Abstract

   Pingliang Red Cattle, a renowned geographical indication product in
   China, is distinguished by its superior meat quality, yet the
   scientific basis for its unique attributes remains underexplored. This
   study integrated metabolomic and transcriptomic analyses to elucidate
   the biochemical and physiological factors underlying the enhanced
   flavor, color stability, and tenderness of Pingliang Red Cattle beef
   compared to Qinchuan and Simmental cattle. Metabolomic profiling
   revealed significantly elevated levels of inosine monophosphate (IMP,
   2.86–3.96× higher) and glutathione (GSH, 2.42–5.43× higher) in
   Pingliang Red Cattle, contributing to intense umami flavor and
   prolonged meat color retention. Notably, ergothioneine (EGT), a potent
   antioxidant, was identified for the first time in Pingliang Red Cattle
   beef, with concentrations 2.55× and 4.25× higher than in Qinchuan and
   Simmental, respectively. Transcriptomic analysis highlighted the
   upregulation of 21 tenderness-related genes (e.g., FABP3, PRDX6, CAST)
   and key enzymes in purine and glutathione metabolism pathways (e.g.,
   PDE4D, ADSL, GGT1), correlating with meat tenderness and the improved
   meat quality. Additionally, Pingliang Red Cattle’s natural forage-rich
   diet and low-density rearing practices were critical in enhancing these
   traits. These findings provide a scientific foundation for Pingliang
   Red Cattle’s premium quality, offering actionable insights for GI
   product branding, quality optimization, and market competitiveness. The
   multi-omics approach established here serves as a paradigm for quality
   assessment and improvement of other GI agricultural products, bridging
   traditional reputation with molecular evidence.

   Keywords: metabolome, transcriptome, meat quality, Pingliang Red Cattle

1. Introduction

   Geographical indication products are often highly favored by consumers
   for their superior quality and unique characteristics. These products
   are typically closely associated with specific geographical regions,
   where the natural environment, traditional craftsmanship, and cultural
   heritage collectively contribute to their distinct attributes [[40]1].
   However, despite their high market potential, many of these products
   suffer from a lack of clear definition regarding their specific
   features, which limits their market promotion and further development.
   The absence of a solid scientific basis to support their uniqueness
   makes it difficult for these products to stand out in a competitive
   market [[41]2]. Pingliang Red Cattle is a well-known geographical
   indication product among the general public [[42]3]. Pingliang Red
   Cattle beef is renowned for its vibrant color, delicious taste, and
   tender texture. The marble score of Pingliang Red Cattle beef is
   greater than or equal to 4.5 and the taste is comparable to that of
   “Japanese Black Cattle” meat [[43]4]. From a nutritional perspective,
   beef is an excellent source of high-quality protein, essential amino
   acids, bioavailable iron (heme iron), zinc, and B vitamins, which play
   crucial roles in muscle maintenance, immune function, and metabolic
   health [[44]5]. According to a previous study, compared to other
   similar types of beef, the Pingliang Red Cattle beef has 18.92% higher
   total content of flavor precursor amino acids; it also contains 11.58%
   more oleic acid and 11.06% more linoleic acid. Additionally, the
   content of the aroma compound 1-octen-3-ol is 55.85% higher than other
   similar types of beef [[45]6].

   However, for a long time, the scientific foundation behind these
   desirable sensory qualities was unclear. This situation not only
   affects the further breeding and quality improvement of the product but
   also hinders brand enhancement and the realization of premium pricing.
   To overcome these challenges, it is necessary to conduct scientific
   research to uncover the specific reasons behind its superior qualities
   and to leverage these scientific findings to strengthen brand building
   and enhance market competitiveness. Such efforts can not only bring
   broader market prospects for Pingliang Red Cattle but also provide a
   successful model for other geographical indication products, helping
   them better showcase their unique value, and aiming to achieve the goal
   of premium quality at a fair price.

   In recent years, the development of omics technologies has provided
   effective tools for uncovering characteristic factors in foods.
   Particularly, metabolomics and transcriptomics technology have emerged
   as robust tools in the agricultural and food research fields, offering
   comprehensive insights into the complex biological processes that
   underpin food safety, quality, and sustainability. Untargeted
   metabolomics, leveraging high-resolution mass spectrometry (HRMS)
   platforms, has been developing rapidly and boasts a broad range of
   applications, it could detect and quantify as many metabolites as
   possible within a biological sample, facilitating the discovery of
   novel metabolic pathways or biomarkers. In recent years, untargeted
   metabolomics analysis has been successfully applied in meat freshness
   evaluation [[46]7], food safety assessment [[47]8], food fraud
   identification [[48]9], monitoring of component changes during food
   processing [[49]10], and so on. Along with the continuous advancements
   in the Next-generation sequencing (NGS) technology, transcriptomics has
   profoundly deepened our comprehension of how genes are expressed within
   organisms, revealing intricate patterns and regulatory mechanisms that
   govern cellular functions [[50]11]. At present, transcriptomics has
   been widely used in many fields, such as disease diagnosis and drug
   discovery [[51]12], crop and animal breed improvement [[52]13] and
   plant-pathogen interactions [[53]14], etc. Additionally, the
   integration of metabolomics with transcriptomics, provide a
   systems-level understanding by analyzing their metabolites and gene
   expression profiles, which offers unprecedented opportunities for
   uncovering the distinctive qualities of Geographical Indication
   products. In recent years, livestock products [[54]15], agricultural
   products [[55]16], microbial fermentation, and pharmacological
   discovery [[56]17] have greatly benefited from combined metabolomic and
   transcriptomic analyses.

   In this study, metabolomic and transcriptomic analysis were integrated
   to uncover the distinctive quality indicators of Pingliang Red Cattle
   and the underlying gene expression mechanism. Results showed that
   Pingliang Red Cattle beef exhibit more intense umami flavor,
   longer-lasting color, and tenderer texture than the Qinchuan and
   Simmental cattle beef. Moreover, this study also made the first
   discovery of the presence of ergothioneine, a bioactive alkaloid and
   potent natural antioxidant, within Pingliang Red Cattle beef. Finally,
   this study analyzed the relationship between these distinctive quality
   traits and environmental feed factors, providing a paradigm for the
   quality exploration of Geographical Indication products.

2. Materials and Methods

2.1. Chemicals and Reagents

   Both formic acid and methanol (chromatographically pure) were purchased
   from Beijing Dima Technology Co., Ltd. (Beijing, China). Acetonitrile
   (chromatographically pure) was obtained from Thermo Fisher Scientific
   (Fair Lawn, NJ, USA). Ultrapure water was obtained using a Milli-Q
   Integral Water Purification System (Millipore, Billerica, MA, USA). A
   TransZolUp Plus RNA Kit (Hangzhou Chuangshi Biotechnology Co., Ltd.,
   Hangzhou, China) was used to extract total RNA. Reverse transcription
   was performed using an IST Strand cDNA Synthesis Kit (Baori Medical
   Biotechnology Co., Ltd., Beijing, China). SYBR FAST qPCR Kit Master Mix
   (2×) Universal (Annolun Biotechnology Co., Ltd., Beijing, China) was
   used for quantitative polymerase chain reaction (qPCR) validation.

2.2. Sample Collection and Preparation

   Pingliang Red Cattle samples were obtained from four certified local
   feeding business from Pingliang city (China), where all cattle were
   free-range grazing without confinement in modern barns, Qinchuan and
   Simmental cattle samples (captive) were used as a control. The
   Simmental cattle were purchased from Horqin Cattle Industry Co., Ltd.
   (Inner Mongolia, China). To enhance the universality and reliability of
   this study, we selected cattle of the same breed from different
   pastures without imposing any deliberate restrictions on feeding
   methods, age, weight, or other variables. This approach could better
   reflect the variability and complexity under actual farming conditions,
   ensuring the broad applicability of the study results. Additionally, we
   concurrently collected and analyzed relevant information on the rearing
   of Pingliang Red Cattle, including feed types, feed consumption rates,
   and rearing methods. All samples (approximately 500 g each) were taken
   from the loin of healthy bulls that had not received antibiotic
   treatment for at least 3 months before sampling. Twenty samples were
   isolated from Pingliang Red Cattle and 15 samples were isolated from
   each group of control cattle (Qinchuan and Simmental). Information on
   samples collected from the three beef cattle breeds is shown in
   [57]Table S1. From each of the three sample groups, four samples were
   randomly selected. These selected samples were then cut into 8–10 g
   pieces using scalpels and stored at –80 °C until they were subjected to
   transcriptome analysis. The remaining meat samples were minced using a
   meat grinder and stored at −20 °C until further analyses.

2.3. Metabolomic Profiling

   Metabolites were extracted as previously described [[58]18], with some
   modifications. Each tenderloin sample (weighing 150 mg, stored at −20
   °C) was placed in a 2 mL centrifuge tube, and 1 mL of pre-frozen
   methanol/water solution (1/1, v/v) was added. The samples were
   ultrasonically treated in a water bath (20 °C) for 10 min, vortexed for
   5 min, and centrifuged at 14,500× g for 20 min (4 °C). The supernatant
   was filtered through a 0.22 µm polytetrafluoroethylene membrane to
   obtain the final product. For each beef sample, quality control (QC)
   samples were prepared by equal pooling to monitor the stability and
   robustness of the data based on a previously described method [[59]19].
   A QC sample was inserted and measured after measuring every eight test
   beef samples during instrumental analysis. All samples were analyzed
   using a UPLC system equipped with a Q-TOF mass spectrometer (TripleTOF
   6600, AB Sciex, Framingham, MA, USA). The solvent system used for UPLC
   comprised mobile phase A of 0.1% formic acid in water and mobile phase
   B of 0.1% formic acid in methanol. Chromatographic separation was
   performed using a C[18] column (Zorbax Eclipse C[18], 3.0 mm × 150 mm,
   1.8 μm, Agilent, Foster City, CA, USA), which was maintained at 40 °C.
   Five µL of the extract was then injected at 4 °C, with the flow rate of
   the mobile phase being 0.3 mL/min. The elution gradient was as follows:
   0.00–1.00 min, 90% A (10% B); 1.00–1.01 min, 90% A (10% B); 1.01–12.00
   min,10% A (90% B); 12.00–12.10 min, 10% A (90% B); and 12.10–17 min,
   90% A (10% B). TOF MS and product ion scanning were carried out
   simultaneously in the range of m/z 50–1000, using information-dependent
   acquisition (IDA) MS. The experimental parameters of TOF MS scanning in
   the positive-ion mode were set as follows: ion source, 5500 V; gas
   temperature, 500 °C; nebulizer gas (GS1), 50 psi; Tissue-Equivalent gas
   (GS2), 50 psi; declustering potential, 80 V; and collision energy (CE),
   10 eV. The MS2 (50–1000 m/z) analysis was conducted at a CE of 35–15
   eV, with the CE ramped up at an interval according to the CE spread
   value.

2.4. RNA Library Construction and RNA-Seq

   Total RNA was extracted from samples in the three groups using a
   standard extraction method [[60]20], and RNA integrity was accurately
   determined using a Bioanalyzer Agilent 2100 system (Agilent
   Technologies, CA, USA). The qualified RNA from each sample was used for
   library construction and RNA-seq by NuoheZhiyuan Biotechnology Co.,
   Ltd. (Beijing, China). The cDNA library was sequenced on an
   IlluminaNovaSeq platform (Beijing, China). Four samples from each group
   were selected as biological replicates.

   To obtain high-quality clean reads, the image data obtained from
   Illumina sequencing were converted into raw reads, and sequencing
   errors, low-quality reads, and sequence connectors in raw reads were
   removed. Q20 and Q30 scores and GC content were calculated to ensure
   the reliability of all subsequent clean data analyses. A reference
   genome index was created using HISAT2 v2.0.5 and then the paired clean
   reads were compared to the reference genome. Reads mapped to each gene
   were then studied, and the fragments per kilobase of transcript per
   million mapped reads of each gene value was calculated according to the
   length of the gene. Differentially expressed genes (DEGs) were
   identified using the DESeq2 method. cDNA obtained from RNA transcripts
   from beef tenderloin tissues in all three groups was used as a qPCR
   template, and the 18S rRNA gene was used as a reference gene to design
   primers. The expression levels of 10 genes were quantified using a
   QuantStudio6 Flex Real-Time PCR system. The genes analyzed and primers
   used are shown in [61]Table 1, and the thermal cycling procedure is
   shown in [62]Table 2. Each gene was analyzed four times by qPCR.

Table 1.

   Primer sequences used for real-time quantitative polymerase chain
   reaction.
   Gene Name (Symbol) Primer Sequences (5′–3′) Product Length (bp)
   Carnitine palmitoyltransferase 1B (CPT1B) CCCAGGGAAGGACACAGATGT 149
   GGATCCTCTGGAACTGCATCTC
   Acyl-CoA dehydrogenase medium chain (ACADM) GCTTGGGAAGTTGATTCTGGTC 224
   TGTTCACGGGCTATAATAAGCC
   Peroxisome proliferator-activated receptor delta (PPARD)
   TGCAAAATCCAGAAGAAGAACC 171
   CTGGGGGTTGTGCTGACTC
   Acyl-CoA oxidase 2 (ACOX2) TTCCTGTCTGGTGCCCAAATA 160
   GACGTTCATAGGCATGTCCATC
   Ubiquitin C (UBC) CCGGACCGGGAGTTCAGT 172
   GGGATGCCTTCTTTTTCTTGTAT
   Stearoyl-CoA desaturase (SCD) CCAGGGCACCCATCAGATAG 162
   TCCAAGGTGGTCTCGACA
   Sorbin and SH3 domain containing 1 (SORBS1) TGTCCTGGAAGGAGGAGACATC 110
   CAGCTGGTATAAAATGCCTTGG
   Aquaporin 7 (AQP7) TCCAAGGTGGTCTCGACA 204
   ACCTATGGTGACTCCGAAGC
   Fatty acid binding protein 3 (FABP3) GCGTTCTCTGTCGTCTTTCC 169
   GATGATTGTGGTAGGCTTGGT
   Glycerol kinase (GK) TAAGGAAATTCTGCAGTCTGTCT 139
   TAACTTGTCCCAGACTACAGTGG
   18S rRNA CGGAACTGAGGCCATGATT 145
   CCTCCGACTTTCGTTCTTGAT
   [63]Open in a new tab

Table 2.

   Program for real-time quantitative PCR polymerase chain reaction.
     Procedure    Temperature (°C) Time
   Pre-incubation        95        5 min
   Amplification
   (40 cycles)           95         3 s
         63             20 s
   Melting curves        95        15 s
                         60        15 s
                         95        15 s
   [64]Open in a new tab

2.5. Statistical Analysis

   MS-DIAL (version 4.90) was used to process the mass spectrum data, and
   the compounds lacking MS/MS (tandem mass spectrometry) data and those
   with response values below 5000 were excluded from the analysis
   ([65]Table S2). Microsoft Office Excel 2021 was used to filter out the
   MS1 peaks with a detection rate below 80%, and relative standard
   deviation (RSD) greater than 30% of QC samples, and perform Student’s
   t-test analysis (after confirming the data met parametric assumptions
   including normal distribution and homogeneity of variance) and fold
   change analysis. The MS2 information of potential markers obtained by
   IDA was imported into MS-FINDER Version 3.52 software ([66]Table S2),
   annotated using three major databases (HMDB, and LipidMaps), and
   analyzed to screen for significantly differential metabolites (SDM) of
   Pingliang Red Cattle. The screening criteria for SDM was fold change
   (FC) of ≥2 or ≤0.5, an adjusted p-value < 0.05, and a Variable
   Importance in Projection (VIP) value > 1. The biological functions and
   metabolic pathways of the SDM were analyzed using the Gene Ontology
   (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases,
   respectively. Principal component analysis (PCA) and orthogonal partial
   least-squares discriminant analysis (OPLS-DA) were performed using the
   SIMCA^® of Umetrics company (Umea, Sweden). A heatmap and volcano plot
   were created using the MetaboAnalyst Software
   ([67]https://www.metaboanalyst.ca/) (accessed on 15 October 2024)
   [[68]21]. DESeq2 software (1.20.0) was used to perform the differential
   expression analysis of two comparison samples; genes with expression
   levels meeting the criteria p ≤ 0.05 and log2 fold change > 1 were
   designated as differentially expressed genes (DEG). The relative
   expression level of gene was calculated using the 2^−ΔΔCT method.

3. Results and Discussion

3.1. Identification of Differential Metabolites

   Metabolomics based on mass spectrometry offers significant advantages
   due to its high selectivity and sensitivity, providing unparalleled
   convenience for identifying metabolites from samples of different
   origins. This provides insights into the distinctive flavors,
   nutritional profiles, and other quality attributes that characterize
   agricultural products originating from this region. Finally, 4876
   original peaks were reserved. The total ion current (TIC) chromatograms
   of six QC samples exhibit high overlap, as shown in [69]Figure S1,
   indicating good stability of the mass spectrometry signals. The high
   stability of the instrument ensures the repeatability and reliability
   of the data, providing a critical foundation for robust analytical
   results. Furthermore, as evident from the provided TIC chromatograms,
   the chromatographic peaks exhibit high resolution and moderate response
   intensity. This suggests that the metabolites extracted from the beef
   samples were well-separated, indicating effective extraction and
   analysis procedures.

   Principal Component Analysis (PCA) is an unsupervised pattern
   recognition method for multidimensional data statistical analysis. This
   method compresses the original data into a several principal components
   to describe the characteristics of the dataset. By performing PCA on
   the samples, we can gain preliminary insights into the overall
   metabolic differences between groups and the variability within each
   group [[70]22]. The PCA results, illustrated in [71]Figure 1a,b,
   demonstrate the separation trends of the metabolome between Pingliang
   Red Cattle and Simmental cattle, and Pingliang Red Cattle and Qinchuan
   cattle, suggesting that there are distinct differences in the metabolic
   profiles between Pingliang Red Cattle and the two other cattle.

Figure 1.

   [72]Figure 1
   [73]Open in a new tab

   Multivariate analysis of beef metabolites across cattle breeds (PL, KL,
   QL). (a) PCA score plot of PL vs. KL metabolites. (b) PCA score plot of
   PL vs. QL metabolites. (c) OPLS-DA score plot showing metabolic
   discrimination. (d) Displacement test of metabolic variations.
   Abbreviations: PL, Pingliang Red Cattle; KL, Simmental; QL, Qinchuan;
   PCA, principal component analysis; OPLS-DA, orthogonal partial least
   squares discriminant analysis.

   Orthogonal partial least-squares discriminant analysis (OPLS-DA)
   combines Orthogonal Signal Correction (OSC) with Partial Least Squares
   Discriminant Analysis (PLS-DA). This approach allows for the filtering
   of variables by removing unrelated variations during modeling, thereby
   enhancing the selection of differential variables [[74]23]. Therefore,
   an OPLS-DA model was constructed based on the metabolite data from the
   three sample groups, as shown in [75]Figure 1c. The results further
   elucidates the distinct differences in metabolic profiles between the
   three groups, and also highlights that the Pingliang Red Cattle exhibit
   noticeable differences at the metabolite level compared to the
   Simmental and Qinchuan cattle. To validate the accuracy of the model,
   we conducted 200 permutation tests, as shown in [76]Figure 1d, all the
   blue Q^2 values and green R^2 values on the left are lower than the
   original points on the right, indicating the validity of the original
   model. Furthermore, the R^2Y value is 0.912 and the Q^2 value is 0.796
   in the model. The R^2Y value approaches 1, the accuracy of the model
   increases, and Q^2 value closer to 1 indicates greater predictive
   capacity in the original model [[77]23]. So, the aforementioned results
   demonstrate that the model possesses excellent reliability and
   predictive capability.

   Based on the Student’s t-test and fold change (FC) analysis, and the
   OPLS-DA model, the Simmental and Qinchuan cattle were used as control
   samples to identify significant differential metabolites (SDM) with
   Pingliang Red Cattle. The screening criteria are as follows: FC of ≥2
   or ≤0.5, an adjusted p-value < 0.05, and a Variable Importance in
   Projection (VIP) value > 1. By combining results of the MS-DIAL (a
   data-independent acquisition method) mirror image, 36 SDM were
   preliminarily screened as candidates distinguishing Pingliang Red
   Cattle and Simmental cattle ([78]Table S3), and 28 SDM were
   preliminarily distinguished between Pingliang Red Cattle and Qinchuan
   cattle ([79]Table S4). Among the SDM identified between Pingliang Red
   Cattle and Simmental cattle, 35 were significantly annotated to 32
   pathways (p < 0.05; [80]Figure 2a, [81]Table S5); those identified SDM
   between Pingliang Red Cattle and Qinchuan cattle, 28 were significantly
   annotated to 18 pathways (p < 0.05; [82]Figure 2b, [83]Table S6). Of
   particular note, purine metabolism and glutathione metabolism are key
   metabolic pathways shared by Pingliang Red Cattle between Simmental and
   Qinchuan cattle, highlighting their significant research importance.
   Moreover, 30 SDM exhibited relatively higher levels in Pingliang Red
   Cattle samples, including 12 amino acids, 5 fatty acids, 5 alkaloids, 3
   nucleotides, 2 purines, and 4 other metabolites ([84]Table S7), such as
   hypoxanthine and its derivatives (IMP), glutathione (GSH),
   ergothioneine (EGT), taurine, linoleic acid and phenylalanine betaine.
   The mirror images and structural formulae of these metabolites are
   shown in [85]Figure 3a–f, respectively. These metabolites play an
   important role in determining the flavor, taste, color, and antioxidant
   properties of meat.

Figure 2.

   [86]Figure 2
   [87]Open in a new tab

   KEGG pathway enrichment analysis of significantly differential
   metabolites (SDM) in different beef cattle breeds. (a) KEGG pathway
   enrichment diagram of SDM in PL and KL. (b) KEGG pathway enrichment
   diagram of SDM in PL and QL. Abbreviations: KEGG, Kyoto Encyclopedia of
   Genes and Genomes; PL, Pingliang Red Cattle; KL, Simmental; QL,
   Qinchuan; SCM, significantly changed metabolites.

Figure 3.

   [88]Figure 3
   [89]Open in a new tab

   Annotation of unique biomarkers in Pingliang Red Cattle. Mirror plots
   and structures of (a) hypoxanthine, (b) glutathione, (c) ergothioneine,
   (d) taurine, (e) linoleic acid, and (f) phenylalanine betaine.

3.2. Identification of DEG and Verification of Transcriptomic Data

   Initially, we evaluated the quality of the RNA from the samples. The
   concentration and integrity assessment of sample RNA confirmed its
   suitable purity and integrity for subsequent sequencing and associated
   experiments ([90]Table S8 and Figure S2). After removing sequencing
   adapters and low-quality reads, an average of 44,089,758 clean reads
   were obtained ([91]Table S9). Sequence comparison showed that the 12
   samples (four samples from each of the three cattle groups) had Q20
   values (sites with base correctness > 99%) greater than 97% and Q30
   values (sites with base correctness > 99% in the test data) greater
   than 94%, demonstrating the high quality of the sequencing data and
   their suitability for use in downstream analyses. To assess the
   reliability and representativeness of the samples, the squared Pearson
   correlation coefficient (R^2) was analyzed, the results were as shown
   in [92]Figure S3. Within-group samples with an R^2 closer to 1 exhibit
   higher similarity in expression patterns, indicating more reliable
   results. Moreover, the within-group R^2 being greater than the
   between-group R^2 demonstrates that the selected samples are
   representative of their respective groups. Then, we performed PCA on
   the gene expression levels in each of the samples, as shown in
   [93]Figure 4a. The PCA results demonstrate a clear and significant
   separation trend among the three groups of samples, highlighting the
   distinct gene expression profiles of Pingliang Red Cattle, Simmental
   Cattle, and Qinchuan Cattle. This separation suggests notable genetic
   expression differences between these breeds. Further clustering heatmap
   results show distinct groupings of gene expression patterns among
   Pingliang Red Cattle, Simmental Cattle, and Qinchuan Cattle, consistent
   with the PCA findings ([94]Figure 4b).

Figure 4.

   [95]Figure 4
   [96]Open in a new tab

   Identification and analysis of DEG among Pingliang Red Cattle,
   Simmental, and Qinchuan cattle. (a) PCA of identified genes. The red,
   green, and blue dots represent PL, KL and QL, respectively. (b) Heatmap
   of DEG. (c) Volcano plot of DEG (PL vs. KL). (d) Volcano plot of DEG
   (PL vs. QL). Red and dark green dots represent significantly
   upregulated and downregulated genes, respectively, and black dots
   represent genes with no significant changes. Abbreviations: PL,
   Pingliang Red Cattle; KL, Simmental; QL, Qinchuan.

   Differentially expressed genes (DEGs) were identified using the DESeq2
   method, applying stringent filtering criteria: FC of ≥2 or ≤0.5 and an
   adjusted p-value < 0.05. These thresholds ensured that only genes with
   significant expression differences between the groups were selected for
   further analysis. In total, 3694 DEG were screened in the comparison
   between Pingliang Red Cattle and Simmental cattle, including 1773
   upregulated genes and 1921 downregulated genes in Pingliang Red Cattle
   samples ([97]Figure 4c). And 2695 DEG were identified in the comparison
   between Pingliang Red Cattle and Qinchuan cattle, including 1459
   upregulated genes and 1506 downregulated genes in Pingliang Red Cattle
   samples ([98]Figure 4d). To gain a deeper understanding of the
   biological roles and pathways associated with the DEGs, Gene Ontology
   (GO) enrichment analysis to categorize gene functions and Kyoto
   Encyclopedia of Genes and Genomes (KEGG) pathway analysis have been
   conducted to explore the involvement of these genes in specific
   metabolic and signaling pathways. These analyses provided valuable
   insights into the molecular mechanisms underlying the observed
   expression differences. The results were shown as [99]Tables S10 and
   S11 and [100]Figure 5a–d. The DEGs between the Pingliang Red Cattle and
   Simmental groups were enriched in 7 biological processes (BP), 16
   cellular components (CC), and 10 molecular functions (MF). The DEGs
   between the Pingliang Red Cattle and Qinchuan groups were enriched in
   12 BP, 7 CC, and 2 MF. Among these, many terms were related to beef
   quality, such as “peptide biosynthetic process” (GO: 0043043) and
   “purine nucleoside triphosphate metabolic process” (GO: 0009144).
   “Cellular components” and “ribosome types” were the main
   representatives of the CC terms, and “binding” and “enzyme activity”
   were dominant among the significantly enriched MFs. In addition, a
   series of KEGG signaling pathways related to beef quality were
   identified, including “purine metabolism” (bta00230), “glutathione
   metabolism” (bta00480), and “the PPAR signaling pathway” (bta03320). We
   therefore further focused on the GO terms and DEG involved in these
   KEGG pathways. The genes phosphodiesterase 4D (PDE4D), peroxiredoxin 6
   (PRDX6), and 5-aminoimidazole-4-carboxamide
   ribonucleotideformyltransferase/IMP cyclohydrolase (ATIC) emerged as
   candidates most likely to have an important role in the taste and
   flavor of beef.

Figure 5.

   [101]Figure 5
   [102]Open in a new tab

   Functional enrichment analysis of differentially expressed genes
   comparing PL with KL and QL groups. (a) Significant GO terms for PL and
   KL. (b) Significant GO terms for PL and QL. (c) Significant KEGG
   pathways in PL and KL. (d) Significant KEGG pathways in PL and KL.
   Abbreviations: PL, Pingliang Red Cattle; KL, Simmental; QL, Qinchuan;
   BP, biological process; CC, cellular components; MF, molecular
   functions; GO, gene ontology; KEGG, Kyoto Encyclopedia of Genes and
   Genomes.

   To verify the reliability of the RNA-seq data, 10 DEG (CPT1B, ACADM,
   SORBS1, ACOX2, UBC, SCD, AQP7, GK, PPARD, and FABP3) identified between
   the cattle groups were randomly selected for qPCR. As shown in
   [103]Figure 6 and [104]Table S12, the expression levels of the selected
   genes showed some deviations, but the overall trends were consistent,
   indicating that the RNA-seq analysis results were accurate and
   reliable.

Figure 6.

   [105]Figure 6
   [106]Open in a new tab

   Real time quantitative PCR verification of randomly selected
   differentially expressed genes. (a) Pingliang Red Cattle and Simmental
   cattle. (b) Pingliang Red Cattle and Qinchuan cattle. Abbreviations:
   CPT1B, carnitinepalmitoyltransferase 1B; ACADM, acyl-CoA dehydrogenase
   medium chain; SORBS1, sorbin and SH3 domain containing 1; ACOX2,
   acyl-CoA oxidase 2; UBC, ubiquitin C; SCD, stearoyl-CoA desaturase;
   AQP7, aquaporin 7; FABP3, fatty acid binding protein 3; GK, glycerol
   kinase; PPARD, peroxisome proliferator activated receptor delta.

3.3. Unveil Pingliang Red Cattle’s Distinctive Quality Traits

   To comprehensively elucidate the unique quality attributes of Pingliang
   Red Cattle, an integrative analysis combining metabolomics,
   transcriptomics, specific feed composition, and rearing environment was
   conducted. Specifically, we identified two key metabolites that are
   closely associated with the umami flavor and color of beef: inosine
   monophosphate (IMP) and glutathione (GSH), as well as one high-content
   bioactive substance, ergothioneine (EGT), in Pingliang Red Cattle beef.
   Furthermore, transcriptomics analysis not only revealed the genetic
   basis behind the higher content of the aforementioned three metabolites
   but also identified a set of genes associated with meat tenderness,
   which was found to be significantly upregulated in Pingliang Red Cattle
   compared to both Qinchuan and Simmental cattle. Through this
   integrative analysis, we comprehensively elucidated the causes behind
   the superior quality of Pingliang Red Cattle, offering solid scientific
   data to support the explanation of their distinctive traits.

3.3.1. IMP and Meat Flavor

   IMP (inosinic acid) is a key umami compound that contributes
   significantly to the flavor characteristics of meat [[107]24,[108]25].
   As the primary nucleotide responsible for meat’s savory taste, IMP and
   its derivatives play an essential role in determining overall flavor
   quality. Our metabolomics analysis demonstrated that the relative IMP
   content in Pingliang Red Cattle samples was 2.86 and 3.96 times higher
   than in Qinchuan and Simmental cattle samples, respectively
   ([109]Figure 7a). These findings suggest that Pingliang Red Cattle
   possess superior flavor potential compared to the other breeds, which
   may translate to enhanced taste attributes. The elevated IMP levels
   could explain the more pronounced umami characteristics observed in
   this cattle breed.

Figure 7.

   [110]Figure 7
   [111]Open in a new tab

   Contents of metabolites as potential markers for beef quality
   determined by liquid chromatography-mass spectrometry in Pingliang Red
   Cattle. (a) Inosine Monophosphate, (b) Glutathione, (c) Ergothioneine,
   (d) Proline Betaine, (e) Phenylalanine Betaine, (f) Ecgonine, (g)
   Ricinine, and (h) Linoleic acid. Asterisks indicate statistically
   significant differences: * p < 0.05, ** p < 0.01, *** p < 0.001.

   Through transcriptomics analysis of the purine metabolism pathway in
   cattle, compared to Simmental and Qinchuan cattle, Pingliang Red Cattle
   exhibited significantly increased transcription levels of 17 and 14 key
   genes, respectively, within the purine metabolism pathway. Among these,
   eight genes were commonly upregulated in both comparisons, including
   PDE4D, GMPS, GART, GMPR, ADCY, ATIC, AMPD, and ADSL. The relative
   expression levels of the eight genes, as depicted in [112]Figure 8a.

Figure 8.

   [113]Figure 8
   [114]Open in a new tab

   Comparison of gene transcript levels in Pingliang Red Cattle,
   Simmental, and Qinchuan cattle samples. (a) Purine metabolism-related
   genes. (b) Glutathione metabolism-related genes. (c) Shear
   force-related genes. Abbreviations: PDE4D, phosphodiesterase 4D; GMPS,
   guanine monophosphate synthase; GART,
   glycinamideribonucleotidetransformylase; GMPR, guanosine monophosphate
   reductase; ADCY, adenylatecyclase; ATIC, 5-aminoimidazole-4-carboxamide
   ribonucleotideformyltransferase/IMP cyclohydrolase; AMPD, adenosine
   monophosphate deaminase; ADSL, adenylosuccinatelyase; IDH2, isocitrate
   dehydrogenase 2; OPLAH, 5-oxoprolinase ATP-hydrolyzing; MGST3,
   microsomal glutathione S-transferase 3; ODC1, ornithine decarboxylase
   1; GGT1, gamma-glutamyltransferase 1; PRDX6, peroxiredoxin 6; RRM1,
   ribonucleotidereductase catalytic subunit M1; SMS, spermine; FABP3,
   fatty acid binding protein 3; PPARD, peroxisome proliferator activated
   receptor delta; FABP7, fatty acid binding protein 7; ACADM, acyl-CoA
   dehydrogenase medium chain. Asterisks indicate statistically
   significant differences: * p < 0.05, ** p < 0.01, *** p < 0.001.

   For example of Phosphodiesterase 4D (PDE4D), the PDE4D has the Enzyme
   Commission (EC) number 3.1.4.53. The expression level of the PDE4D gene
   in Pingliang Red Cattle was 9.66 times higher than that in Qinchuan
   cattle and 16.04 times higher than in Simmental cattle. This
   substantial upregulation suggests enhanced cAMP-specific
   phosphodiesterase activity, which plays a crucial role in regulating
   intracellular signaling pathways related to muscle development and
   metabolism.

   The formation of umami substances is regulated by several enzymes,
   including AMP deaminase (AMPD), adenylosuccinatelyase (ADSL), and ATIC.
   AMPD1 catalyzes adenosine monophosphate hydrolysis and deamination to
   produce IMP, whereas ADSL catalyzes the initial synthesis and
   circulation of purine nucleotides and is the rate-limiting enzyme for
   IMP formation. ATIC is a biofunctional enzyme with aminoimidazole
   carbamoyl transferase and catalytic activities that regulates IMP
   synthesis [[115]26]. Therefore, we have elucidated the intrinsic
   reasons for the superior flavor of Pingliang Red Cattle beef from both
   transcriptional and metabolic perspectives.

3.3.2. GSH and Meat Color

   Glutathione (GSH) is a tripeptide with γ-amide bonds and a sulfhydryl,
   composed of glutamic acid, cysteine, and glycine. GSH is a crucial
   antioxidant that plays a vital role in protecting cells from oxidative
   stress and damage. Beyond its antioxidant role, GSH is involved in
   several metabolic processes: such as (1) detoxification, conjugates
   with toxins and carcinogens, facilitating their excretion from the body
   [[116]27]. (2) DNA synthesis and repair, contributing to genomic
   stability. (3) Immune function, facilitate the proliferation of
   lymphocytes and the production of cytokines [[117]28]. Research has
   found that beef treated with a combination of GSH and a mixed gas of CO
   and O[3] exhibited significantly better results in terms of total
   bacterial count, total volatile basic nitrogen (TVBN), and redness
   value (a*) during refrigerated storage compared to the group treated
   only with the mixed gas of CO and O[3] [[118]29]. The study results
   indicate that the presence of antioxidants can effectively maintain the
   bright color and remove off-flavors.

   The metabolomics analysis results showed that the relative content of
   glutathione (GSH) in Pingliang Red Cattle samples is 5.43 times and
   2.42 times higher than that in Qinchuan cattle and Simmental cattle,
   respectively, and the differences are statistically significant
   ([119]Figure 7b).

   Meanwhile, the transcriptomic analysis of the glutathione metabolism
   pathway in beef samples revealed that, compared to Qinchuan cattle and
   Simmental cattle, Pingliang Red Cattle samples had eight genes with
   extremely significant upregulation in their relative expression levels,
   and these expression levels were positively correlated with the
   glutathione content in the beef ([120]Figure 8b). These genes are
   specifically identified as: IDH2, OPLAH, MGST3, ODC1, GGT1, PRDX6,
   RRM1, and SMS. The relative expression levels of these genes were more
   than two-fold higher in Pingliang Red Cattle compared to both Qinchuan
   cattle and Simmental cattle. Waerp et al. reported that upregulation of
   MGST3 expression, which has GSH transferase activity and can catalyze
   various GSH-dependent reactions during GSH synthesis, is closely
   related to meat color. Glutathione may improve muscle water retention,
   meat color, and meat quality by reducing the damage of free radical
   reaction to cell structure and the oxidation degree of myoglobin
   [[121]30]. So to improve glutathione content in meat, targeted
   intervention measures can be taken. Studies have shown that the
   glutathione level in slaughtered muscles is increased by supplementing
   the diet with hydroxymethionine 14 days before slaughter, and the
   oxidation of meat after slaughter is alleviated [[122]31].

3.3.3. DEGs and Meat Tenderness

   Meat tenderness is a key indicator of beef quality and determines
   consumer satisfaction with the meat. Beef tenderness is categorized
   based on shear stress values as follows: tender and liked (<4.0 kg),
   acceptable (4.1–5.2 kg), and tough and disliked (>5.3 kg). The shear
   stress of the outer ridge of Pingliang Red Cattle beef is 2.72 kg,
   which is 19.58% lower than that of A3-grade snowflake beef purchased in
   the market [[123]32]. The tenderness of beef is influenced by multiple
   factors, with cattle genotype playing a significant role alongside
   various environmental and management conditions. Research indicates
   that genetic factors contribute substantially to meat tenderness,
   though this characteristic is also notably affected by post-slaughter
   handling, feeding regimes, and processing techniques [[124]33]. Genetic
   factors that determine beef tenderness mainly include specific variants
   of myogenic determination (MYOD), calmodulin kinase (CaMK), calpain
   (CAPN), calpain inhibitor (CAST), fatty acid binding protein (FABP),
   and PRDX6. In the study, transcriptomic analysis results show that,
   compared to Qinchuan and Simmental cattle, Pingliang Red Cattle have 21
   significantly upregulated DEGs related to beef tenderness. These DEGs
   are as follows: FABP3, PPARD, FABP7, ACADM, PRDX6, ACSL1, PLIN2, CPT1B,
   SORBS1, PDPK1, ACOX2, ACSL3, UBC, SCD, CPT2, AQP7, CD36, ACADL, GK,
   ME1, CAST. Additionally, the relative expression levels of FABP3,
   PPARD, FABP7, and ACADM in Pingliang Red Cattle were significantly
   higher compared to Qinchuan and Simmental cattle ([125]Figure 8c).
   Notably, the expression level of FABP3 in Pingliang Red Cattle was
   12.81-fold times higher than that in Simmental cattle. CAST activity is
   highly correlated with meat tenderness, and its average contribution to
   muscle shear force reaches up to 28.9%; thus, it has been proposed as a
   candidate gene for studying meat tenderness traits [[126]34]. FABP
   proteins belong to a family of homologous small-molecule cytoplasmic
   proteins. Nine FABP members have been identified to date, among which
   heart- and adipocyte-specific members are considered candidate genes
   for studying intramuscular fat. FABP3 polymorphisms significantly
   correlate with water content, tenderness, muscle scores, and
   intramuscular fat content [[127]35]. PRDX6, an important antioxidant
   defense enzyme in animals, can regulate beef tenderness during
   postmortem ripening by regulating the level of oxidative stress in
   cells, thus serving as potential a biomarker of beef tenderness
   [[128]36]). Notably, due to the varying expression levels of PRDX6
   among cattle breeds, it can serve as a biomarker for beef tenderness.
   Specifically, PRDX6 exhibits a significant positive correlation with
   another well-known protein biomarker for beef tenderness, u-calpain
   (CAPN1). This suggests that PRDX6 plays an important role in
   determining meat quality and could be used as an indicator for
   assessing beef tenderness across different cattle breeds [[129]34].

3.4. Bioactive Compounds

   Ergothioneine (EGT) is a rare amino acid that was first discovered in
   1909, which was initially identified in ergot fungi (Claviceps
   purpurea), and EGT has since been found in various organisms, including
   bacteria, plants, and animals. EGT has many physiological functions,
   such as scavenging free radicals, detoxification, maintaining DNA
   biosynthesis, supporting normal cell growth, and enhancing cellular
   immunity [[130]37]. Both the European Food Safety Agency and US Food
   and Drug Administration have affirmed the importance of EGT in
   functional foods and support that it can be safely added to pregnant
   women’s diet and children’s food [[131]38]. Metabolomic analysis
   results show that the EGT relative content in Pingliang Red Cattle
   samples is significantly higher compared to Qinchuan and Simmental.
   Specifically, the EGT relatively content in Pingliang Red Cattle is
   2.55 times higher than that in Qinchuan cattle and 4.25 times higher
   than that in Simmental cattle ([132]Figure 7c). Subsequently, a
   confirmatory method for L-ergothioneine in beef samples was established
   using UPLC-Q-TOF-MS. The method was tested on solutions of
   L-ergothioneine standard and three types of beef samples. The results
   indicated that under positive ion scanning mode, compounds with primary
   ions (m/z 230.0954) and secondary ions were detected from beef samples
   of Red Pingliang cattle, Simmental cattle, and Qinchuan cattle. These
   findings matched the retention times, primary ions, and secondary ions
   of the L-ergothioneine standard solution, as shown in [133]Figure 3c.
   Following this, according to the chromatographic and mass spectrometric
   conditions described in [134]Section 2.3, quantitative analysis of
   L-ergothioneine in three types of beef samples was conducted using a
   standard solution of L-ergothioneine. A corresponding calibration curve
   was established: y = 347.08x + 302.09 (R^2 = 0.9926). This curve
   demonstrated a good linear relationship for L-ergothioneine within the
   concentration range of 1.0 to 200.0 μg/L ([135]Figure S4). To calculate
   the concentration of L-ergothioneine using a standard curve, the
   average concentration of L-ergothioneine in Pingliang Red Cattle was
   16.6 μg/kg, in Simmental cattle it was 3.93 μg/kg, and in Qinchuan
   cattle it was 6.53 μg/kg. Based on this results, there were significant
   differences in the L-ergothioneine content among the three cattle
   breeds.

   Although no experimental studies have shown that animals can synthesize
   ergothioneine (EGT), it is believed that they may absorb and retain it
   in their cells. Furthermore, transcriptomic analysis of samples from
   three cattle breeds did not identify any genes related to the synthesis
   or metabolism of EGT, nor their expression. This suggests that animals
   can only digest and absorb EGT, indicating that the EGT found in beef
   originates from the feed consumed by the animals. To determine the
   source of this component, we further investigated the feed sources of
   Pingliang Red Cattle. In contrast to the intensive feeding practices
   for Simmental and Qinchuan cattle, the feed composition for Pingliang
   Red Cattle typically consists of coarse feed (70%) and fine feed (30%).
   The coarse feed is derived from a wide variety of sources, including
   multiple plant stems, leaves, and natural pasture grasses, which serve
   as the primary contributors of ergothioneine.

   The alkaloids detected in the meat are completely or partially derived
   from the digestion, absorption, and retention of these feeds by the
   animals. To understand the alkaloid content in the natural forage
   consumed by the cattle, our research team conducted a comprehensive
   analysis over two consecutive years. A total of 70 samples of natural
   forage were analyzed for their alkaloid content. The results indicated
   an overall detection rate of 42.86% for alkaloids, with 11 out of the
   15 types of alkaloids being detectable. The total alkaloid content in
   the samples ranged from 10.40 to 106.99 g/kg. Further research has
   revealed that, in addition to EGT mentioned earlier, the bioactive
   alkaloids retained in Pingliang Red Cattle meat samples include:
   D-proline betaine, phenylalanine betaine, anabasine, ricinine, butyryl
   carnitine, and hematoporphyrin.

   Our investigation also revealed that Pingliang Red Cattle are raised at
   low stocking densities with relatively low levels of intensification.
   This management practice contributes to higher animal welfare
   standards, as evidenced by the animals’ behavior. The improved welfare
   conditions have a positive influence on beef quality to some extent.

4. Conclusions

   This study identified significant biochemical differences in Pingliang
   Red Cattle compared to Simmental and Qinchuan cattle, including: (1)
   elevated levels of flavor-related compounds (IMP and GSH), (2)
   upregulation of tenderness-associated genes, and (3) higher content of
   the bioactive compound ergothioneine (EGT). While these biochemical
   markers suggest potential implications for meat quality attributes, we
   acknowledge that actual sensory characteristics would require
   verification through dedicated evaluation methods. Through integrated
   metabolomics and transcriptomics analysis combined with feeding
   condition assessment, we have systematically characterized the
   distinctive biochemical profile of Pingliang Red Cattle. These findings
   provide technical references for understanding the molecular basis of