Abstract

Background

   A key factor used to evaluate the quality of meat is intramuscular fat
   (IMF) content, which has a strong association with the flavor of meat.
   To date, however, the influence of IMF content on the volatile profiles
   of ovine meat has not been clarified.

Results

   We used headspace solid-phase microextraction gas chromatography-mass
   spectrometry (HS-SPME-GC/MS) to study the aroma characteristics of
   low-IMF (L-IMF) and high-IMF (H-IMF) content in muscle tissue and to
   identify volatile compounds in L-IMF and H-IMF groups. Thirteen of 557
   volatile compounds identified were significantly different between the
   groups (Odor Activity Values > 1, VIP values > 1), of which aldehydes,
   ketones, and esters were the predominant volatile compounds. We
   conducted RNA-seq analysis and found 985 differentially expressed genes
   (DEGs) in the L-IMF and H-IMF groups. The weighted gene coexpression
   network analysis (WGCNA) showed that DEGs associated with muscle aroma
   characteristics were enriched in lipolysis regulation in adipocytes,
   PPAR signaling, AMPK signaling, arachidonic acid metabolism, and the
   biosynthesis of unsaturated fatty acids (UFAs). These pathways played a
   role in regulating energy metabolism and muscle lipid metabolism
   functions in Tibetan sheep. The combined results of DEGs, WGCNA, and
   gene-aroma compound interactions revealed ADIPOQ, FABP4, FADS2, GPD2,
   HSL, LEP, PEPCK and PLIN4 to be potential candidate genes affecting IMF
   and aroma compounds content.

Conclusion

   In this study, we established profiles of the transcriptome and
   volatile compounds of L-IMF and H-IMF ovine muscle tissues. The
   findings suggested that HSL, FABP4 and PLIN4, along with the lipolysis
   regulation in adipocytes and PPAR signaling pathways, contributed to
   the difference in lipid metabolism within muscle tissues, and
   biosynthesis of the UFA pathway may influence the production of
   volatile compound precursors in muscles. These results offer important
   information about regulating genes that pertain to ovine meat flavor
   compounds.

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s12864-025-11997-w.

   Keywords: Tibetan sheep, Intramuscular fat, Flavor compound,
   Differentially expressed genes, Lipid deposition

Introduction

   Tibetan sheep live about 3,000 m in the Qinghai-Tibet Plateau. These
   sheep account for China’s largest population of primitive sheep.
   Tibetan sheep mutton is known for its good tenderness and rich aroma,
   containing high-quality proteins, amino acids, and polyunsaturated
   fatty acids (FAs) [[32]1]. Volatile compounds, as crucial attributes of
   the sensory quality of mutton, are increasingly emphasized by breeding,
   processing industries, and consumers as key indicators of the
   commercial value of mutton [[33]2]. These volatile compounds are
   derived from the Maillard reaction, lipid-Maillard interaction, and
   lipid oxidation degradation. They include acids, alcohols, aldehydes,
   furans, hydrocarbons, ketones, nitrogen- and sulfur-containing
   compounds, and pyrazines [[34]3–[35]5].

   Intramuscular fat (IMF) is strongly associated with meat quality
   attributes and flavor formation [[36]2, [37]6]. Liu et al. [[38]7]
   identified 61 different lipid subclasses that may produce volatile
   aroma compounds. IMF is rich in lipids such as triglycerides,
   glycerophospholipids, and fatty acyls, playing primarily two roles in
   flavor formation: acting as a solvent and precursor to volatile
   compounds [[39]8]. Volatile compounds in meat products tend to dissolve
   preferentially in the lipid fraction, thereby influencing their release
   and perception [[40]9]. The hydrolysis of triglycerides and
   glycerophospholipids is responsible for the production of FAs. After
   undergoing nonenzymatic oxidation, these FAs produce volatile
   compounds, including ketones, alcohols, aldehydes, carbonyl compounds,
   hydroxy epoxides, and ketone acid epoxides [[41]7, [42]10]. The thermal
   oxidation and decomposition of ω−3 and ω−6 polyunsaturated FAs
   generates the volatile aroma compounds that are characteristic of
   mutton [[43]11, [44]12]. Oleic acid primarily generates aldehydes with
   fruity, sweet, and fatty aromas, such as octanal, nonanal, heptanal,
   and (E,E)−2,4-decadienal [[45]13]. After linoleic acid completes
   thermal oxidation, it degrades into 1-octen-3-ol (mushroom), hexanal
   (green and grassy), 2-pentylfuran (vegetable and earthy), (E)−2-octenal
   (grilled meat and peanut cake), (E)−2-nonenal (cucumber and fatty), and
   pentanal (spicy odor) as well as other volatile compounds [[46]14,
   [47]15]. The production of benzaldehyde (bitter almond aroma),
   (E,Z)−2,4-heptadienal (cucumber), and (E,E)−2,4-heptadienal (grilled
   meat) is associated closely with alpha-linolenic acid oxidation
   [[48]10, [49]13, [50]15]. Therefore, an appropriate deposition of IMF
   can ensure that mutton products possess good quality.

   Some studies to date have examined the factors that form and influence
   the flavor quality of mutton. Most of this research, however, has
   identified volatile compounds [[51]3] and focused on the impact of
   factors such as breed [[52]4], processing conditions [[53]16], diet
   [[54]15, [55]17], and gender [[56]18]. Studies on the mechanisms by
   which lipids affect meat flavor quality are mainly concentrated on in
   vitro simulation verification and the impact of adding single or
   multiple lipids on volatile flavor compounds [[57]11, [58]19, [59]20].
   An understanding of the relationship between volatile compounds and IMF
   is lacking significant research to examine mutton flavor quality and
   its regulatory mechanisms. This study utilized HS-SPME-GC/MS to
   quantitatively characterize the volatile characteristics and chemical
   components sampled with high and low IMF levels muscle tissues in
   Tibetan sheep. Using RNA-seq technology, an expression profiles
   analysis was conducted on the two groups of muscle tissues, and key
   candidate genes regulating the metabolism of volatile compounds were
   identified. The results of our study may reveal the relationship
   between volatile compounds and DEGs and clarify the impact of IMF on
   the flavor quality of mutton.

Materials and methods

Animals and muscle sampling

   The Animal Ethics Committee of Qinghai University (Ethics approval file
   No. QUA-2020-0709) reviewed and approved this animal study. We randomly
   selected six Tibetan ewes that were two years old and that had normal
   development and growth. The ewes belonged to the same natural grazing
   flock, which was located at an altitude of 3,000 m above sea in Qilan
   County, Qinghai Province, China. The ewes were raised under the same
   environmental conditions, with natural light, and were provided the
   same nutrition, with free access to food and water. According to
   Islamic practice, the sheep were exsanguinated humanely. We immediately
   collected samples of muscle tissue from 10 portions of the left half of
   the carcass. The sampling parts included the triceps brachii muscle
   (m.TB), supraspinatus muscle (m.S), deltoid muscle (m.D), Longissimus
   thoracis et lumborum muscle (m. LTL), latissimus dorsi muscle (m.LD),
   psoas major muscle (m.PM), trapezius thoracalis muscle (m.TT), biceps
   femoris muscle (m.BF), quadriceps femoris muscle (m.QF), and
   gastrocnemius muscle (m.G), Figure [60]S1 details the anatomical
   locations of the sampled muscle tissues. We used physiological saline
   and RNase-free water to rinse the samples. We collected samples for
   RNA-seq, reverse transcription-quantitative PCR (RT-qPCR), and volatile
   compounds detection in cryotubes, which were immediately stored in
   liquid nitrogen. At the same time, we cooled the other sample for 24 h
   at 4 °C, which were saved for additional IMF content analysis.

IMF measurement and HS-SPME-GC/MS analysis

   Following the Chinese “National Food Safety Standard—Determination of
   Fat in Foods” (GB 5009.6–2016), we used the Soxhlet extraction method
   to determine the IMF content of the muscle samples. The IMF was
   expressed as the percent of the meat’s weight when wet.

   To stain neutral lipids, we used BODIPY 493/503 and followed the method
   reported by Li et al. [[61]21]. Specifically, the frozen tissue
   sections were thawed at room temperature, followed by incubation with
   diluted Bodipy (1:200 dilution in PBS, Thermo Fisher Scientific, D3922)
   for 20 min at room temperature to label neutral lipids. Subsequently,
   slides were washed three times in PBS (pH 7.4) for 5 min each on an
   orbital shaker to remove unbound dye. Nuclei were counterstained by
   incubating with DAPI (1 µg/mL, Sigma-Aldrich, D9542) for 10 min in the
   dark, followed by three additional PBS washes (5 min each). To preserve
   fluorescence, the sections were mounted using anti-fade mounting medium
   and coverslipped. Imaging was performed using the IX53 inverted
   fluorescence microscope (Olympus, Tokyo, Japan) with the following
   settings: DAPI (excitation: 330–380 nm, emission: 420 nm), Bodipy
   (excitation: 465–495 nm, emission: 515–555 nm), and CY3 (excitation:
   510–560 nm, emission: 590 nm). To calculate the area of BODIPY
   staining, we randomly collected four images for each field of
   observation. Furthermore‌, we performed H&E staining on the
   aforementioned frozen sections to confirm the location of lipid
   droplets within the muscle fibers, following the experimental
   procedures described by Zhang et al.[22].

   According to the research by Yang et al. [[62]23], the volatile
   compounds in muscle samples were pre-treated and extracted via a
   headspace solid-phase micro extraction system. The volatile compounds
   were separated and analyzed according to the method described by Yang
   et al. [[63]24], with the GC/MS conditions detailed in Table [64]S1. We
   used the NIST v2.3 databases, as reported [[65]13, [66]25, [67]26], to
   conduct mass spectra matching and identify the volatile compounds. We
   used an in-house developed library based on commercial standards to
   compare retention times (R.T.) and we matched the retention indices
   (R.I.) to the authentic compounds. We completed quantification
   according to the internal standard method, as reported in Oin et al.
   [[68]27]. We normalized each volatile compound’s concentration to align
   with 3-Hexanone-2,2,4,4-d4. We used the Odor Activity Value (OAV) to
   identify the volatile compounds that contributed the most to the flavor
   of ovine meat. We set the threshold to OAV great than 1.

RNA extraction and RNA-Seq analysis

   RNA extraction, transcriptome sequencing, and data analysis were
   performed according to the approach described by Wang et al. [[69]28].
   RNA quality was assessed using 2100 Bioanalyzer (Agilent Technologies,
   Palo Alto, CA, USA), only high-quality RNA samples (RIN ≥ 6.0) was used
   to construct sequencing library. We generated individual complementary
   DNA (cDNA) libraries for five low-IMF (L-IMF) groups and five high-IMF
   (H-IMF) groups. We used an Illumina Novaseq6000 platform (Illumina, San
   Diego, CA, United States) for sequencing according to Guangzhou Gene
   Denovo Biotechnology Co., Ltd. (Guangzhou, China). We filtered the raw
   reads stored in the FASTQ file format [[70]28] to obtain high-quality
   clean reads for annotation. We mapped the results to the reference
   genome NCBI_GCF_016772045.2
   ([71]https://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/016/772/045/GCF_01677
   2045.2_ARS-UI_Ramb_v3.0/) using HISAT2 (v2.2.4). We calculated fragment
   per kilobase of transcript per million mapped reads (FPKM) to normalize
   each annotated gene’s expression abundance and variations using
   StringTie v1.3.1. To compare the expression levels between L-IMF and
   H-IMF groups, we used DESeq v2.0 [[72]29]. We defined differentially
   expressed genes (DEGs) as those with a p-value < 0.01 and |Log[2] (fold
   change)| ≥0.5. In this study, we performed KEGG pathway enrichment
   analysis of the DEGs using DAVID ([73]http://david.abcc.ncifcrf.gov/)
   with all KEGG-annotated gene from the reference genome (Ovis aries,
   NCBI_GCF_016772045.2) as background. To determine the relationship
   between the gene expression and different volatile compounds [[74]30],
   we performed WGCNA analysis using the R package ‘WGCNA’ [[75]31].

Validation of DEGs by reverse transcription-quantitative PCR

   The RT-qPCR was utilized for verifying RNA-Seq findings following the
   protocol outlined by Hao et al. [[76]32]. We selected β-tubulin and
   β-actin as the internal references to normalize the DEG expression