The experimental protocol was reviewed and approved by the Animal Care and Use Ethics Committee, Faculty of Veterinary Medicine, Universitas Airlangga, Indonesia (Approval No. 1.KEH.039.03.2024). All procedures involving KUB chickens were conducted in accordance with the Indonesian national guidelines for the ethical care and use of animals in research, the institutional animal welfare requirements of Universitas Airlangga, and the principles outlined in the ARRIVE 2.0 guidelines.

Throughout the 90-day experimental period, the birds were housed under appropriate environmental, hygienic, and husbandry conditions, with unrestricted access to clean drinking water and nutritionally adequate feed in accordance with their assigned dietary treatments. The chickens were observed regularly for general health, behavior, feed and water consumption, mobility, signs of pain or distress, injury, and adverse responses to the experimental diets. All handling, weighing, sampling, and husbandry procedures were performed by trained personnel using methods designed to minimize fear, discomfort, restraint time, and unnecessary stress.

Predefined humane considerations were applied throughout the experiment. Any bird showing persistent, severe distress; inability to access feed or water; marked loss of body condition; serious injury; or clinical deterioration that could not be promptly alleviated was to be evaluated by qualified veterinary personnel and either removed from the experiment or, when necessary, humanely euthanized. At the end of the feeding trial, the chickens were slaughtered using the humane procedure authorized in the approved institutional protocol. Tissue collection was performed only after confirmation of death. Muscle samples were collected aseptically for meat quality assessment and molecular analysis, immediately processed or preserved as required, and stored under appropriate conditions.

The number of birds was limited to the minimum considered necessary to address the study objectives while permitting the planned statistical analyses. All reasonable measures were taken to refine the experimental procedures, reduce bird use, and minimize pain and distress. No procedures were performed solely for purposes unrelated to the approved scientific objectives.

This study was conducted from June 5 to September 6, 2024 at the Faculty of Veterinary Medicine, Airlangga University.

A total of 25 one-day-old male KUB chicks were randomly assigned to five dietary treatments (n = 5 per treatment). The experiment followed a 5 × 3 factorial arrangements comprising five dietary treatments and three TGFβ2 genotypes (CC, TC, and TT), with genotypes determined post hoc by molecular analysis. The dietary treatments were: F0P0 (control, no enzymes), F4P0 (400 mg/kg phytase), F4P3 (400 mg/kg phytase + 300 mg/kg protease), F4P5 (400 mg/kg phytase + 500 mg/kg protease), and F4P7 (400 mg/kg phytase + 700 mg/kg protease). Enzyme inclusion levels were selected based on commercial recommendations and previous studies in native chickens. Birds were housed individually and fed ad libitum twice daily, with free access to water. All procedures were approved by the institutional ethics committee and conducted in accordance with animal welfare guidelines.

Body weight (BW) was recorded individually using a digital hanging scale with an accuracy of 0.01 g at day 1 and at the end of the experimental period (90 days of age). Birds were weighed weekly to monitor growth performance. Average daily gain (ADG) was calculated using the following formula: ADG (g/day) = (IBW − BW) / days, where BW represents BW (g). Feed intake (FI) was calculated as the difference between feed offered and feed refused for each bird, adjusted for any feed spillage. Daily FI (DFI) was determined by dividing total feed consumption by the number of feeding days. Feed conversion ratio (FCR) was calculated as: FCR = Total FI (g) / Total BW gain (g)

Meat quality parameters were analyzed including water-holding capacity (WHC), pH, texture, and cooking loss (CL), which were measured using 5 g of raw meat samples as the initial weight, with two replicates for each treatment. Meat color was evaluated using a colorimeter (12 mm aperture, U 59730-30, Cole-Parmer International Inc., Pittsford, NY, USA) on first use at the same measurement point for each sample. Meat color characteristics were then expressed in Lightness (L*), Redness (a*), and Yellowness (b*) values [12].

Muscle samples weighing approximately 0.5-1.0 g of lean tissue were taken and placed into sterile cryovial tubes. Following instant flash freezing in liquid nitrogen, the samples were stored at −80 °C until DNA extraction and additional molecular analysis. The specific primer sequences used for the chicken TGFβ2 target gene KUB (GenBank accession no. X58071) were as follows: forward primer, 5’-GCC ATA GGT TCA GTG CAA G-3’; reverse primer, 5’-TGA CAG AAG CTC TCA AGC C-3’. The manufacturer's instructions (Promega, Madison, WI, USA) on first use were followed to generate the polymerase chain reaction (PCR) reaction mixture, which included 12.5 μL of master mix, 1 μL of forward primer, 1 μL of reverse primer, 3 μL of DNA template, and 7.5 μL of distilled water. The ideal conditions for amplification were 5 min of initial denaturation at 95°C, 30 cycles of denaturation at 95°C for 30 seconds, annealing at 54°C for 30 seconds, extension at 72°C for 30 seconds, and a final extension at 72°C for 7 min.

PCR-RFLP analysis involved digesting 20 μL of PCR product with 1 μL of the restriction enzyme AluI for 3 h at 37°C. Following digestion, the pieces were separated using 1.5% agarose gel electrophoresis, stained with ethidium bromide, and examined under an ultraviolet lamp. Additionally, 20 μL of the PCR product was sent to Macrogen (Seoul, Korea) for sequencing analysis on first use [13].

DEGs were subjected to hierarchical clustering analysis using the Gplots package in the R software environment (R Foundation for Statistical Computing, Vienna, Austria). Following gene identification, STRING functional protein association networks were used to build a regulatory enrichment network, and Cytoscape was used to visualize it. The web-based program DAVID was used for Gene Ontology (GO) enrichment analysis, and the significance threshold for relevant GO categories was set at p < 0.05. Additionally, KOBAS 3.0 was used to conduct Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis [14].

The Benjamini-Hochberg method was used to calculate adjusted p-values for DEG analysis and KEGG pathway enrichment. For growth performance and meat quality traits, data were analyzed using a two-way analysis of variance to evaluate the main effects of TGFβ2 genotype and enzyme supplementation level, as well as their interaction (genotype × enzyme level). When significant effects were detected (p < 0.05), treatment means were compared using Duncan’s Multiple Range Test. All statistical analyses were performed using SPSS version 20.0. Results are presented as mean ± standard deviation (SD).

Over 80% of the reads were successfully mapped to the reference genome of the chicken (Gallus gallus). After the mapping process, the expression level of each gene was calculated as the average, and genes showing significant DEGs in KUB chickens were identified (Figure 1A). The three biological replicates of KUB chickens showed very similar expression patterns, according to hierarchical clustering analysis, suggesting that the sequencing data were highly biologically reproducible. A volcano plot visualized the distribution of DEGs identified using DESeq2, with the X-axis representing changes in gene expression (log2 Fold Change) and the Y-axis indicating statistical significance (-log10 Adjusted p-value). Genes with significantly higher expression are indicated by red dots, whereas genes with significantly lower expression are indicated by blue dots (Figure 1B). Overall, these results indicate that KUB chickens have several candidate genes that warrant further exploration to elucidate the regulatory mechanisms of gene expression related to muscle tissue growth and development (Figure 1C).

The results in Table 1 show no significant difference between the three TGFβ2 genotypes in improving the performance of KUB chickens. However, there was a significant difference (p < 0.05) in BW in the F4P7 treatment with the TC genotype, but no significant difference was found between the C and T genotypes. The dominant BW of KUB chickens with the CT genotype in the F4P7 treatment was 1586 g/bird, 1584 g/bird, and 1582 g/bird, respectively. The TC genotype in the F4P5 treatment was 1537 g/bird, followed by the CC genotype at 1535 g/bird and the TT genotype at 1532 g/bird. The increase in BW is suspected to be related to the KUB chicken genotype.

Table 1 also shows significant differences between the three TGFβ2 genotypes in carcass weight (CW). Meanwhile, there was a significant difference in CW in the control treatment genotypes, namely TT 848 g/head, 852 g/head, and 850 g/head. The F4P7 treatment with the dominant TC genotype was higher at 1198 g/head, followed by the CC genotype at 1194 g/head, and then the TT genotype at 1190 g/head. The data obtained from this study are presented in Table 1. The TGFβ2 genotype did not differ significantly in FI. However, in the F4P5 treatment, the CT genotype had a higher FI compared to the other treatments, namely 3881 g/head/90 days, followed by the CC genotype at 3880 g/head/90 days. In general, Table 1 shows a significant difference between the three TGFβ2 genotypes in the FCR in the F4P3 treatment of TC genotypes 2.70, CC 2.72, and TT 2.76. While there was no significant difference with the F4P7 treatment, a high FCR was detected in the CC genotype control treatment, namely 4.21.

Dietary multi-enzyme affected various meat quality traits, such as pH, texture, CL, and WHC (Table 2), as well as color parameters (Table 3). Meat quality analysis showed that pH and texture effects were not significantly different (p > 0.05) between the multienzyme addition treatment and the treatment without multienzyme in the feed, but were significantly different (p < 0.05) in CL and WHC, as seen in Table 2. Color parameters were significantly higher (p < 0.05) in the multienzyme feed group compared to the control group, and there was a significant difference (p < 0.05) between the lower (F4P0) and higher (F4P3, F4P5 and F4P7) multienzyme concentrations; consequently, the results for other different analyses were based on the Lightness of KUB chicken meat. Redness of KUB chicken was significantly increased (p < 0.05) between the control group (F0P0) and the multienzyme treatments (F4P3, F4P5, and F4P7). The results of the analysis of the yellowness of KUB chicken meat showed a significant decrease (p < 0.05) in the F4P5 and F4P7 multienzyme groups when compared with the F4P0 and F4P3 groups; then, the group with the highest dose of multienzyme (F4P7) decreased significantly (p < 0.05) when compared with the control group (F0P0) without multienzymes in the feed.

Table 4 presents the genotypic distribution of the TGFβ2 gene coding region, as determined by distinct restriction patterns obtained from enzymatic digestion. The frequencies of the CC, TC, and TT genotypes were 20%, 56%, and 24%, respectively. Moreover, these genotype proportions showed highly significant differences (p > 0.05). According to the Hardy–Weinberg equilibrium, the allele frequencies for C and T were 0.48 and 0.52, respectively. Chi-square analysis revealed a strong association between the chicken strain and the AluI restriction enzyme. Such variations could be attributed to factors including gene flow across populations, interbreeding among groups, and differences in sample size.

Amplification was performed to identify SNPs by DNA sequencing (Figure 2). SNPs were detected using the F4P3 primer pair, whereas no polymorphisms were observed with the TGFβ2 primer pair (Figure 3). Two SNPs were identified within the exon 1 region (c.103C/T and c.99G/A). The g.1079C/T substitution altered the codon from GGC to GGT, representing a synonymous mutation that still encodes glycine (Figure 4). DNA sequencing successfully genotyped the c.103C/T and c.99G/A SNPs, with the g.103C/T locus exhibiting CC and CT genotypes, while the g.99G/A locus showed GG and GA genotypes.

In this study, two SNPs were identified for the TGFβ2 gene target (Figure 3). SNP identification was performed by comparing the KUB chicken sequence and two local chicken sequences published in NCBI (Acc. X58071). Consequently, two SNPs were identified in the TGFβ2 gene target. One SNP was detected in the F4P5 treatment, and another in the F4P7 treatment (Figure 4). However, these SNPs were not confirmed among the F0P0, F4P0, and F4P3 samples. In fact, two other SNPs were found in the same gene target (Figure 3).

Previous research has identified several genes and signaling pathways that play key roles in muscle fiber formation and pectoral muscle growth across various chicken breeds [14]. However, molecular mechanisms underlying muscle development in the selectively bred KUB chicken remain insufficiently explored. The results of this study indicate that DEGs exhibit considerable variation in expression across developmental stages. GO enrichment analysis indicates that most DEGs are involved in biological processes related to muscle tissue growth, cell differentiation, and contractile function [15]. Furthermore, significant differences in TGFβ2 gene expression levels were observed in KUB chickens, suggesting that this gene exhibits a specific regulatory pattern in muscle development.

Importantly, the KEGG and GO enrichment analyses of DEGs showed notable enrichment in several biological pathways, including focal adhesion, extracellular matrix (ECM)–receptor interaction, and cysteine and methionine metabolism, all of which support increased metabolic activity and muscle tissue development in KUB. Phytase improves phosphorus bioavailability, while protease enhances amino acid digestibility, both of which are critical for protein synthesis and satellite cell activity. Enhanced nutrient supply may therefore potentiate TGFβ2-regulated muscle development pathways, particularly in specific genotypes [16]. The observed genotype-enzyme-level interaction indicates that the growth response to multienzyme supplementation varies with TGFβ2 genotype, supporting the presence of a nutrigenetic mechanism in KUB chickens [17]. This integrative gene-nutrition interaction highlights the potential of TGFβ2 as a candidate marker for genetic selection and for precision feeding strategies in indigenous-derived tropical poultry systems [18].

Physiologically, improved growth performance in livestock is highly dependent on genetic activity that regulates the proliferation, differentiation, and maturation of skeletal muscle cells. One key gene that coordinates these processes is TGFβ2 [19]. This gene directly affects growth efficiency and production performance by modulating cell repair processes and preserving the delicate balance between growth and tissue regeneration [20]. Furthermore, TGFβ2 is involved in regulating physiological processes related to nutrient utilization and energy homeostasis, thereby indirectly influencing FI and energy conversion in chickens.

Research shows that genotypic variation in the TGFβ2 gene is significantly associated with differences in BW gain and final weight in chickens, indicating a genetic link between the gene’s expression pattern and growth performance [21]. Individuals with genotypes that express TGFβ2 more actively tend to exhibit higher growth efficiency without significant increases in feed consumption, as well as better physiological resistance to stress and disease [22]. These findings suggest that TGFβ2 may contribute to improved feed conversion efficiency by optimizing muscle growth metabolism. Therefore, chickens with elevated TGFβ2 expression may possess superior genetic merit for growth performance and overall productivity [23].

The influence of multienzyme supplementation on meat physicochemical properties, including pH, texture, CL, and water holding capacity (WHC) as presented in Table 2 and on meat color parameters such as lightness (L*), redness (a*), and yellowness (b*) as shown in Table 3, was systematically evaluated. The pH value is a critical determinant of meat quality, as it directly affects protein denaturation, CL, and textural characteristics. A negative correlation was observed between pH and both lightness (L*) and yellowness (b*), whereas redness (a*) exhibited a positive correlation with pH. Analysis of breast meat from chickens fed multienzyme-supplemented diets demonstrated a decrease in yellowness (b*) and a corresponding increase in redness (a*), suggesting improved meat coloration. The correlation coefficients between color attributes (L*, a*, b*) and pH were highly significant, indicating a strong relationship between enzymatic treatment and color stability. Changes in meat pH and texture are often linked to microbial contamination, such as Salmonella infection, which can decrease pH and compromise texture [24]. A lower pH, associated with higher redness (a*) values, may also influence consumer perception of freshness and overall meat quality, as optimal chicken meat typically exhibits enhanced lightness and redness [25].

CL value decreased significantly as multienzyme dosage increased, whereas both texture and WHC improved markedly (Table 2). This finding is consistent with previous reports indicating that multienzyme supplementation enhances WHC by promoting improved moisture retention within muscle tissues [26]. CL was calculated as the percentage difference between raw and cooked sample weights and was lowest in the SRC group, which also exhibited the highest WHC. This suggests that multienzyme supplementation may enhance meat juiciness by reducing water loss during cooking [27]. The improvement may be attributed to the higher multienzyme concentration in the feed, which facilitates the formation of a denser, more cohesive muscle structure [28]. Furthermore, the type and dosage of multienzymes significantly affected meat texture, likely through enzymatic and biochemical modifications influencing muscle pigment oxidation and structural protein integrity. These changes were associated with improved color characteristics and textural quality [26]. Correlation matrix analysis revealed that CL was inversely related to texture parameters, including chewiness and firmness, suggesting a potential interaction between structural integrity and moisture retention capacity [27].

Crossbreeding is an important mechanism that contributes to increased genetic diversity and heterozygosity in livestock populations [29]. The process of gene flow between populations introduces new alleles into the gene pool, thereby enriching genetic variation and increasing heterozygosity at both the population and subpopulation levels [30]. In the context of the TGFβ2 gene, previous studies have reported two major alleles (T and C) and three genotype combinations (TT, CT, and CC) in a chicken population [31]. A similar polymorphism pattern was also found in local Indonesian chickens, including KUB chickens, with two alleles (T and C) and three genotypes (TT, CT, and CC) [32]. However, the frequency distribution showed dominance of the T allele over the C allele, likely influenced by inbreeding practices and strong selection pressure on the TGFβ2 gene. Changes in the frequency of the TGFβ2 allele in KUB chickens may reflect an adaptive response to environmental conditions and nutritional factors provided during rearing [33]. Feed supplementation with enzymes such as phytase and protease has been reported to affect the bioavailability of essential amino acids and minerals, thereby altering the transcriptional activity and expression of the TGFβ2 gene [34].

Certain alleles of the TGFβ2 gene, such as the T allele, are thought to be associated with higher nutrient utilization efficiency and increased muscle protein deposition, while the C allele may be associated with a more conservative metabolic strategy under conditions of nutrient deficiency or excess [35]. Thus, nutritional selection pressure arising from variations in feed composition can influence allele frequency dynamics in the KUB chicken population. These results indicate a synergistic interaction between genetic factors (TGFβ2 gene polymorphisms) and nutritional factors (enzyme content and nutrient availability), which collectively contribute to metabolic adaptation, growth efficiency, and the physiological performance of KUB chickens in tropical environments.

The hypothalamus, which primarily expresses the TGFβ2 gene, is crucial for regulating poultry FI and energy balance. Furthermore, this gene is expressed in other tissues, including erectile tissue, where TGFβ2 is reported to contribute to physiological mechanisms underlying erectile function [36]. The candidate gene approach has been widely used to identify genomic regions associated with important traits in livestock, including specific genetic markers for selection and breeding. Several previous studies reported the presence of four variations in the TGFβ2 gene promoter region in chickens [37], as well as five SNPs in the 5’UTR region and one SNP in the exon region (c.129A/G) [38]. Another study also identified four additional SNPs in the 3’UTR region using PCR-SSCP and DNA sequencing. In this study, analysis of a 387-bp fragment of the TGFβ2 gene in KUB chickens revealed no significant polymorphisms. However, when the gene sequence was compared with GenBank (NCBI) reference data, two new SNPs were identified in the 5'UTR region. This difference indicates that the TGFβ2 gene in KUB chickens has unique sequence variations compared with other poultry species, and thus could be used as a molecular marker for the genetic identification of KUB chickens.

Furthermore, this study also identified two SNPs in the exon region, as well as one additional SNP detected after comparison with the GenBank TGFβ2 sequence. The two identified point mutations, c.103C/T and c.99G/A, are unique variations not found in the other two reference sequences in GenBank. This finding aligns with a previous report that also identified SNPs in the exon region of the TGFβ2 gene in chickens [38]. That study reported one synonymous substitution at position c.93G/A (GG→AG) and one nonsynonymous substitution at position 293G/A, which causes an amino acid change from glycine to arginine. Overall, these results strengthen the evidence of genetic diversity in TGFβ2 in KUB chickens and provide an important scientific basis for further research on the function of this gene in the physiological performance of local poultry. In addition, three restriction enzymes, RsaI, AluI, and KpnI, were successfully used for genotyping by PCR-RFLP [39]. These three enzymes recognize specific sequences GT’AC, G’GTAC_C, and G_GTAC’C, respectively. The use of restriction enzymes as a molecular analysis tool has proven effective for evaluating genetic diversity and gene function in vivo, and similar methods have also been applied in genotyping the TGFβ2 gene in other local chicken populations [40].