Skip to main content
PMC home page
Journal of Animal Science logoLink to Journal of Animal Science
. 2019 Apr 8;97(6):2505–2514. doi: 10.1093/jas/skz118

Meta-regression analysis to predict the influence of branched-chain and large neutral amino acids on growth performance of pigs1

Henrique S Cemin 1,✉, Mike D Tokach 1, Steve S Dritz 2, Jason C Woodworth 1, Joel M DeRouchey 1, Robert D Goodband 1
PMCID: PMC6541811  PMID: 30959521

Abstract

A meta-analysis was conducted to evaluate the effects of branched-chain amino acids (BCAA), their interactions, and interactions with large neutral amino acids (LNAA) to develop prediction equations for growth performance of pigs. Data from 25 papers, published from 1995 to 2018, for a total of 44 trials and 210 observations were recorded in a database. Diets were reformulated using the NRC (2012) loading values to estimate nutrient concentrations. The response variables were average daily gain (ADG), average daily feed intake (ADFI), and gain-to-feed ratio (G:F). The predictor variables tested included average body weight (BW), crude protein, neutral detergent fiber, Ile:Lys, Leu:Lys, Val:Lys, BCAA:Lys, Ile:Leu, Val:Leu, Ile:Val, (Ile+Val):Leu, Trp:Lys, Leu:Trp, Ile:Trp, Val:Trp, BCAA:Trp, Met:Lys, Leu:Met, Ile:Met, Val:Met, BCAA:Met, His:Lys, Leu:His, Ile:His, Val:His, BCAA:His, Thr:Lys, Leu:Thr, Ile:Thr, Val:Thr, BCAA:Thr, (Phe+Tyr):Lys, Leu:(Phe+Tyr), Ile:(Phe+Tyr), Val:(Phe+Tyr), BCAA:(Phe+Tyr), LNAA:Lys, Leu:LNAA, Ile:LNAA, Val:LNAA, and BCAA:LNAA. Amino acids were expressed on standardized ileal digestible basis. The MIXED procedure of SAS (SAS Institute Inc., Cary, NC) was used to develop the equations. The inverse of squared SEM was used to account for heterogeneous errors using the WEIGHT statement. Models were selected with a step-wise manual forward selection. In order to be included in the final model, predictor variables had to be statistically significant (P < 0.05) and provide an improvement of at least 2 points in Bayesian information criterion. The optimum equations were: ADG, g = – 985.94 + (15.2499 × average BW (kg)) – (0.08885 × average BW × average BW) + (1.063 × Leu:Lys) + (20.2659 × Ile:Lys) – (0.1479 × Ile:Lys × Ile:Lys) + (9.2243 × (Ile+Val):Leu) – (0.03321 × (Ile+Val):Leu × (Ile+Val):Leu) – (0.4413 × Ile:Trp); G:F, g/kg = 648.3 – (6.2974 × average BW (kg)) + (0.02051 × average BW × average BW) + (0.5396 × Ile:Lys) + (1.7284 × Val:Lys) – (0.00795 × Val:Lys × Val:Lys) – (1.7594 × Met:Lys); and ADFI, kg = predicted ADG/predicted G:F. Overall, the prediction equations suggest that increasing Leu:Lys negatively impacts ADG due to a reduction in G:F and ADFI caused by insufficient levels of other BCAA and LNAA relative to Leu. According to the model, the addition of Val, Ile, and Trp, alone or in combination, has the potential to counteract the negative effects of high dietary Leu concentrations on growth performance.

Keywords: branched-chain amino acids, growth, prediction, swine

Introduction

Low crude protein (CP), amino acid (AA) fortified diets are a well-established practice to reduce diet cost and nitrogen excretion. Currently, Lys, Thr, Met, Trp, and Val are commonly used in swine diets to replace a portion of intact protein sources, such as soybean meal. After Val, Ile is probably the next limiting AA in low CP, AA-fortified diets (Figueroa et al., 2003). Of the three branched-chain amino acids (BCAA), Val and Ile are frequently limiting and thus their amount is minimized. In contrast, Leu is usually in excess in corn-based diets due to its high concentration in corn and corn by-products, such as distillers dried grains with solubles (DDGS).

The BCAA are structurally similar and share the first steps of their catabolism. Therefore, excess of a BCAA, particularly Leu, may result in increased degradation of the others (Harper et al., 1984; Brosnan and Brosnan, 2006). Moreover, BCAA and the other large neutral amino acids (LNAA), such as Trp, Tyr, Phe, Met, Thr, and His, share the same brain transporters (Pardridge, 1977; Fernstrom, 2005). Therefore, excess BCAA may decrease the AA absorption and subsequent utilization as neurotransmitter precursors that play an important role in appetite regulation. For instance, Trp is a precursor of serotonin, Tyr forms catecholamines, and His is essential for the synthesis of histamine, all of which are involved in feed intake regulation (Henry et al., 1992; Kurose and Terashima, 1999; Fernstrom, 2013).

Although data on the effects of high Leu diets on pig performance are inconsistent (Cisneros et al., 1996; Hyun et al., 2007; Morales et al., 2016), we hypothesize that practical corn-soybean meal diets supplemented with high-feed grade AA, especially with inclusion of DDGS, can potentially create imbalances and negatively impact growth performance. Therefore, the objective of this study was to evaluate prediction equations to quantify the effects of dietary BCAA, their interactions, and the potential interactions between BCAA and LNAA for nursery and growing-finishing pigs.

MATERIALS AND METHODS

Database

A literature search similar to that described by Nitikanchana et al. (2015) and Flohr et al. (2018) was conducted to directly or indirectly evaluate the effects of BCAA on growth performance. The initial screening process included trials that directly evaluated BCAA as part of the treatment structure and also those that indirectly created diet formulations with variation in one or more of the BCAA, such as Ile, Val, and Leu requirement studies. In addition, trials that assessed the inclusion of corn by-products (corn germ meal, corn gluten meal, DDGS) were considered. All data were derived exclusively from peer-reviewed publications. In order to be included in the final database, trials had to provide treatment means and SEM or SD, pigs had to have ad libitum access to feed and water, and enough diet formulation detail to recreate the diet formulation. Trials that used blood products were eliminated based on preliminary model evaluation based on model fit and residual analysis. The reason for this is not clear but could be driven by the AA profile of blood products, high in Leu and low in Ile, that results in a substantial imbalance between BCAA.

Data from each trial were recorded in a spreadsheet template. Data included average daily gain (ADG), average daily feed intake (ADFI), gain-to-feed ratio (G:F), average body weight (BW), SEM, replicates per treatment, and number of pigs per pen. For papers that reported feed efficiency as feed-to-gain ratio (F/G), the inverse proportion was calculated based on the ADG and ADFI values provided. In some cases, growth performance was reported by period. In those situations, each period was identified in the database as a trial. For trials that reported only overall performance but diet formulation by period, the average nutrient content was pooled across periods to be used with overall growth performance.

The final database contained data from 25 papers, published from 1995 to 2018, that comprised 44 trials for a total of 210 observations. Dietary treatments from each trial were reformulated using a spreadsheet-based formulator with NRC (2012) nutrient loading values for ingredients to achieve a common basis for dietary nutrient concentrations. All AA were expressed on a standardized ileal digestible basis. Dietary factors evaluated included CP, neutral detergent fiber, Ile:Lys, Leu:Lys, Val:Lys, BCAA:Lys, Ile:Leu, Val:Leu, Ile:Val, (Ile+Val):Leu, Trp:Lys, Leu:Trp, Ile:Trp, Val:Trp, BCAA:Trp, Met:Lys, Leu:Met, Ile:Met, Val:Met, BCAA:Met, His:Lys, Leu:His, Ile:His, Val:His, BCAA:His, Thr:Lys, Leu:Thr, Ile:Thr, Val:Thr, BCAA:Thr, (Phe+Tyr):Lys, Leu:(Phe+Tyr), Ile:(Phe+Tyr), Val:(Phe+Tyr), BCAA:(Phe+Tyr), LNAA:Lys, Leu:LNAA, Ile:LNAA, Val:LNAA, and BCAA:LNAA.

Statistical Analysis

The MIXED procedure of SAS (SAS Institute Inc.) was used to develop the regression equations. The meta-regression was conducted as described by Flohr et al. (2018). Using the method of maximum likelihood, potential variables were selected by evaluating single variable equations. The statistical significance for inclusion of terms in the models was determined at P < 0.05. The WEIGHT statement of PROC MIXED was used to account for heterogeneous errors as the inverse of squared SEM (St-Pierre, 2001). Trial was included as a random effect. The single variable model with the lowest Bayesian information criterion (BIC) was selected and additional terms were added in a step-wise manual forward selection. In order to be included in the model, a reduction of 2 points or more in BIC was required (Kass and Raftery, 1995). Linear and quadratic terms as well as interactive effects were evaluated. If the quadratic or interactive terms were statistically significant and provided an improvement in BIC, the single order variables were also included in the model. Once the model with the lowest BIC was obtained, the method of residual maximum likelihood was used to obtain the parameter estimates. These methods were used to develop regression equations to predict ADG and G:F. Once ADG and G:F models were obtained, predicted values were used to estimate ADFI by dividing predicted ADG by predicted G:F. The adequacy of the models was examined by evaluating studentized residuals and residuals vs. predicted values plots.

RESULTS AND DISCUSSION

A summary of the publications used in the analysis is presented in Table 1. The values describe average initial and final BW and ranges of Leu:Lys, Ile:Lys, Val:Lys, and Trp:Lys. The studies in the final database comprised an average BW range of 6.5 to 126.7 kg, 82 to 715% Leu:Lys, 46 to 103% Ile:Lys, 38 to 124% Val:Lys, and 16 to 27% Trp:Lys. The inspection of residuals vs. predicted values relative to the line of equality (Figure 1) suggests that the predictions were precise and not biased. The studentized residuals plots indicate that normality assumption was met and no evidence for outliers and heteroscedasticity was observed.

Table 1.

Summary of publications used in the meta-regression to predict growth performance from branched-chained and large neutral amino acids1,2

Publication Trials Average initial BW, kg Average final BW, kg Range of SID Leu:Lys Range of SID Ile:Lys Range of SID Val:Lys Range of SID Trp:Lys
Lewis and Nishimura (1995) 1 67.0 80.0 102 70 38–93 19
Cisneros et al. (1996) 1 80.8 110.6 229–715 81 99 19–24
Mavromichalis et al. (2001) 1 10.7 17.0 82–127 50–63 39–88 19–20
Whitney et al. (2004) 4 8.2 16.0 130–176 66–78 71–87 19–22
Dean et al. (2005) 3 82.3 114.3 160–251 48–69 64–124 19–24
Whitney et al. (2006) 1 28.3 115.1 171–251 71–89 81–109 20–21
Hyun et al. (2007) 1 73.3 126.7 204–594 77 91 16–24
Linneen et al. (2008) 4 52.3 89.2 145–221 70–87 76–101 20–22
Weber et al. (2010) 1 30.8 54.6 92–139 49–64 52–71 19–27
Xu et al. (2010) 3 53.4 84.3 162–239 70–73 80–93 19–20
Cromwell et al. (2011) 3 61.2 89.8 176–383 81–103 90–133 23–27
Gaines et al. (2011) 3 16.1 28.1 110–145 58–74 54–79 18–22
Castilha et al. (2012) 1 15.4 30.3 111 46–67 67 16
Lee et al. (2012) 1 68.1 95.2 137–218 63–74 76–91 16–20
Waguespack et al. (2012) 1 21.8 42.2 132 61 61–73 21
Bergstrom et al. (2014) 1 35.1 125.0 186–247 71–81 83–100 17–19
Graham et al. (2014) 1 68.9 125.6 167–244 70–82 80–101 18–19
Liu et al. (2015) 4 60.0 81.8 113–138 52–56 54–79 18–23
Morales et al. (2016) 1 31.8 46.8 94–140 57–78 69–96 24
Overholt et al. (2016) 3 58.4 87.4 133–230 60–78 66–96 17–20
Clark et al. (2017) 1 6.5 9.8 104 57 53–87 20
Coble et al. (2017) 1 105.7 125.4 165–209 62–66 75–81 17–19
Huepa et al. (2017) 1 11.1 24.8 91–142 51 64 17
Lazzeri et al. (2017) 1 15.0 31.1 111 46–75 68 17
Xu et al. (2018) 1 8.8 20.1 113 53 60–90 21
Open in a new tab

1For trials that reported growth performance by period, each period was identified in the database as a trial.

2Standardized ileal digestible (SID) amino acid ranges presented are based on reformulated diets using NRC (2012) nutrient loading values.

Figure 1.

Figure 1.

Open in a new tab

Plots of actual vs. predicted values relative to the line of equality and studentized residual of average daily gain (ADG) and gain-to-feed ratio (G:F). (a) and (b) Plots for ADG, and (c) and (d) plots for G:F.

For ADG, the model with average BW as a single predictor had the lowest BIC value (2258.3) and it was a significant predictor of ADG (P < 0.01). Therefore, average BW was selected as the first predictor variable and other variables were subsequently added to the model. The step-wise inclusion of Leu:Lys (linear term; P < 0.01, BIC = 2244.7), Ile:Lys (linear and quadratic terms; P < 0.01, BIC = 2238.6), (Ile+Val):Leu (linear and quadratic terms; P < 0.01, BIC = 2214.6), and Ile:Trp (linear term; P < 0.01, BIC = 2208.6) resulted in improvements in BIC. Other variables did not further improve BIC. Therefore, the model with average BW, Leu:Lys, Ile:Lys, (Ile+Val):Leu, and Ile:Trp was selected as the final equation with a BIC of 2208.6 (Table 2).

Table 2.

Regression equations to predict growth performance of grow-finish pigs1

Variable2 Equation3 BIC4
ADG, g = – 985.94 + (15.2499 × average BW (kg)) – (0.08885 × average BW × average BW) + (1.063 × Leu:Lys) + (20.2659 × Ile:Lys) – (0.1479 × Ile:Lys × Ile:Lys) + (9.2243 × (Ile+Val):Leu) – (0.03321 × (Ile+Val):Leu × (Ile+Val):Leu) – (0.4413 × Ile:Trp) 2,208.6
G:F, g/kg = 648.3 – (6.2974 × average BW (kg)) + (0.02051 × average BW × average BW) + (0.5396 × Ile:Lys) + (1.7284 × Val:Lys) – (0.00795 × Val:Lys × Val:Lys) – (1.7594 × Met:Lys) 1,985.8
ADFI, kg = Predicted ADG ÷ Predicted G:F —
Open in a new tab

1Model adjusted for heterogeneous errors using the inverse of squared SEM.

2ADG = average daily gain. G:F = gain-to-feed ratio. ADFI = average daily feed intake.

3Average BW = average body weight. Amino acids ratios to lysine are expressed on standardized ileal digestible basis.

4BIC = Bayesian information criterion.

For the G:F model, average BW was the single predictor with the lowest BIC (2023.5). The step-wise inclusion of Val:Lys (linear and quadratic terms; P < 0.01, BIC = 2000.2), Met:Lys (linear term; P < 0.01, BIC = 1990.7), and Ile:Lys (linear term; P < 0.01, BIC = 1985.8), resulted in improvements in BIC. The inclusion of other variables to the model did not further improve BIC. The final equation for G:F included average BW, Ile:Lys, Val:Lys, and Met:Lys (BIC = 1985.8; Table 2).

Surprisingly, we could not develop an equation to predict ADFI with the current database. Average BW was the single predictor with the lowest BIC, but the step-wise inclusion of other predictor variables did not improve BIC. Therefore, ADFI estimates were obtained by dividing predicted ADG by predicted G:F, in a procedure similar to Flohr et al. (2018).

The equations suggest that increasing Leu:Lys ratio has a negative effect on predicted performance. This agrees with the well-described metabolism of BCAA (Harper et al., 1984; Brosnan and Brosnan, 2006). While most AA are metabolized in the liver, BCAA are transported to the skeletal muscle to be degraded. Through the action of branched-chain amino-transferase (BCAT), BCAA are reversibly converted to α-keto acids. The α-keto acids of Ile, Leu, and Val are α-keto-β-methylvalerate, α-ketoisocaproate, and α-ketoisovalerate, respectively. In the next step, α-keto acids are transported to the liver, where they are irreversibly decarboxylated by branched-chain α-keto acid dehydrogenase complex (BCKD). Both enzymes BCAT and BCKD are common to all three BCAA. Stimulation of enzymatic activity by one of the BCAA will increase the catabolism of all BCAA, possibly creating antagonisms among them.

Antagonisms among BCAA are expected to be especially detrimental when the excess of a BCAA occurs while the others are marginally meeting or below the requirement. In practice, this situation commonly occurs in swine diets based on corn and/or corn by-products supplemented with feed-grade AA. Because of the AA profile of these ingredients and the increasing use of feed-grade AA in the swine industry, it is common to have diets with high concentrations of Leu while Ile and Val are close to the requirement estimate. This becomes an important issue if we consider that the most potent stimulator of BCAT and BCKD is Leu and its α-keto acid (Harper et al., 1984). In a recent study, Wessels et al. (2016a) fed nursery pigs with 1, 2, or 4% dietary Leu (approximately 100, 200, and 400% SID Leu:Lys) and observed that increasing Leu increased the activity of BCKD and reduced Ile and Val in plasma and several other tissues. This indicates that Leu stimulates not only its own catabolism but also of the other BCAA. As a consequence, Wessels et al. (2016a) observed that ADFI and ADG decreased by approximately 9 and 23% for pigs fed diets with 2 and 4% Leu, respectively, with no impact on G:F.

Excessive Val and Ile seems to have little effect on increasing the catabolism of the other BCAA (D’Mello and Lewis, 1970), which may be partially explained by their lower affinity to BCAT and BCKD compared with Leu (Harper et al., 1984). Moreover, early research (Benton et al., 1956) showed that Leu utilization for growth was decreased by high levels of Val or Ile only when Leu was growth-limiting, which is not a common circumstance in practical swine diets. Therefore, while Leu is undoubtedly the critical BCAA in swine diets regarding antagonisms, Val and Ile seem to require more careful manipulation of AA composition to cause antagonisms (Harper et al., 1984), which seems unrealistic in practical swine diets.

Contrary to Leu, the equations indicate that Ile and Val have positive effects on predicted performance and can potentially overcome the negative effects of high concentrations of Leu. Although both AA present diminishing returns due to the quadratic terms, Ile seems to have smaller capacity to counteract the negative effects of Leu compared with Val. This is mainly driven by the quadratic term for Ile:Lys and in the negative linear term for Ile:Trp, which indicates that higher Ile relative to Trp results in decreased ADG. The ability of Val and Ile to overcome the negative impact of high levels of Leu is not a new concept. Harper et al. (1954) demonstrated that Ile partially recovered decreased growth caused by high Leu diets in rats. Later, Benton et al. (1956) demonstrated that a combination of Ile and Val was needed to fully restore growth performance in rats fed high Leu diets. Although, our meta-regression quantifies the magnitude of the effect under practical conditions of diet formulation with corn by-products and these predictions seem to agree well with the biochemical pathway data reported earlier. Recently, Millet et al. (2015) conducted a study with 10- to 45-kg pigs to evaluate the effects of SID Leu (1.08 or 2.13%), Val (0.70 or 0.78%), Ile (0.60 or 0.73%), and Trp (0.20 or 0.25%), which resulted in ratios to Lys of approximately 97 to 192% SID Leu:Lys, 63 to 70% SID Val:Lys, 54 to 66% SID Ile:Lys, and 18 to 22.5% SID Trp:Lys. Increasing Leu greatly decreased ADG, ADFI, and G:F and Val effectively counteracted the ADG and G:F reductions and partially recovered ADFI. The addition of Ile or Trp did not provide further improvements in growth performance. Thus, it seems that in diets with high Leu concentrations, more Val is needed to ensure optimal growth performance, which may indicate a different Val requirement depending on Leu level (Millet et al., 2015). In a study with nursery pigs, Gloaguen et al. (2011) provided diets with 60 or 70% Val:Lys and 111 or 165% Leu:Lys and observed that pigs fed diets with high Leu:Lys ratios had decreased growth performance, but the reduction was lessened with higher Val:Lys.

On the basis of our model, a similar inference can be made for Ile. However, our equations indicate that Ile alone cannot overcome the negative impact of high levels of Leu on performance and needs to be added in combination with Val and/or Trp, which seems to agree with early research (Harper et al., 1954, 1984; Benton et al., 1956). The potential counteractive effects of Ile and Val in high Leu diets may be explained by transport competition into the brain. Hjelle et al. (1978), using an in vitro model, observed that the transport of Leu through the blood–brain barrier is inhibited by Val. This finding was confirmed by Hargreaves and Pardridge (1988), who observed that Val and, to a lesser extent, Ile, Tyr, Phe, and Trp significantly decreased the transport of Leu into the brain.

The ratio between Ile and Trp is the final component of the ADG prediction equation. As this ratio increases, due to a reduction in Trp and/or an increase in Ile to very high levels, predicted performance is negatively impacted. The LNAA enter the brain via a shared transporter (Pardridge et al., 1977). Increasing levels of one of the LNAA raises its brain uptake and decreases the uptake of the other LNAA (Fernstrom, 2013). Carlsoon and Lindqvist (1978) observed lower Trp concentration in the brain of rats after intraperitoneal injection of Ile, Leu, or Val. Thus, it seems that Ile:Trp in the equations can be at least partially explained by the competition between Ile and Trp for brain uptake, which ultimately results in reduced serotonin activity in the brain and decreased feed intake (Henry et al., 1992). It is well described that low dietary Trp has profound effects on feed intake (Ettle and Roth, 2004; Gonçalves et al., 2015), and high Leu seems to be negatively correlated with brain Trp and serotonin (Wessels et al., 2016a, 2016b). Rogers et al. (1967) observed that in addition to Val and Ile, the inclusion of the LNAA (Trp, Phe, and Thr) was required to recover growth depression caused by a high Leu diet in rats. However, the specific interactions between Ile and Trp is not fully understood. Interestingly, the ratios of Leu or Val to Trp were not found to be significant predictors of performance in either model. However, it is important to note this only means the magnitude was not sufficiently large to be captured, it does not exclude the possibility that Leu:Trp or Val:Trp affect growth performance.

The Met:Lys ratio was a significant predictor of G:F but not ADG. It is important to note that the trials selected for this meta-regression were not designed as Met requirement studies, thus the range of Met:Lys was fairly narrow (i.e., 27 to 56% SID Met:Lys). The NRC (2012) requirement estimate for SID Met:Lys is 29%. On the basis of our equation, a decrease in G:F is observed as SID Met:Lys increases. Edmonds and Baker (1987) observed that high levels of Met result in decreased ADFI and ADG but not G:F in young pigs and Edmonds et al. (1987) observed lower ADG, ADFI, and G:F in weanling pigs fed diets with 4% dl-Met. The relationship between Met and the BCAA is not fully understood. Surprisingly, Langer and Fuller (2000) observed that an excess of Leu or all BCAA improved N utilization in pigs fed Met-limiting diets. In a subsequent study, Langer et al. (2000) suggested that this may be driven by BCKD, which catabolizes BCAA keto acids but also Met keto acids. The addition of excess BCAA would affect Met catabolism by supplying alternative substrates to BCKD, resulting in more Met available for protein synthesis. However, it is important to note that, in our model, there was no evidence for interactive terms between Met and any BCAA.

A hypothetical scenario that simulates a practical situation is shown in Table 3, where grower pig diets based on corn, soybean meal, and DDGS were formulated with low or high addition of feed-grade AA to manipulate the BCAA levels. The diet with low feed-grade AA has a greater Leu:Lys ratio compared with the diet with high inclusion of feed-grade AA. However, the predicted ADG is greater for the low than the high feed-grade AA diet because Ile:Lys and Val:Lys are also increased in the low feed-grade AA diet. The predicted ADG of the high feed-grade AA diet can be restored to the same level of the low feed-grade AA diet by supplementing Val alone or a combination of Val, Ile, and Trp.

Table 3.

Hypothetical scenario for prediction of average daily gain (ADG), gain-to-feed ratio (G:F), and average daily feed intake (ADFI) of 75 kg pigs based on branched-chain and large neutral amino acid concentrations1

Ingredient, % Low AA High AA High AA+Val High AA+Val, Ile High AA+Val, Ile, Trp
Corn 53.71 65.91 65.69 65.72 65.76
Soybean meal 14.42 1.55 1.56 1.56 1.56
Distillers dried grains with solubles 30.00 30.00 30.00 30.00 30.00
Calcium carbonate 0.98 0.90 0.90 0.90 0.90
Monocalcium phosphate — 0.10 0.10 0.10 0.10
Salt 0.25 0.25 0.25 0.25 0.25
l-Lysine HCl 0.22 0.63 0.63 0.63 0.63
dl-Methionine — 0.02 0.02 0.02 0.02
l-Threonine — 0.16 0.16 0.16 0.16
l-Tryptophan — 0.07 0.07 0.07 0.09
l-Valine — — 0.20 0.11 0.06
l-Isoleucine — — — 0.07 0.06
Vitamin–mineral premix 0.43 0.43 0.43 0.43 0.43
Calculated analysis
 Crude protein, % 19.7 15.2 15.3 15.3 15.2
 SID lysine 0.82 0.82 0.82 0.82 0.82
 SID isoleucine:lysine 78 52 52 60 58
 SID leucine:lysine 211 173 173 173 173
 SID valine:lysine 93 67 91 79 74
 SID tryptophan:lysine 18.9 18.9 18.9 18.9 21.3
 SID isoleucine:tryptophan 413 275 275 319 274
 SID methionine:lysine 37 32 32 32 32
Predicted ADG2, g 911 852 911 911 911
 % relative to low AA diet — 93.5 100 100 100
Predicted G:F3, g/kg 360 343 355 354 351
 % relative to low AA diet — 95.3 98.6 98.3 97.5
Calculated ADFI4, kg 2.53 2.48 2.57 2.57 2.60
 % relative to low AA diet — 98.0 101.6 101.6 102.7
Open in a new tab

1Diets were formulated with the NRC (2012) nutrient loading values to meet or exceed the nutrient requirements.

2ADG, g = – 985.94 + (15.2499 × average BW (kg)) – (0.08885 × average BW × average BW) + (1.063 × Leu:Lys) + (20.2659 × Ile:Lys) – (0.1479 × Ile:Lys × Ile:Lys) + (9.2243 × (Ile+Val):Leu) – (0.03321 × (Ile+Val):Leu × (Ile+Val):Leu) – (0.4413 × Ile:Trp).

3G:F, g/kg = 648.3 – (6.2974 × average BW (kg)) + (0.02051 × average BW × average BW) + (0.5396 × Ile:Lys) + (1.7284 × Val:Lys) – (0.00795 × Val:Lys × Val:Lys) – (1.7594 × Met:Lys).

4Calculated ADFI, kg = predicted ADG/predicted G:F.

Our equations for ADG and G:F suggest that ADFI is reduced in high Leu diets. Other research also suggests that ADFI is decreased when pigs are fed diets with high Leu or imbalanced BCAA concentrations (Harper et al., 1984; Gloaguen et al., 2011; Millet et al., 2015; Wessels et al., 2016a; Meyer et al., 2017). A possible explanation for this response is the excessive stimulation of the mammalian target for rapamycin (mTOR), a signaling pathway that stimulates protein synthesis and cell growth in the presence of adequate nutrients (Schmelze and Hall, 2000). Leucine is a potent stimulator of mTOR and consequently plays an important role in protein synthesis (Suryawan et al., 2011). However, excessive mTOR signaling caused by high levels of Leu seems to impact growth performance through a strong inhibitory effect on feed intake (Cota et al., 2006). Interestingly, the effect on feed intake is not observed for Val (Cota et al., 2006). The available literature also supports the negative effects of increasing Leu on G:F (Gloaguen et al., 2011; Millet et al., 2015; Duan et al., 2016).

However, a few studies did not observe evidence for changes in any growth performance variable even at moderate to high Leu concentrations (Cisneros et al., 1996; Hyun et al., 2007; Morales et al., 2016). Although reasons are not well-defined, those diets were also well above the Val and Ile requirement estimates. According to our model, this could indicate that Val and Ile could be potentially counteracting the negative effect of high Leu on growth performance.

In our prediction equations, the combination of AA used to minimize the negative impact of Leu can be changed by the nutritionist to optimize feed cost depending on current AA costs. Other hypothetical scenarios with corn–soybean meal diets or different amounts of DDGS produce similar results. The implementation of prediction equations in a diet formulation is a straightforward process that would allow the nutritionist to visualize the predicted performance based on BCAA and LNAA concentrations. Field research trials are required to validate the predicted growth performance provided by the models. If proven correct, the prediction equations provide an important tool that allows nutritionists to formulate practical diets with high inclusion levels of feed-grade AA and overcome the negative effects of excessive Leu:Lys.

Conflict of interest statement. The authors declare that they have no conflict of interest.

Footnotes

Contribution no. 19-244-J from the Kansas Agricultural Experiment Station. This work was supported by the USDA National Institute of Food and Agriculture, Hatch Funding project 1007039.

LITERATURE CITED

  1. Benton D. A., Harper A. E., Spivey H. E., and Elvehjem C. A.. . 1956. Leucine, isoleucine and valine relationships in the rat. Arch. Biochem. Biophys. 60:147–155. doi: 10.1016/0003-9861(56)90406-4 [DOI] [PubMed] [Google Scholar]
  2. Bergstrom J. R., J. L. Nelssen M. D. Tokach S. S. Dritz R. D. Goodband, and DeRouchey J. M.. 2014. The effects of feeder design and dietary dried distillers’ grains with solubles on the performance and carcass characteristics of finishing pigs. J. Anim. Sci. 92:3591–3597. doi: 10.2527/jas.2014-7686 [DOI] [PubMed] [Google Scholar]
  3. Brosnan J. T., and Brosnan M. E.. . 2006. Branched-chain amino acids: enzyme and substrate regulation. J. Nutr. 136:207–211. doi: 10.1093/jn/136.1.207S [DOI] [PubMed] [Google Scholar]
  4. Castilha L. D., Pozza P. C., Nunes R. V., Lazzeri D. B., Somensi M. L., and Pozza M. S. S.. . 2012. Levels of digestible isoleucine on performance, carcass traits and organs weight of gilts (15–30 kg). Cienc. Agrotec. 36:446–453. doi: 10.1590/S1413-70542012000400009 [DOI] [Google Scholar]
  5. Carlsoon A. and Lindqvist M.. . 1978. Dependence of 5-HT and catecholamine synthesis on concentrations of precursor amino-acids in rat brain. Naunyn Schmiedebergs Arch. Pharmacol. 303:157–164. doi: 10.1007/BF00508062 [DOI] [PubMed] [Google Scholar]
  6. Cisneros F., Ellis M., Baker D. H., Easter R. A., and McKeith F. K.. . 1996. The influence of short-term feeding of amino acid-deficient diets and high dietary leucine levels on the intramuscular fat content of pig muscle. Anim. Sci. 63:517–522. doi: 10.1017/S1357729800015411 [DOI] [Google Scholar]
  7. Clark A. B., Tokach M. D., DeRouchey J. M., Dritz S. S., Goodband R. D., Woodworth J. C., Touchette K. J., and Bello N. M.. . 2017. Modeling the effects of standardized ileal digestible valine to lysine ratio on growth performance of nursery pigs. Trans. Anim. Sci. 1:448–457. doi: 10.2527/tas2017.0049 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Coble K. F., DeRouchey J. M., Tokach M. D., Dritz S. S., Goodband R. D., and Woodworth J. C.. . 2017. Effects of distillers dried grains with solubles and added fat fed immediately before slaughter on growth performance and carcass characteristics of finishing pigs. J. Anim. Sci. 95:270–278. doi: 10.2527/jas2016.0679 [DOI] [PubMed] [Google Scholar]
  9. Cota D., K. Proulx K. A. Smith S. C. Kozma G. Thomas S. C. Woods, and Seeley R. J.. 2006. Hypothalamic mtor signaling regulates food intake. Science 312:927–930. doi: 10.1126/science.1124147 [DOI] [PubMed] [Google Scholar]
  10. Cromwell G. L., M. J. Azain O. Adeola S. K. Baidoo S. D. Carter T. D. Crenshaw S. W. Kim D. C. Mahan P. S. Miller, and Shannon M. C.; North Central Coordinating Committee on Swine Nutrition 2011. Corn distillers dried grains with solubles in diets for growing-finishing pigs: a cooperative study. J. Anim. Sci. 89:2801–2811. doi: 10.2527/jas.2010-3704 [DOI] [PubMed] [Google Scholar]
  11. Dean D. W., L. L. Southern B. J. Kerr, and Bidner T. D.. 2005. Isoleucine requirement of 80- to 120-kilogram barrows fed corn-soybean meal or corn-blood cell diets. J. Anim. Sci. 83:2543–2553. doi: 10.2527/2005.83112543x. [DOI] [PubMed] [Google Scholar]
  12. D’Mello J. P. F., and Lewis D.. . 1970. Amino acid interactions in chick nutrition. 2. Interrelationships between leucine, isoleucine and valine. Br. Poult. Sci. 16:607–615. doi: 10.1080/00071667008415821 [DOI] [PubMed] [Google Scholar]
  13. Duan Y. H., Zeng L. M., Li F. N., Li Y. H., Tan B. E., Ji Y. J., Kong X. F., Tang Y. L., Zhang Y. Z., and Yin Y. L.. . 2016. Effects of dietary branched-chain amino acid ratio on growth performance and serum amino acid pool of growing pigs. J. Anim. Sci. 201694:129–134. doi: 10.2527/jas2015-9527 [DOI] [Google Scholar]
  14. Edmonds M. S., and Baker D. H.. . 1987. Amino acid excesses for young pigs: effects of excess methionine, tryptophan, threonine or leucine. J. Anim. Sci. 64:1664–1671. doi:10.2527/jas1987.6461664x [DOI] [PubMed] [Google Scholar]
  15. Edmonds M. S., H. W. Gonyou, and Baker D. H.. 1987. Effect of excess levels of methionine, tryptophan, arginine, lysine or threonine on growth and dietary choice in the pig. J. Anim. Sci. 65:179–185. doi:10.2527/jas1987.651179x [DOI] [PubMed] [Google Scholar]
  16. Ettle T., and Roth F. X.. . 2004. Specific dietary selection for tryptophan by the piglet. J. Anim. Sci. 82:1115–1121. doi: 10.2527/2004.8241115x [DOI] [PubMed] [Google Scholar]
  17. Fernstrom J. D. 2005. Branched-chain amino acids and brain function. J. Nutr. 135(6 Suppl):1539S–1546S. doi: 10.1093/jn/135.6.1539S [DOI] [PubMed] [Google Scholar]
  18. Fernstrom J. D. 2013. Large neutral amino acids: dietary effects on brain neurochemistry and function. Amino Acids 45:419–430. doi: 10.1007/s00726-012-1330-y [DOI] [PubMed] [Google Scholar]
  19. Figueroa J. L., A. J. Lewis P. S. Miller R. L. Fischer, and Diedrichsen R. M.. 2003. Growth, carcass traits, and plasma amino acid concentrations of gilts fed low-protein diets supplemented with amino acids including histidine, isoleucine, and valine. J. Anim. Sci. 81:1529–1537. doi: 10.2527/2003.8161529x [DOI] [PubMed] [Google Scholar]
  20. Flohr J. R., S. S. Dritz M. D. Tokach J. C. Woodworth J. M. DeRouchey, and Goodband R. D.. 2018. Development of equations to predict the influence of floor space on average daily gain, average daily feed intake and gain: feed ratio of finishing pigs. Animal 12:1022–1029. doi: 10.1017/S1751731117002440 [DOI] [PubMed] [Google Scholar]
  21. Gaines A. M., D. C. Kendall G. L. Allee J. L. Usry, and Kerr B. J.. 2011. Estimation of the standardized ileal digestible valine-to-lysine ratio in 13- to 32-kilogram pigs. J. Anim. Sci. 89:736–742. doi: 10.2527/jas.2010-3134 [DOI] [PubMed] [Google Scholar]
  22. Gloaguen M., N. Le Floc’h L. Brossard R. Barea Y. Primot E. Corrent, and van Milgen J.. 2011. Response of piglets to the valine content in diet in combination with the supply of other branched-chain amino acids. Animal 5:1734–1742. doi: 10.1017/S1751731111000760 [DOI] [PubMed] [Google Scholar]
  23. Gonçalves M. A., S. Nitikanchana M. D. Tokach S. S. Dritz N. M. Bello R. D. Goodband K. J. Touchette J. L. Usry J. M. DeRouchey, and Woodworth J. C.. 2015. Effects of standardized ileal digestible tryptophan: lysine ratio on growth performance of nursery pigs. J. Anim. Sci. 93:3909–3918. doi: 10.2527/jas.2015-9083 [DOI] [PubMed] [Google Scholar]
  24. Graham A. B., Goodband R. D., Tokach M. D., Dritz S. S., DeRouchey J. M., and Nitikanchana S.. . 2014. The effects of medium-oil dried distillers grains with solubles on growth performance, carcass traits, and nutrient digestibility in growing–finishing pigs. J. Anim. Sci. 92:604–611. doi: 10.2527/jas2013-6798 [DOI] [PubMed] [Google Scholar]
  25. Hargreaves K. M., and Pardridge W. M.. . 1988. Neutral amino acid transport at the human blood-brain barrier. J. Biol. Chem. 263:19392–19397. [PubMed] [Google Scholar]
  26. Harper A. E., Benton D. A., Winje M. E., and Elvehjem C. A.. . 1954. Leucine-isoleucine antagonism in the rat. Arch. Biochem. Biophys. 51:523–524. doi: 10.1016/0003-9861(54)90509-3 [DOI] [PubMed] [Google Scholar]
  27. Harper A. E., R. H. Miller, and Block K. P.. 1984. Branched-chain amino acid metabolism. Annu. Rev. Nutr. 4:409–454. doi: 10.1146/annurev.nu.04.070184.002205 [DOI] [PubMed] [Google Scholar]
  28. Henry Y., B. Sève Y. Colléaux P. Ganier C. Saligaut, and Jégo P.. 1992. Interactive effects of dietary levels of tryptophan and protein on voluntary feed intake and growth performance in pigs, in relation to plasma free amino acids and hypothalamic serotonin. J. Anim. Sci. 70:1873–1887. doi:10.2527/1992.7061873x [DOI] [PubMed] [Google Scholar]
  29. Hjelle J. T., J. Baird-Lambert G. Cardinale S. Specor, and Udenfriend S.. 1978. Isolated microvessels: the blood-brain barrier in vitro. Proc. Natl. Acad. Sci. USA. 75:4544–4548. doi:10.1073/pnas.75.9.4544 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Huepa L. M. D., Fachinello M. R., Esteves L. A. C., Paula V. R. C., Ferreira S. L., Pasquetti T. J., Castilha L. D., Vasconcellos R. S., and Pozza P. C.. . 2017. Leucine levels in low protein diets for pigs in the initial phase. Cienc. Agr. 38:3829–3840. doi: 10.5433/1679-0359.2017v38n6p3829 [DOI] [Google Scholar]
  31. Hyun Y., Kim J. D., Ellis M., Peterson B. A., Baker D. H., and McKeith F. K.. . 2007. Effect of dietary leucine and lysine levels on intramuscular fat content in finishing pigs. Can. J. Anim. Sci. 87:303–306. doi: 10.4141/CJAS06042 [DOI] [Google Scholar]
  32. Kass R. E. and Raftery A. E.. . 1995. Bayes factors. J. Am. Stat. Assoc. 90:773–795. doi: 10.1080/01621459.1995.10476572 [DOI] [Google Scholar]
  33. Kurose Y., and Terashima Y.. . 1999. Histamine regulates food intake through modulating noradrenaline release in the para-ventricular nucleus. Brain Res. 828:115–118. doi:10.1016/S0006-8993(99)01339-6 [DOI] [PubMed] [Google Scholar]
  34. Langer S., and Fuller M. F.. . 2000. Interactions among the branched-chain amino acids and their effects on methionine utilization in growing pigs: effects on nitrogen retention and amino acid utilization. Br. J. Nutr. 83:43–48. doi: 10.1017/S0007114500000076 [DOI] [PubMed] [Google Scholar]
  35. Langer S., P. W. Scislowski D. S. Brown P. Dewey, and Fuller M. F.. 2000. Interactions among the branched-chain amino acids and their effects on methionine utilization in growing pigs: effects on plasma amino- and keto-acid concentrations and branched-chain keto-acid dehydrogenase activity. Br. J. Nutr. 83:49–58. doi:10.1017/S0007114500000088 [PubMed] [Google Scholar]
  36. Lazzeri D. B., Castilha L. D., Costa P. B., Nunes R. V., Pozza M. S. S., and Pozza P. C.. . 2017. Standardized ileal digestible (SID) isoleucine requirement of barrows (15- to 30-kg) fed low crude protein diets. Cienc. Agr. 38:3283–3294. doi: 10.5433/1679-0359.2017v38n5p3283 [DOI] [Google Scholar]
  37. Lee J. W., F. K. McKeith, and Stein H. H.. 2012. Up to 30% corn germ may be included in diets fed to growing-finishing pigs without affecting pig growth performance, carcass composition, or pork fat quality. J. Anim. Sci. 90:4933–4942. doi: 10.2527/jas.2012-5129 [DOI] [PubMed] [Google Scholar]
  38. Lewis A. J., and Nishimura N.. . 1995. Valine requirement of the finishing pig. J. Anim. Sci. 73:2315–2318. doi:10.2527/1995.7382315x [DOI] [PubMed] [Google Scholar]
  39. Linneen S. K., DeRouchey J. M., Dritz S. S., Goodband R. D., Tokach M. D., and Nelssen J. L.. . 2008. Effects of dried distillers grains with solubles on growing and finishing pig performance in a commercial environment. J. Anim. Sci. 86:1579–1587. doi: 10.2527/jas.2007-0486 [DOI] [PubMed] [Google Scholar]
  40. Liu X. T., W. F. Ma X. F. Zeng C. Y. Xie P. A. Thacker J. K. Htoo, and Qiao S. Y.. 2015. Estimation of the standardized ileal digestible valine to lysine ratio required for 25- to 120-kilogram pigs fed low crude protein diets supplemented with crystalline amino acids. J. Anim. Sci. 93:4761–4773. doi: 10.2527/jas.2015-9308 [DOI] [PubMed] [Google Scholar]
  41. Mavromichalis I., B. J. Kerr T. M. Parr D. M. Albin V. M. Gabert, and Baker D. H.. 2001. Valine requirement of nursery pigs. J. Anim. Sci. 79:1223–1229. doi:10.2527/2001.7951223x [DOI] [PubMed] [Google Scholar]
  42. Meyer F., C. Jansen van Rensburg, and Gous R. M.. 2017. The response of weaned piglets to dietary valine and leucine. Animal 11:1279–1286. doi: 10.1017/S1751731116002834 [DOI] [PubMed] [Google Scholar]
  43. Millet S., M. Aluwé B. Ampe, and De Campeneere S.. 2015. Interaction between amino acids on the performances of individually housed piglets. J. Anim. Physiol. Anim. Nutr. (Berl). 99:230–236. doi: 10.1111/jpn.12227. [DOI] [PubMed] [Google Scholar]
  44. Morales A., N. Arce M. Cota L. Buenabad E. Avelar J. K. Htoo, and Cervantes M.. 2016. Effect of dietary excess of branched-chain amino acids on performance and serum concentrations of amino acids in growing pigs. J. Anim. Physiol. Anim. Nutr. (Berl). 100:39–45. doi: 10.1111/jpn.12327 [DOI] [PubMed] [Google Scholar]
  45. Nitikanchana S., S. S. Dritz M. D. Tokach J. M. DeRouchey R. D. Goodband, and White B. J.. 2015. Regression analysis to predict growth performance from dietary net energy in growing-finishing pigs. J. Anim. Sci. 93:2826–2839. doi: 10.2527/jas.2015-9005 [DOI] [PubMed] [Google Scholar]
  46. NRC 2012. Nutrient requirements of swine. 11th rev. ed.National Academy Press, Washington, DC. [Google Scholar]
  47. Overholt M. F., Lowell J. E., Arkfeld E. K., Grossman I. M., Stein H. H., Dilger A. C., and Boler D. D.. . 2016. Effects of pelleting diets without or with distillers’ dried grains with solubles on growth performance, carcass characteristics, and gastrointestinal weights of growing–finishing barrows and gilts. J. Anim. Sci. 94:2172–2183. doi: 10.2527/jas2015-0202 [DOI] [PubMed] [Google Scholar]
  48. Pardridge W. M. 1977. Kinetics of competitive inhibition of neutral amino acid transport across the blood-brain barrier. J. Neurochem. 28:103–108. doi:10.1111/j.1471-4159.1977.tb07714.x [DOI] [PubMed] [Google Scholar]
  49. Rogers Q. R., R. I. Tannous, and Harper A. E.. 1967. Effects of excess leucine on growth and food selection. J. Nutr. 91:561–572. doi: 10.1093/jn/91.4.561 [DOI] [PubMed] [Google Scholar]
  50. Schmelze T., and Hall M. N.. . 2000. TOR, a central controller of cell growth. Cell 2:253–262. doi: 10.1016/S0092-8674(00)00117-3 [DOI] [PubMed] [Google Scholar]
  51. St-Pierre N. R. 2001. Integrating quantitative findings from multiple studies using mixed model methodology. J. Dairy Sci. 84:741–755. doi: 10.3168/jds.S0022-0302(01)74530-4 [DOI] [PubMed] [Google Scholar]
  52. Suryawan A., R. A. Orellana M. L. Fiorotto, and Davis T. A.. 2011. Triennial growth symposium: leucine acts as a nutrient signal to stimulate protein synthesis in neonatal pigs. J. Anim. Sci. 89:2004–2016. doi: 10.2527/jas.2010-3400 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Waguespack A. M., T. D. Bidner R. L. Payne, and Southern L. L.. 2012. Valine and isoleucine requirement of 20- to 45-kilogram pigs. J. Anim. Sci. 90:2276–2284. doi: 10.2527/jas.2011-4454 [DOI] [PubMed] [Google Scholar]
  54. Weber T. E., S. L. Trabue C. J. Ziemer, and Kerr B. J.. 2010. Evaluation of elevated dietary corn fiber from corn germ meal in growing female pigs. J. Anim. Sci. 88:192–201. doi: 10.2527/jas.2009-1896 [DOI] [PubMed] [Google Scholar]
  55. Wessels A. G., H. Kluge F. Hirche A. Kiowski A. Schutkowski E. Corrent J. Bartelt B. König, and Stangl G. I.. 2016a. High leucine diets stimulate cerebral branched-chain amino acid degradation and modify serotonin and ketone body concentrations in a pig model. PLoS One 11:e0150376. doi: 10.1371/journal.pone.0150376 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Wessels A. G., Kluge H., Hirche F., Kiowski A., Bartelt J., Corrent E., and Stangl G. I.. . 2016b. High leucine intake reduces the concentration of hypothalamic serotonin in piglets. J. Anim. Sci. 94:26–29. doi: 10.2527/jas2015-9728 [DOI] [Google Scholar]
  57. Whitney M. H., and Shurson G. C.. . 2004. Growth performance of nursery pigs fed diets containing increasing levels of corn distiller’s dried grains with solubles originating from a modern midwestern ethanol plant. J. Anim. Sci. 82:122–128. doi:10.2527/2004.821122x [DOI] [PubMed] [Google Scholar]
  58. Whitney M. H., G. C. Shurson L. J. Johnston D. M. Wulf, and Shanks B. C.. 2006. Growth performance and carcass characteristics of grower-finisher pigs fed high-quality corn distillers dried grain with solubles originating from a modern midwestern ethanol plant. J. Anim. Sci. 84:3356–3363. doi: 10.2527/jas.2006-099 [DOI] [PubMed] [Google Scholar]
  59. Xu G., Baidoo S. K., Johnston L. J., Bibus D., Cannon J. E., and Shurson G. C.. . 2010. Effects of feeding diets containing increasing content of corn distillers dried grains with solubles to grower-finisher pigs on growth performance, carcass composition, and pork fat quality. J. Anim. Sci. 88:1398– 1410. doi: 10.2527/jas.2008-1404 [DOI] [PubMed] [Google Scholar]
  60. Xu Y. T., X. K. Ma C. L. Wang M. F. Yuan, and Piao X. S.. 2018. Effects of dietary valine:lysine ratio on the performance, amino acid composition of tissues and mrna expression of genes involved in branched-chain amino acid metabolism of weaned piglets. Asian-Australas. J. Anim. Sci. 31:106–115. doi: 10.5713/ajas.17.0148 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Journal of Animal Science are provided here courtesy of Oxford University Press

RESOURCES