC (J). n = 6 replicates per treatment. FGF21 = fibroblast growth factor 21; MyHC = myosin heavy chain; PSCs = porcine skeletal muscle satellite cells.

Fig. 7.

The reduction of FGF21 weakens the ERK1/2-mTORC1 pathway in differentiated PSCs. The PSCs transiently transfected with siRNA-FGF21 or negative control were induced to myogenic differentiation for 6 days. (A) Representative immunoblotting images. (B-E) Ratio of phosphorylated ERK1/2 to total ERK1/2 (B), phosphorylated mTOR to total mTOR (C), phosphorylated Raptor to total Raptor (D) and phosphorylated S6 to total S6 (E). (F) PGC-1α relative protein expression. n = 6 replicates per treatment. ERK1/2 = extracellular signal-regulated kinase 1/2; mTOR = mammalian target of rapamycin; S6 = ribosomal protein S6; PGC-1α = peroxisome proliferator-activated receptor-γ coactivator-1α; PSCs = porcine skeletal muscle satellite cells.

3.7. ERK1/2 signaling is critical for the FGF21-induced increase in oxidative myofibers in PSCs

To further confirm that FGF21 promotes a switch to an oxidative myofiber-type phenotype via activation of the ERK1/2 pathway, we used PD184352, a MEK inhibitor that selectively blocks the ERK1/2 cascade, to assess the effects of ERK1/2 inhibition on oxidative myofiber marker expression in FGF21-overexpression PSCs. Expectedly, the FGF21 protein levels were significantly higher in pcDNA3.1-FGF21 and pcDNA3.1-FGF21 + IN groups than those in the corresponding control groups (P < 0.01) (Fig. 8A and B). It was noteworthy that the increased expression levels of MyHC-7 and MyHC-2 as well as the decreased expression level of MyHC-4 in the FGF21-overexpressed group were prevented by ERK1/2 inhibitor treatment (Fig. 8C). Moreover, compared to the pcDNA3.1-FGF21 group, the inhibitory effect of FGF21 on fast-MyHC protein expression was diminished in the pcDNA3.1-FGF21 + IN group (Fig. 8D and E). However, the stimulatory effect of FGF21 on slow-MyHC protein expression was abrogated by ERK1/2 inhibitor treatment (Fig. 8D and F). Consistently, we acquired the same results in immunofluorescence analysis (Fig. 8G and H). The phosphorylation levels of ERK1/2 and mTOR, and its associated protein Raptor as well as downstream target S6 were increased by the FGF21 overexpression but reduced by the ERK1/2 inhibitor treatment (Fig. 8I). These findings suggest that ERK1/2 mediates muscle fiber-type shift toward a slow-oxidative phenotype in FGF21-overexpression PSCs.

Fig. 8.

ERK1/2 signaling pathway is critical for the FGF21-induced increase in oxidative myofibers in PSCs. The PSCs transiently transfected with pcDNA3.1-FGF21 plasmid or empty pcDNA3.1 vector were treated with ERK1/2 inhibitor (2 μmol/L) and then induced to myogenic differentiation for 6 days. (A and B) Representative images of immunoblotting (A) and relative protein content of FGF21 (B). (C) Relative mRNA expression of MyHC-7, MyHC-2, MyHC-1 and MyHC-4. (D-F) Representative images of immunoblotting (D) and relative protein content of Fast-MyHC and (E) Slow-MyHC (F). (G and H) Representative photographs of fast-MyHC (G) and slow-MyHC (H) immunofluorescence staining. (I) Representative images of immunoblotting and the ratio of phosphorylated ERK1/2 to total ERK1/2, phosphorylated mTOR to total mTOR, phosphorylated Raptor to total Raptor and phosphorylated S6 to total S6. n = 6 replicates per treatment. Mean values with different letters are significantly different (P < 0.05). FGF21 = fibroblast growth factor 21; MyHC = myosin heavy chain. ERK1/2 = extracellular signal-regulated kinase 1/2; mTOR = mammalian target of rapamycin; S6 = ribosomal protein S6. PSCs = porcine skeletal muscle satellite cells.

