The Animal Ethics Committee of the Banjarmasin City Food Security, Agriculture, and Fisheries Service approved the experimental procedures (approval number: 520/624/DKP3/X11/2021). All sampling procedures complied with established animal welfare guidelines. Gentle restraint, appropriately sized needles, and rapid venipuncture performed by trained personnel were used to minimize pain and stress during blood collection. The puncture site was cleaned before and after sampling to reduce discomfort and promote healing. Slaughter was conducted at a licensed slaughterhouse in accordance with Indonesian regulations for humane handling and slaughter of cattle.

The experiment was conducted from December 2022 to June 2023. A total of 95 Bali cattle bulls from a single population were included, and all available animals were used as representative samples. The animals weighed 250–350 kg and were 18–36 months of age. Bali cattle were sourced from Kupang, East Nusa Tenggara Province, Indonesia, and transported by ship to the Basirih slaughterhouse in Banjarmasin. Upon arrival, cattle were maintained under an intensive management system for 2 weeks before slaughter in South Banjarmasin, South Kalimantan. During this period, forage was provided at 10% of body weight, concentrate at 2%, and water ad libitum. Animals were restrained using a cattle chute to minimize movement during blood collection by a trained veterinarian.

Blood samples were collected from the jugular vein using a Venoject with a sterile disposable 21-gauge needle (Becton, Dickinson and Company, Franklin Lakes, NJ, USA). Approximately 5 mL of blood was transferred into vacuum tubes containing 1.5 mL ethylenediaminetetraacetic acid (EDTA). Samples were immediately placed on ice and stored at 4°C until DNA extraction. Genomic DNA was extracted using the Geneaid Genomic DNA Mini Kit (Geneaid Biotech Ltd., Taipei, Taiwan) according to the manufacturer’s instructions. Extracted DNA was stored at −20°C. DNA purity was assessed spectrophotometrically, with acceptable A260/280 ratios of 1.8–2.2 and concentrations ≥20 ng/μL.

Primer sequences were designed using Primer3 and BLAST primer tools. The forward (5′-CCC TCC ATC ATT GTA ATC ACT-3′) and reverse (5′-GGA CAA CGT ATC CAG CAG AAA-3′) primers for the FABP4 gene amplified a fragment spanning intron 2 to exon 4. The expected PCR product length was 721 base pairs (bp). Amplification was performed using an AB System thermocycler (Applied Biosystems, Foster City, CA, USA) under the following conditions: pre-denaturation at 95°C for 1 min, followed by 35 cycles of denaturation at 95 °C for 15 s, annealing at 57°C for 15 s, extension at 72°C for 10 s, and a final extension at 72°C for 3 min.

PCR products were separated on a 1% agarose gel stained with Florosafe (1^st BASE, Singapore, Singapore) and visualized using a UV transilluminator (Bio-Rad, Hercules, CA, USA). Sequencing was performed at 1st BASE Laboratory Services (Selangor, Malaysia) using an ABI PRISM system with the BigDye Terminator kit v3.1 (Applied Biosystems). Sequencing quality was manually examined using FinchTV, BioEdit, and Molecular Evolutionary Genetic Analysis software. The first and last 30 bp were excluded because of low peak quality, and only SNPs with clear chromatogram peaks were retained. Sequence alignment was conducted against the Ensembl Bos taurus RefSeq transcript ENSBTAG00000037526, and SNP positions were determined using the Bos taurus ARS-UCD2.0: CM008181.2 reference genome assembly.

In vivo carcass and meat traits were measured using a 7.1-MHz ultrasound transducer applied to the left rib area. A portable ultrasound device (SIUI CTS-800, SIUI Medical Systems, Shantou, China) was used to measure LDT, BFT, MS, and IMF at the 12th–13th rib position (Figure 1A) using transverse and longitudinal scans [25, 26]. IMF sonograms were obtained 3 days before slaughter. Hair at the scanning site was shaved, and ultrasound gel was applied to ensure adequate transmission of sound waves. Images were displayed in real time, saved, and analyzed using ImageJ software (National Institutes of Health, Bethesda, MD, USA). ImageJ was used to calibrate scale units to millimeters. Measurements of BFT, LDT (Figure 1B), and IMF (Figure 1C) were recorded at three points per animal. IMF values were subjected to simple linear regression to estimate the IMF percentage. Marbling was scored according to AUS-MEAT and MSA standards, with scores ranging from 0 to 9 (AUS-MEAT, 2018).

Approximately 250 g of tenderloin muscle was collected from the right side of each carcass about 1 h post-mortem and stored at −20°C. From this sample, 40 g was used for FA analysis. Lipids were extracted using a chloroform–methanol solution, followed by transesterification to produce fatty acid methyl esters (FAMEs). FAMEs were extracted, centrifuged, dried in hexane, resuspended in chloroform to remove impurities, and purified by solid-phase extraction.

FAMEs (1 μL) were analyzed using gas chromatography equipped with a flame ionization detector. Separation was achieved on a cyanopropyl methyl sil capillary column (60 m × 0.25 mm, 0.25 μm film thickness) with nitrogen as the carrier gas at 30 mL/min. Injector and detector temperatures were set at 220°C and 240°C, respectively. The oven temperature program was 125°C for 5 min, then increased to 225°C at 10, 5, and 3°C/min, with holding times of 5, 10, and 7 min, respectively. The split ratio was 1:80, the injection volume was 1 μL, and the linear velocity was 23.6 cm/s. FAMEs were quantified as relative percentages using a mixed FAME standard for retention-time and peak-area comparisons. Quality control was ensured by standard injections before sample analysis. Fatty acids were grouped into saturated fatty acids (SFA), monounsaturated fatty acids (MUFA), and polyunsaturated fatty acids (PUFA) [27]. Total unsaturated fatty acids (UFA) were calculated as the sum of MUFA and PUFA. Analyses were performed in duplicate, and instruments were calibrated according to the manufacturer’s guidelines.

Allele and genotype frequencies, observed and expected heterozygosity, and Hardy–Weinberg equilibrium were calculated using PopGen 1.32 software. Polymorphic information content (PIC) was calculated using the formula:

where p_i and p_j represent allele frequencies at a given locus.

Associations between FABP4 genotypes and carcass traits, meat quality traits, and FA composition were analyzed using a general linear model (GLM) in SAS version 9.4 (SAS Institute Inc., Cary, NC, USA). Duncan’s multiple-range test was used for post hoc comparisons. Statistical significance was set at p < 0.05. The GLM was expressed as:

Y_ij = μ + G_i + e_ij,

where Y_ij is the phenotypic observation, μ is the overall mean, G_i is the genotype effect, and e_ij is the random error. Carcass and meat traits, including FA composition, were corrected to 36 months of age, body weight, and similar environmental conditions using the equation:

X_i(corrected) = (X̅standard / X̅observation) × X_i(observation)

where X_i(corrected) is the corrected value, X̅standard is the standard group mean, X̅observation is the observation group mean, and X_i(observation) is the observed value.

Genetic variation in FABP4 was identified by SNP analysis of a 721 bp DNA sequencing product. SNPs were detected and characterized by examining double peaks in sequencing chromatograms using FinchTV software (Figure 2). Four novel SNPs were identified in FABP4 from the sequencing map. Three SNPs (g.4631T>C, g.4724T>C, and g.4769G>A) were located in intron 3, whereas one SNP (g.5002C>T) was located in exon 4 (Table 1). Among these variants, g.5002C>T was a nonsynonymous substitution resulting in an amino acid change from valine to alanine, whereas the remaining three SNPs were synonymous and did not alter the amino acid sequence. Analysis of allele and genotype frequencies indicated that the AA genotype was more frequently observed for g.4724T>C, g.4769G>A, and g.5002C>T, whereas the BB genotype predominated for g.4631T>C (Table 2).

The associations between FABP4 SNPs and carcass and meat quality traits are summarized in Table 3. The SNP g.5002C>T showed a significant association with LDT (p < 0.05), suggesting a potential role in muscle development. In contrast, no significant associations were detected between FABP4 SNPs and BFT, MS, or IMF in Bali cattle. Notably, animals carrying the CT genotype at g.5002C>T exhibited significantly greater LDT compared with those carrying the CC and TT genotypes.

The relationship between FABP4 polymorphisms and FA composition in Bali cattle is presented in Table 4. None of the identified FABP4 SNPs showed a significant association with FA composition. Although FABP4 polymorphisms were hypothesized to influence FA profiles, the results indicate that these variants did not contribute significantly to FA variation in Bali cattle. This suggests that other genetic factors or environmental influences may play a more dominant role in determining FA composition. Further studies are required to elucidate the interactions between FABP4, other candidate genes, and non-genetic factors involved in the regulation of meat quality traits in Bali cattle.

Genetic diversity is a key determinant of adaptability and long-term sustainability in livestock populations. Based on allele and genotype frequency data, the AA genotype was more frequently observed for SNPs g.4724T>C, g.4769G>A, and g.5002C>T, whereas the BB genotype predominated for g.4631T>C (Table 2). Adequate genetic variation is essential for breeding programs aimed at improving performance and ensuring food security [28, 29]. Preservation of genetic diversity reduces the risks associated with inbreeding and genetic uniformity, thereby maintaining resilience, fitness, and protection against inbreeding depression [30].

All SNPs identified in this study were polymorphic, with major allele frequencies <99% and minor allele frequencies >1% [31]. The observed heterozygosity (Ho) was 50% for all SNPs, indicating relatively low heterozygosity in the Bali cattle population. This pattern may reflect inbreeding, genetic drift, and a small effective population size, which can increase homozygosity and allele fixation [32]. Reduced genetic variation may negatively affect overall health and adaptability, increasing susceptibility to diseases and environmental challenges [30]. All SNPs conformed to Hardy–Weinberg equilibrium, suggesting stable allele and genotype frequencies that were not significantly influenced by selection, migration, mutation, or genetic drift [33]. This equilibrium indicates population stability and limited disturbance from breeding practices or environmental pressures [34]. In addition, all SNPs in FABP4 exhibited low PIC values, consistent with the limited genetic variability commonly reported in indigenous breeds with narrow genetic bases [35, 36].

The FABP4 gene has been widely investigated for its role in regulating meat quality traits in cattle, particularly those related to lipid metabolism and muscle development. Genes involved in these pathways influence key traits such as IMF deposition and muscle thickness. In the present study, SNP g.5002C>T showed a significant association with LDT (p < 0.05), indicating its potential involvement in muscle development. This nonsynonymous substitution results in an amino acid change that may alter FABP4 protein function, supporting its candidacy as a genetic marker for selective breeding in Bali cattle.

Previous studies have reported significant associations between FABP4 SNPs and MS and BFT, particularly for variants located in exon 3 [37]. Gao et al. [38] observed associations between FABP4 diversity and LDT as well as ultrasonographic IMF in Qinchuan cattle. Associations with hump height and eye muscle area have also been reported in northern Australian crossbred cattle [39], while Shin et al. [40] demonstrated relationships between FABP4 variants and MS in Hanwoo cattle. In contrast, the lack of association between SNPs and MS, IMF, and BFT in the present study may be attributed to unbalanced genotype distributions. For example, only one AA genotype was observed for SNP g.4769G>A, precluding robust association analysis. Differences in population structure, sample size, and genetic background across studies can lead to variable allele and genotype frequencies and influence statistical outcomes [41].

The absence of significant associations between FABP4 polymorphisms and FA composition in Bali cattle contrasts with findings reported by Bayraktar et al. [24], who demonstrated strong associations between FABP4 variants and milk composition and FA profiles in Native Southern Yellow cattle. These observations highlight tissue- and breed-specific functions of the FABP gene family, suggesting that the influence of FABP4 on lipid metabolism may be more pronounced in milk than in muscle or adipose tissues of Bali cattle.

Other studies have reported significant relationships between FABP4 variants and specific FAs. Hoashi et al. [42] identified associations between FABP4 polymorphisms and palmitoleic acid (C16:1) and linoleic acid (C18:2n6c) in Qinchuan cattle. Palmitoleic acid has been linked to improved metabolic health, including reduced insulin resistance [43, 44], while linoleic acid is known for its cardiovascular benefits [45, 46]. Yin et al. [47] identified six FABP4 SNPs associated with fat content and MS in Yellow Yanbian cattle, and Oh et al. [48] reported associations with oleic acid (C18:1n9c) and monounsaturated FA content. The lack of association observed in Bali cattle may reflect the polygenic nature of FA composition, limited genetic diversity within the population, and strong environmental influences such as diet, management system, and physiological status, which can mask genetic effects [49, 50].

The SNP g.5002C>T in FABP4 represents a promising candidate for MAS in Bali cattle. However, validation in larger and geographically diverse populations under varying environmental conditions is required before practical implementation. Following validation, incorporation of this SNP into national breeding programs could facilitate the identification of animals with favorable genotypes, improve meat quality efficiency, and accelerate genetic improvement in Bali cattle [51].