The study protocol and all procedures involving the collection and use of bovine skeletal muscle tissue were reviewed and approved by the Institutional Animal Care and Use Committee of the Faculty of Medicine, Khon Kaen University, Thailand (approval no. AEMDKKKU 014/2022). Skeletal muscle tissues were obtained from adult male Pon Yang Kham Thai-French crossbred cattle (Bos taurus) slaughtered at the Pon Yang Kham Livestock Breeding Cooperative, Sakon Nakhon, Thailand, as part of routine commercial slaughter; no animals were slaughtered specifically for the purposes of this study. Tissue collection and handling were conducted in accordance with the Good Manufacturing Practice standards of the Pon Yang Kham Livestock Breeding Cooperative Limited. The collected tissues were then used to isolate, expand, phenotypically characterize, and differentiate MDSCs in vitro. All applicable experimental procedures were conducted in accordance with the approved institutional protocol, and the study was reported in accordance with the ARRIVE 2.0 guidelines, where applicable.

The study was conducted from December 2022 to July 2023. Skeletal muscle samples were collected from adult male Pon Yang Kham Thai-French crossbred cattle slaughtered at the Pon Yang Kham Livestock Breeding Cooperative, Sakon Nakhon, Thailand [27, 28]. Subsequent isolation, culture, characterization, and experimental analyses of muscle-derived cells were performed at the Department of Biochemistry, Faculty of Medicine, Khon Kaen University, Thailand.

This experimental study was designed to develop and validate a simplified protocol for isolating and expanding CD29-positive MDSCs from adult Thai-French crossbred cattle and to characterize the spatial heterogeneity and lineage-specific differentiation of their constituent SC and FAP populations. Skeletal muscle tissues from three donor animals were used for protocol validation and assessment of viable cell yield and CD29-positive enrichment efficiency. Detailed subpopulation characterization and functional assays were subsequently conducted using three independent spatial isolates (BNS1, BNS2, and BNS3) obtained from distinct locations within the chuck roll of a single donor animal. Thus, BNS1–BNS3 represented within-animal spatial replicates rather than independent biological donors. The experimental workflow comprised tissue collection and processing, MDSC isolation and expansion, flow cytometric characterization and FACS-based separation of SCs and FAPs, proliferation assays, lineage-specific differentiation, gene expression analysis, immunofluorescence staining, and mycoplasma contamination testing.

Skeletal muscle tissue was collected from three adult male Pon Yang Kham Thai-French crossbred cattle (Bos taurus) (N = 3 animals; age, 2.5–3 years; body weight, 532–650 kg) slaughtered at the Pon Yang Kham Livestock Breeding Cooperative, Sakon Nakhon, Thailand [27, 28]. Tissue samples from all three animals were used to validate the isolation protocol and determine viable cell yield per gram of tissue and CD29-positive enrichment efficiency. For detailed subpopulation characterization, differentiation assays, and SC:FAP heterogeneity analysis, three independent spatial isolates (BNS1, BNS2, and BNS3) were obtained from three distinct locations within the chuck roll of a single donor (Animal 3; N = 1 animal; n = 3 spatial isolates; age, 34 months; body weight, 532 kg; slaughtered on July 11, 2023). Therefore, BNS1–BNS3 represented within-animal spatial replicates and not independent animals.

Muscle tissue was obtained from the chuck roll, a commercial designation for a boneless forequarter subprimal (IMPS/NAMP item 116A) comprising Musculus longissimus dorsi, Musculus rhomboideus, Musculus spinalis dorsi, Musculus complexus, Musculus multifidus dorsi, Musculus serratus ventralis, Musculus subscapularis, and Musculus splenius. Tissue blocks were excised from three distinct anatomical locations within this subprimal. The subprimal was not seam-dissected into individual muscles at the time of collection; therefore, the specific muscle of origin of each tissue block was not recorded.

At the slaughterhouse, tissue samples were placed in sterile 50-mL centrifuge tubes, rinsed once with 75% ethanol and twice with sterile normal saline solution (NSS) with vigorous shaking, and cut into 0.1–0.2 cm³ pieces. The samples were maintained in ice-cold NSS until further processing or cryopreservation. For cryopreservation, dissected tissue samples were stored in fetal bovine serum (FBS) containing 10% dimethyl sulfoxide (DMSO) at −80°C.

Muscle-derived cells were isolated using a modified previously described protocol [6]. Tissue fragments were enzymatically dissociated in 0.2% collagenase type I (C0130, ≥125 CDU/mg solid; Sigma-Aldrich, St. Louis, MO, USA) prepared in serum-free Dulbecco's modified Eagle medium with high glucose (DMEM-HG; 25 mM glucose; Gibco BRL, Carlsbad, CA, USA). Samples were incubated at 37°C for 1.5 h under continuous rotation at 20 rpm using a MACSmix™ Tube Rotator (Miltenyi Biotec, Bergisch Gladbach, Germany).

The cells were washed once with DMEM supplemented with 20% FBS (20% FBS-DMEM-HG) and filtered through a 200-µm nylon mesh to remove undigested tissue debris. The resulting cell pellet was treated with red blood cell (RBC) lysis buffer containing 15 mM NH₄Cl, 1 mM KHCO₃, and 10 µM ethylenediaminetetraacetic acid (EDTA) for 5 min and subsequently washed twice with phosphate-buffered saline (pH 7.4). Viable cells were quantified using the trypan blue exclusion method (Gibco BRL).

Isolated cells were seeded at an initial density of 1.8–3.8 × 10⁵ cells/well in 6-well plates and cultured in basic MDSC culture medium (BMCM), consisting of 20% FBS-DMEM-HG supplemented with 5 ng/mL basic fibroblast growth factor (bFGF; R&D Systems, Minneapolis, MN, USA) and 2.5 µM p38 mitogen-activated protein kinase (MAPK) inhibitor SB203580 (Selleck Chemicals, Houston, TX, USA) to support proliferation and delay spontaneous differentiation [29]. Cells were passaged at approximately 60% confluence for subsequent experiments.

Cell-surface marker expression was evaluated using allophycocyanin (APC)-conjugated anti-human CD29 (clone TS2/16; BioLegend, San Diego, CA, USA), fluorescein isothiocyanate (FITC)-conjugated anti-sheep CD31 (clone CO.3E1D4; Bio-Rad, Hercules, CA, USA), FITC-conjugated anti-sheep CD45 (clone 1.11.32; Bio-Rad), and phycoerythrin (PE)-conjugated anti-human CD56 (clone MEM-188; BioLegend). Cells were expanded for 28–48 days before FACS.

Flow cytometry was performed using a FACSCanto™ II system (BD Biosciences, Franklin Lakes, NJ, USA), and the resulting data were analyzed using FlowJo™ software version 10.4 (Tree Star, Ashland, OR, USA). CD29⁺/CD56⁺ SCs and CD29⁺/CD56⁻ FAPs were isolated by FACS using a BD FACSAria™ III Cell Sorter (BD Biosciences). The approximate cell inputs and sorted yields were as follows: BNS1, approximately 7 × 10⁵ input cells, 1 × 10⁴ sorted FAPs, and 1.4 × 10⁵ sorted SCs; BNS2, approximately 1.4 × 10⁶ input cells, 4.7 × 10⁵ sorted FAPs, and 2.4 × 10⁵ sorted SCs; and BNS3, approximately 7 × 10⁵ input cells, 9.3 × 10³ sorted FAPs, and 4.1 × 10⁵ sorted SCs.

Sorted SCs were maintained in BMCM, whereas FAPs were cultured in minimum essential medium alpha (MEM-α; Gibco BRL) supplemented with 15% FBS (15% FBS-MEM-α). Both cell types were cultured with 1× antibiotic-antimycotic at 37°C in a humidified atmosphere containing 5% CO₂.

All functional assays, including proliferation and myogenic and adipogenic differentiation, were performed using early-passage cells (P2–P8 post-sorting). Proliferative capacity was evaluated using a colorimetric cell counting assay. SCs and FAPs were seeded at 3 × 10³ cells/well in 96-well plates and cultured for 24, 48, and 72 h. At each time point, SuperKine™ Maximum Sensitivity Cell Counting Kit-8 (CCK-8; Boster Biological Technology, Atlanta, GA, USA) solution was added, and the cells were incubated for 2 h at 37°C. Absorbance at 450 nm (OD₄₅₀) was measured using an EZ Read 2000 microplate reader (Biochrom, Cambridge, UK), and the data were analyzed using Galapagos software (Biochrom). Baseline OD₄₅₀ was recorded 6 h after plating and designated as day 0. Proliferative capacity was expressed as the fold change relative to the day 0 value.

Myogenic differentiation of SCs was performed using a protocol adapted from a previously described method [29]. SCs were seeded on plates coated with 50 µg/mL rat-tail collagen type I (A10483-01, lot no. 531799; Invitrogen, Carlsbad, CA, USA) in 20% FBS-DMEM-HG and cultured for 24 h to promote attachment. The cells were subsequently transferred to myogenic differentiation medium (MDM), consisting of DMEM with low glucose (DMEM-LG; 5.6 mM glucose) supplemented with 2% FBS and 1 µg/mL insulin (I6634; Sigma-Aldrich). The medium was replaced every 2 days, and the cells were harvested on day 7 for analysis.

Adipogenic differentiation of FAPs was performed using a method adapted from Dohmen et al. [6]. FAPs were pre-cultured overnight in 15% FBS-MEM-α and subsequently transferred to adipogenic differentiation medium (ADM), consisting of 1% FBS-MEM-α supplemented with 1 µg/mL insulin. During the first 3 days, ADM was supplemented with three adipogenic inducers: 0.5 mM 3-isobutyl-1-methylxanthine (IBMX; I5879), 100 µM indomethacin (I7378), and 1 µM dexamethasone (D1756) (all from Sigma-Aldrich). The medium was replaced every 2 days, and the cells were analyzed on day 14.

For lineage-specific differentiation of unfractionated CD29-positive MDSCs, a protocol adapted from Naraoka et al. [9] was used. Cells were pre-cultured for 3 days in BMCM on collagen-coated or uncoated plates. Myogenic differentiation was induced by culturing the cells in MDM-2 (2% FBS-DMEM-HG supplemented with 1 µg/mL insulin) for 7 days. For adipogenic differentiation, cells were sequentially cultured in MDM-2 for 7 days followed by ADM for 14 days.

Total RNA was extracted using TRIzol™ reagent (Invitrogen) according to the manufacturer's protocol. RNA (0.5 µg) was reverse-transcribed using ReverTra Ace™ qPCR RT Master Mix with gDNA Remover (Toyobo, Osaka, Japan). Quantitative real-time polymerase chain reaction (qPCR) was performed using a LightCycler® 480 system (Roche Diagnostics, Mannheim, Germany) with 100 ng complementary DNA (cDNA), 1 µM primers, and SYBR Green I Master Mix (Roche). Melting-curve analysis was performed to confirm qPCR product specificity. GAPDH served as the internal reference gene.

For each cell isolate and experimental condition, RNA was extracted from three independent culture replicates (Batches 1–3), and each RNA sample was analyzed in technical duplicate. Only amplification curves with crossing point (Cp) values before cycle 35 were considered detectable. Samples showing no amplification or amplification at or after cycle 35 were classified as undetectable and excluded from relative expression calculations. Genes for which fewer than three batches yielded detectable amplification are identified in the corresponding figure legends, together with the number of contributing batches. Relative gene expression was calculated using the 2^−ΔCp method, where ΔCp = Cp_target − Cp_GAPDH. Primer sequences are presented in Supplementary Table 1 [6, 29].

Cells were fixed with 4% paraformaldehyde for 15 min, permeabilized with 0.2% Triton X-100 for 10 min, and blocked with 5% FBS for 1 h at room temperature. For detection of the myogenic marker, cells were incubated overnight at 4°C with mouse anti-myosin heavy chain antibody (1:500; MY-32; Sigma-Aldrich), followed by Alexa Fluor™ 647-conjugated goat anti-mouse immunoglobulin G secondary antibody (1:500; Invitrogen). Nuclei were counterstained with Hoechst 33342 (Thermo Fisher Scientific, Waltham, MA, USA).

For visualization of intracellular lipids, cells were stained with 2 µM BODIPY™ 493/503 (Sigma-Aldrich) at 37°C for 15 min in the dark, fixed with 4% paraformaldehyde, and counterstained with Hoechst 33342. Fluorescence images were acquired using a ZEISS LSM 710 confocal microscope with ZEN software (Carl Zeiss Microscopy GmbH, Jena, Germany).

All cell cultures were evaluated for mycoplasma contamination using the MycoAlert® Mycoplasma Detection Kit (LT07-318; Lonza, Switzerland) according to the manufacturer's enzymatic protocol. All tested cultures were negative for mycoplasma contamination.

Three biological isolates (BNS1, BNS2, and BNS3) derived from a single donor animal (Animal 3) were used to assess spatial/intra-donor heterogeneity. For gene expression analyses, each isolate and experimental condition comprised three independent culture batches (Batches 1–3), with each batch analyzed in technical duplicate. The mean of the technical duplicates was calculated for each batch, and the independent batches served as the units of statistical analysis.

Data are presented as the mean ± standard deviation. Normality was assumed based on the experimental design and small number of replicates. Differences between experimental conditions were evaluated using Student's t-test for pairwise comparisons and one-way analysis of variance (ANOVA) for multiple-group comparisons. p < 0.05 was considered statistically significant. No correction for multiple comparisons was applied because each gene was analyzed as an independent biological marker rather than as part of a family-wise hypothesis. All statistical analyses were performed using GraphPad Prism version 10 (GraphPad Software, Boston, MA, USA).

Muscle-derived cell isolation and expansion: Muscle tissue from the chuck roll of adult Pon Yang Kham Thai-French crossbred cattle was dissected into fragments with an average weight of 58.2 ± 4.3 mg and subjected to enzymatic digestion with collagenase, followed by size-based filtration, RBC lysis, and washing. The isolation protocol consistently yielded 5.13 ± 1.23 × 10⁵ viable cells/g of tissue, with cell viability exceeding 80% across all isolates (Supplementary Table 2). During the pre-sorting expansion phase, cells proliferated 3–6-fold within 10 days under BMCM conditions and were subcultured once to twice per week. Three to four passages were sufficient to obtain the cell numbers required for flow cytometric analysis and FACS-based sorting.

To preserve tissue resources, dissected muscle samples were cryopreserved in FBS containing 10% DMSO at −80°C. After thawing, the cryopreserved tissues remained viable and yielded muscle-derived cell numbers comparable to those from freshly processed tissues, indicating the feasibility of tissue cryopreservation for subsequent experiments. Long-term cryopreservation stability beyond 100 days was not evaluated in this study.

Subpopulation identification and FACS-based isolation of SCs and FAPs: CD29-positive MDSCs were characterized by flow cytometry using a panel of cell-surface markers. Following pre-sorting culture, >95% of the cells expressed CD29 across all isolates, whereas CD31- and CD45-positive cells were not detected, indicating effective depletion of endothelial and hematopoietic cell populations (Supplementary Figure 1). Within the CD29-positive population, two major subpopulations were distinguished according to CD56 expression: CD29⁺/CD56⁺ SCs and CD29⁺/CD56⁻ FAPs [9, 29].

The cell isolation and sorting workflow is summarized in Figure 1A. The relative proportions of SCs and FAPs varied markedly among the tissue isolates (Figure 1B). The SC:FAP ratios ranged from 4.3:95.1 in the FAP-dominant BNS1 isolate to 96.0:3.8 in the SC-enriched BNS3 isolate, with the remaining cells representing CD29-negative populations. These findings demonstrated pronounced spatial variation in SC:FAP composition among isolates obtained from a single bovine donor.

Following flow cytometric characterization, SCs and FAPs were purified by FACS using a conservative gating strategy that excluded boundary events between the CD29⁺/CD56⁺ and CD29⁺/CD56⁻ gates to minimize cross-contamination between the two subpopulations. Consequently, the percentages of cells recovered in the sorted fractions were slightly lower than their corresponding proportions in the pre-sorting populations. BNS1 yielded a predominantly FAP fraction, accounting for 88.5% of the pre-sorting population. In contrast, BNS3 yielded a predominantly SC-enriched fraction (85.9%). BNS2 contained both subpopulations, comprising 68.4% FAPs and 21.6% SCs, enabling comparative analyses of the two cell types derived from the same tissue isolate (Figure 1C). No apparent morphological differences were observed between sorted SCs and FAPs by phase-contrast microscopy (Figure 1D). Representative FACS gating plots are presented in Supplementary Figures 2–4.

Proliferative capacity: The proliferative capacity of SCs and FAPs was evaluated across multiple passages using the CCK-8 assay. Both cell types proliferated robustly during early passages, with cell yield increasing approximately 6–8-fold over the first four passages (passages 7–8 after initial isolation). Proliferative capacity subsequently declined with increasing passage number, and cell division was rarely observed at passages 15–16 (Figures 2A–D).

The PDT of BNS1-derived FAPs increased from 21.8 h at passage 4 to 27.0 h at passage 6 and 29.8 h at passage 8. For BNS2-derived FAPs, the corresponding PDTs were 23.6, 32.7, and 39.1 h, whereas BNS2-derived SCs showed PDTs of 19.8, 24.9, and 25.5 h at passages 4, 6, and 8, respectively. BNS3-derived SCs showed PDTs of 21.1 h at passage 4, 23.1 h at passage 6, and 23.6 h at passage 8. Thus, a passage-dependent increase in PDT and a decline in proliferative capacity were observed across the evaluated isolates and cell types, consistent with previously reported proliferative limitations of primary adult muscle stem cells during prolonged serum-dependent culture [30–32].

Adipogenic differentiation of FAPs: FAPs derived from BNS1 and BNS2 were cultured in ADM for 14 days to evaluate their adipogenic differentiation capacity. Both isolates showed increased intracellular lipid accumulation following adipogenic induction, as demonstrated by BODIPY™ 493/503 staining (Figure 3A). At the transcriptional level, differentiated FAPs showed decreased expression of the FAP-associated marker PDGFRA; however, detectable amplification was obtained in only one of three independent batches. In contrast, expression of the adipocyte-associated gene CIDEC increased across all three batches in both BNS1- and BNS2-derived FAPs (Figures 3B and C).

Expression of FABP4, a marker associated with mature adipocyte lipid metabolism [33], increased in BNS2-derived FAPs but remained undetectable in BNS1-derived FAPs following adipogenic induction. BNS1-derived FAPs also showed comparatively lower lipid-droplet accumulation than BNS2-derived FAPs. Together, these findings demonstrated isolate-dependent differences in adipogenic differentiation-associated responses.

Myogenic differentiation of SCs: SCs derived from BNS2 and BNS3 were cultured in MDM on collagen type I-coated plates for 7 days. Following myogenic induction, the cells formed elongated, multinucleated myotube-like structures with strong MyHC immunoreactivity compared with undifferentiated controls (Figure 4A). Gene expression analysis showed that the myogenic differentiation-associated genes Myf5, MyHC, and MyoG were undetectable in undifferentiated SCs but became detectable following myogenic induction. Conversely, the SC-associated genes Pax3 and Pax7 were detected only in undifferentiated SCs (Figures 4B and C).

Pax7 expression was consistently detected across the evaluated SC isolates, whereas Pax3 expression varied between isolates. Pax3 was detected in all three independent RNA batches from BNS2 but in only one of three batches from BNS3, consistent with previously reported inter-isolate variability in bovine MDSCs [6, 9, 29].

Lineage-committed subpopulations drive differentiation responses in heterogeneous CD29-positive MDSCs

To determine the lineage-specific differentiation capacity of unfractionated CD29-positive populations, BNS2 cells were cultured under myogenic or adipogenic induction conditions without prior FACS-based separation of the constituent subpopulations (Figure 5A). Under adipogenic induction conditions, the cells exhibited prominent lipid accumulation with minimal MyHC immunoreactivity (Figure 5B). Conversely, under MDM-2 conditions, MyHC immunoreactivity increased markedly while lipid accumulation was suppressed, irrespective of whether the culture surface was collagen-coated (Figure 5C). Similar condition-dependent responses were reproduced using cryopreserved BNS2 cells.

These reciprocal differentiation responses demonstrated that heterogeneous CD29-positive MDSC populations contained cells capable of responding selectively to myogenic and adipogenic induction conditions. In conjunction with the FACS-defined CD56-positive SC and CD56-negative FAP subpopulations, these findings support an association between subpopulation composition and the lineage-specific differentiation behavior of unfractionated CD29-positive MDSCs [9, 25].

This study provides evidence that bovine CD29-positive MDSCs are better interpreted as heterogeneous mixtures of lineage-biased progenitor subpopulations than as a uniformly bipotent progenitor pool. This interpretation is supported by the marked inter-isolate variation in the relative proportions of SCs and FAPs and by the distinct lineage-specific responses observed after myogenic and adipogenic induction. Accordingly, the apparent dual-lineage differentiation capacity of unfractionated CD29-positive populations may be explained, at least in part, by their spatially variable cellular composition rather than by intrinsic multipotency of individual cells.

In addition to this biological finding, the present study established a reproducible and field-compatible protocol for isolating CD29-positive MDSCs from adult Pon Yang Kham Thai-French crossbred cattle. The protocol yielded 5.13 ± 1.23 × 10⁵ viable cells/g of tissue, achieved >95% CD29-positive enrichment after pre-sorting expansion, and supported robust proliferation during early passages. The capacity to recover and expand progenitor cells from adult slaughterhouse-derived skeletal muscle is relevant to veterinary biotechnology and cultured meat development, where accessible and standardized cell sources are required.

A major finding of this study was the pronounced variation in SC:FAP composition among spatial isolates. Individual isolates ranged from FAP-dominant to SC-enriched populations, indicating substantial heterogeneity within the sampled chuck roll tissue. SCs and FAPs have distinct but complementary biological roles, with SCs contributing primarily to myogenesis and FAPs contributing to adipogenic and stromal processes [26, 34, 35]. In livestock, these functions are particularly relevant because muscle growth and intramuscular fat deposition contribute to important meat quality characteristics, including marbling [26, 34, 35].

The observed variation may reflect differences in local tissue composition and microenvironment. Previous studies show that muscle stem cells and progenitor populations can vary by age, tissue context, and muscle characteristics [26, 34, 35]. However, the present experimental design does not allow separation of the relative contributions of genetic background, age, nutritional status, or the local muscle microenvironment. In addition, because the chuck roll was not seam-dissected and the individual muscle of origin of each isolate was not recorded, intermuscular variation may also have contributed to the observed SC:FAP differences.

Mechanistically, pathways such as Wnt and transforming growth factor-β (TGF-β) regulate myogenic and adipogenic lineage behavior in experimental systems [36, 37]. Whether comparable signaling mechanisms contribute to spatial variation in bovine SC and FAP abundance remains uncertain. Future studies should investigate whether lineage bias in adult bovine muscle is associated with stable molecular or epigenetic differences or remains responsive to extrinsic cues during in vitro expansion.

The present protocol prioritizes simplicity, accessibility, and compatibility with adult slaughterhouse-derived tissue. Compared with previously reported approaches, the workflow incorporated short-term culture enrichment before phenotypic characterization and FACS, thereby achieving high CD29-positive enrichment without density-gradient centrifugation [9]. Differences in reported cell yield among studies should nevertheless be interpreted cautiously because donor age, tissue source, digestion conditions, and enrichment strategy can substantially influence recovery. Adult skeletal muscle may also contain fewer expandable progenitor cells than fetal or juvenile tissue [26, 34, 35].

Supplementary Table 3 provides a direct comparison with representative bovine MDSC isolation methods, including those described by Naraoka et al. [9] and Dohmen et al. [6]. The present protocol yielded 5.13 ± 1.23 × 10⁵ viable cells/g of tissue and achieved >95% CD29-positive enrichment using skeletal muscle obtained from market-weight adult cattle aged 2.5–3 years. In contrast, several previously reported protocols have relied on fetal or juvenile tissues, which may not reflect the practical cell sources available through commercial slaughter systems. The present workflow also used single-step enzymatic digestion and did not require density-gradient centrifugation. Furthermore, the successful recovery of viable cells from cryopreserved tissue supports the potential use of tissue banking for subsequent experiments, although cryopreservation stability beyond 100 days was not evaluated.

Both SCs and FAPs showed strong proliferative capacity during early passages, followed by a progressive decline at later passages. This pattern is consistent with the finite proliferative lifespan commonly observed in primary adult muscle-derived progenitor cells [30–32]. The gradual increase in population doubling time across passages further supports the importance of limiting prolonged expansion when cells are intended for downstream differentiation experiments.

The culture medium contained 2.5 µM of the p38 MAPK inhibitor SB203580, a lower concentration than that used in several previous studies [25, 29, 30, 38]. Under the conditions used here, cells retained both proliferative and lineage-specific differentiation capacity. These findings indicate that 2.5 µM SB203580 was compatible with the present culture system. However, because the study did not directly compare different SB203580 concen-trations, it cannot establish that this concentration is superior, more cost-effective, or biologically optimal. Direct dose-response experiments are needed to determine the concentration that best balances expansion and maintenance of differentiation potential.

The differentiation-associated responses of isolated FAPs and SCs further supported their functional distinction. FAPs from BNS1 and BNS2 accumulated intracellular lipids after adipogenic induction and showed increased CIDEC expression. In BNS2-derived FAPs, FABP4 expression also increased, whereas FABP4 remained undetectable in BNS1-derived FAPs despite lipid accumulation. Because FABP4 is associated with adipocyte differentiation and lipid metabolism [33, 39, 40], this isolate-dependent difference may indicate variation in the extent of adipogenic differentiation. However, the present data do not establish a definitive temporal sequence between CIDEC and FABP4 expression. The interpretation that these markers represent different stages of adipogenic progression is biologically plausible and consistent with previous studies [41, 42], but it should be validated with time-course experiments and additional adipogenic markers.

SCs derived from BNS2 and BNS3 formed multinucleated myotube-like structures and showed MyHC immunoreactivity following myogenic induction. The myogenic differentiation-associated genes Myf5, MyHC, and MyoG became detectable after induction, whereas the SC-associated genes Pax3 and Pax7 were detected in undifferentiated cells. Pax7 was detected more consistently than Pax3 across the evaluated SC isolates. Previous studies have also described variability in satellite cell-associated marker expression in bovine muscle-derived populations [6, 9, 25, 29]. These findings further emphasize that characterization of individual isolates may be more informative than assuming uniform marker expression across all bovine MDSC populations.

The reciprocal lineage responses observed in unfractionated CD29-positive populations provide important context for interpreting their apparent bipotency. Under adipogenic induction, the cells showed prominent lipid accumulation with minimal MyHC expression, whereas myogenic induction increased MyHC expression and suppressed lipid accumulation. In parallel, FACS identified distinct CD29⁺/CD56⁺ SC and CD29⁺/CD56⁻ FAP subpopulations within the bulk CD29-positive fraction.

Taken together, these findings support a model in which the differentiation behavior of bulk CD29-positive MDSCs is influenced by the relative abundance and selective response of pre-existing lineage-biased subpopulations. This interpretation is consistent with single-cell transcriptomic evidence showing that bovine SCs and FAPs constitute transcriptionally distinct populations before in vitro manipulation [31]. However, the present study did not perform single-cell lineage tracing; therefore, intrinsic bipotency at the level of individual CD29-positive cells cannot be completely excluded. The findings instead demonstrate that population level dual-lineage differentiation alone is insufficient to establish intrinsic single-cell multipotency.

The spatial variation in SC:FAP composition has practical implications for cultured meat production because uncontrolled differences in starting-cell composition may affect subsequent biomass expansion, myogenic differentiation, adipogenic differentiation, and tissue composition. If bulk MDSC preparations differ substantially in their baseline SC:FAP ratios, downstream responses may vary even when the same culture and induction conditions are applied.

Accordingly, characterization of subpopulation composition before large-scale expansion may improve process reproducibility. Potential strategies include prospective enrichment, standardized SC:FAP profiling, or culture conditions designed to accommodate defined mixtures of lineage-committed progenitors. However, the present study did not directly evaluate large-scale manufacturing performance, batch-to-batch variability across independent donors, or final cultured meat characteristics. Therefore, the suggestion that baseline SC:FAP composition may contribute to inconsistencies reported in cultured meat production should be considered a testable hypothesis rather than a demonstrated causal mechanism.

The present findings also have potential implications for donor selection and cell-bank development. Rapid assessment of SC:FAP composition before large-scale culture could help identify tissue sources with predictable myogenic or adipogenic potential. Genetic or epigenetic screening may also be useful in this context, but the present study did not evaluate these approaches, and they require independent validation before they can be recommended for routine cell sourcing.

Successful translation of cultured meat technologies will require scalable cell expansion platforms, serum-free or chemically defined culture systems, standardized upstream processing, and careful control of manufacturing costs. Although the present workflow provides a practical method for isolating cells from adult slaughterhouse-derived tissue, its compatibility with large-scale bioreactor expansion and food-grade production remains to be established.

Subpopulation heterogeneity may represent both a biological and a process-control variable. Variation in SC:FAP ratios could potentially influence biomass yield, final tissue composition, and differentiation performance. Consequently, subpopulation-aware quality control may be useful during process development. Nevertheless, this study did not directly examine the effects of SC:FAP composition on media consumption, manufacturing cost, bioreactor performance, or final product quality; future scale-up studies should evaluate these effects experimentally.

Several limitations should be considered when interpreting these findings. First, detailed subpopulation characterization and functional analyses were based on three spatial isolates obtained from a single donor animal. Therefore, the observed SC:FAP ratios cannot be generalized to the broader Pon Yang Kham Thai-French crossbred cattle population. Although the within-donor design was useful for demonstrating spatial heterogeneity, larger studies involving genetically and phenotypically diverse animals are required to quantify inter-animal variability and establish population level distributions of SC:FAP composition.

Second, although BNS1, BNS2, and BNS3 were obtained from distinct anatomical positions within the chuck roll, the subprimal was not seam-dissected and the specific muscle of origin of each isolate was not recorded. Because SC and FAP abundance may vary with muscle characteristics and fiber-type composition, intermuscular differences may have contributed to the observed heterogeneity. Future studies should therefore use individually identified muscles and standardized anatomical sampling locations.

Third, the culture system relied on FBS, which limits direct translation to food-grade cultured meat production because of ethical, regulatory, reproducibility, and scalability considerations [13, 23, 43–45]. Development and validation of serum-free or chemically defined media will therefore be necessary for subsequent translational applications.

Fourth, the findings require validation in larger and more diverse donor populations before breed-level conclusions can be drawn. Longitudinal studies should also evaluate whether SC:FAP composition remains stable during expansion and whether donor age, sex, genotype, nutritional status, or muscle type systematically influences subpopulation abundance.

Fifth, the gene expression panel was designed primarily for lineage identification and did not include key upstream regulators such as PPARG for adipogenic commitment or MYOD1 for early myogenic activation [46, 47]. The authors also did not perform protein-level validation by Western blotting, enzyme-linked immunosorbent assay, or other quantitative approaches. In addition, quantitative assessment of myotube fusion, lipid-droplet accumulation, and contractile function would strengthen functional characterization. Future studies should therefore incorporate broader transcriptomic or proteomic profiling, validated reference genes, fusion-index analysis, quantitative lipid measurements, and functional contractility assays.

Sixth, the PDGFRA findings shown in Figure 3 were based on detectable amplification in only one of three independent RNA batches for both BNS1- and BNS2-derived FAPs. In that batch, PDGFRA was abundantly detected, whereas amplification was absent in the remaining two runs. The >11-Cp difference among runs, together with the identical failure pattern in two independently derived FAP isolates analyzed within the same runs, suggests that technical assay failure may have contributed to the missing amplification [48]. Because the remaining cDNA was exhausted, the failed runs could not be repeated. Consequently, the apparent reduction in PDGFRA expression after adipogenic induction should be considered preliminary and requires independent confirmation.

Seventh, warm ischemia time between slaughter and tissue processing was not formally recorded. This factor may affect cell viability, recovery, and subsequent culture performance, and documenting it would improve reproducibility between laboratories [49]. Future studies should therefore standardize and report the interval between slaughter, tissue collection, cooling, and processing.

Overall, future work should prioritize multi-donor validation, standardized anatomical sampling, serum-free culture systems, broader molecular and functional characterization, long-term cryopreservation studies, and scale-up experiments using controlled SC:FAP compositions. These approaches will be important for determining whether subpopulation-aware cell sourcing can improve the reproducibility and industrial feasibility of bovine cultured meat production.