Abstract
This study investigated the effects of dietary protein levels on muscle protein breakdown (MPB) in lactating sows using 3-methyl-2H-histidine (3MH) as an isotopic tracer. Objectives also include first step of evaluation among candidate exponential models and second step of comparison between non-linear regression and traditional curve peeling method for the dynamics of 3MH. A total of eight sows were randomly assigned to two treatment: a standard protein diet (SP; 18.0% CP) or a low-protein diet (LP; 15.0% CP + crystalline amino acids). Plasma 3MH kinetics were modeled using single-, two-, and three-pool models, and comparison between nonlinear regression and curve peeling method in R program or Excel, respectively. The ANOVA in R program was finally performed to compare MPB between dietary treatment. Results showed that the LP diet increased MPB (P < 0.05), suggesting escalated mobilization of muscle reserves to sustain lactation. Model evaluation also revealed that three-exponential model (three-pool model) with nonlinear regression approach provided the most accurate characterization of 3MH dynamics. These findings highlight the trade-off between reduced dietary protein supply and increased maternal muscle catabolism, underscoring the importance of precision diet formulation for lactating sows and paving the way for the balanced diet with appropriate dietary CP level with crystallized amino acids.
Keywords: Exponential model, Non-linear regression, Curve peeling method, Muscle protein breakdown, Low protein diet, Lactating sows
1. Introduction
China is the world’s dominant pig producer, and efficient pig production is vital to ensure national food security. Among all parts of the swine production chain, the efficient milk production of sows, especially lactating sows is the key to enhancing the quality and efficiency of pig production. Research on lactating sows is of significant importance for both human health and agricultural production.
1.1 Lactating sows as model animals for study of human health
In human health research, the selection of model animals often have a crucial influence on the reliability and application value of the research outcome. In recent years, pigs are attracting more and more attention as a biomedical research model with huge potential owing to their resemblance with human. (Lunney,et.al, 2021). From a physiological perspective, porcine and human exhibit significant comparability in protein metabolism. Taking lactating sows as an example, it’s similarity with human female during their lactation lie mainly in their shared ability to regulate the dynamic balance between protein synthesis and breakdown through their bodies, prioritizing the nutritional needs of lactation. Thus, Lactating sows often face high levels of protein turnover and metabolic challenges, owing to the need to meet nutritional requirements for milk production and piglet growth.
1.2 Importance of lactating sows and low protein diet
Soybean meal is a vital protein ingredient in pig industry. However, the significant disparity between domestic soybean production and the rising demand of soybean meal within China has led to a notable reliance on imported soybeans. This dependence renders the domestic supply chain highly vulnerable to international market change. China’s soybean imports reached at 100.33 million tons in the year 2020, which was 5.11 times the national production volume, and accounted for 48.27% of the country’s total agricultural imports (Hu, 2021).
Under such circumstances, low protein diet technique has been proposed, and has garnered increasing attention in the field of animal nutrition. These diets offer notable benefits, including conservation of protein resources, reduction in feed costs, and significant decreases in nitrogen emissions (Chamberlin, 2017). Research indicates that 1% of reduction in dietary crude protein coupled with appropriate amino acid supplementation, can decrease 2.3% use of soybean meal (Liu et al., 2021), and reduce nitrogen output by at least 8% (Chamberlin, 2017). In fact, protein requirement for pig is fundamentally amino acids. Therefore, achieving an appropriate amino acid profile, based on the ideal protein concept, by supplementing crystalline amino acids (CAAs) ensured a reduction in total dietary protein without compromising animal performance (Zhu et al., 2022). Conventional swine diets are mostly composed of cereal grains and soybean meal, wherein lysine is generally recognized as the first limiting amino acid due to its low concentration in grain-based ingredients (Hasan et al., 2020). Thus, the technique of low-protein diets that’s supplemented with crystalline amino acids become a promising nutritional manipulation to reduce the use of soybean meal while achieving the required lysine levels in swine. These diets enable a decrease in total dietary crude protein while still fulfilling the essential amino acid requirements for swine growth and performance.
Sows, especially during lactating have a more stringent requirement for protein and amino acid. Studies have shown that a lack of protein would result in impaired lactation performance and subsequent reproduction (Hasan et al., 2020). When dietary protein supply falls short of requirements, sows compensate for lactation by mobilizing body reserves such as muscle protein serving as source of amino acids for milk production. This mobilization temporarily sustained milk production but has the cost of accelerated muscle loss and deteriorated body condition (Tokach et al., 2019). The long-term consequences also extend beyond lactation itself: excessive muscle mobilization impairs post-lactation recovery, delays estrus resumption, and ultimately compromises reproductive performance. Thus, exploring the ideal protein diet for sows during lactation through analyzing the dynamics of their muscle protein catabolism is important.
1.3 Isotopic 3-[methyl-²H₃]histidine and muscle protein breakdown
Estimation of muscle protein breakdown (MPB) is meaningful for optimizing lactation nutrition. Isotope tracer technique has always been an effective and powerful tool to quantify MPB. The 3-[methyl-²H₃]histidine or 3MH has proven to be a ideal biomarker for MPB (Davydova et al., 2021). Why was 3MH selected instead of more abundant amino acids commonly involved in protein metabolism? For most amino acids like lysine, leucine, and glycine, they would be reutilized readily after liberation from MPB. Reutilization include new protein synthesis, oxidized for energy, or diverted into metabolic pathways in other tissues. As a result, their circulating or excreted levels represent the sum of dietary intake, synthesis, and catabolism, confounding their use as specific markers of muscle protein breakdown. In comparison, 3MH is a unique amino acid derivative arised from post-translational methylation of peptide-bound histidine residues in actin and myosin (Davydova et al., 2021). No aminoacyl-tRNA exists for 3MH, and therefore it cannot be reincorporated into protein synthesis once liberated (Davydova et al., 2021). The myosin and actin, which account for 50–65% of total muscle protein mass are the principal sources of 3MH in vivo (Thomsen et al., 2022).
1.4 Dynamics of isotope 3-[methyl-²H₃]histidine described by mathematical model
In this study, model selection plays an important part owing to the changebility of muscle protein turnover kinetics. Specifically, the discrepancy between the sarcoplasmic proteins’ rapid degradation and myofibrillar proteins’ slower turnover (Moore et al., 2009). This study also aims to develop and compare appropriate candidate models for MPB.
The objectives of this study were to:
1) develop and select candidate mathematical model for dynamics of MPB (Single pool vs. Two pool vs. Three pool model);
2) compare pros and cons in methodology of non-linear regression and curve peeling; 3) evaluate effect of low protein diet on MPB in lactating sows.
2. Methodology
2.1 Dietary treatments
The nutritional requirements of lactating sows is being estimated by the NRC (2012). A control or standard protein diet (SP) was formulated using corn and soybean meal and consequently contained 18% CP. The experimental diet or low protein diet (LP) was formulated to lower dietary protein level to 15% CP and improve AA balance. Major ingredients and calculated nutrient level of diets are in Table 1.
Table 1. Feed composition and nutrient content of diets (as-fed)
2.2 Animals and feeding
Eight gestating sows with parity from 2 to 4 were moved to farrowing crates. Then, they were randomly put into one of two diets: the standard protein (SP) group (4 sows) and the low protein (LP) group (4 sows). The sows were fed three equal meals every day, at 7:00 AM, 1:00 PM, and 7:00 PM. Both the sows and their piglets could drink water whenever they wanted. In addition, each litter of piglets was adjusted to have exactly 10 piglets per sow.
Figure 1. Timeline of lactation and bolus injection of 3MH.
2.3 Ear vein catheterization
Ear vein catheters were inserted on d 15-16. Piglets were removed and sows were sedated. Ears were aseptically scrubbed with 10% betadine. A 61-cm medical micro-tube was used. A tourniquet firstly distended ear veins, then a 14G needle was inserted, followed by the push of tube for 30 cm to the external jugular vein. The tube entry were sutured, gauze covered, and elastic tape wrapped; a 120-cm extension tube was added. The whole procedure took 45-90 min/sow. After sows stood, a wide elastic bandage protected the pouch. Catheters were checked twice everyday and flushed with heparinized saline.
Figure 2. The ear vein of lactating sow and blood sampling from ear vein catheter.
2.4 Isotope preparation and bolus infusion
Tracers were first weighed, dissolved in saline, and was sterilized by filtration (Millipore Steriflip). A stock solution of 3-[methyl-2H3]histidine (containing 181.4 μmol of the compound dissolved in 20 mL of saline) was made for bolus injection. The bolus dose of 3-[methyl-2H3]histidine (abbreviated as 3MH) was calculated by assuming a 20% pool size of free 3MH in sows (Rathmacher et al., 1993; Trottier, 1995). Finally, the 20 mL saline solution was administered through the ear vein catheter.
2.5 Blood sampling
Blood samples were harvested through the ear vein catheter at: 0 (right after the bolus infusion of 3MH), 0.08, 0.16, 0.25, 0.5, 0.75, 1, 2, 3, 4, 5, 6, 7, 8, 24, 34, 48, 58, and 72 hours. Blood sample was were spun under at 1,500 × g at 4°C for 5 minutes, and then was separated into 1.5-milliliter microcentrifuge tubes.
2.6 Isotope analysis
The 3-[methyl-²H₃]histidine in the plasma samples were analyzed using HESI LC-MS (Marini, 2011).
2.7 Calculation
The dose of 3MH isotope was estimated by 20% of the free 3MH pool extracellularly (Rathmacher et al., 1993). The extracellular water space was assumed to be 20% of the body weight and the 3MH concentration in the extracellular space to be 20 nmoles/mL (Rathmacher et al., 1993).
Raw data were presented as tracer to tracee ratio (TTR). AUC, Ra and half-life (tail curve rate constant) were calculated.
Muscle protein breakdown rate (%/d) was calculated as follows:
Where total protein bound 3MH pool = 3.8742 µmol × muscle protein (g)
muscle protein (kg) = 8.21% × sow BW (kg) × 8.21% (Rathmacher et al., 1996).
Muscle protein breakdown, g/d, was calculated as follows:
2.8 Statistical analysis
Curve peeling was performed in Microsoft Excel to fit the three-exponential model to experimental data, enabling the extraction of six parameters (Figure 3).
Nonlinear regression (nls) with different exponential models (e.g. single exponential model, two exponential model and three exponential model) was conducted in R software (4.5.0) to fit the experimental data. Scatter plots were generated to visualize the raw data and fitted distinct exponential curves.
Variance analysis (ANOVA) was performed in R software (4.5.0) to compare differences in target indices (e.g., 3MH Ra, AUC, muscle protein breakdown) between SP and LP groups, with statistical significance set at P < 0.05 and tendencies at 0.05 <P ≤ 0.1.
Figure 3. Procedures of curve peeling method in Excel sheet
3. Results
3.1 One-pool model vs. two-pool model vs. three-pool model
Figure 4 showed the comparison among single-pool, two-pool and three-pool models. Single-pool model mathematically corresponds to single exponential function, while two-pool and three-pool models corresponds to two-exponential and three-exponential model. Scatter plot indicated that three-exponential model fit the raw data the best, suggesting that three-pool model is the most appropriate model to depict dynamics of 3MH in lactating sows.
Figure 4. Comparison among single pool model, two pool model and three pool model
3.2 Non-linear model vs. curve peeling
Figure 5 showed the comparison between non-linear model (NLIN program) between curve peeling. The six parameters derived respectively from NLIN program and curve peeling were also shown to the left. The six parameters of triple-exponential model from non-linear regression and curve peeling are comparable. Curves using these two sets of parameters were both close to the scatter plot of raw data.
Figure 5. Comparison between non-linear model and curve peeling
3.3 Muscle protein breakdown
The dynamics of 3MH and MPB were presented in Table 4. Rate of appearance (Ra; μmol/h) of 3MH in the plasma of sows fed SP and LP diets were different (P < 0.05). The MPB per kg of body weight (g/d·kg⁻¹; P < 0.05) and fractional MPB (%/d; P < 0.05) of sows fed SP and LP diets were significantly different.
Table 4. Dynamics of 3-[methyl-²H₃]histidine for sows with standard protein (SP; 18% CP; n = 4) or low protein (LP; 14% CP + CAA; n = 4) diets¹
4. Discussion
Swine production is a pillar industry of Chinese animal husbandry, among which the efficient feeding of lactating sows is particularly important. Due to the high dependence on soybean meal import and the unstable supply of protein feed ingredients, the low protein diet technique has been adopted.
Lactation represents the most nutritionally demanding phase of a sow’s life, as the production of milk proteins such as casein requires a continuous and substantial supply of amino acids(Trottier, 1995). When dietary protein intake is inadequate, sows compensate for the increased demand for milk protein synthesis primarily by mobilizing protein reserves, particularly from skeletal muscle (Tokach et al., 2019). This adaptive response involves alterations in whole-body protein turnover and enhanced muscle protein catabolism, ultimately impairing subsequent reproductive performance(Dourmad et al.,1998; Tokach et al., 2019). When dietary protein intake is inadequate, milk protein synthesis can be largely supported by nutrients derived directly from feed, thereby maintaining muscle structures and keeping muscle protein degradation rate (MPDR) relatively low. Thus, maintaining sufficient dietary protein intake can suppress unnecessary MPDR, allowing sows to preserve muscle mass while still meeting the amino acid demands of milk synthesis. Consequently, studying MPDR provides valuable insight into the physiological trade-offs between dietary protein, body mobilization, and milk production. It is possible to appropriately reduce the dietary protein content without affecting the lactation performance of sows, thereby identifying the optimal low protein level suitable for lactating sows.
In this study, 3MH is used as a tracer of muscle protein degradation in lactating sows. The 3MH has been proven as an effective biomarker to estimate muscle protein breakdown in multiple species (Trottier, 1995; Auclair et al., 1997; Hare et al., 2019). Quantitative studies further show that, about 80% of excreted 3MH originates from actin and 20% from myosin, with the latter varying by fiber type, being most abundant in myosin of white skeletal fibers and lowest in red skeletal and smooth muscle (Ballard and Thomas, 1983). Critically, because 3MH is exclusively generated during the breakdown of myofibrillar proteins and cannot be metabolically reused, its appearance in circulation provides a direct and specific measure of muscle protein degradation (Kochlik et al., 2019). These properties make 3MH an irreplaceable biomarker for assessing muscle proteolysis, in contrast to conventional amino acids whose concentrations reflect a range of metabolic processes beyond catabolism.
Although urinary sampling for the excretion of 3MH has often been considered a convenient non-invasive approach for measurement, blood sampling is employed in this research. Indeed, urinary collection is less invasive and logistically easier than repeated blood sampling, and urinary 3MH excretion has proven its close correlation with muscle proteolysis rates in several animals, such as cattle (Hare et al., 2019) and rats(Auclair et al., 1997). However, this method has proven to be invalid in porcine (Trottier, 1995), owing to the metabolic incorporation of free 3MH in to balenine— a dipeptide retained in muscle—and thereby decoupling urinary 3MH concentration from the true rate of muscle protein breakdown as the pathway reduced the amount of free 3MH excreted in urine(Trottier, 1995). Consequently, urinary excretion cannot provide a quantitative reflection of muscle proteolysis in sows. In contrast, measuring isotopic 3MH appearance in blood offers a direct and accurate assessment of muscle protein breakdown that is unaffected by downstream modification of 3MH such as balenine formation (Trottier, 1995). Although less convenient, blood sampling provides essential accuracy for measuring the protein breakdown rate in lactating sows.
Figure 6. Dipeptide balenine (β-alanyl-3MH), the acid labile component containing 3MH (Trotter, 1995)
Precise calculation of MPB in swine is important for explaining mechanisms of nutrient partitioning, optimizing growth performance of swine and improving lean tissue accretion. However, the fidelity of estimated parameters, particularly fractional degradation rates (k), dependent upon both the kinetic model structure and the computational approach employed in data analysis. Single-exponential models (y=A₁e⁻ᵏ¹ˣ), assuming kinetic homogeneity, fail to capture the multi-compartmental nature of muscle protein metabolism, leading to oversimplified distribution and clearance.
In contrast, double- (y=A₁e⁻ᵏ¹ˣ+A₂e⁻ᵏ²ˣ) and triple-exponential (y=A₁e⁻ᵏ¹ˣ+A₂e⁻ᵏ²ˣ+A₃e⁻ᵏ³ˣ) models allow for the resolution of discrete turnover components (e.g., k₁, k₂, k₃), corresponding to fast, intermediate, and slow-degrading protein fractions, respectively. However, two-pool model (double-exponential model) only partially captured the dual dynamics of plasma turnover and muscle sequestration. The three pool model (triple-exponential model) accounted for all major fates of 3MH, including an intermediate compartment representing transient passage through metabolic tissues (Rathmacher et al., 1993). Thus, the three pool model (triple-exponential model) aligned most closely with physiological reality, reflecting the rapid appearance of free plasma 3MH, its slower release from muscle-bound pools, and temporary processing in hepatic or renal tissues. Our mathematic evaluation confirmed its superiority by comparing among single exponential, two exponential and three exponential function (Figure 4). By minimizing bias and yielding estimates that more accurately represent true muscle protein degradation, the three-pool model establishes itself as the optimal tool for assessing MPB. Importantly, this methodological refinement ensures that tracer-based measurements can be translated into biologically meaningful insights.
When fitting the three-exponential model of 3-methylhistidine (3MH) kinetics, curve peeling and nonlinear regression served distinct but complementary roles. Analytical methodology further influences parameter reliability. Traditional methods, most commonly using Excel-based curve peeling, a sequential graphical approach, allowed factoring of the composite curve into rapid, intermediate, and slow phases. While conceptually useful, its reliance on visual extrapolation made it prone to error when pool turnover rates overlapped, limiting its value for precise quantification of MPB. Its strength, however, was in providing reasonable initial parameter estimates, which improved convergence reliability in nonlinear regression.
Nonlinear regression, by contrast, employing numerical optimization to directly fit multi-exponential functions to raw data, on the other hand, minimizes residual variance and enables objective estimation of pool-specific turnover rates. Nonlinear regression, optimized all six parameters simultaneously through least-squares minimization, eliminating subjectivity and yielding far greater accuracy and precision in MPB estimates. Yet it depended on good starting values to avoid local minima. Comparative analyses indicate that nonlinear regression yields more robust and reproducible estimates for myofibrillar protein degradation, whereas curve peeling often inflates fast-phase kinetics due to subjective curve segmentation (Wolfe, 2011). Together, the methods proved complementary: curve peeling served as a preparatory tool, while nonlinear regression ensured robust and physiologically valid MPDR assessment.
In this study, MPB of sows fed LP diet was higher compared to control or SP diet. Previous research reported that reducing crude protein by 10–15% compared with standard diets, while balancing EAAs such as lysine and methionine, prevents the excessive muscle mobilization observed under severe protein restriction (Sobotka, et al., 2025). Note that in this study, dietary protein was lowered from (SP diet) to (LP diet), meaning that dietary protein was decreased by more than 15%. Thus, reducing dietary protein from 18% to 14% was too aggressive for lactating sows, and muscle protein mobilization became inevitable. In fact, low-proteins diets enacted to meet the essential amino acid (EAA) requirements in different circumstances can exert beneficial effects on muscle protein metabolism in lactating sows by stabilizing MPB. By doing so, the diet provide more balanced amino acid supply for nutritional requirements of lactating sows, thereby reducing the metabolic cost of excess protein catabolism, including hepatic urea synthesis, while maintaining adequate substrate availability for milk protein synthesis (Zhang, et al., 2020). Previous reports also suggest that the implementation of low protein diets with optimized amino acid balance would considerably reduce protein breakdown rate in sows (Trottier et al.,1995), and probably the reduction of dietary protein in these study was moderate.
The absolute and fractional rates of muscle protein degradation observed in this study were higher than those reported previously in the literature (Trottier, 1995). Specifically, estimated MPB in multiparous lactating sows (13.70–19.81 μmol/kg/d; 4.32–6.24%/d) exceeded values reported by Trottier (1995) for lactating gilts (12.0 μmol/kg/d; 3.4%/d) using the same 3MH tracer method. This discrepancy suggests that parity plays an important role, as multiparous sows may mobilize body protein more readily than gilts to support the greater amino acid demand associated with higher milk yield. While this higher mobilization highlights the physiological cost of sustaining lactation under reduced dietary protein supply, it does not diminish the overall benefits of employing a lower-protein feeding strategy. Rather, these findings illustrate that the level of crude protein reduction applied in this study may have been too aggressive, thereby pushing sows to rely more heavily on muscle reserves. Importantly, this outcome provides valuable insight into the balance required for defining an optimal protein diet one that avoids unnecessary catabolism while still maintaining the environmental and economic advantages of lowering crude protein. By clarifying the trade-offs between dietary protein supply, milk production, and body mobilization, the results of this study contribute to refining nutritional strategies that sustain sow productivity and reproductive performance while promoting feed efficiency and sustainability.
5. Conclusions
In this study, we investigated muscle protein breakdown (MPB) in lactating sows subjected to different dietary protein levels using the isotopic tracer 3-methylhistidine (3MH). Our results demonstrate that
1) The three-pool model (three-exponential model) is the most appropriate compared to single-pool or two-pool model.
2) Parameters estimated by non-linear regression or curve peeling method are comparable. Although non-linear regression is superior, the curve peeling method is a feasible way to estimate initial value for non-linear regression.
3) a LP diet (15% CP) increases MPB rates, suggesting that agressive reduction in crude protein further compromises muscle protein reserves during lactation. This finding highlighted the importance of an optimal protein diet that balances both production performance and metabolic needs (i.e., lactation).
Although the study involved a limited number of subjects and yielded outcomes that seems diverge from the initial expectation of low protein diet research, it provides valuable insights into the development of feeding strategies that simultaneously lower feeding costs and precisely meet the protein requirements of lactating sows. Further research could focus on the effect on the lactating sows with a moderate reduction in crude protein.
6. References
Auclair D, Garrel DR, Chaouki Zerouala A, Ferland LH. Activation of the ubiquitin pathway in rat skeletal muscle by catabolic doses of glucocorticoids. Am J Physiol. 1997;272(3 Pt 1):C1007-C1016. doi:10.1152/ajpcell.1997.272.3.C1007
Ballard FJ, Tomas FM. 3-Methylhistidine as a measure of skeletal muscle protein breakdown in human subjects: the case for its continued use. Clin Sci (Lond). 1983;65(3):209-215. doi:10.1042/cs0650209
Chamberlin, D. P. (2017). Impacts of reducing dietary crude protein with crystalline amino acid supplementation on lactating sow performance, nitrogen utilization and heat production. Michigan State University. Animal Science.
Dourmad JY, Noblet J, Etienne M. Effect of protein and lysine supply on performance, nitrogen balance, and body composition changes of sows during lactation. J Anim Sci. 1998;76(2):542-550. doi:10.2527/1998.762542x
Davydova E, Shimazu T, Schuhmacher MK, et al. The methyltransferase METTL9 mediates pervasive 1-methylhistidine modification in mammalian proteomes. Nat Commun. 2021;12(1):891. doi:10.1038/s41467-020-20670-7
Hare KS, Wood KM, Fitzsimmons C, Penner GB. Oversupplying metabolizable protein in late gestation for beef cattle: effects on postpartum ruminal fermentation, blood metabolites, skeletal muscle catabolism, colostrum composition, milk yield and composition, and calf growth performance. J Anim Sci. 2019;97(1):437-455. doi:10.1093/jas/sky413
Hasan MS, Crenshaw MA, Liao SF. Dietary lysine affects amino acid metabolism and growth performance, which may not involve the GH/IGF-1 axis, in young growing pigs1. J Anim Sci. 2020;98(1):skaa004. doi:10.1093/jas/skaa004
Hu, X. (2021). Research on potential risks and countermeasures of China's soybean imports [D]. China Agricultural University
Kochlik B, Stuetz W, Pérès K, et al. Associations of Plasma 3-Methylhistidine with Frailty Status in French Cohorts of the FRAILOMIC Initiative. J Clin Med. 2019;8(7):1010. doi:10.3390/jcm8071010
Liu SY, Macelline SP, Chrystal PV, Selle PH. Progress towards reduced-crude protein diets for broiler chickens and sustainable chicken-meat production. J Anim Sci Biotechnol. 2021;12(1):20. doi:10.1186/s40104-021-00550-w
Lunney JK, Van Goor A, Walker KE, Hailstock T, Franklin J, Dai C. Importance of the pig as a human biomedical model. Sci Transl Med. 2021;13(621):eabd5758. doi:10.1126/scitranslmed.abd5758
Moore DR, Tang JE, Burd NA, Rerecich T, Tarnopolsky MA, Phillips SM. Differential stimulation of myofibrillar and sarcoplasmic protein synthesis with protein ingestion at rest and after resistance exercise. J Physiol. 2009;587(Pt 4):897-904. doi:10.1113/jphysiol.2008.164087
Marini JC. Quantitative analysis of 15N-labeled positional isomers of glutamine and citrulline via electrospray ionization tandem mass spectrometry of their dansyl derivatives. Rapid Commun Mass Spectrom. 2011;25(9):1291-1296. doi:10.1002/rcm.5007
NRC. Nutrient requirements of swine. 11th ed. Washington, DC: National Academy Press; 2012.
Rathmacher JA, Link G, Nissen S. Measurement of 3-methylhistidine production in lambs by using compartmental-kinetic analysis. Br J Nutr. 1993;69(3):743-755. doi:10.1079/bjn19930075
Rathmacher JA, Nissen SL, Paxton RE, Anderson DB. Estimation of 3-methylhistidine production in pigs by compartmental analysis. J Anim Sci. 1996;74(1):46-56. doi:10.2527/1996.74146x
Sobotka W, Drażbo A. The Effect of Dietary Protein Restriction in Phase Feeding Systems on Nitrogen Metabolism and Excretion in Pig Production. Animals (Basel). 2025;15(11):1521. doi:10.3390/ani15111521
Trottier NL. Protein metabolism in the lactating sow. PhD Diss, University of Illinois, Urbana–Champaign. 1995.
Tokach MD, Menegat MB, Gourley KM, Goodband RD. Review: Nutrient requirements of the modern high-producing lactating sow, with an emphasis on amino acid requirements. Animal. 2019;13(12):2967-2977. doi:10.1017/S1751731119001253
Thomsen, E. M., Sonne, A., Bertelsen, M., Vermeulen, E., Ganswindt, A., Wang, T., & Ochala, J. (2022). Fine‐regulation of myosin relaxed states in vertebrate muscle. The FASEB Journal, 36.
Wolfe, R. R. . (2011). Stable isotope tracers applied to measuring rates of protein synthesis and breakdown in muscle: principles and applications. Journal of Biomolecular Techniques Jbt 22(Suppl), S12.
Zhu C, Yang J, Wu Q, et al. Low Protein Diet Improves Meat Quality and Modulates the Composition of Gut Microbiota in Finishing Pigs. Front Vet Sci. 2022;9:843957. doi:10.3389/fvets.2022.843957
Zhang S, Johnson JS, Qiao M, Trottier NL. Reduced protein diet with near ideal amino acid profile improves energy efficiency and mitigate heat production associated with lactation in sows. J Anim Sci Biotechnol. 2020;11:4. doi:10.1186/s40104-019-0414-x
