Effects of dietary valine/isoleucine ratio and slaughter weight on the composition of pork flavor precursors

Abstract

This study aimed to investigate the effects of dietary valine/isoleucine (V/I) ratio and slaughter weight on the profile of pork flavor precursors. Metabolomics and volatilomic profiling were performed on meat harvested from a 2 × 2 factorial feeding study. Thirty-six Duroc × Landrace × Yorkshire castrated male pigs with around 75 kg body weight (BW) were randomly allocated to two dietary treatments, involving V/I (1.23 and 2.60 at 75 to 100 kg BW, 1.24 and 2.39 at 100 to 135 kg BW for the normal dietary valine/isoleucine ratio defined N [V/I] and high dietary valine/isoleucine ratio defined H [V/I], respectively) and two slaughter weights (lower weight [LW] 100 kg and higher weight [HW] 130 kg). Each group consisted of six replicates of three pigs per replicate. The trial lasted for 63 d. Dietary V/I significantly affected several metabolic pathways in muscle, including alanine, aspartate and glutamate metabolism, fructose and mannose metabolism, and amino sugar and nucleotide sugar metabolism. On the other hand, contents of acylcarnitine (ACar; P < 0.001) and lysophosphatidylethanolamine (LPE; P = 0.036) were significantly higher in the meat of the HW pigs than those of the LW pigs. Among 48 volatile organic compounds (VOCs) identified in pork, the concentrations of seven VOCs were altered by dietary V/I, that of 17 VOCs by slaughter weight, and acetoin and 1-heptanol by both factors (P < 0.05). In addition, increased dietary V/I altered 53 hydrophilic compounds and 10 VOCs in LW pigs, and altered 39 hydrophilic compounds and six VOCs in HW pigs (P < 0.05). In conclusion, the increased dietary V/I significantly altered flavor precursors of pork, but this effect decreased as the slaughter weight increased.

1. Introduction

Flavor is an important edible characteristic of meat that directly affects consumer acceptance (Arshad et al., 2018). The development of meat flavor depends on a chain of chemical reactions during processing or cooking. Specifically, the degradation of fatty acids and Maillard reaction produce various flavor precursors that further react to give rise to many volatile organic compounds (VOCs), such as ketones, aldehydes, and heterocyclic compounds (Whitfield and Mottram, 1992). In particular, unsaturated fatty acids such as oleic acid, linolenic acid, and linoleic acid are key components in the formation of aroma compounds in meat (Sohail et al., 2022). Several factors including genetics, feed, age, live animal handling, and subsequent meat handling affect the production of these VOCs in the meat (Aaslyng and Meinert, 2017; Lu et al., 2008). Dietary amino acid composition can also affect the content of VOCs in meat (Ma et al., 2020). For example, supplementation with 1% arginine reduced aldehyde content and increased ketone content in lamb meat (Dou et al., 2023), whereas dietary supplementation of glutamate and arginine increased aldehyde concentrations in pork (Guo et al., 2019). The branched-chain amino acids (BCAA) play a role in muscle energy metabolism and protein synthesis, and BCAA supplementation ameliorated the poor growth performance of pigs (Kamei et al., 2020; Zhang et al., 2021b). However, their effects on meat flavor have not been reported. Our previous studies found that dietary supplementation of isoleucine and valine increased intramuscular fat (IMF) content and reduced drip loss of pork (Luo et al., 2018; Xu et al., 2020). Furthermore, a reduced dietary valine/isoleucine (V/I) ratio significantly increased water holding capacity (WHC) of pork (Xu et al., 2022). Pork with poor WHC easily loses water and has a high level of protein oxidation (Traore et al., 2012), which may affect the Maillard reaction during cooking. Changes in muscle free amino acid content can influence meat taste, and VOCs production through Maillard reactions (Liu et al., 2020). Therefore, it was hypothesized that dietary V/I affects the contents of metabolites in pork and the production of flavor precursors during cooking.

Slaughter weight is an important variable that affects the productivity and profitability of finishing pigs. To increase profitability, slaughter weights have risen over the past 10 years (Wu et al., 2017). China is the largest pork producer, accounting for more than 40% of global pork yield (Yan et al., 2023). In China, the slaughter weight have increased from 110 kg to over 130 kg (Xu et al., 2022), to improve flavour and reduce the costs for piglets. The increase in slaughter weight of finishing pigs could be detrimental to the efficiency of pork production, such as increased average daily feed intake (ADFI) and back fat thickness, but decreased feed efficiency. Previously, we also showed that slaughter weight had a significant effect on carcass traits, meat quality, and fatty acid concentrations in pork, and the effect of dietary V/I on meat quality was influenced by slaughter weight (Xu et al., 2022). Nevertheless, it is unknown whether dietary V/I has an impact on pork flavor precursors and whether this effect is influenced by slaughter weight.

Therefore, this study aimed to explore the effects of dietary V/I and slaughter weight on the composition of flavor precursors of pork, including lipidomic metabolites, hydrophilic metabolites, and VOCs, using metabolomics and volatilomic profiling.

2. Materials and methods

2.1. Animal ethics statement

All the procedures employed in this study were approved by the Institutional Animal Care and Use Committee of China Agriculture University (No. SKLAB-B-2010–003).

2.2. Experimental design

Thirty-six Duroc × Landrace × Yorkshire castrated male pigs with an initial body weight (BW) of about 75 kg were randomly allocated to two dietary treatments. Each treatment involved six replicates, and three pigs per replicate (pen). Experimental diets were formulated based on corn-soybean meal with the addition of crystalline amino acids to meet or exceed the nutritional requirements of pigs with a BW ranging from 75 to 100 kg and 100 to 135 kg (NRC, 2012), except that the levels of valine and isoleucine were provided according to experiment design. Low protein diets were formulated with specified ratios of valine/isoleucine. To prevent the interference on pig feed intake from dietary supplementation of crystal amino acids and minerals, the sweetener was added in the diets. Dietary V/I that met the nutritional requirements of NRC (2012) for pigs was defined as N (V/I), and the higher V/I was defined as H (V/I). That is, pigs with BW ranging from 75 to 100 kg were fed diets with N (V/I) of 1.23 and H (V/I) of 2.60, respectively, and pigs with BW ranging from 100 to 135 kg with N (V/I) = 1.24 and H (V/I) = 2.39 as well. Details of animal management were described previously (Xu et al., 2020). Ingredient composition and nutritional levels of diets are shown in Table 1. Dietary crude protein levels were analyzed according to the procedures of AOAC (2006) method 984.13 using a Kjeldahl nitrogen determination system (Kjeldahl 8100, FOSS Analytical A/S, Hillerød, Denmark). The contents of amino acids in diets were analyzed by High-performance liquid chromatography (HPLC, L-8900, Hitachi High-Tech Corp., Tokyo, Japan) according to the method described previously (Heinrikson and Meredith, 1984). Standardized ideal digestible (SID) amino acid, digestible energy, and metabolizable energy levels were calculated according to NRC (2012). During feeding trials, pigs had ad libitum access to the diets and clean drinking water for 63 d. At the end of each experimental phase, the pigs were fasted overnight for 12 h with free access to water, and the individual weight was measured. The feed intake of each group was recorded every week. Based on these data, the average daily gain (ADG), ADFI, and feed conversion rate (FCR) were calculated for each group.

Table 1.

The composition and nutrient content of the experimental diets provided for finishing pigs (% as-fed basis).

Items	75–100 kg BW		100–135 kg BW
	N (V/I)1	H (V/I)1	N (V/I)1	H (V/I)1
Ingredients
Corn	88.70	88.70	90.75	90.75
Soybean meal	2.90	2.90	1.50	1.50
Wheat bran	2.50	2.50	2.26	2.26
Soybean oil	1.80	1.80	1.80	1.80
Limestone	0.96	0.96	0.96	0.96
Dicalcium phosphate	0.40	0.40	0.40	0.40
Salt	0.35	0.35	0.35	0.35
L-Lysine HCl	0.63	0.63	0.52	0.52
DL-Methionine	0.12	0.12	0.07	0.07
L-Threonine	0.23	0.23	0.19	0.19
L-Tryptophan	0.08	0.08	0.07	0.07
L-Isoleucine	0.14	0.01	0.10	0.00
L-Valine	0.15	0.30	0.10	0.23
L-Histidine HCl	0.10	0.10	0.06	0.06
L-Phenylalanine	0.09	0.09	0.04	0.04
L-Alanine	0.18	0.16	0.16	0.13
50% Choline chloride	0.08	0.08	0.08	0.08
Sweetener	0.08	0.08	0.08	0.08
Phytase2	0.01	0.01	0.01	0.01
Premix3	0.50	0.50	0.50	0.50
Total	100.00	100.00	100.00	100.00
Analyzed nutrient levels
Crude protein	10.39	10.98	9.97	10.11
Lysine	0.79	0.78	0.70	0.61
Methionine + cysteine	0.45	0.45	0.39	0.39
Threonine	0.52	0.49	0.46	0.42
Tryptophan	0.17	0.17	0.15	0.15
Isoleucine	0.43	0.30	0.38	0.26
Leucine	1.00	0.93	0.97	0.86
Valine	0.59	0.66	0.50	0.63
Calculated nutrient levels4
DE, MJ/kg	14.19	14.19	14.24	14.24
ME, MJ/kg	13.92	13.92	13.81	13.81
Crude protein	10.83	10.83	10.01	10.01
Standardized ileal digestible amino acids4
Lysine	0.73	0.73	0.61	0.61
Methionine + cysteine	0.42	0.42	0.36	0.36
Threonine	0.46	0.46	0.40	0.40
Tryptophan	0.13	0.13	0.11	0.11
Isoleucine	0.39	0.25	0.33	0.23
Leucine	0.89	0.89	0.87	0.87
Valine	0.48	0.65	0.41	0.55
Valine/Isoleucine	1.23	2.60	1.24	2.39

BW = body weight; DE = digestible energy; ME = metabolizable energy; N (V/I) = normal dietary valine/isoleucine ratio; H (V/I) = high dietary valine/isoleucine ratio.

¹N (V/I) means that dietary valine/isoleucine (V/I) met the nutritional requirements of NRC (2012) for pigs, and H (V/I) is higher V/I.

²Branded as Maite 5000, is purchased from VTR Biotech Co., Ltd., Guangdong, China.

³The premix provided the following per kg of diets: vitamin A, 6000 IU; vitamin D₃, 2400 IU; vitamin E, 20 IU; vitamin K₃, 2 mg; vitamin B₁, 0.96 mg; vitamin B₂, 4 mg; vitamin B₆, 2 mg; vitamin B₁₂, 0.012 mg; biotin, 0.04 mg; folic acid 0.40 mg; pantothenic acid 11.2 mg; nicotinic acid 22 mg; Cu, 120 mg; Fe, 76 mg; Mn, 12 mg; Zn, 76 mg; I, 0.24 mg; Se, 0.40 mg.

⁴The values were calculated using data provided by NRC (2012).

2.3. Sample collection

While pig BW reached about 100 and 130 kg, one pig close to the assigned BW was selected from each pen, transported to the local abattoir and humanely slaughtered with electric stunning and exsanguination. At 30 min postmortem, approximately 100 g of chops were sampled from the M. longissimus thoracis of the 10th to 12th ribs on the left side of each carcass, and quickly immersed in liquid nitrogen and stored at −80 °C for the following analysis. After the experiment, the remaining pigs were also transported to the local abattoir and humanely slaughtered by electric stunning and exsanguination.

2.4. Metabolomics analysis

Pork samples (25 mg) were homogenized in the extraction solution (methanol:water = 3:1, with isotope-labeled internal standard; 500 μL) at 35 Hz for 4 min and sonicated for 5 min in an ice-water bath, and repeated 3 times. Then, the samples were incubated at −40 °C for 1 h and centrifuged at 13,800 × g for 15 min at 4 °C. The supernatant was transferred into a fresh glass vial for further analysis.

Metabolomics was performed using a HPLC-mass spectrometer/mass spectrometer (LC-MS/MS), an UHPLC system (Vanquish, Thermo Fisher Scientific Inc., Waltham, MA, USA) equipped with a UPLC HSS T3 column (2.1 mm × 100 mm, 1.8 μm), and coupled with an Orbitrap Exploris 120 mass spectrometer (Orbitrap MS, Thermo Fisher Scientific Inc., Waltham, MA, USA). In running, mobile phase A of the aqueous phase contained 5 mmol/L ammonium acetate and 5 mmol/L acetic acid, and mobile phase B was acetonitrile. The injection volume of samples was 2 μL and the temperature of auto-sampler was 4 °C.

An Orbitrap Exploris 120 mass spectrometer was used to collect primary and secondary mass spectrometry data under the control of the acquisition software (Xcalibur, Thermo Fisher Scientific Inc., Waltham, MA, USA). The ESI source conditions were set as follows: sheath gas flow rate as 50 arbitrary units (a.u.), aux gas flow rate as 15 a.u., capillary temperature as 320 °C, full MS resolution as 60,000, MS/MS resolution as 15,000, collision energy as 10/30/60 in NCE mode, the spray voltage as 3.8 kV in the positive ion mode or −3.4 kV in the negative ion mode. The ProteoWizard software (ProteoWizard Software Foundation, Apache License 2.0, hosted on SourceForge) was applied to convert the raw data into mzML format, followed by peak extraction, alignment, and integration with R based on the XCMS. An in-house MS2 database was applied in metabolite annotation. The cutoff for annotation was set at 0.3.

2.5. Volatile compound analysis

Volatile compounds of meat (200 ± 5 mg) with an internal standard of 2-octanol (10 mg/L dH₂O; 10 μL) were extracted by solid-phase microextraction (SPME) as follows: incubated at 60 °C, preheated for 15 min, incubated for 30 min and desorbed for 4 min, then analyzed on a gas chromatograph-mass spectrometry (GC–MS) using a GC system (Agilent 7890, Agilent Technologies Inc., Santa Clara, CA, USA) coupled with a mass spectrometer (Agilent 5977B, Agilent Technologies Inc.). The system used a DB-Wax (30 m × 250 μm × 0.25 μm; Agilent Technologies Inc.) and injected in Splitless Mode. The carrier gas used in the system was helium, with the front inlet purge flow of 3 mL/min and the flow rate of the gas through the column of 1 mL/min. The heating program started at an initial temperature of 40 °C, maintained for 4 min. Then, the temperature was increased to 245 °C at a rate of 5 °C/min and remained at 245 °C for 5 min. The temperatures of quad, ion source, injection and transfer line were 150, 230, 250, and 250 °C, respectively. In electron impact mode, the energy was −70 eV and the solvent delay time was 0 min. The mass spectrometer scanned a mass-to-charge (m/z) range of 20 to 400.

Raw data peaks were processed using Chroma TOF software (v4.3x, LECO), and the resulting spectra were identified by searching against the NIST database. The identification results with forward and reverse match factors both > 80 are reported. The calculation formula for VOCs is as follows:

$$\text{VOC}\left(\mu \mathrm{g}/\text{kg}\right)=P\left(V\right)/P\left(I\right)\times C\left(I\right)\text{,}$$

where P(V), peak area of the VOCs of pork; P(I), peak area of the internal standard substance; C(I), concentration of the internal standard substance.

The odor thresholds of volatile compounds were obtained by referring to the relevant literatures. Volatile compounds with an odor activity value (OAV) greater than 1 were identified as characteristic volatile compounds (Yang et al., 2022). The calculation formula is as follows:

$$\text{OAV}=\left(\text{Concentration}\text{of}\text{the}\text{compound}\right)/\left(\text{Odor}\text{threshold}\text{of}\text{the}\text{compound}\right)\text{.}$$

2.6. Statistical analysis

Principal component analysis (PCA) and orthogonal partial least squares discrimination analysis (OPLS-DA) were applied for multivariate analysis of metabolomics and volatile compounds using SIMCA (Sartorius Stedim Data Analytics AB, Umea, Sweden). Pathway analysis was conducted using MetaboAnalyst 6.0 online (https://www.metaboanalyst.ca/). GraphPad Prism 8.0 (GraphPad Inc., La Jolla, CA, USA) was used to draw bar charts and pie charts, and to perform Pearson's correlation analysis between lipids and VOCs (OAV > 1). The changes in flavor precursor substances in pork were analyzed using two-way analysis of variance (two-way ANOVA) in the general linear model (GLM) by SPSS 23.0 (SPSS Inc., Chicago, IL, USA). The model was as follows:

$$Y_{ijk}=\mu +{\alpha }_i+{\beta }_j+{\left(\alpha \beta \right)}_{ij}+e_{ijk}\text{,}$$

where Y_ijk is the dependent variable; μ is the overall mean; α_i and β_j are the fixed treatment effects; (αβ)_ij is the interaction effect between two fixed treatment effects; e_ijk is random error.

In the 2 × 2 factorial study, the analysis of all traits and descriptions included dietary V/I (N [V/I] and H [V/I]) and slaughter weight (lower weight [LW] 100 kg and higher weight [HW] 130 kg) as fixed factors and coupled with the interactions. Random term included replicate in the models. Furthermore, the Student's two-tailed t-test was employed to analyze the effects of dietary V/I on VOCs and hydrophilic compounds in pork under two BWs, respectively. Data were expressed as mean and standard error of the mean (SEM). Significance was defined at P < 0.05.

3. Results

3.1. Growth performance

As shown in Table 2, increased dietary V/I did not alter growth performance in pigs (P > 0.05).

Table 2.

Effects of dietary ratio of V/I supplementation and slaughter weight on growth performance of pigs.

Items	Treatments1		SEM	P-value
	N (V/I)	H (V/I)
About 105 kg
ADG, g/d	701	738	52.6	0.498
ADFI, g/d	2509	2639	163.6	0.445
FCR	3.59	3.59	0.136	0.934
About 130 kg
ADG, g/d	923	823	54.4	0.114
ADFI, g/d	3356	3171	151.3	0.267
FCR	3.64	3.88	0.197	0.265

ADG = average daily gain; ADFI = average daily feed intake; FCR = feed conversion ratio (average daily feed intake per average daily gain); SEM = standard error of the mean; N (V/I) = normal dietary valine/isoleucine ratio; H (V/I) = high dietary valine/isoleucine ratio.

¹N (V/I) means that dietary valine/isoleucine (V/I) met the nutritional requirements of NRC (2012) for pigs, and H (V/I) is higher V/I (n = 6).

3.2. Change in metabolic profiles in pork

3.2.1. Lipophilic metabolites

In this study, 291 lipids of 33 classes were detected in all pork samples, including 170 lipids in the negative ion mode, and 121 lipids in the positive ion mode using LC-MS/MS (Table S1). According to the PCA result, the lipid composition of pork tended to be separated between two slaughter weights (Fig. S1A), whereas dietary V/I treatment did not alter the lipid compositions of the pork (Fig. S1B).

The amounts of different types of lipids were summed individually (Fig. 1A and B). Changes in the top 10 major lipids were analyzed at different slaughter weights using two-way ANOVA. As shown in Fig. 1C, the content of acylcarnitine (ACar) increased by 47.06% (P < 0.001) and lysophosphatidylethanolamine (LPE) increased by 50.37% (P = 0.036) in the meat of HW pigs compared to LW pigs.

Fig. 1.

Lipophilic metabolites in pork of 100 kg (LW) and 130 kg (HW) pigs. (A and B) Percentages of lipid species. (C) Changes in the top 10 major lipids. (D) Orthogonal partial least squares discrimination analysis (OPLS-DA) score plots of the confirmed lipids. (E) Number of differential lipids. (F–I) Representative lipid molecules with VIP > 1.0 and P < 0.05 in both positive and negative ion modes. LW, pigs with a slaughter weight of 100 kg; HW, pigs with a slaughter weight of 130 kg. LW = lower weight; HW = higher weight; PA = phosphatidic acid; PI = phosphatidylinositol; ACar = acylcarnitine; FAHFA = fatty acid esters of hydroxy fatty acids; PC = phosphatidylcholine; TAG = triacylglycerol; Cer = ceramide; LPC = lysophosphatidylcholine; PG = phosphatidylglycerol; PE = phosphatidylethanolamine; LPE = lysophosphatidylethanolamine. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, based on two-way ANOVA (n = 12).

To better understand the effect of slaughter weight on lipid composition in pork, OPLS-DA (Fig. 1D) and two-way ANOVA were employed to analyze all identified lipids in the meat of LW and HW pigs. The criteria for a significant difference in lipids was a P < 0.05 (based on two-way ANOVA) and variable importance in projection (VIP) value > 1.0 (based on OPLS-DA). Thirty differential lipids affected by slaughter weight were observed (VIP >1.0, P < 0.05; Fig. 1E). Compared to LW pigs, 16 lipids in the meat of HW pigs, mainly including long-chain ACar and phosphatidylinositol (PI), were significantly down-regulated, while 14 lipids, mainly including short and medium chain ACar, lysophosphatidylcholine (LPC) and LPE, were up-regulated (Fig. 1F–I).

3.2.2. Hydrophilic metabolites

A total of 823 hydrophilic compounds were detected in pork. Among these, 59 compounds were significantly affected by dietary V/I (P _V/I < 0.05), 79 compounds were significantly affected by slaughter weight (P _SW < 0.05) and 45 compounds were significantly influenced by the interaction between dietary V/I and slaughter weight (P _{V/I × SW} < 0.05) (Table S2). An obvious separation between N (V/I) and H (V/I) treatments based on OPLS-DA was shown in Fig. 2A. There were 59 differential hydrophilic metabolites with VIP > 1.0 (based on OPLS-DA) and P < 0.05 (based on two-way ANOVA) identified between the N (V/I) and H (V/I) treatments under both the positive and negative ion modes (Table S3). These differential hydrophilic metabolites were significantly enriched in fructose and mannose metabolism (D-[+]-mannose and mannose 6-phosphate), alanine, aspartate and glutamate metabolism (D-aspartic acid and L-glutamic acid) and amino sugar and nucleotide sugar metabolism (D-[+]-mannose and mannose 6-phosphate) pathways (P < 0.05 and impact > 0; Fig. 2B and Table S4).

Fig. 2.

Hydrophilic metabolites in the muscle of different slaughter weights and dietary valine/isoleucine (V/I) treatments. (A) Orthogonal partial least squares discrimination analysis (OPLS-DA) score plots of individual hydrophilic metabolites and (B) pathway analysis in dietary N (V/I) and H (V/I) treatments. (C) OPLS-DA score plots of individual hydrophilic metabolites and (D) pathway analysis in 100 kg (LW) and 130 kg (HW) pigs. (E) OPLS-DA score plots of individual hydrophilic metabolites and (F) pathway analysis in LW-N (V/I) and LW-H (V/I) groups. (G) OPLS-DA score plots of individual hydrophilic metabolites and (H) pathway analysis in HW-N (V/I) and HW-H (V/I) groups. N (V/I), H (V/I), LW and HW, n = 12; LW-N (V/I), LW-H (V/I), HW-N (V/I) and HW-H (V/I), n = 6. LW, pigs with a slaughter weight of 100 kg; HW, pigs with a slaughter weight of 130 kg; LW/HW-N (V/I), pigs fed an N (V/I) diet until reaching a live weight of 100 kg (for LW) or 130 kg (for HW); LW/HW-H (V/I), pigs fed an H (V/I) diet until reaching a live weight of 100 kg (for LW) or 130 kg (for HW). LW = lower weight; HW = higher weight; N (V/I) = the normal dietary valine/isoleucine ratio; H (V/I) = high dietary valine/isoleucine ratio; TCA cycle = citrate cycle.

An obvious separation of hydrophilic compounds between LW and HW pigs was observed based on OPLS-DA (Fig. 2C). A total of 78 significant differential hydrophilic metabolites were identified between LW and HW pigs under both positive and negative ion modes, with VIP >1.0 (based on OPLS-DA) and P < 0.05 (based on two-way ANOVA) (Table S5). The results of the pathway analysis showed that differential hydrophilic metabolites were significantly enriched in alanine, aspartate and glutamate metabolism (L-aspartic acid, D-aspartic acid, and succinic acid), and citrate cycle (TCA cycle; succinic acid and L-malic acid) pathways (P < 0.05 and impact >0; Fig. 2D and Table S6).

As shown in Fig. 2E and G, profiles of hydrophilic metabolites detected in meat were significantly separated between dietary V/I treatments at both slaughter weights of pigs. In LW pigs, 53 differential hydrophilic metabolites with VIP > 1.0 (based on OPLS-DA) and P < 0.05 (based on Student's t-test) were identified in the meat of pigs receiving two dietary treatments of V/I (Table S7), while in HW pigs, 39 hydrophilic metabolites with VIP > 1.0 (based on OPLS-DA), P < 0.05 (based on Student's t-test) were identified in the meat of pigs receiving two dietary V/I treatments (Table S8). Metabolic pathway analysis showed that in LW pigs, differential hydrophilic metabolites between the different dietary V/I treatments were significantly enriched in fructose and mannose metabolism (D-[+]-mannose and mannose 6-phosphate) and amino sugar and nucleotide sugar metabolism (D-[+]-mannose and mannose 6-phosphate) (P < 0.05 and impact > 0; Fig. 2F and Table S9). In HW pigs, differential hydrophilic metabolites between the different dietary V/I treatments were significantly enriched in primary bile acid biosynthesis only, and cholesterol and cholic acid were enriched in this pathway (P < 0.05 and impact > 0; Fig. 2H and Table S10).

3.3. Changes in VOCs in pork

In this study, a total of 48 VOCs were identified in all groups, including alcohols (10), aldehydes (5), esters (11), hydrocarbons (19), ketones (1), and others (2). Among them, 7 VOCs were significantly affected by dietary V/I (P _V/I < 0.05), and 17 VOCs were significantly affected by slaughter weight (P _SW < 0.05) (Table 3). Furthermore, the content of acetoin, as a ketone, was also influenced by the interaction between dietary V/I and slaughter weight (P _{V/I × SW} = 0.034). Additionally, it was screened out a total of 4 VOCs with OAV > 1 and identified them as characteristic volatile compounds (Table 4).

Table 3.

Composition of the volatile compounds identified in meat of LW and HW pig fed N (V/I) and H (V/I) diets (μg/kg)¹.

Compound names	RT, min	100 kg (LW)		130 kg (HW)		SEM	P-value
		N (V/I)	H (V/I)	N (V/I)	H (V/I)		SW	V/I	SW × V/I
Alcohols
2-Methyl-2-propanol	3.13	11.05	10.2	6.67	5.82	1.601	0.013	0.601	0.998
Ethanol	3.63	238.91	154.34	168.78	144.49	30.225	0.201	0.087	0.331
1-Methoxy-2-propanol	8.78	25.45	4.87	5.32	5.93	10.195	0.361	0.339	0.311
1-Butanol	9.27	5.08	3.14	3.63	2.98	0.583	0.180	0.038	0.284
3-Methyl-3-butenol	12.18	1.54	1.03	0.58	0.8	0.303	0.065	0.634	0.241
1-Pentanol	12.29	5.42	9.96	8.18	20.07	4.064	0.129	0.057	0.377
1-Hexanol	15.21	2.59	1.78	1.58	3.04	0.644	0.847	0.617	0.092
1-Octen-3-ol	17.74	4.04	7.47	6.57	19.17	4.469	0.127	0.088	0.317
1-Heptanol	17.97	0.43	0.52	0.52	1.05	0.145	0.045	0.041	0.142
1-Dodecanol	29.57	0.96	0.76	0.63	0.47	0.121	0.019	0.148	0.862
Aldehydes
Hexanal	7.01	12.31	28.25	25.8	85.22	18.514	0.072	0.055	0.254
Heptanal	10.03	0.23	0.41	0.36	0.93	0.172	0.072	0.041	0.278
Nonanal	16.05	1.01	1.22	1.2	1.82	0.261	0.147	0.124	0.430
Pentadecanal	30.56	0.16	0.29	0.22	0.26	0.033	0.776	0.015	0.190
Benzaldehyde	19.32	0.69	0.89	0.72	1.17	0.117	0.202	0.011	0.292
Esters
Methyl acetic ester	2.34	5.26	1.93	2.06	1.69	0.943	0.083	0.063	0.132
Ethyl acetic ester	2.88	2.85	5.55	2.5	4.97	1.555	0.768	0.112	0.941
Hexanoic acid, ethenyl ester	14.19	0.38	0.88	0.64	1.48	0.289	0.155	0.032	0.552
Lactic acid, ethyl ester	14.78	4.01	4.1	3.64	3.66	0.825	0.630	0.946	0.967
Sulfurous acid, 2-ethylhexyl hexyl ester	16.25	0.4	0.4	0.37	0.32	0.019	0.004	0.206	0.205
Methyl salicylate	25.25	0.08	0.08	0.07	0.07	0.018	0.662	0.845	0.906
Triisobutyl phosphate	28.40	18.84	21.99	35.26	38.03	4.901	0.003	0.553	0.970
Dodecyl acrylate	29.90	0.88	0.69	0.49	0.46	0.145	0.043	0.451	0.570
Tributyl phosphate	32.62	1.37	2.31	2.93	3.03	0.443	0.018	0.256	0.355
1,2-Benzenedicarboxylic acid, bis(2-methylpropyl) ester	39.56	0.32	0.3	0.29	0.35	0.035	0.685	0.571	0.260
Dibutyl phthalate	41.98	2.2	2.64	2.37	2.78	0.267	0.562	0.125	0.964
Hydrocarbons
Hexane	1.59	2.02	1.33	0.72	0.81	0.480	0.074	0.538	0.425
Cyclohexane	1.84	0.98	0.95	1.09	0.98	0.174	0.703	0.69	0.832
4-Methylheptane	1.97	3.39	2.17	1.66	1.37	0.549	0.032	0.184	0.406
2,4-Dimethylheptane	2.21	1.66	0.91	0.70	0.68	0.294	0.058	0.204	0.221
2,4-Dimethyl-1-heptene	2.82	1.78	1.19	0.86	0.6	0.237	0.005	0.088	0.500
2,2-Dimethyldecane	3.98	3.28	1.34	5.66	5.26	1.391	0.035	0.411	0.584
2,2,4,6,6-Pentamethylheptane	4.80	6.51	5.58	4.68	1.77	0.956	0.008	0.058	0.312
2,3,6,7-Tetramethyloctane	5.16	7.78	6.57	5.81	3.03	1.101	0.021	0.085	0.482
Toluene	5.82	89.92	59.67	41.59	34.75	11.887	0.006	0.134	0.337
2,6-Dimethyloctane	6.06	3.61	3.14	2.72	1.02	0.562	0.015	0.068	0.289
2,3,4-Trimethyldecane	6.43	7.15	6.12	5.27	2.12	1.036	0.010	0.058	0.318
Ethylbenzene	8.09	21.61	11.77	11.03	11.91	3.053	0.103	0.158	0.094
P-xylene	8.51	34.94	16.81	17.88	19.83	5.386	0.208	0.149	0.077
O-xylene	9.84	24.72	10.8	13.29	13.98	3.744	0.284	0.093	0.065
Limonene	10.23	1.47	0.59	1.17	1.22	0.314	0.613	0.199	0.158
Hexadecane	11.89	0.95	0.93	0.49	0.19	0.215	0.011	0.464	0.498
Styrene	12.10	0.75	0.35	0.63	0.68	0.127	0.405	0.183	0.093
2,3,6-Trimethylheptane	18.87	4.85	4.26	5.07	5.01	0.417	0.257	0.443	0.533
Tridecane	21.34	0.41	0.33	0.32	0.28	0.033	0.066	0.081	0.571
Ketones
Acetoin	13.06	496.77	26.9	24.77	9.67	99.995	0.024	0.025	0.034
Others
2-Methoxy-2-methylpropane	1.73	2.14	2.84	2.49	2.63	0.241	0.763	0.097	0.263
Longifolene	20.45	0.58	0.35	0.41	0.28	0.139	0.416	0.216	0.743

RT = retention time; LW = lower weight; HW = higher weight; SW = lower weight vs. higher weight; N (V/I) = normal dietary valine/isoleucine ratio; H (V/I) = high dietary valine/isoleucine ratio; SEM = standard error of the mean; n = 6.

¹LW, pigs with a slaughter weight of 100 kg; HW, pigs with a slaughter weight of 130 kg; N (V/I) means that dietary valine/isoleucine (V/I) met the nutritional requirements of NRC (2012) for pigs, and H (V/I) is higher V/I.

Table 4.

Volatile compounds (μg/kg) in cooked longissimus thoracis muscle of finishing pigs with odor activity value (OAV > 1).

Compound names	Odor description1	Thresholds	100 kg (LW)		130 kg (HW)		SEM	P-value
			N (V/I)2	H (V/I)2	N (V/I)2	H (V/I)2		SW	V/I	SW × V/I
1-Octen-3-ol	Mushroom∗	1.5	4.04	7.47	6.57	19.17	4.469	0.127	0.088	0.317
Hexanal	Green, fresh, grass∗	5.0	12.31	28.25	25.8	85.22	18.514	0.072	0.055	0.254
Nonanal	Citrus, green, citronella, grass∗	1.1	1.01	1.22	1.2	1.82	0.261	0.147	0.124	0.430
Acetoin	Milky#	14.0	496.77	26.9	24.77	9.67	99.995	0.024	0.025	0.034

LW = lower weight; HW = higher weight; SW = lower weight vs. higher weight; N (V/I) =normal dietary valine/isoleucine ratio; H (V/I) = high dietary valine/isoleucine ratio; SEM = standard error of the mean; n = 6.

¹Odor thresholds in water and the odor description from the literature, ∗Sohail et al. (2022), ^#Wang et al. (2022).

²LW, pigs with a slaughter weight of 100 kg; HW, pigs with a slaughter weight of 130 kg; N (V/I) means that dietary valine/isoleucine (V/I) met the nutritional requirements of NRC (2012) for pigs, and H (V/I) is higher V/I.

3.3.1. Variation in aroma compound profiles induced by dietary treatment

Dietary V/I significantly influenced VOCs (Fig. 3A). A high ratio of valine/isoleucine decreased the content of ketones in pork compared to N (V/I) (Fig. 3B). There were seven VOCs with VIP > 1.0 (based on OPLS-DA) and P < 0.05 (based on two-way ANOVA) significantly affected by dietary V/I. Among them, 1-heptanol, heptanal, pentadecanal, benzaldehyde, hexanoic acid, and ethenyl ester were up-regulated, and 1-butanol and acetoin were down-regulated in H (V/I) compared to N (V/I) (Fig. 4A).

Fig. 3.

Changes in volatile organic compounds (VOCs) in pork. (A) Orthogonal partial least squares discrimination analysis (OPLS-DA) score plots of VOCs and (B) concentration of volatile categories in dietary N (V/I) and H (V/I) treatments. (C) OPLS-DA score plots of VOCs and (D) concentration of volatile categories in 100 kg (LW) and 130 kg (HW) pigs. (E) OPLS-DA score plots of VOCs and (F) concentration of volatile categories in LW-N (V/I) and LW-H (V/I) groups. (G) OPLS-DA score plots of VOCs and (H) concentration of volatile categories in HW-N (V/I) and HW-H (V/I) groups. N (V/I), H (V/I), LW and HW, n = 12; LW-N (V/I), LW-H (V/I), HW-N (V/I) and HW-H (V/I), n = 6. LW, pigs with a slaughter weight of 100 kg; HW, pigs with a slaughter weight of 130 kg; LW/HW-N (V/I), pigs fed an N (V/I) diet until reaching a live weight of 100 kg (for LW) or 130 kg (for HW); LW/HW-H (V/I), pigs fed an H (V/I) diet until reaching a live weight of 100 kg (for LW) or 130 kg (for HW). LW = lower weight; HW = higher weight; N (V/I) = the normal dietary valine/isoleucine ratio; H (V/I) = high dietary valine/isoleucine ratio. ∗P < 0.05.

Fig. 4.

Changes in candidate volatile organic compounds (VOCs) in pork. (A) Significant different VOCs affected by dietary V/I. (B) Venn diagrams of differential metabolites in different slaughter weights and dietary V/I. (C and D) Significantly different VOCs affected by slaughter weight. (E and F) Significant different VOCs affected by dietary V/I in 100 kg (LW) pigs. (G) Significant different VOCs affected by dietary V/I in 130 kg (HW) pigs. N (V/I), H (V/I), LW and HW, n = 12; LW-N (V/I), LW-H (V/I), HW-N (V/I) and HW-H (V/I), n = 6. LW, pigs with a slaughter weight of 100 kg; HW, pigs with a slaughter weight of 130 kg; LW/HW-N (V/I): pigs fed an N (V/I) diet until reaching a live weight of 100 kg (for LW) or 130 kg (for HW); LW/HW-H (V/I): pigs fed an H (V/I) diet until reaching a live weight of 100 kg (for LW) or 130 kg (for HW). LW = lower weight; HW = higher weight; N (V/I) = normal dietary valine/isoleucine ratio; H (V/I) = high dietary valine/isoleucine ratio. ∗P < 0.05, ∗∗P < 0.01.

3.3.2. Difference of aroma compounds between different slaughter weights

It was observed that an obvious separation of VOC profiles between two slaughter weights based on OPLS-DA (Fig. 3C). The contents of hydrocarbons and ketones were significantly lower and esters were significantly higher in the meat of HW pigs compared to LW pigs (Fig. 3D). There were 17 VOCs significantly affected by slaughter weight based on OPLS-DA and two-way ANOVA. Compared with the LW pigs, contents of 1-heptanol, 2,2-dimethyldecane, tributyl phosphate, and triisobutyl phosphate were significantly up-regulated, and contents of 2-methyl-2-propanol, 1-dodecanol, sulfurous acid, 2-ethylhexyl hexyl ester, dodecyl acrylate, 4-methylheptane, 2,4-dimethyl-1-heptene, 2,2,4,6,6-pentamethylheptane, 2,3,6,7-tetramethyloctane, toluene, 2,6-dimethyloctane, 2,3,4-trimethyldecane, hexadecane, and acetoin were significantly down-regulated in the meat of HW pigs (VIP > 1.0, P < 0.05) (Fig. 4C and D). A correlation analysis between lipids and VOCs (OAV > 1) was performed based on Pearson's correlation coefficient. The results showed that Acetoin was significantly positively correlated with long-chain ACar (Fig. S2).

Furthermore, the effect of dietary V/I on VOCs in pork at the two slaughter weights was analyzed, respectively. Dietary V/I significantly influenced VOCs of pork in LW pigs (Fig. 3E). In LW pigs, increased dietary V/I increased the content of aldehydes and decreased the content of alcohols and ketones in pork (P < 0.05) (Fig. 3F). Moreover, there were 10 VOCs significantly affected by dietary V/I based on OPLS-DA and Student's t-test analyses in LW pigs, and among them, contents of 1-pentanol, 1-octen-3-ol, hexanal, pentadecanal, benzaldehyde, hexanoic acid, ethenyl ester, and 2-methoxy-2-methylpropane in meat were up-regulated. And contents of 1-butanol, limonene, and acetoin in meat were down-regulated in H (V/I) compared to N (V/I) in LW pigs (VIP > 1.0, P < 0.05) (Fig. 4E and F).

In HW pigs, dietary V/I also had a significant effect on VOCs of the meat (Fig. 3G). In HW pigs, increased dietary V/I had no significant effect on the types of VOCs (Fig. 3H), but there were six VOCs significantly affected by dietary V/I based on OPLS-DA and Student's t-test analyses. Among them, 1-heptanol content was up-regulated. And 2,2,4,6,6-pentamethylheptane, 2,3,6,7-tetramethyloctane, 2,6-dimethyloctane, 2,3,4-trimethyldecane, and hexadecane contents were down-regulated in H (V/I) compared to N (V/I) in HW pigs (VIP > 1.0, P < 0.05) (Fig. 4G).

As shown in Fig. 4B, two kinds of differential VOCs, acetoin and 1-heptanol, were affected by slaughter weight and dietary V/I. Among the seven differential VOCs treated with different dietary V/I, five were the common differential VOCs in LW pigs offered different V/I diets, and one was the common differential VOCs in HW pigs offered different V/I diets.

4. Discussion

Branched-chain amino acids are involved in adipogenesis and glucogenesis (Green et al., 2016; Zhang et al., 2021a). Increased V/I of the diet significantly increased IMF content, but did not affect fatty acid composition of pork (Xu et al., 2022). Accordingly, in the present study, dietary V/I did not alter the growth in growing-fattening pigs and the lipid metabolite profile of pork. Postmortem muscular glycogen depletion and proteolysis affect contents of various hydrophilic flavor precursors in pork, including free amino acids and peptides, and sugar-derived metabolites (Williamson et al., 2014). Free amino acids endow cooked meat with the bitterness, umami, and sweetness, while reducing sugars such as mannose produce the aroma during meat cooking (Meinert et al., 2009). In this study, the mannose content in pork was increased, which might facilitate the increase of some VOCs in pigs receiving H (V/I) diets compared to pigs receiving N (V/I) diets. Previously, it was showed that increasing the dietary V/I significantly reduced the concentrations of umami amino acids in pork (Xu et al., 2022). Glutamic acid is the main source of umami in food, which makes food more flavorful (Yamamoto and Inui-Yamamoto, 2023). Consistently, it was found that increased dietary V/I decreased the content of glutamic acid in pork by altering alanine, aspartate and glutamate metabolism. The decreased contents of glutamic acid may reduce the umami of pork.

The oxidation of fat, especially fat in meat, forms VOCs such as fatty hydrocarbons, aldehydes, and ketones during cooking, providing important fatty flavors for cooked meat (Bleicher et al., 2022). With respect to VOCs, it was found that the content of the ketone compound, acetoin, in pork significantly decreased as dietary V/I increased. Acetoin possesses a pleasant creamy yogurt odor and a fatty butter taste, and is mainly used as a flavor enhancer in the food industry (Xiao and Lu, 2014). Meanwhile, increased dietary V/I reduced the content of 1-butanol in pork. However, the aroma threshold of 1-butanol (460 μg/kg) is much higher than the levels detected in pork (Sohail et al., 2022). Thus, further investigation is needed to determine whether the observed variation in 1-butanol content contributes to the overall flavor of pork. Besides, increased dietary V/I also increased the content of other important aroma compounds in pork, such as almond-flavored benzaldehyde, green and aldehyde-flavored heptanal, and floral-flavored 1-heptanol (Sohail et al., 2022). These changes in the concentrations of the VOCs may alter the flavor of pork. However, the overall flavor of pork was not comprehensively evaluated using sensory evaluation, which was the limitation of this study.

