Introduction
Analyzing the transcriptome of a living individual can display almost any deviation of the organism from its normal state. Diseases can be detected and, with SARS-CoV-2 infection as its most famous example in recent years, transcriptome analysis can also be used to monitor the recovery from infection (Gong and An, 2025). Aging can also be displayed (Wang et al., 2024a), and unborn individuals show specific transcriptome profiles at different stages of development (reviewed in Assou et al., 2011; Llobat, 2020). Although the death of an individual is accompanied by the final cessation of all vital functions, it is still possible to detect transcriptomic activity postmortem (Scott et al., 2020). It even appears that gene expression does not decrease steadily, but also changes at different times, as some genes are upregulated or more abundant several hours after death (Pozhitkov et al., 2017; Zhu et al., 2017; Javan et al., 2020). In the context of animal carcasses used for meat production, the conversion of muscle to meat includes several biochemical and physical changes (Ouali et al., 2006). Therefore, it is not surprising that gene expression has been the subject of several studies in this context (Bowker et al., 2004; Tang et al., 2010; Zhao et al., 2019; Zequan et al., 2022; Wang et al., 2024b). While only a few studies have focused on changes in gene expression during slaughter and subsequent aging of meat (Fontanesi et al., 2011), several factors can influence the maturation and quality of pork, including stunning method (Velarde et al., 2000; Channon et al., 2002). As meat maturation is not a static process (reviewed in Scheffler and Gerrard, 2007), comparing transcriptomic profiles at 45 min and 36 h postmortem might allow differentiation between the immediate effects of the stunning procedure and its impact on subsequent time-dependent transcriptomic changes and the impact on meat quality. Within the European Union, most slaughter plants work with high concentrations of CO2 to keep pigs unconscious and insensible to pain during slaughter, as required by Council Regulation (EC) No 1099/2009. Stunning with high concentrations of CO2 induces a deep and long-lasting unconsciousness through hypercapnia (reviewed in Terlouw et al., 2016), but simultaneously provokes aversive behavior of the animals (Raj and Gregory, 1995; Dalmau et al., 2010). To improve animal welfare during the controlled atmosphere stunning process, inert gases like argon (Ar) and nitrogen (N2) have been investigated for their applicability in the stunning process (Troeger et al., 2004a; Llonch et al., 2012; Atkinson et al., 2020). Although these gases contribute to a reduction of the aversive behavior, in some studies, a reduction in meat quality was reported. For instance, a higher incidence of pale, soft, and exudative (PSE) meat was reported when stunning with N2 (Atkinson et al., 2020), or higher core temperatures in the M. semimembranosus (SM) were reported when stunning with Ar (Troeger et al., 2004b). However, the influence of inert gases for stunning on the muscle transcriptome of pigs has not been investigated. A study in broiler chicken revealed a downregulation of JNK1 in the breast muscle when the animals were stunned with 40% CO2 compared to 79% CO2 (Xu et al., 2018). Therefore, this study aimed to firstly assess whether differences in meat quality resulting from the use of different gases for stunning are reflected in the muscle transcriptome at 2 stages of maturation, and secondly, characterize the temporal dynamics of gene expression in the M. longissimus thoracis et lumborum (LTL) during meat maturation.
Animals, Materials, and Methods
Animals, stunning, and sampling
The investigations presented in this study were part of the first experiment described in Gelhausen et al. (2025). Specifically, a subset of 120 (64 male castrated and 56 female) crossbreed pigs ([German Edelschwein × German Landrace] × Pietrain) was selected from the 400 pigs involved in experiment one. These pigs originated from 2 different conventional farms (Lower Saxony, Germany) with straw husbandry and had an average slaughter weight of 105 ± 8 kg. All pigs were tested homozygously negative for the C1843T point mutation of the RYR1 gene. Stunning and slaughtering were carried out in a conventional slaughterhouse in Thuringia, Germany, where a Butina Dip-Lift system was retrofitted with a new and patented gassing system by Air Liquide Germany GmbH (Krefeld, Germany). Each gondola was loaded with 2 pigs, which were almost distributed equally by their origin to the used gas mixture. In the first experiment described by Gelhausen et al. (2025), 10 gas mixtures were used for stunning: Ar, a N2-Ar mixture as well as mixtures of the inert gases with 10%, 20% and 30% of CO2 in the mixture (residual O2 < 1%), a CO2 atmosphere with a residual O2 concentration < 1% and an at least 90% CO2 control atmosphere with a residual O2 concentration about 2% to reflect the concentration predominantly used in practice in Germany.
For the investigation of the transcriptional profile, only pigs stunned with Ar, N2-Ar, or the CO2 control group were taken into account, with a steady gas exposure time, resulting in 30 pigs per stunning condition (Table 1). For gene expression analysis, these animals were randomly assigned to 3 pools per stunning condition, each consisting of samples from 10 pigs (Table 2). Due to the study design described in Gelhausen et al. (2025), the different inert gas mixtures were tested on 4 separate slaughter days. The Ar and N2-Ar measurements series were each carried out on 2 of these days. However, the CO2 control group was included on each day. In order to capture the potential slaughter day effect, a separate CO2 control group for each inert gas mixture, measured on the same day of slaughter, was used. This resulted in 4 conditions: Ar = 95% Ar, CO2 (Ar) = 90% CO2, N2-Ar = 70% N2, 29% Ar and CO2 (N2) = 90% CO2. In the case of N2-Ar, the first 16 pigs were stunned with a proportion of 80% N2 and 20% Ar, which was adjusted to 29% Ar due to gas stability issues. All pigs entered the abattoir’s normal slaughter routine. After stunning, the pigs were bled, scalded in a tunnel system, dehaired, and afterwards burned to remove any remaining hair. Carcasses were chilled after evisceration, splitting, weighing, and classification at 2–7°C in cold rooms.
Number and distribution of pigs (n) stunned by gas mixture and slaughter day, with Ar = argon; CO2 (Ar) = carbon dioxide control group measured on the same day as argon; N2-Ar = nitrogen-argon mixture; and CO2 (N2) = carbon dioxide control group measured on the same day as the nitrogen-argon mixture. Each gas mixture was applied for its respective exposure time
| Gas mixture | ||||
|---|---|---|---|---|
| Ar | CO2 (Ar) | N2-Ar | CO2 (N2) | |
| Exposure time (sec) | 240 | 180 | 240 | 180 |
| Day 1 | 12 | 16 | ||
| Day 2 | 16 | 18 | ||
| Day 3 | 18 | 14 | ||
| Day 4 | 14 | 12 | ||
| Sum | 30 | 30 | 30 | 30 |
Overview of planned and analyzed pools per gas mixture used for stunning and postmortem time point of sample collection, with Ar = argon; CO2 (Ar) = carbon dioxide control group measured on the same day as argon; N2-Ar = nitrogen-argon mixture; and CO2 (N2) = carbon dioxide control group measured on the same day as the nitrogen-argon mixture
| Gas mixture | Time point | Planned number of pools | Animals per pool | Total animals planned | Pools included in analysis | Animals represented in analysis |
|---|---|---|---|---|---|---|
| Ar | 45 min | 3 | 10 | 30 | 3 | 30 |
| Ar | 36 h | 3 | 10 | 30 | 3 | 30 |
| CO2 (Ar) | 45 min | 3 | 10 | 30 | 2 | 20 |
| CO2 (Ar) | 36 h | 3 | 10 | 30 | 2 | 20 |
| N2-Ar | 45 min | 3 | 10 | 30 | 3 | 30 |
| N2-Ar | 36 h | 3 | 10 | 30 | 3 | 30 |
| CO2 (N2) | 45 min | 3 | 10 | 30 | 3 | 30 |
| CO2 (N2) | 36 h | 3 | 10 | 30 | 3 | 30 |
| Total | 24 | 240 | 22 | 220 |
Sampling and meat quality measurement
After classification, muscle samples were collected 45 min postmortem from the LTL of the 120 pigs, above the 14th rib on the left side of the warm carcass. After a chilling duration of 36 h, a piece of LTL with bones was dissected from the carcass near the area where the first sample was taken. At each time point, 2 samples per pig were fine sectioned to 125 mm3 slices for RNA isolation, to avoid cross-contamination, and placed in 1.5 ml RNAlater™ (Thermo Fischer Scientific, Waltham, Massachusetts, USA). The samples were stored for 24 h at 4°C and were then transferred to −20°C for long-term storage. For meat quality monitoring, pH and temperature (T, °C) were measured simultaneously with a portable pH meter (HI98163, Hanna Instruments, Woonsocket, USA) following a 2-point calibration at pH 4 and pH 7. Both parameters were measured in the LTL between the 13th and 14th rib of the left half, and in the SM before and after chilling. Temperature compensation for the pH measurements was automatically performed by the device. The threshold for pale, soft, and exudative (PSE) meat was set for < 5.8 (Ryu et al., 2005; Mörlein et al., 2007).
RNA isolation
Total RNA isolation was performed with the RNeasy Plus Universal Mini Kit (Qiagen N.V., Hilden, Germany). Around 20 mg frozen muscle tissue was transferred into a 2 mL tube and mixed with 900 μL QIAzol Lysis Reagent (Qiagen N.V., Hilden, Germany) and 5 μL Reagent DX (Qiagen N.V., Hilden, Germany). Additionally, 1.4 mm Ceramic Beads (Biolabproducts GmbH, Bebensee, Deutschland) were added to the mixture, and tubes were loaded into a Bead Ruptor Elite (Omni International, Kennesaw, GA, USA) and processed in 3 cycles with 4.5 m/s for 15 s and a dwell time of 10 s. Afterwards, the samples were processed according to the manufacturer´s specifications and eluted in RNAse-free water. Isolated RNA samples were evaluated for their quality based on their RNA integrity number (RIN) by microfluidic capillary electrophoresis on an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA) using the RNA 6000 Nano kits according to the manufacturer´s instructions. The RINs for the samples at 45 min postmortem ranged from 7 to 8.3, with a mean of 7.6. After 36 h, RINs ranged from 5.9 to 7.8, with a mean of 6.9. After this and before sequencing, samples were pooled.
RNA sequencing
For sequencing, 24 equimolar pools were generated, including 3 pools of 10 animals for both time points of the 4 gas mixtures investigated, representing 30 animals per condition and time point (45 min and 36 h, Table 2). The 10 animals per pool were randomly selected within the used gas mixture and the time point of sample collection. However, sample composition for the pools after 36 h was the same as for 45 min, resulting in 3 prepared pools per condition for Ar, N2-Ar, and both CO2 groups (3 × 3 comparison). Nevertheless, errors in sample preparations occurred at both time points for the CO2 (Ar) group. Consequently, 2 groups had to be excluded, and subsequent analyses were carried out with 2 pools each, resulting in a 2 × 2 comparison over time for the CO2 (Ar) group and a 3 × 2 comparison for Ar versus CO2 (Ar). Sequencing procedure, including quality control, library preparation, and sequencing, was carried out by BGI Genomics Co., Ltd., Shenzhen, China. The library type was DNBSEQ Eukaryotic Strand-specific mRNA library, and sequencing was performed with a DNBSEQ™ platform. Paired-end reads with a read length of 100 bp were produced. Raw sequencing reads were filtered and trimmed with SOAPnuke with the following settings: -n 0.001 -l 20 -q 0.4 --adaMR 0.25 --ada_trim --minReadLen 100 (Chen et al., 2018).
Transcriptome analysis
RNA sequencing reads of the pooled samples were aligned to the Sus scrofa 11.1 reference genome version GCF_000003025.6 using HiSat2 version 2.1.0 with default settings (Kim et al., 2015). Splice sites were derived from the Gene Transfer Format file. FeatureCounts from the Subread package (Version 2.0.0) was used to count exon spanning reads (Lawrence et al., 2013). Analysis of the differentially expressed genes (DEG) was carried out in R, version 4.4.0 (R Core Team, 2024), with DESeq2 (Version 1.38.3) (Love et al., 2014), using default settings, including Benjamini-Hochberg adjustment P values. Repeated measurements were included in the model for the comparisons over time by “design = ∼pool+treatment.” Results were visualized with the R package EnhancedVolcano (Version 1.16.0) (Blighe et al., 2018) and pheatmap (Version 1.0.12) (Kolde, 2018). Genes were considered differentially expressed with an absolute Log2Foldchange (Log2 FC) > 1 and with an adjusted P value cutoff < 0.01 for the comparisons between the 2 time points and with an adjusted P value cutoff < 0.05 between the used gas mixtures.
Functional analysis
The integrated Biological ID Translator (bitr) function of the R package clusterProfiler (Version 4.6.2) (Yu et al., 2012) was used to convert gene symbols to Entrez IDs. The same package was used for the following gene cluster comparison. Gene ontology (GO) enrichment analysis was performed with the enrichGO function for cell component (CC), molecular function, and biological pathways (BP) with pAdjustMethod = “fdr,” pvalueCutoff = 1, qvalueCutoff = 0.25, readable = TRUE, minGSSize = 10. For Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis with the enrichKEGG function, settings were set to pvalueCutoff = 1, pAdjustMethod = “BH,” minGSSize = 10, qvalueCutoff = 0.25, use_internal_data = FALSE). Plots were created with the dotplot function of the enrichplot package (Version 1.24.0) (Yu, 2024).