4. Discussion

Skeletal muscle fibers show a large degree of structural and functional plasticity to adapt to stimuli such as endurance training, nutritional interventions, environmental factors, and metabolic diseases (Smith et al., 2023). Studies suggest that increasing the proportion of muscle oxidative myofibers exerts beneficial role in the prevention and management of metabolic disorders throughout the body (Duan et al., 2017). In pig production, the LP diet with balanced AAs is receiving considerable attention for its promising high-quality meat producing benefits, which is likely through an increase of oxidative myofibers-related marker expression (MyHC-7 and MyHC-2) in the muscle (Li et al., 2018; Zhou et al., 2021). The early stage of skeletal muscle development is a vital period for setting the long-term growth trajectory in later life. Thus, the underlying mechanism of how LP diets affect the formation of slow oxidative myofiber should be elucidated, especially during the early growth stage, aimed to provide a comprehensive understanding of stage-specific nutritional strategies for accelerating metabolic health and meat quality. In this study, we showed that the promotion effects of LP diets on myofiber switching towards an oxidative phenotype could be mediated by the FGF21-ERK1/2 signaling pathway in pigs, which may be a potential target for nutrition interventions aimed at optimizing skeletal muscle function and metabolism.

Dietary protein intake or protein status could be transmitted to the brain in a myriad of ways, most notably the hormones produced by liver and skeletal muscle which represent dominant sites of AAs catabolism and biosynthesis (Morrison and Laeger, 2015). Importantly, previous research has identified FGF21 as a novel protein-specific signal that responds to low protein intake, which plays an important role in coordinating the metabolism and growth during periods of reduced protein intake (Laeger et al., 2014). Additionally, recent study showed that FGF21 treatment increased the relative abundance of plasma proteinogenic AAs in female mice, and the topmost enriched metabolic pathways were related to AA metabolism (Larson et al., 2024). FGF21 acts as an endocrine hormone that functions on distinct target tissues, which displays a broad range of beneficial metabolic effects on hyperglycemia, fatty liver, energy expenditure, and insulin sensitivity (BonDurant et al., 2017; Flippo and Potthoff, 2021). Skeletal muscle is a key site of FGF21 action in the body, and several studies suggest that FGF21 exerts a pivotal physiological role in facilitating muscle fiber type transformation in mice (Liu et al., 2017; Luo et al., 2023). In this study, we found that FGF21 was not only rapidly and robustly induced in liver, but its expression in LD muscle was also markedly enhanced by LP diets in pigs. Additionally, our results showed that the LP diet promoted a transition of myofibers from the glycolytic to the oxidative phenotype in pigs, as evidenced by the increased oxidative myofibers marker (slow MyHC, myoglobin, and PGC-1α) expression, proportion of oxidative myofiber and the oxidized enzyme (SDH) activity, and the decreased glycolytic myofiber marker (fast MyHC) expression, as well as glycolytic enzyme (LDH) activity, which is similar with previous study (Li et al., 2018). These data collectively show that FGF21 may play a pivotal role in the LP-induced transition of myofiber types in pigs. In this study, to further explore the potential beneficial effect of FGF21 on oxidative myofiber expression, we examined whether FGF21 overexpression or knockdown affects the muscle fiber type transitions in PSCs. We found that overexpression of FGF21 significantly enhanced the PGC-1α protein expression, a marker of oxidative muscle fiber type characteristics, along with the increased slow MyHC protein expression, while FGF21 knockdown significantly reduced their expression. These results indicate that the FGF21 signaling is crucial in regulating the formation of slow oxidative myofibers.

It has been proven that FGF21 controls cell survival, proliferation, and differentiation by activating multiple cellular signaling cascades, including the mitogen-activated protein kinase (MAPK) axis (Fisher and Maratos-Flier, 2016). As a part of the MAPK, ERK1/2 is predominantly expressed in skeletal muscle, which is involved in regulating a myriad of cellular responses to extracellular stimuli, such as growth factors, hormones and cytokines (Michailovici et al., 2014). Here, we found that the LP diet-induced increased proportion of oxidative myofibers in pigs was largely prevented by ERK1/2 inhibitor administration, which suggests that the ERK1/2 pathway may participate in the effect of LP diet on oxidative myofibers expression. The activation of ERK1/2 by FGF21 in myoblasts acts in FGF21-mediated anti-myotube atrophy effects (Ma et al., 2023). Moreover, several compelling evidence in vivo and in vitro support an important role for the ERK1/2 pathway in slow oxidative muscle fiber specification (Boyer et al., 2019; Meissner et al., 2011). Muscle ERK1/2 sustained activation induces slow oxidative fiber-type switching that protects against muscular dystrophy in mice (Boyer et al., 2019). Activated ERK1/2 enhances the expression of the slow muscle fiber-specific gene MyHC Ⅰ/β in C2C12 myotubes (Meissner et al., 2011). In this study, to explore potential mechanism, we examined whether the induction of FGF21 on oxidative muscle fiber phenotype is mediated by activating ERK1/2 pathway in PSCs. Results suggested that overexpression of FGF21 enhanced the ERK1/2 phosphorylation, while knockdown of FGF21 suppressed its phosphorylation. Further, by inhibiting ERK1/2 in differentiated PSCs, we found that ERK1/2 inactivation suppressed the promotion effect of FGF21 on oxidative myofiber expression. These findings indicate that ERK1/2 pathway is pivotal in regulating FGF21's effects on the proportion of oxidative myofibers.

The mTOR acts as the catalytic subunit of mTOR complex 1 (mTORC1), and is pivotal in determining muscle fiber type and metabolic capacity (Cunningham et al., 2007; Zhou et al., 2024). Consistently, the present study showed higher activity of mTORC1 in the muscle of LP-fed pigs, as evidenced by the increased expression level of p-mTOR, p-Raptor (a scaffolding protein of mTORC1) and p-S6 (a downstream target of mTORC1), compared to the that of the NP group. Additionally, increasing evidence suggests that the positive effect of ERK1/2 on cellular metabolism is mainly mediated by activating its downstream mTORC1 pathway (Carriere et al., 2011; Zhou et al., 2016). Based on this, we speculate that the mTORC1 acts as a master mediator of FGF21-induced myofiber remodeling. The present study showed that overexpression of FGF21 enhanced the activity of mTORC1, while knockdown of FGF21 suppressed its activity. Further, we also found that ERK1/2 inactivation abolished the regulatory effects of FGF21 on mTORC1 activity. These findings suggest that ERK1/2-dependent activation of mTORC1 mediates FGF21-induced muscle fiber remodeling toward an oxidative phenotype. Additionally, it has been demonstrated that the mTORC1 acts to boost FGF21-mediated metabolic regulation in vivo (Minard et al., 2016). Considering that the beneficial effects of the mTORC1 in regulating skeletal muscle function and oxidative metabolism (Cunningham et al., 2007; You et al., 2021), it is possible that FGF21 can directly target and activate the mTORC1 pathway to exert beneficial roles in muscle fiber type composition, which needs to be studied further. Moreover, as the pivotal role of intestinal microbiota in dietary nutrients and the maintenance of host health (Fan et al., 2023; Wang et al., 2024), more interest will be focused on identifying whether intestinal microbiota mediate the FGF21 adaptive response to low protein intake in pigs.

5. Conclusion

In conclusion, our study revealed that activation of the FGF21-ERK1/2 signaling pathway is a key link mediating the effect of LP diets on myofiber switching toward an oxidative phenotype in pigs. These findings provide new evidence for the role of the FGF21-ERK1/2 signaling pathway in LP diet-induced muscle fiber type transitions, which could contribute to the development of nutritional solutions for maintaining skeletal muscle fitness and improving meat quality.

Credit Author Statement

Shuo Li: Writing – original draft, Methodology, Investigation. Pengbo Liang: Investigation. Bo Wang: Writing – review & editing. Jun Chen: Investigation. Jinming You: Writing – review & editing, Conceptualization. Tiande Zou: Writing – review & editing, Funding acquisition, Conceptualization.

Declaration of competing interest

We declare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our work, and there is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as influencing the content of this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article: