The experimental protocol involving caring, handling, and treatment of pigs was approved by the Mississippi State University Institutional Animal Care and Use Committee. Twenty crossbred young growing barrows (Yorkshire $\times$ Landrace; initial BW 8.1 $\pm$ 0.9 kg) purchased from a local commercial farm were transferred to an environment-controlled swine barn at the Leveck Animal Research Center of Mississippi Agricultural and Forestry Experiment Station. Upon arrival, pigs were randomly assigned to 5 feeding pens and fed a commercial nursery diet until their BW reached 23.6 $\pm$ 2.4 kg, during which time the pigs were allowed ad libitum access to the diet and fresh water. Pigs were then randomly assigned to 20 individual pens, and further randomly allotted to 2 dietary treatments according to a completely randomized experimental design with pig serving as experimental unit.

A corn and soybean meal based diet (Diet 1, a Met-restricted diet) was formulated to meet or exceed the recommended requirements for energy, crude protein, essential AA, minerals, and vitamins, except for Met [4, 15]. Diet 2 (a Met-adequate diet) was produced by supplementing a commercial product of crystalline DL-Met (99% purity; Evonik Operations GmbH, Hanau-Wolfgang, Germany) to Diet 1 at the expense of corn to meet the requirement of pigs for Met [4, 15]. The diet composition and the calculated nutrient contents are both shown in Table 1 (Ref. [16]), which demonstrates that Diet 2 was a Met-adequate diet while Diet 1 was deficient in SID Met by roughly 40.5%.

To confirm the contents of major nutrients, samples of the two diets were submitted to the Essig Animal Nutrition Laboratory at Mississippi Agricultural and Forestry Experiment Station for proximate and energy analyses, and to the Evonik’s chemical laboratory at Hanau-Wolfgang, Germany for AA analysis. For proximate analysis, the contents of dry matter, crude protein, crude fat, crude fiber, and ash were determined according to AOAC International (2000) [17] official methods 9340.01, 2001.11, 920.39, 92.09, and 924.05, respectively. Gross energy was determined using a Parr 1261 Isoperibol Bomb Calorimeter (Parr Instrument Company, Moline, IL, USA). Amino acids were analyzed using ion-exchange chromatography [18, 19]. Tryptophan was analyzed by high-performance liquid chromatography with fluorescence detection [20]. The determined compositions of selected nutrients contained in the diets are shown in Table 2 (Ref. [16]), which indicate that Diet 1 was a diet deficient in total Met by roughly 35.1%.

During the four-week feeding trial, pigs had ad libitum access to the experimental diets and fresh water. All feeders, waterers, and pigs were checked at least twice a day (0600 to 2100 hr) to ensure proper function of the facilities and healthy animal behavior. Feed refusals and spillage were collected and immediately returned to the feeders or reserved and weighed for feed intake calculation. Pigs’ BW were measured immediately before, and also at the end of, the four-week feeding trial. The average daily gain (ADG), average daily feed intake (ADFI), and gain to feed ratio (G:F) were then calculated accordingly.

Immediately before the beginning and at the end of the feeding trial, blood samples (approximately 10 mL/pig) were collected by jugular venipuncture of individual pigs [in a non-fasting state as in an industry setting where pigs are not fasted] in early morning (between 0600 and 0800 hr). The remaining feeds in all the feeders, however, were removed approximately 30 to 60 minutes before the blood collection. Blood samples were kept on ice immediately after the collection until plasma was separated within 30 to 60 minutes through centrifugation at 800 $\times$ g and 4 ${}^{\circ }$C for 16 min. Plasma samples in 500-$\mu$L aliquots were then stored at –80 ${}^{\circ }$C until laboratory analysis of nutrient metabolites including AA.

After blood collection, a muscle sample (about 200 mg/pig) was collected from the middle portion (the left side) of longissimus dorsi muscle of each pig using our standard aseptic biopsy protocol [21]. All muscle samples collected were snap frozen in liquid nitrogen, and then transferred to a –80 ${}^{\circ }$C freezer for storage until gene expression analyses.

The concentrations of plasma free AA were determined at the analytical laboratories of Ajinomoto Heartland, Inc. (Chicago, IL, USA) and Evonik Operations GmbH (Hanau-Wolfgang, Germany) using the official standard high-performance liquid chromatography methods [17]. The principles and procedures of the methods were briefly described by Regmi et al. [22] previously.

Batch analysis using the automated ACE Alera Clinical Chemistry System (Alfa Wassermann, West Caldwell, NJ, USA) was performed at the College of Veterinary Medicine Diagnostic Laboratory of Mississippi State University for determination of the concentrations of six representative plasma metabolites with six respective ACE reagents (Alfa Wassermann), and these six metabolites are urea nitrogen, albumin, total protein, glucose, triglycerides, and total cholesterol. The principles and procedures of these laboratory analyses were briefly described by Regmi et al. [23] previously.

The myogenic gene expression was analyzed by following our previously reported protocols [24]. Briefly, the total RNA was extracted from approximately 50 mg of muscle sample per pig using TRIzol Reagent (Invitrogen Corporation, Carlsbad, CA, USA) following the manufacturer’s instructions. Briefly, a frozen sample was homogenized in a 15-mL polypropylene centrifuge tube using a Polytron mixer (0.5 mL TRIzol per 50 mg tissue), and the homogenate was transferred to a 1.5-mL micro-centrifuge tube. Chloroform (400 $\mu$L/tube) was used to separate RNA from DNA and proteins, and then the total RNA was precipitated with isopropyl alcohol (at a ratio of 1:1) and washed with 750 $\mu$L of 75% ethanol. The resulted RNA was air-dried, dissolved in 60 $\mu$L RNase-free water, and stored in a freezer at –80 ${}^{\circ }$C. The purity and concentration of the total RNA samples were checked by using an ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA).

First-strand cDNA was reverse-transcribed from 1 $\mu$g of total RNA by using QuantiTect Reverse Transcription Kit (Qiagen, Valencia, CA, USA). The semi-quantitative polymerase chain reaction (PCR) analysis was performed using the Rotor-Gene SYBR Green PCR Kit with the Rotor-Gene Q System (Qiagen), followed by melting curve analysis to verify the specificity and identity of the PCR products [24]. The thermal cycling parameters were 95 ${}^{\circ }$C for 5 min, followed by 40 cycles of 95 ${}^{\circ }$C for 5 s and 60 ${}^{\circ }$C for 10 s. Primers for the selected myogenesis-related genes, as well the hypoxanthine phosphoribosyl transferase 1 (Hprft1; as the endogenous control) gene, were manufactured at Integrated DNA Technologies (Coralville, IA, USA) with the primer sequences being adopted from Yang et al. [24]. The Hprft1 gene was used for normalization of the potential variations caused during the sample preparation [25].

The comparative $\mathrm{\Delta }\mathrm{\Delta }$C${}_{\text{T}}$ method was used for mRNA quantity calculation. Briefly, the raw quantity of a given gene was normalized against the raw quantity of Hprft1 reference gene of a given sample obtained from the Rotor-Gene Q System, and then the normalized level of the given gene of each sample in the Diet 1 group was expressed as a quantity relative to the mean of the normalized quantities of the given gene in the Diet 2 samples [24].

Data were subjected to statistical analysis with Student t-test using the SAS software (version 9.4; SAS Institute Inc., Cary, NC) with pigs being the experimental units. A P-value less than or equal to 0.05 was considered as having a significant difference between treatment means, and a P-value between 0.05 and 0.10 as having a tendency to be different. Each value of the measurements is presented as mean $\pm$ standard deviation (SD).

As shown in Table 3, there was no difference in the initial BW between the two dietary treatment groups (P $>$ 0.10). At the end of the feeding trial, even though the final BW of the pigs fed Diet 1 and Diet 2 were not significantly different (P $>$ 0.10), the ADG of the pigs fed Diet 1 was significant lower (P 0.01) than that of the pigs fed Diet 2. There was no difference (P $>$ 0.10) existed in the ADFI between the two groups. Thus, the G:F of the pigs fed the Met-restricted diet was lower (P 0.01) than that of the pigs fed the Met-adequate diet. These results indicated that a restriction of dietary SID Met by 40.5% did not affect pig feed intake, but significantly decreased their G:F and ADG. This result of compromised growth performance (i.e., reduced G:F and ADG) is in line with several previous researches, such as those conducted by Bell et al. [7], Chung et al. [6], Ly et al. [9], and Humphrey et al. [16], who all reported that dietary Met restriction or deficiency can compromise pig growth performance.

In order to further understand how dietary Met restriction could affect nutrient metabolism in growing pigs, the concentrations of plasma free AA were analyzed. As shown in Table 4, prior to the 4-week feeding trial there was no differences (P $>$ 0.10) between the two dietary treatment groups in the plasma concentrations of nearly all free AA, except for aspartate. The plasma aspartate concentration tended to be lower in the pigs fed Diet 1 than the pigs fed Diet 2 (P = 0.07). As shown in Table 5, after the 4-week feeding trial, the plasma Met concentration was significantly lower in the pigs fed Diet 1 than the pigs fed Diet 2 (P 0.01), and so were the plasma concentrations of Cys and taurine (P 0.05). The plasma concentrations of histidine and glycine, however, were greater (P 0.05) in the pigs fed Diet 1 than fed Diet 2, while the plasma concentrations of lysine and asparagine had tendencies (0.05 P 0.10) to be greater in the pigs fed Diet 1 than Diet 2. The plasma concentrations of other 17 AAs were not different (P $>$ 0.11) between the two dietary treatment groups (Table 5).

The shift of plasma AA profile in pigs fed two different levels of dietary Met is generally in agreement with the results of Li et al. [26] and Tian et al. [27]. Li et al. [26] reported that the plasma concentrations of Met (numerically) and taurine were lower, while the plasma concentrations of lysine (numerically) was higher, in the sows fed a Met adequate vs. a Met excess diet. No data for Cys and glycine from these sows were reported by Li et al. [26]. Tian et al. [27] reported that the serum concentrations of Met and Cys (numerically) was lower, while the serum concentrations of lysine, histidine (numerically), and glycine were higher, in young growing pigs fed a Met deficient than fed a Met adequate diet. No data for taurine were reported by Tian et al. [27].

Obviously, the lower concentrations of plasma Met in pigs fed diets deficient or low in Met can be attributed to the limited dietary supply of Met in these studies. Given the fact that Cys and taurine are products of Met metabolism [28], the lower Cys and taurine concentrations in the plasma of pigs fed a Met restricted diet might be due to the limited dietary Met supply as well. The higher plasma concentrations of lysine, histidine, and glycine associated with the low dietary Met supply might be attributed to the fact that lysine, histidine, glycine, asparagine, and Met share the B${}^{0,+}$ AA transport system in small intestine [26, 29]. Lysine, histidine, asparagine, and Met also share the b${}^{0,+}$ transport system in small intestine [26, 29]. In addition, the reduced ADG (Table 3) due to the limited supply of Met might also be a reason for the higher plasma concentrations of lysine, histidine, and glycine, because the body protein biosynthesis may be limited and these three AA became extra in the Met restricted pigs.

A high plasma glycine concentration indicates that it was possible that less glycine was utilized for glutathione synthesis due to insufficient Met supply in the diet. Given the fact that both glutathione and taurine (derived from Cys) are essential antioxidants in pig body [1], the reduced plasma concentrations of Met, taurine, and possibly glutathione, might also be responsible for the reduced growth performance of pigs fed the Met-restricted diet. Although the parameters related to oxidative status were not measured in this study, the change in plasma AA concentration indicates that dietary Met restriction may negatively affect pig’s antioxidant capacity, health status, and, in consequence, growth performance.

In addition to the free AA, the plasma concentrations of six major metabolites were also analyzed. As shown in Table 6, before the feeding trial, there were no differences (P $>$ 0.10) between the two dietary treatment groups in the plasma concentrations of any one of the metabolites. After the feeding trial, however, the plasma urea nitrogen concentration was greater (P 0.01) in pigs fed Diet 1 than pigs fed Diet 2, while the plasma albumin concentration tended to be greater (P 0.09) in pigs fed Diet 1 than Diet 2. The plasma concentrations of other metabolites, including total protein, total cholesterol, glucose, and triglycerides, were not different (P $>$ 0.10) between the two treatment groups. The result of high plasma urea nitrogen concentration associated with the Met-restricted diet obtained in this study is in consistent with the results of some previous studies conducted by Shen et al. [5] and Tian et al. [27], who reported increased concentrations of plasma urea nitrogen with decreased dietary Met content in nursery and starter pigs, respectively.

The plasma urea nitrogen concentration is a reliable indicator of AA utilization efficiency by the animal [30, 31, 32]. The ideal amounts and ratios of all proteinogenic AA are essential for efficient nitrogen utilization, minimal AA catabolism, maximal protein synthesis, and thus the least plasma urea nitrogen concentration [11, 23, 33]. Data from this study showed that the plasma urea nitrogen concentration was greater in the Met-restricted pigs than in the Met-adequate pigs. The probable reason for this finding is that in the Met-restricted group, Met was the first limiting AA, and after the Met was exhausted, all other AA became “extra” and were catabolized to produce more urea nitrogen [30].

The plasma albumin concentration is also a good indicator of the effectiveness of dietary AA utilization, and of the liver capacity of protein synthesis [23, 34, 35]. The present study is in consistent with the result of Tian et al. [27] for young growing pigs. Our result is also supported by Litvak et al. [36] who reported that a low-level Met diet increased serum albumin concentration in growing pigs. It may also be speculated that increased hepatic albumin synthesis reflects increased need for endogenous Met to support the immune response since albumin is used to transfer Met [36].

There was no difference in the plasma total protein concentrations between the pigs fed Diet 1 and Diet 2 (P $>$ 0.10), indicating that the dietary Met restriction did not affect the plasma total protein content. This result is consistent with the findings of Chattopadhyay et al. [37], Meng et al. [38], and Remus et al. [39], who reported that the plasma total protein concentrations were not different in animals with different content of Met supply. The unchanged plasma total protein concentration but decreased BW gain may indicate a homeostatic control of plasma total protein concentration and a non-preferential utilization of AA for muscle protein syntheses.

There was no significant difference (P $>$ 0.10) between the two treatment groups in the plasma concentrations of glucose, triglycerides, and total cholesterol at the end of the feeding trial. This finding is in agreement with the results of Saeid et al. [40], who also found in broilers that Met alone has no regulating effect on plasma concentrations of glucose, triglycerides and cholesterol. The results of the current study and that of Saeid et al. [40] may indicate that the restriction of dietary Met had no effects on energy and lipid metabolism.

Before the feeding trial, the mRNA levels of the selected myogenic genes were similar (P $>$ 0.10) between the Diet 1 and Diet 2 pigs (Table 7). After the four-week feeding trial, however, the pigs fed with Diet 1 tended to have lower levels of Myf6 (P 0.08) and Mef2D (P 0.09) mRNA. There were no differences in the mRNA levels of other 6 myogenic genes between the two groups of pigs (P $>$ 0.10). These results indicate that dietary Met restriction may reduce the expression of Myf6 and Mef2D genes, which might reduce the expression levels of the corresponding proteins in the skeletal muscle of growing pigs.

As it is known, the MRF family are highly conserved and collectively expressed in skeletal muscle lineages [10, 14]. Assisted by the MEF2 family, MRFs coordinate the activities of a host of co-activators and co-repressors, resulting in a tight control of gene expression during myogenesis [41, 42]. Either MyoD or Myf5 is sufficient for skeletal muscle formation [43], but MyoG and Myf6 are directly involved in myotube differentiation [10, 44]. Therefore, the present data imply that Met may affect the myotube or muscle cell differentiation, but not the muscle cell formation. The functions of different MEF2 isoforms (activating the muscle structural genes) are difficult to distinguish because they are expressed in distinct but overlapping patterns [2, 45]. The reason why only Mef2D, but not the other isoforms, was affected by Met in this study are unknown.

A study conducted in pigs by Li et al. [2] showed that dietary Met supplementation increased the mRNA expression levels of MyoG, Mef2A, and Mef2D, but not of Myf6 (as in this study). The discrepancy between the results of Li et al. [2] and of this study might be mainly due to the difference in the animal models used. The low birth weight piglets were used by Li et al. [2] without reporting breed and sex, and there was no difference in the growth performance between the control and the Met supplemented pigs [2]. In this study, the muscle samples were collected when pigs were around 80 d of age, while Li et al. [2] collected their muscle samples when pigs were 180 d of age.