Statistical analysis of meat quality
Statistical evaluation of meat quality traits was conducted in R, version 4.4.0 (R Core Team, 2024). First parameters were tested for normal distribution with the Shapiro.test function of the integrated stats package. Testing for variance homogeneity was carried out with the leveneTest function of the car (version 3.1-3) package (Fox and Weisberg, 2019). If parameters fulfilled normal distribution and variance homogeneity, an analysis of variance with the aov function, followed by a pairwise comparison with the TukeyHSD function, was carried out. Both functions are included in the stats package. If only the normal distribution were fulfilled, the functions welch_anova_test and games_howel_test for pairwise comparison from the rstatix (version 0.7.2) (Kassambara, 2023) package were used. If none of the requirements were fulfilled, a Kruskal-Wallis test was carried out via the kruskal.test function (stats package) followed by the dunnTest function (FSA package; version 0.9.6; Ogle et al., 2025) for pairwise comparison.
Results
Differences in meat quality can be observed when pigs are stunned with inert gas mixtures, compared to high-concentration CO2 stunning (Atkinson et al., 2020). Therefore, we investigated the transcriptome of the LTL to find evidence on a transcriptional level at 2 time points of meat maturation. To the best of our knowledge, little is known about the dynamics of transcriptional profiles in pork maturation.
Differentially expressed genes between gases
A total of 26 genes were differentially expressed between the different gas mixtures (Figure 1). None of these genes were differentially expressed in all comparisons; however, the direction of expression (up or down) was similar in the different groups, except for LOC110258827, LOC102167466, LOC102167597, LOC102167964, LOC110258346, LOC110258349, and LOC110258347, which tend to be upregulated in the Ar groups and downregulated in the N2 groups compared to CO2 (Figure 2). When comparing Ar and CO2 after 45 min postmortem, CISH was the only gene that was significantly differentially expressed, and it was only observed when comparing Ar and the CO2 control group. Out of these 26 genes, HGFAC and ACBD7 were found to be differentially expressed in more than one group. Most DEGs were found for the comparison of Ar and CO2 as well as for N2 and CO2 at 45 min after 36 h, with 10 significant genes each.
Volcano lots of differential gene expression between the inert gases and CO2 in the M. longissimus thoracis et lumborum of pigs at different time points. (A) Ar vs CO2 (45 min), (B) Ar vs CO2 (36 h), (C) N2-Ar vs CO2 (45 min), (D) N2-Ar vs CO2 (36 h). significance thresholds: absolute Log2 Foldchange > 1. adjusted P value < 0.05).
Heatmap of the differentially expressed genes (significance thresholds: absolute Log2 Foldchange > 1, P adjusted < 0.05) between the inert gases and CO2. Stars mark the genes that fulfilled the significance threshold in the comparison.
Differentially expressed genes between time points
The comparisons between the different time points (36 h vs 45 min) postmortem for each gas are shown in Figure 3 (A–D). With an adjusted P value < 0.01 and an absolute Log2 FC > 1, DEGs were filtered for each comparison. In the muscle tissue of the pigs that were stunned with N2 616, DEGs were found, of which 270 were up, and 346 were downregulated 36 h postmortem compared to 45 min postmortem. The comparison between time points within the CO2 (N2) group resulted in 410 DEGs, of which 239 were upregulated and 171 downregulated after 36 h. The most DEGs, with a total of 1958, were found for the Ar group, where 1519 were upregulated and 439 downregulated after 36 h of death, while the CO2 (Ar) control group had the lowest number of DEGs with a total of 216, with 100 genes upregulated and 116 downregulated genes. In our gene set enrichment analysis, pathways involving the extracellular matrix were found in each GO comparison (Figure 4 A-D). The Venn diagram (Figure 5) showed that all comparisons between time points had 112 DEGs in common, which were regulated in the same direction across all gas mixtures used. KEGG pathway analysis revealed them to be involved in the cytoskeleton in muscle cells, protein digestion and absorption, and ECM-receptor interaction (Figure 6). Out of these genes, 24 were upregulated in the muscle 36 h after death (Table 3, Figure 7).
Volcano Plots of differential gene expression between 36 h and 45 min postmortem in the M. longissimus thoracis et lumborum of pigs stunned with different gas mixtures. (A) Ar, (B) CO2 (Ar), (C) N2-Ar, (D) CO2 (N2). significance thresholds: absolute Log2 Foldchange > 1, adjusted P value < 0.01.
Gene set enrichment analysis results for (A) gene ontology cell component, (B) gene ontology molecular function, (C) gene ontology biological processes, and (D) gene ontology KEGG pathways.
Venn diagram of concordant and discordant differentially expressed genes between the group comparisons (significance thresholds: absolute Log2 Foldchange > 1, adjusted P value < 0.01).
Gene set enrichment analysis results for gene ontology KEGG pathways of the 112 differentially expressed genes (significance thresholds: absolute Log2 Foldchange > 1, adjusted P value < 0.01) in all comparisons between 36 h and 45 min.
Upregulated genes in the M. longissimus thoracis et lumborum of pigs at 36 h after death compared to 45 min stunned with different gas mixtures, with Ar = argon; CO2 (Ar) = carbon dioxide control group measured at the same day as argon; N2-Ar = nitrogen-argon mixture; and CO2 (N2) = carbon dioxide control group measured on the same day as the nitrogen-argon mixture
| Ar | CO2 (Ar) | N2-Ar | CO2 (N2) | |||||
|---|---|---|---|---|---|---|---|---|
| Gene Name | LFC | P adj. | LFC | P adj | LFC | P adj | LFC | P adj |
| NRN1 | 5.67 | 0.00 | 4.02 | 0.00 | 5.13 | 0.00 | 4.37 | 0.00 |
| HOXC10 | 2.57 | 0.00 | 2.32 | 0.00 | 1.43 | 0.00 | 3.20 | 0.00 |
| USP44 | 2.28 | 0.00 | 1.30 | 0.00 | 1.72 | 0.00 | 1.86 | 0.01 |
| FASTKD1 | 2.15 | 0.00 | 1.14 | 0.00 | 1.55 | 0.00 | 1.81 | 0.00 |
| DNAJC21 | 1.95 | 0.00 | 1.01 | 0.00 | 1.32 | 0.00 | 1.24 | 0.00 |
| HOMER1 | 1.95 | 0.00 | 1.44 | 0.00 | 1.09 | 0.00 | 1.27 | 0.00 |
| EIF1AY | 1.91 | 0.00 | 2.02 | 0.00 | 1.47 | 0.01 | 2.11 | 0.00 |
| JUNB | 1.82 | 0.00 | 2.16 | 0.00 | 1.11 | 0.00 | 2.61 | 0.00 |
| BIRC3 | 1.81 | 0.00 | 1.17 | 0.00 | 1.45 | 0.00 | 1.40 | 0.00 |
| SMIM10L1 | 1.78 | 0.00 | 1.09 | 0.00 | 1.48 | 0.00 | 1.34 | 0.00 |
| SLTM | 1.73 | 0.00 | 1.12 | 0.00 | 1.22 | 0.00 | 1.24 | 0.00 |
| ZC3H15 | 1.73 | 0.00 | 1.14 | 0.00 | 1.35 | 0.00 | 1.20 | 0.01 |
| MBNL1 | 1.70 | 0.00 | 1.25 | 0.00 | 1.06 | 0.01 | 1.30 | 0.00 |
| ROCK2 | 1.68 | 0.00 | 1.25 | 0.00 | 1.01 | 0.01 | 1.22 | 0.01 |
| NRIP1 | 1.68 | 0.00 | 1.17 | 0.00 | 1.22 | 0.00 | 1.36 | 0.01 |
| SEC62 | 1.67 | 0.00 | 1.13 | 0.00 | 1.29 | 0.00 | 1.38 | 0.01 |
| EIF2S2 | 1.65 | 0.00 | 1.05 | 0.00 | 1.14 | 0.00 | 1.30 | 0.00 |
| CUL5 | 1.62 | 0.00 | 1.06 | 0.00 | 1.11 | 0.00 | 1.17 | 0.00 |
| TMEM106B | 1.61 | 0.00 | 1.07 | 0.00 | 1.17 | 0.00 | 1.25 | 0.01 |
| LUC7L3 | 1.58 | 0.00 | 1.02 | 0.00 | 1.29 | 0.00 | 1.21 | 0.00 |
| RWDD1 | 1.56 | 0.00 | 1.01 | 0.00 | 1.11 | 0.00 | 1.45 | 0.00 |
| PPP1R3A | 1.43 | 0.00 | 1.04 | 0.00 | 1.01 | 0.00 | 1.27 | 0.00 |
| DNAJB4 | 1.39 | 0.00 | 1.27 | 0.00 | 1.20 | 0.00 | 1.26 | 0.00 |
| CEP85L | 1.34 | 0.00 | 1.46 | 0.00 | 1.22 | 0.00 | 1.54 | 0.00 |
Abbreviations: LFC = Log2 Foldchange; P adj. = adjusted P value.
Heatmap of the 112 differentially expressed genes (significance thresholds: absolute Log2 Foldchange > 1, adjusted P value < 0.01) in all comparisons between 36 h and 45 min.
Meat quality measurements
The meat quality of all pigs involved in this study was monitored in terms of pH and temperature at 45 min and 36 h postmortem. None of the pigs showed a pH45 value below 6, and no statistical differences were observed between the stunning groups. However, the Ar groups showed statistically lower T45 values in LTL and SM (Table 4).
Means and SD of corresponding meat quality parameters in M. longissimus thoracis et lumborum (LTL) and M. semimembranosus (SM) of the pigs stunned with the respective gas, with Ar = argon; CO2 (Ar) = carbon dioxide control group measured at the same day as argon; N2-Ar = nitrogen-argon mixture; and CO2 (N2) = carbon dioxide control group measured on the same day as the nitrogen-argon mixture
| Gas mixtures used for stunning | ||||
|---|---|---|---|---|
| Meat quality parameter | Ar | CO2 (Ar) | CO2 (N2) | N2-Ar |
| LTL | ||||
| T45 (°C) | 35.2 ± 1.5a | 36.3 ± 1.2b | 37 ± 0.8b | 36.7 ± 1.7b |
| pH45 | 6.5 ± 0.2 | 6.5 ± 0.1 | 6.5 ± 0.2 | 6.4 ± 0.2 |
| TU (°C) | 5.6 ± 1.1a | 4.5 ± 0.3b | 4.8 ± 0.4b | 4.9 ± 0.8b |
| pHU | 5.62 ± 0.07a | 5.61 ± 0.06a | 5.58 ± 0.05ab | 5.56 ± 0.04b |
| SM | ||||
| T45 (°C) | 36.2 ± 0.9a | 37 ± 0.8b | 37.5 ± 0.6c | 37.5 ± 1.1bc |
| pH45 | 6.5 ± 0.2 | 6.6 ± 0.1 | 6.6 ± 0.2 | 6.5 ± 0.2 |
| TU (°C) | 5.5 ± 0.9a | 4.4 ± 0.3b | 4.7 ± 0.5b | 4.7 ± 0.6b |
| pHU | 5.56 ± 0.03a | 5.56 ± 0.04a | 5.55 ± 0.04ab | 5.53 ± 0.04b |
Abbreviations: T = temperature; indices 45 and U mark the time points of investigation.
Different letters mark significant differences (P < 0.05) within rows.
Discussion
Gene expression differences between gas mixtures used for stunning
Deviations in meat quality have been reported in pigs stunned with inert gases compared to CO2 (Troeger et al., 2004b; Llonch et al., 2012; Atkinson et al., 2020). To identify potential transcriptional changes in the LTL muscle, the transcriptome was analyzed for DEGs resulting from the use of different gases for stunning. The comparisons between inert gases and CO2 at 45 min and 36 h postmortem showed 26 genes overall that were found to be differentially expressed in at least one of the comparisons between inert gas and CO2 stunning (Figure 2). Among these genes, CISH, JUNB, ACBD7, FOS, and ATF3 were also identified in the study by Zequan et al. (2022), who compared the transcriptome of LTL and SM exhibiting PSE characteristics to normal pork. They reported a downregulation of JUNB, ABCD7, FOS, and ATF3 in LTL with PSE characteristics. In our study, these genes were likewise downregulated in the inert gas groups, while CISH was upregulated. Although our subset of genes displayed a similar expression pattern, our findings nevertheless contrast with those of Zequan et al. (2022) for certain genes, as none of the animals in our study showed a pH45 < 5.8 and were therefore not classified as PSE meat. Additionally, no statistically significant differences were found for pH45 in SM and LTL of the pigs stunned with different gas mixtures. However, some of these genes found by Zequan et al. (2022), namely JUNB, FOS, as well as ATF3, were also found to be involved in the regulation of the intramuscular fat content in Iberian pigs (Muñoz et al., 2018). These genes, as well as EGR1 and FOSB, were downregulated in pigs with a higher intramuscular fat content (Muñoz et al., 2018), indicating a potentially higher intramuscular fat content in the Ar or N2 groups in the study presented herein. Given that intramuscular fat content or backfat thickness were not measured in the presented study, it cannot be proven whether the observed differences in gene expression are due to variation in intramuscular fat among the studied pigs. However, it must be taken into account that these DEGs appeared to be significant inconsistently across the different comparisons. As the same animals were investigated at the 2 time points, we would expect the same genes to be expressed across all comparisons if fat content is the primary driver of the differential expression of these genes. Additionally, the expression pattern of CISH does not fit this previous assumption. CISH encodes a cytokine-inducible SH2-containing protein and has often been associated with adipose tissue. Naser et al. (2022) showed that a knockout of the CISH gene in male mice led to a significant reduction in fat mass. In pigs, CISH was upregulated in individuals with a significantly lower backfat thickness (Ropka-Molik et al., 2014). In the presented study, CISH was upregulated in the pigs stunned with Ar compared to CO2 at both postmortem time points, which does not support the assumption of a higher fat content in the Ar groups. It becomes evident that, besides meat quality and carcass characteristics, further factors must be considered to explain the observed differences in gene expression. The genes JUNB, FOS, FOSB, ATF3, EGR1, as well as IER5 belong to the so-called immediate early genes, and respond very quickly to external and internal cell stimuli (reviewed in Bahrami and Drabløs, 2016). For example, heat-shocked cells increased their expression of IER5 and FOS when they were incubated at 42.5°C compared to 37°C (Ishikawa and Sakurai, 2015), while ATF3 was described to be upregulated in the brown fat tissue of mice exhibiting cold stress (Verma et al., 2020). In our study, a significantly higher TU in the LTL was observed in the Ar group compared to the other groups, which could affect the expression of ATF3. However, the immediate early response genes JUNB, FOS, EGR1, as well as IER5, were differentially expressed in the comparison of N2-Ar and CO2 after 45 min, despite no difference in temperature between these conditions. We therefore conclude that the differences in gene expression may be partially driven by temperature differences, although other unmeasured stimuli are also likely to contribute to the observed transcriptional patterns. We therefore recommend against using these genes as reliable markers for meat quality traits in expression studies. Furthermore, our findings indicate that the use of inert gases on pigs does not alter the expression of genes involved in the meat maturation process.
Gene expression differences between time points
The process by which muscle matures into meat is crucial for the subsequent use of the product. Several studies have been carried out to investigate differences in meat quality on a transcriptional level (et al., 2010; Zhao et al., 2019; Zequan et al., 2022; Wang et al., 2024b). While most studies focus on the early postmortem stages of meat maturation, Fontanesi et al. (2008) reported that porcine RNA remains stable up to 24 h postmortem, but degrades drastically at 48 h. In the study presented here, samples were taken from the LTL at 45 min and 36 h postmortem. As expected, RIN values decreased over time after death, but the samples were still suitable for downstream analysis, as none of the generated pooled samples failed in the sequencing process. The dynamics of the postmortem transcriptome have already been described by Pozhitkov et al. (2017) in zebrafish and in the liver and the brain of mice. In zebrafish, the overall abundance of total RNA decreased strongly between 12 and 24 h postmortem, which was not the case in mouse brain samples, and suggested a tissue-specific mRNA abundance over time, as some genes had the highest abundance at 24 h postmortem in both species. In pigs, Fontanesi et al. (2011) performed a similar experiment with 3 female pigs at 20 min, 2 h, 6 h, and 24 h postmortem, but reported no different expression profiles between the time points. These results are in contrast to the findings here, where several DEGs were found between 45 min and 36 h postmortem, respectively. A total of 112 genes were differentially expressed in all 4 comparisons, where most of them were upregulated at 45 min postmortem (Figure 7). Some of these genes, namely APOE, ADIPOQ, and PLIN1, have been described to be involved in lipid metabolism in pigs (Passols et al., 2023), while another subset of genes encoding collagens COL1A1, COL3A1, COL4A1, COL4A2, COL5A1, COL5A3, COL6A2, COL6A3, and COL15A1, were downregulated at 36 h postmortem. KEGG pathway analysis revealed several of them to be involved in the cytoskeleton in muscle cells, protein digestion and absorption, and ECM-receptor interaction. Collagens and the expression of their genes have been studied in several contexts of meat quality. The expression of COL1A1, COL4A1, and COL6A3 is correlated with higher drip loss (Ponsuksili et al., 2008), while COL1A1, COL5A1, and COL15A1 were upregulated in Berkshire pigs with a higher intramuscular fat content. While most of the DEGs were downregulated 36 h postmortem, 24 genes were upregulated (Table 2). Protein phosphatase-1 regulatory subunit 3A (PPP1R3A) is one of these genes, which was upregulated 36h postmortem. This subunit of protein phosphatase-1 (PP-1) plays a crucial role in the glycogen metabolism of muscle cells by activation of glycogen synthase, leading to increased cellular glycogen stores, while reducing glycogen phosphorylase activity, when the glucose-6-phosphate content is high (Lerín et al., 2003). Following death, muscle glycogen breakdown is the central metabolism, crucial for the development of meat maturation defects (reviewed in Scheffler and Gerrard, 2007). It is therefore likely that these ongoing processes affect PPP1R3A expression. However, our results reveal PPP1R3A to be upregulated after 36 h compared to 45 min postmortem. However, we suggest that the upregulation observed at 36 h might not represent a true postmortem increase of expression, but rather a relative decrease at 45 min. This suggestion emphasizes the potential value of knowing antemortem expression levels, which remain unknown in this study, as these would allow a more precise understanding of the gene’s regulation. Nevertheless, PPP1R3A appears to be of considerable interest in the context of postmortem glycolysis, as its expression has previously been associated with altered ultimate pH (pHu) values in chicken muscle (Beauclercq et al., 2017). Our results reveal several genes to be differentially expressed between postmortem time points in the LTL of slaughter pigs. In particular, genes related to lipid metabolism, cell structures, and glycogen metabolism were upregulated in the early postmortem period. Furthermore, our results support previous findings, indicating ongoing transcriptional activity after death (Tolbert et al., 2018), with evidence of increased activity for certain transcripts at longer postmortem intervals. As some of these genes have already been the subject of investigations into meat quality traits, we recommend taking changes of the respective transcripts during sampling into account to avoid biasing the results.
Statistical limitations
Differential gene expression analysis was performed using 3 pooled replicates per condition, except for the CO2 (Ar) group, which included only 2 pooled replicates per time point. Pooling samples has been shown to reduce biological variability within groups, for instance, resulting from sex (Zhang et al., 2013), increasing the detectability of biological affected processes (Takele Assefa et al., 2020). However, using a low number of replicates is considered to reduce the power of a statistical model. To address this limitation, DESeq2 (Love et al., 2014) was used, which has been demonstrated to perform robustly even with only a few replicates (Schurch et al., 2016). It has been shown by Schurch et al. (2016) that DESeq2 controls its false discovery rate to be below 5%, regardless of the replicate number, including 2 and 3 replicates, albeit with reduced sensitivity. Consequently, some DEGs may not have been detected, resulting in the lowest number of DEGs in the comparison over time for CO2 (Ar) and only one DEG between Ar and the CO2 (Ar) control. Nevertheless, since no significant differences in pH45 between the groups were observed, a high magnitude of DEGs was not expected. Furthermore, applying an absolute log2FC threshold greater than 0.5 increases the true positive rates, and in this study, a conservative threshold of 1 was used (Schurch et al., 2016; Takele Assefa et al., 2020). Despite the low number of replicates, the combination of pooled sampling, DESeq2, and a stringent log2FC cutoff ensured a reliable first overview of biologically relevant expression changes, although further studies are needed to uncover potentially hidden processes.
Conclusion
In the study presented here, we were able to describe the trajectories of the transcription profiles over the initial time period of meat maturation in pigs. Although most of the transcripts were downregulated at 36 h postmortem, 24 genes were found to be upregulated compared to 45 min. The comparison between different stunning gases only revealed a few genes to be differentially expressed between the groups. We hypothesize that these genes are affected by several ongoing processes during meat maturation and are therefore not altered by the gas used for stunning.
Ethics Approval
This study was approved by the Ethics Committee of the “Thueringer Landesamt für Verbraucherschutz” (TLV), file number: 22-2684-O4-BFI-21-001.
Funding
The project “Testing Inert Gases in order to Establish Replacements for high concentration CO2 stunning for pigs at the time of slaughter” (TIGER) was supported by funds of the Federal Ministry of Agriculture, Food and Regional Identity (BMELH) based on a decision of the Parliament of the Federal Republic of Germany via the Federal Office for Agriculture and Food (BLE) under the innovation support program (grant number 2817803B18). The project was financially supported with additional funds from Verband der Fleischwirtschaft e.V., from QS Science Funds of the QS Qualität and Sicherheit GmbH and from Fördergesellschaft für Fleischforschung e.V. (Kulmbach, Germany).
Author contributions
JG: Writing—original draft, Methodology, Investigation, Formal analysis; NFP: Writing—review & editing, Investigation; CFG: Writing—review & editing, Data curation, Formal analysis; JK: Writing—review & editing, Methodology, Conceptualization; IW: Writing—review & editing, Project administration, Methodology, Conceptualization; DM: Funding acquisition, Conceptualization; JT: Writing—review & editing, Supervision, Funding acquisition, Conceptualization.
Acknowledgments
We acknowledge support by the Open Access Publication Funds of the Göttingen University. We further thank the “Gesellschaft für wissenschaftliche Datenverarbeitung mbH Göttingen” (GWDG) for the access to their high-performance computing cluster and the board for financial support of early career scientists of the Department of Animal Science, Georg – August- University Göttingen.
Declaration of Interest
The authors declare no conflicts of interest.
Literature Cited
Assou, S., I. Boumela, D. Haouzi, T. Anahory, H. Dechaud, J. de Vos, and S. Hamamah. 2011. Dynamic changes in gene expression during human early embryo development: from fundamental aspects to clinical applications. Hum. Reprod. Update 17:272–290. doi: https://doi.org/10.1093/humupd/dmq036.
Atkinson, S., B. Algers, J. Pallisera, A. Velarde, and P. Llonch. 2020. Animal welfare and meat quality assessment in gas stunning during commercial slaughter of pigs using hypercapnic-hypoxia (20% CO2 2% O2) compared to acute hypercapnia (90% CO2 in Air). Animals. 10:2440. doi: https://doi.org/10.3390/ani10122440.
Bahrami, S. and F. Drabløs. 2016. Gene regulation in the immediate-early response process. Advances in biological regulation. 62:37–49. doi: https://doi.org/10.1016/j.jbior.2016.05.001.
Beauclercq, S., C. Hennequet-Antier, C. Praud, E. Godet, A. Collin, S. Tesseraud, S. Métayer-Coustard, M. Bourin, M. Moroldo, F. Martins, S. Lagarrigue, E. Le Bihan-Duval, and C. Berri. 2017. Muscle transcriptome analysis reveals molecular pathways and biomarkers involved in extreme ultimate pH and meat defect occurrence in chicken. Sci. Rep. 7:6447. doi: https://doi.org/10.1038/s41598-017-06511-6.
Blighe, K., S. Rana, and M. Lewis. 2018. EnhancedVolcano: Publication-ready volcano plots with enhanced colouring and labeling. R package version 1.22.0. doi: https://doi.org/10.18129/B9.bioc.EnhancedVolcano.
Bowker, B. C., C. Botrel, D. R. Swartz, A. L. Grant, and D. E. Gerrard. 2004. Influence of myosin heavy chain isoform expression and postmortem metabolism on the ATPase activity of muscle fibers. Meat Sci. 68:587–594. doi: https://doi.org/10.1016/j.meatsci.2004.05.010.
Channon, H. A., A. M. Payne, and R. D. Warner. 2002. Comparison of CO(2) stunning with manual electrical stunning (50 Hz) of pigs on carcass and meat quality. Meat Sci. 60:63–68. doi: https://doi.org/10.1016/s0309-1740(01)00107-3.
Chen, Y., Y. Chen, C. Shi, Z. Huang, Y. Zhang, S. Li, Y. Li, J. Ye, C. Yu, Z. Li, X. Zhang, J. Wang, H. Yang, L. Fang, and Q. Chen. 2018. SOAPnuke: a MapReduce acceleration-supported software for integrated quality control and preprocessing of high-throughput sequencing data. GigaScience 7:1–6. doi: https://doi.org/10.1093/gigascience/gix120.
Dalmau, A., P. Rodríguez, P. Llonch, and A. Velarde. 2010. Stunning pigs with different gas mixtures: aversion in pigs. Anim. Welfare 19:325–333. doi: https://doi.org/10.1017/S096272860000172X.
European Council Regulation No 1099/2009 2009. European Council Regulation No 1099/2009. COUNCIL REGULATION (EC) No 1099/2009 of 24 September 2009 on the protection of animals at the time of killing. https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32009R1099. (Accessed 10 March 2025).https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32009R1099
Fontanesi, L., M. Colombo, F. Beretti, and V. Russo. 2008. Evaluation of postmortem stability of porcine skeletal muscle RNA. Meat Sci. 80:1345–1351. doi: https://doi.org/10.1016/j.meatsci.2008.06.014.
Fontanesi, L., G. Galimberti, D. G. Calò, M. Colombo, A. Astolfi, S. Formica, and V. Russo. 2011. Microarray gene expression analysis of porcine skeletal muscle sampled at several postmortem time points. Meat Sci. 88:604–609. doi: https://doi.org/10.1016/j.meatsci.2011.02.001.
Fox, J. and S. Weisberg. 2019. An R companion to applied regression, 3rd ed. Sage, Thousand Oaks.
Gelhausen, J., T. Friehs, N.-F. Paul, T. Krebs, J. Mörlein, J. Knöll, I. Wilk, J. Tetens, and D. Mörlein. 2025. Stunning pigs with inert gases at low residual oxygen does not compromise meat quality. Meat Sci. 228:109898. doi: https://doi.org/10.1016/j.meatsci.2025.109898.
Gong, Z. and H. An. 2025. Integrated transcriptomic analysis of COVID-19 stages and recovery: insights into key gene signatures, immune features, and diagnostic biomarkers through machine learning. Frontiers in Genetics 1599867. doi: https://doi.org/10.3389/fgene.2025.1599867.
Ishikawa, Y. and H. Sakurai. 2015. Heat-induced expression of the immediate-early gene IER5 and its involvement in the proliferation of heat-shocked cells. FEBS J. 282:332–340. doi: https://doi.org/10.1111/febs.13134.
Javan, G. T., E. Hanson, S. J. Finley, S. D. Visonà, A. Osculati, and J. Ballantyne. 2020. Identification of cadaveric liver tissues using thanatotranscriptome biomarkers. Scientific Rep. 10:6639. doi: https://doi.org/10.1038/s41598-020-63727-9.
Kassambara, A. 2023. rstatix: Pipe-friendly framework for basic statistical tests. R package version 0.7.2. doi: https://doi.org/10.32614/CRAN.package.rstatix.
Kim, D., B. Langmead, and S. L. Salzberg. 2015. HISAT: a fast spliced aligner with low memory requirements. Nat. Methods 12:357–360. doi: https://doi.org/10.1038/nmeth.3317.
Kolde, R. 2018. pheatmap: Pretty heatmaps. R package version 1.0.12. doi: https://doi.org/10.32614/CRAN.package.pheatmap.
Lawrence, M., W. Huber, H. Pagès, P. Aboyoun, M. Carlson, R. Gentleman, M. T. Morgan, and V. J. Carey. 2013. Software for computing and annotating genomic ranges. PLoS Comput. Biol. 9:e1003118. doi: https://doi.org/10.1371/journal.pcbi.1003118.
Lerín, C., E. Montell, T. Nolasco, C. Clark, M. J. Brady, C. B. Newgard, and A. M. Gómez-Foix. 2003. Regulation and function of the muscle glycogen-targeting subunit of protein phosphatase 1 (GM) in human muscle cells depends on the COOH-terminal region and glycogen content. Diabetes 52:2221–2226. doi: https://doi.org/10.2337/diabetes.52.9.2221.
Llobat, L. 2020. Embryo gene expression in pig pregnancy. Reprod. Domest. Anim. 55:523–529. doi: https://doi.org/10.1111/rda.13647.
Llonch, P., P. Rodríguez, M. Gispert, A. Dalmau, X. Manteca, and A. Velarde. 2012. Stunning pigs with nitrogen and carbon dioxide mixtures: effects on animal welfare and meat quality. Animal 6:668–675. doi: https://doi.org/10.1017/S1751731111001911.
Love, M. I., W. Huber, and S. Anders. 2014. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15:550. doi: https://doi.org/10.1186/s13059-014-0550-8.
Mörlein, D., G. Link, C. Werner, and M. Wicke. 2007. Suitability of three commercially produced pig breeds in Germany for a meat quality program with emphasis on drip loss and eating quality. Meat Sci. 77:504–511. doi: https://doi.org/10.1016/j.meatsci.2007.04.030.
Muñoz, M., J. M. García-Casco, C. Caraballo, M. Á. Fernández-Barroso, F. Sánchez-Esquiliche, F. Gómez, M. C. Del Rodríguez, and L. Silió. 2018. Identification of candidate genes and regulatory factors underlying intramuscular fat content through longissimus dorsi transcriptome analyses in heavy Iberian pigs. Frontiers in Genetics 9:608. doi: https://doi.org/10.3389/fgene.2018.00608.
Naser, W., S. Maymand, L. R. Rivera, T. Connor, C. Liongue, C. M. Smith, K. Aston-Mourney, D. R. McCulloch, S. L. McGee, and A. C. Ward. 2022. Cytokine-inducible SH2 domain containing protein contributes to regulation of adiposity, food intake, and glucose metabolism. FASEB J. 36:e22320. doi: https://doi.org/10.1096/fj.202101882R.
Ogle, D. H., J. C. Doll, A. Powell Wheeler, and A. Dinno. 2025. FSA: Simple fisheries stock assessment methods. R package version 0.9.6. doi: https://doi.org/10.32614/CRAN.package.FSA.
Ouali, A., C. H. Herrera-Mendez, G. Coulis, S. Becila, A. Boudjellal, L. Aubry, and M. A. Sentandreu. 2006. Revisiting the conversion of muscle into meat and the underlying mechanisms. Meat Sci. 74:44–58. doi: https://doi.org/10.1016/j.meatsci.2006.05.010.
Passols, M., F. Llobet-Cabau, C. Sebastià, A. Castelló, J. Valdés-Hernández, L. Criado-Mesas, A. Sánchez, and J. M. Folch. 2023. Identification of genomic regions, genetic variants and gene networks regulating candidate genes for lipid metabolism in pig muscle. Animal 17:101033. doi: https://doi.org/10.1016/j.animal.2023.101033.
Ponsuksili, S., E. Jonas, E. Murani, C. Phatsara, T. Srikanchai, C. Walz, M. Schwerin, K. Schellander, and K. Wimmers. 2008. Trait correlated expression combined with expression QTL analysis reveals biological pathways and candidate genes affecting water holding capacity of muscle. BMC Genomics 9:367. doi: https://doi.org/10.1186/1471-2164-9-367.
Pozhitkov, A. E., R. Neme, T. Domazet-Lošo, B. G. Leroux, S. Soni, D. Tautz, and P. A. Noble. 2017. Tracing the dynamics of gene transcripts after organismal death. Open Biol. 7:160267. doi: https://doi.org/10.1098/rsob.160267.
R Core Team. 2024. R: A language and environment for statistical computing. R version 4.4.0.
Raj, A. B. M. and N. G. Gregory. 1995. Welfare implications of the gas stunning of pigs 1. Determination of aversion to the initial inhalation of carbon dioxide or argon. Anim. Welfare 4:273–280. doi: https://doi.org/10.1017/S096272860001798X.
Ropka-Molik, K., K. Zukowski, R. Eckert, A. Gurgul, K. Piórkowska, and M. Oczkowicz. 2014. Comprehensive analysis of the whole transcriptomes from two different pig breeds using RNA-Seq method. Animal Genet. 45:674–684. doi: https://doi.org/10.1111/age.12184.
Ryu, Y. C., Y. M. Choi, and B. C. Kim. 2005. Variations in metabolite contents and protein denaturation of the longissimus dorsi muscle in various porcine quality classifications and metabolic rates. Meat Sci. 71:522–529. doi: https://doi.org/10.1016/j.meatsci.2005.04.034.
Scheffler, T. L. and D. E. Gerrard. 2007. Mechanisms controlling pork quality development: The biochemistry controlling postmortem energy metabolism. Meat Sci. 77:7–16. doi: https://doi.org/10.1016/j.meatsci.2007.04.024.
Schurch, N. J., P. Schofield, M. Gierliński, C. Cole, A. Sherstnev, V. Singh, N. Wrobel, K. Gharbi, G. G. Simpson, T. Owen-Hughes, M. Blaxter, and G. J. Barton. 2016. How many biological replicates are needed in an RNA-seq experiment and which differential expression tool should you use? RNA 22:839–851. doi: https://doi.org/10.1261/rna.053959.115.
Scott, L., S. J. Finley, C. Watson, and G. T. Javan. 2020. Life and death: A systematic comparison of antemortem and postmortem gene expression. Gene 731:144349. doi: https://doi.org/10.1016/j.gene.2020.144349.
Sieczkowska, H., A. Zybert, E. Krzecio, K. Antosik, M. Koćwin-Podsiadła, M. Pierzchała, and P. Urbański. 2010. The expression of genes PKM2 and CAST in the muscle tissue of pigs differentiated by glycolytic potential and drip loss, with reference to the genetic group. Meat Sci. 84:137–142. doi: https://doi.org/10.1016/j.meatsci.2009.08.038.
Takele Assefa, A., J. Vandesompele, and O. Thas. 2020. On the utility of RNA sample pooling to optimize cost and statistical power in RNA sequencing experiments. BMC Genomics 21:312. doi: https://doi.org/10.1186/s12864-020-6721-y.
Tang, R., B. Yu, K. Zhang, X. Guo, G. Tian, Z. Huang, X. Chen, and D. Chen. 2010. Effects of nutritional level on pork quality and gene expression of micro-calpain and calpastatin in muscle of finishing pigs. Meat Sci. 85:768–771. doi: https://doi.org/10.1016/j.meatsci.2010.04.002.
Terlouw, C., C. Bourguet, and V. Deiss. 2016. Consciousness, unconsciousness and death in the context of slaughter. Part I. Neurobiological mechanisms underlying stunning and killing. Meat Sci. 118:133–146. doi: https://doi.org/10.1016/j.meatsci.2016.03.011.
Tolbert, M., S. J. Finley, S. D. Visonà, S. Soni, A. Osculati, and G. T. Javan. 2018. The thanatotranscriptome: Gene expression of male reproductive organs after death. Gene 675:191–196. doi: https://doi.org/10.1016/j.gene.2018.06.090.
Troeger, K., U. Machold, M. Moje, and M. Behrschmidt. 2004a. Gasbetäubung von Schweinen : ein Vergleich von Kohlendioxid, Argon und einer Stickstoff-Argon-Mischung bezüglich der Schlachtkörper- und Fleischqualität. 1. Problemstellung, Material und Methodik. Fleischwirtschaft 84:104–106.
Troeger, K., U. Machold, M. Moje, and M. Behrschmidt. 2004b. Gasbetäubung von Schweinen: Ein Vergleich von Kohlendioxid, Argon und einer Stickstoff-Argon-Mischung bezüglich der Schlachtkörper-und Fleischqualität-2. Ergebnisse. Fleischwirtschaft 84:117–121.
Velarde, A., M. Gispert, L. Faucitano, X. Manteca, and A. Diestre. 2000. The effect of stunning method on the incidence of PSE meat and haemorrhages in pork carcasses. Meat Sci. 55:309–314. doi: https://doi.org/10.1016/s0309-1740(99)00158-8.
Verma, N., L. Perie, and E. Mueller. 2020. The mRNA levels of heat shock factor 1 are regulated by thermogenic signals via the cAMP-dependent transcription factor ATF3. J. Biol. Chem. 295:5984–5994. doi: https://doi.org/10.1074/jbc.RA119.012072.
Wang, L., Y. Zhou, Y. Wang, and T. Shan. 2024a. Integrative cross-species analysis reveals conserved and unique signatures in fatty skeletal muscles. Scientific Data 11:290. doi: https://doi.org/10.1038/s41597-024-03114-5.
Wang, W., D. Wang, X. Zhang, X. Liu, X. Niu, S. Li, S. Huang, X. Ran, and J. Wang. 2024b. Comparative dorsi muscle reveal potential genes affecting meat trait in Chinese indigenous Xiang pig. Scientific Rep. 14:8486. doi: https://doi.org/10.1038/s41598-024-58971-2.
Xu, L., H. Zhang, H. Yue, S. Wu, H. Yang, Z. Wang, and G. Qi. 2018. Gas stunning with CO2 affected meat color, lipid peroxidation, oxidative stress, and gene expression of mitogen-activated protein kinases, glutathione S-transferases, and Cu/Zn-superoxide dismutase in the skeletal muscles of broilers. J. Anim. Sci. Biotechno. 9:37. doi: https://doi.org/10.1186/s40104-018-0252-2.
Yu, G. 2024. enrichplot: Visualization of functional enrichment result. R package version 1.24.0. doi: https://doi.org/10.18129/B9.bioc.enrichplot.
Yu, G., L.-G. Wang, Y. Han, and Q.-Y. He. 2012. clusterProfiler: An R package for comparing biological themes among gene clusters. Omics: A Journal of Integrative Biology 16:284–287. doi: https://doi.org/10.1089/omi.2011.0118.
Zequan, X., S. Yonggang, X. Heng, W. Yaodong, M. Xin, L. Dan, Z. Li, D. Tingting, and W. Zirong. 2022. Transcriptome-based analysis of early post-mortem formation of pale, soft, and exudative (PSE) pork. Meat Sci. 194:108962. doi: https://doi.org/10.1016/j.meatsci.2022.108962.
Zhang, J., C. Zhou, J. Ma, L. Chen, A. Jiang, L. Zhu, S. Shuai, J. Wang, M. Li, and X. Li. 2013. Breed, sex and anatomical location-specific gene expression profiling of the porcine skeletal muscles. BMC Genetics 14:53. doi: https://doi.org/10.1186/1471-2156-14-53.
Zhao, X., C. Wang, Y. Wang, H. Lin, H. Wang, H. Hu, and J. Wang. 2019. Comparative gene expression profiling of muscle reveals potential candidate genes affecting drip loss in pork. BMC Genetics. 20:89. doi: https://doi.org/10.1186/s12863-019-0794-0.
Zhu, Y., L. Wang, Y. Yin, and E. Yang. 2017. Systematic analysis of gene expression patterns associated with postmortem interval in human tissues. Scientific Rep. 7:5435. doi: https://doi.org/10.1038/s41598-017-05882-0.