In previous studies, increased slaughter weight significantly affected the production of VOCs in pork (Ba et al., 2019). Phospholipids play a key role in the development of meat aromas and have a much greater impact on flavor generation than triacylglycerols (TAG) (Shahidi and Hossain, 2022). Li et al. (2021) reported that in heavier pigs with a slaughter weight of 150.11 ± 2.43 kg, phosphatidylcholine (PC) and phosphatidylethanolamine (PE) levels were significantly lower than in pigs with less slaughter weight (111.63 ± 0.99 kg). In the present study, PC and PE levels did not change as slaughter weight increased, whereas LPE and ACar levels significantly increased. LPE, a deacylated product of PE hydrolysis is catalyzed by phospholipase A1/A2 (Yamamoto et al., 2022). The increase in LPE levels may be attributed to the degradation of PE. Particularly, LPC and LPE that contain unsaturated fatty acid bonds are supposed to serve as important aroma retainers (Zhou et al., 2024). Therefore, it was deduced that the changes of VOCs may be related to the enhanced phospholipids catabolism in muscle as the slaughter weight increased in finishing pigs.

Primary oxidation products mainly occurred via autoxidation, and then were further degraded into volatile secondary lipid oxidation products with odor–active properties, including hydrocarbons and alcohols (Shahidi and Hossain, 2022). ACar is an intermediate metabolite of β-oxidation of fatty acids. Activated fatty acids are catalyzed with carnitine by carnitine palmitoyl transferase I (CPT I) to form ACar, and then ACar crosses the inner mitochondrial membrane catalyzed by carnitine acylcarnitine translocase (CACT). Finally, ACar is converted back to Acyl-CoA catalyzed by carnitine palmitoyl transferase Ⅱ (CPT Ⅱ) for β-oxidation in the mitochondrial matrix (Qu et al., 2016). In addition, ACar is an important indicator of fatty acid oxidation and mitochondrial function. Accumulation of long-chain ACar results from mitochondrial overload and incomplete oxidation of fatty acids in muscle (Koves et al., 2008). In the present study, contents of short-chain and medium-chain ACar, such as ACar (4:0), ACar (6:0), ACar (7:0), and ACar (6:1) increased as the slaughter weight increased, while the content of most long-chain ACar, such as ACar (18:3), ACar (14:1), ACar (16:2), and ACar (20:4) decreased. This is consistent with the results of the decrease in PUFA content as slaughter weight increased.

Meanwhile, it was found that the differential hydrophilic metabolites in muscle were significantly enriched in the TCA cycle pathway as slaughter weight increased. Meanwhile L-malic acid content decreased, but succinic acid content increased. Therefore, the increase in ACar content may be caused by the increased slaughter weight that affected the mitochondrial function of the muscle. In addition, higher slaughter weight significantly reduced the ratio of PUFA/saturated fatty acid (SFA) in pork (Xu et al., 2022), which might account for the lower contents of hydrocarbons and ketones in pork of HW pigs compared to LW pigs. Hydrocarbons have a high threshold of taste, and the change of hydrocarbons has a low effect on the overall flavor of meat (Qi et al., 2017). Furthermore, the content of acetoin was affected by the interaction between dietary V/I and slaughter weight. Either increased slaughter weight or dietary V/I led to a decrease in acetoin content in pork, which might reduce the creamy odor of pork. Furthermore, increased slaughter weight reduced aspartic acid and succinic acid contents, which may affect the umami taste of meat (Wang et al., 2022). Meanwhile, it has been demonstrated that increasing the slaughter weight of pigs can increase the amount of aldehyde flavor precursors in pork (Ba et al., 2019). Therefore, it was deduced that the slaughter weight was an important factor influencing pork flavor.

It is notable that these changes in pork flavor relating compounds induced by the dietary V/I differed between two slaughter weights of pigs. For example, in LW pigs, VOCs, such as limonene with fresh and sweet flavor (Shin et al., 2024), benzaldehyde with almond-like flavor (Sohail et al., 2022) and 1-octen-3-ol with mushroom-like odor which had a characteristic contribution to the aroma of meat (Cui et al., 2023), were significantly changed by H (V/I) diet. Meanwhile, in HW pigs, dietary H(V/I) raised the content of 1-heptanol with floral flavor (Sohail et al., 2022) and lowered the content of hydrocarbons, such as 2,2,4,6,6-pentamethylheptane. However, considering high aroma threshold (Qi et al., 2017; Sohail et al., 2022), the changes in the content of these VOCs might have a relatively little impact on the overall flavor of pork in HW pigs. This suggested that dietary V/I had a relatively small impact on lipids and hydrophilic compounds in meat of HW pigs. It was further speculated that the impact of dietary treatments on the overall flavor of pork becomes less as slaughter weight increases. This finding was consistent with previous studies showing that pigs with heavier slaughter weight exhibited greater resistance to dietary treatments (Apple et al., 2009; Xu et al., 2022). Meanwhile, it could be deduced that the reduced flavor changes under higher slaughter weights might be related to the reduced levels of PUFAs and long-chain ACars. Specifically, the content of PUFAs, serving as precursors for flavor compounds, directly influences the production of flavor-related molecules (Shahidi et al., 2022). This deduction requires validation in the future using a larger sample size, which is a limitation of this study.

5. Conclusion

Dietary V/I significantly altered the profiles of lipid metabolites, hydrophilic metabolites, and the content of VOCs in pork. However, slaughter weight had a more dominant effect on the composition of lipid metabolites in pork than dietary V/I. In particular, ACar and LPE contents were significantly higher in pork of HW pigs than from LW pigs. Both dietary V/I and slaughter weight altered the flavor precursor profiles and VOCs composition in pork, but the impact of dietary V/I on pork flavor weakened as slaughter weight increased.

Credit Author Statement

Mingyang Tan: Writing – original draft, Visualization, Investigation, Formal analysis. Qiuyu Zeng: Writing – original draft, Investigation, Formal analysis. Enfa Yan: Investigation, Formal analysis. Xiangze Liu: Investigation. Jingdong Yin: Writing – review & editing, Supervision, Resources, Methodology, Funding acquisition, Data curation, Conceptualization.

Declaration of competing interest

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

Appendix A. Supplementary data

The following are the Supplementary data to this article: