Introduction
Skeletal muscle is a heterogeneous tissue composed of many individual multinucleated contractile muscle cells termed myofibers that can be classified into various types based on their morphology, histochemistry, enzymatic activity, and biochemical characteristics (Brooke and Kaiser 1970). One of the most common methods of classification of myofiber type involves identifying the expression of specific myosin heavy chain (MyHC) isoforms (Hoh 2023). In adult mammals, skeletal myofibers are typically categorized into 4 main types: Type I (slow oxidative), IIA (fast oxidative), IIX (fast oxidative and glycolytic), and IIB (fast glycolytic) (Schiaffino and Reggiani 1994).
Myofiber type composition in cattle is an important determinant of meat tenderness and flavor (Calkins et al., 1981; Komiya et al., 2020; Li et al., 2023; Picard and Gagaoua, 2020). The myofiber type composition of bovine muscle differs most notably between muscle groups with distinct physiological demands (Kirchofer et al., 2002; Zhang et al., 2014). For example, the psoas major muscle (tenderloin) has a relatively higher proportion of type I myofibers when compared to the longissimus dorsi (e.g., loin) and semitendinosus (e.g., eye of round), in which type II myofibers predominate (Hwang et al., 2010; Joo et al., 2017; Kim et al., 2016; Lang et al., 2020). Myofiber type composition in beef cattle can be influenced by age (Jurie et al., 1999), breed (Wegner et al., 2000), genetic selection (Ozawa et al., 2000; Picard et al., 2006), method of housing (Jurie et al., 1998), sex (Johnston et al., 1981), nutritional state (Greenwood et al., 2009; Picard et al., 1995; Seideman and Crouse 1986), and administration of anabolic steroids and/or beta-adrenergic agonists (Clancy et al., 1986; Ebarb et al., 2016; Ebarb et al., 2017; Gonzalez et al., 2009; Kellermeier et al., 2009; Vestergaard et al., 1994).
Several different techniques have been developed to identify myofibers of various types, each with its advantages and limitations (Sawano and Mizunoya, 2022). Among these, immunohistochemical staining with myofiber type–specific primary antibodies against MyHC isoforms has become the most widely used method (Schiaffino 2018). In the last decade, there has been a surge in the availability of purpose-built semi or fully automated software options for the image analysis of immunofluorescent muscle cross-sections. Most of these software options can also characterize myofiber type. However, all this software was originally developed and validated using rodent tissue samples (Babcock et al., 2020; Bergmeister et al., 2016; Bergmeister et al., 2017; Danckaert et al., 2023; Desgeorges et al., 2019; Encarnacion-Rivera et al., 2020; Kastenschmidt et al., 2019; Mayeuf-Louchart et al., 2018; Rahmati and Rashno, 2021; Smith and Barton, 2014; Waisman et al., 2021; Wen et al., 2018; Zhang et al., 2025). In contrast, image analysis is generally still performed manually in studies of bovine muscle histology. Such manual image analysis is labor-intensive, time-consuming, and potentially prone to investigator bias. To address these limitations, commercially available software such as Noesis Visilog (Meunier et al., 2010) and Agilent Gen5 (Fuerniss and Johnson, 2023) have been applied to the analysis of immunofluorescent staining of bovine muscle samples. The only freely available software that has been specifically validated for use on bovine muscle samples is the deep learning-based tool MyoV (Gu et al., 2024). However, MyoV is designed for the analysis of brightfield images of hematoxylin and eosin (H&E)-staining and, as such, cannot identify myofiber type profile or quantify myofiber type–specific size (Gu et al., 2024).
In the current study, we established and validated a rapid high-throughput method for analysis of bovine myofiber morphology, including myofiber type profile and myofiber type–specific myofiber size. This method combines automated immunofluorescent microscopy and high-content image analysis using MuscleBos software, a new customized bovine-specific version of the open source and freely available MuscleJ plugin in ImageJ/FIJI (Mayeuf-Louchart et al., 2018). Ultimately, this novel methodology should enable the development of new valuable insights into skeletal myofiber morphology in bovine species and its impact on meat characteristics.
Materials and Methods
Animals and housing
Data for this study were collected, and procedures were performed at the Purdue University Animal Sciences Research and Education Center dairy unit from September 2022 through February 2023 as part of a larger study (Gouveia et al., 2024). All animal procedures were approved by the Purdue University Institutional Animal Care and Use Committee (protocol #2109002197). In this prior study, longissimus dorsi muscle depth was measured in multiparous Holstein dairy cows (∼4 years old) that had completed 2.8 ± 1.0 (mean ± SD) lactations, with previous 305-d milk yield of 11,382 ± 1,338.3 kg, and body weight of 749 ± 72.2 kg at the time of enrollment via ultrasonography at 42 d before expected calving and cows were assigned based on baseline muscle depth to either the high muscle group (HM, ≥4.6 cm, n = 26) or low muscle group (LM, ≤4.6 cm, n = 22). Cows were further randomized to receive dietary supplementation with control (CON) or branched-chain volatile fatty acid (BCVFA) supplementation from 42 d before expected calving until parturition. For more information regarding the animals and housing, please refer to Gouveia et al. (2024).
Skeletal muscle biopsies
At approximately 21 d before expected calving and 21 d in milk, a biopsy sample was obtained from the longissimus dorsi muscle between the 9th and 13th rib of each cow by the same trained individual. Cattle were restrained in a livestock handling chute, hair was clipped, and the underlying skin was cleaned thoroughly. The area was cleaned with 70% isopropyl alcohol, scrubbed with povidone-iodine, and then cleaned once more with 70% isopropyl alcohol. A local anesthetic (2% lidocaine, 5 mL) was then injected into the intercostal area. Under aseptic conditions, an incision was made with a #10 scalpel blade through the hide and adipose tissue to expose the underlying muscle tissue. A 5 mm Bergstrom needle was then used to collect muscle biopsy samples, cutting in a circular pattern to ensure a representative sample of the area. The collected muscle tissue was washed in a beaker containing sterile 1 × phosphate-buffered saline (PBS). The tissue sample obtained was confirmed as muscle via a float test, in which muscle tissue sinks while adipose tissue floats. Following sample collection, the surgical area was once again cleaned with 70% isopropyl alcohol gauze before being patted dry with sterile dry gauze. After drying, the incision was closed using tissue adhesive and sprayed with an aerosol bandage. The incision sites and animals were monitored for a week following the procedure, with no complications or health issues occurring from the surgery.
Immunofluorescence analysis of bovine myofiber type
A portion of each bovine longissimus dorsi biopsy sample was oriented longitudinally on a plastic support, covered with optimal cutting temperature (OCT) compound (Fisher Scientific, 23-730-571), and rapidly frozen in isopentane cooled in liquid nitrogen. OCT-embedded muscle biopsy samples were stored at −80°C until further analysis. Muscle tissue cross-sections (10 μm) were cut in a cryostat (Leica CM1950) at −20°C and adhered to SuperFrost Plus slides (Fisher Scientific, 12-550-15). Slides were air dried at room temperature, tissue sections were encircled with hydrophobic PAP pen (Vector, H-4000), and then blocked for 1 h at room temperature in 10% goat serum (Invitrogen, 10000C) prepared in phosphate-buffered saline (PBS). Slides were incubated overnight in a humidified chamber at 4°C with primary antibodies diluted in blocking buffer. Primary antibodies tested included those against dystrophin [Developmental Studies Hybridoma Bank (DSHB), MANDYS1(3B7)s, MIgG2a, 1:10], Laminin 1 + 2 (Abcam, ab7463, Rabbit, 1:100), MyHC type I (DSHB, BA-D5c, MIgG2b, 1:100), MyHC type IIA (DSHB, SC-71c, MIgG1, 1:100), MyHC type IIB (DSHB, BF-F3c, MIgGM, 1:100), MyHC type IIX (DSHB, 6H1s, MIgM, 1:10), and all but type MyHC IIX (DSHB, BF-35c, MIgG1, 1:100). The following day slides were washed 3 times for 5 min each in PBS before incubating with appropriate Alexa Fluor conjugated secondary antibodies (diluted 1:500 in PBS) including Goat Anti-Rabbit IgG (H+L) Alexa Fluor 350 (Invitrogen, A-11046), Goat Anti-Mouse IgG2a Alexa Fluor 647 (Invitrogen, A-21241), Goat Anti Mouse IgG1 Alexa Fluor 488 (Invitrogen, A-21121), Goat Anti-Mouse IgG2b Alexa Fluor 555 (Invitrogen, A-21147), and Goat Anti-Mouse IgM Alexa Fluor 647 (A-21238). Following further washing in PBS, slides were mounted with coverslips using MOWIOL Fluorescence Mounting Medium and air dried at room temperature for 24 h.
Immunofluorescent microscopy
Each muscle biopsy sample was imaged with a 10 × Plan Fluorite objective using an automated fluorescent microscope equipped with a motorized stage and DAPI, FITC, Texas Red, and CY5 filters operating in upright configuration (Echo Revolution). This system enables the automated capture of multiple adjacent 10 × fields of view across the sample using a motorized stage, which were automatically aligned and merged (stitched) by the instrument’s software to generate a single large composite image of the entire muscle biopsy cross-section. Pseudo-colors were applied to aid visual differentiation of various myofiber types. Stitched images of each fluorescent channel were exported as 16-bit TIFF files and then merged as Multi-Channel TIFF images using ImageJ software before image analysis.
High-throughput image analysis with MuscleBos software
Multi-channel 16-bit TIFF images were analyzed using the MuscleJ 1.0.2 plugin for FIJI/ImageJ with custom modifications. MuscleJ is an open-source plugin for ImageJ/FIJI originally designed and validated for analysis of mouse skeletal muscle tissue sections (Mayeuf-Louchart et al., 2018). The MuscleJ plugin’s “Myofiber morphology” function uses fluorescent images of myofiber border staining (e.g., dystrophin, laminin, or wheat germ agglutinin) to automatically segment, count, and measure the cross-sectional area (CSA) of each individual myofiber within a muscle cross-section. Additionally, MuscleJ includes a “Myofiber type” function that automatically quantifies the signal intensity of MyHC-specific antibody staining within each myofiber boundary relative to a threshold value to determine myofiber type profile and myofiber type–specific CSA. While initially designed and validated for use on mouse skeletal muscle tissue cross-sections, here we customized and validated MuscleJ 1.0.2 to enable rapid high-throughput image analysis of bovine muscle cross-sections. We named this new bovine-specific MuscleJ version MuscleBos. MuscleBos is open-source and freely available for download from GitHub (https://github.com/jmarkwor/MuscleBos), which also hosts the complete step-by-step user guide, an example image, and tutorial documentation.
Statistical analysis
Statistical analysis was performed in GraphPad Prism 10. The results obtained from MuscleBos, both with and without type IIX myofiber-specific CSA exclusion thresholds applied, were compared to those obtained via manual measurements of percentage myofiber type and myofiber type–specific CSA on the same source images via one-way analysis of variance (ANOVA) followed by pairwise Holm-Šídák post hoc tests. For the application of the customized bovine MuscleJ plugin to the characterization of differences in myofiber morphology between dairy cows with differing amounts of longissimus dorsi muscle mass throughout the transition period, data were analyzed by two-way ANOVA followed by pairwise Holm-Šídák post hoc tests. Differences between groups were considered statistically significant at a cutoff of p < 0.05.
Results
The bovine longissimus dorsi muscle lacks type IIB myofibers
Muscle biopsy cross-sections were first stained with a primary antibody against laminin together with a cocktail of mouse monoclonal antibodies against MyHC type I (BA-D5), type IIA (SC-71), and type IIB (BF-F3) (Figure 1). This primary antibody cocktail is commonly used in studies of rodent myofiber type (Bloemberg and Quadrilatero 2012). This staining revealed an apparent lack of type IIB myofibers within the bovine longissimus dorsi muscle (Figure 1). In rodent skeletal muscle, the SC-71 primary antibody reacts specifically with type IIA myofibers and does not stain IIX or IIB myofibers (Markworth et al., 2021; Markworth et al., 2020; Markworth et al., 2017; Sharma et al., 2025). In contrast, we found that the SC-71 antibody appeared to react strongly with all non-type I (BA-D5neg) myofibers in bovine muscle biopsy cross-sections (Figure 1).
Identification of type I vs. II myofibers in bovine skeletal muscle: Cross-sections (10 μm) of bovine longissimus dorsi muscle biopsy samples were stained with a primary antibody against laminin (ab7463, Rabbit, 1:100) to label the myofiber boundaries, together with a cocktail of mouse monoclonal antibodies against MyHC type I (BA-D5c, MIgG2b. 1:100), MyHC type IIA (SC-71c, MIgG1, 1:100), and MyHC type IIB (BF-F3c, MIgM, 1:100). Primary antibody staining was visualized using Alexa Fluor-conjugated secondary antibodies (1:500) including Goat Anti-Rabbit Alexa Fluor 350 (to detect laminin), Goat Anti-Mouse IgG1 Alexa Fluor 488 (to detect SC-71), Goat Anti-Mouse IgG2b Alexa Fluor 555 (to detect BA-D5), and Goat Anti-Mouse IgM Alexa Fluor 647 (to detect BF-F3). Laminin, BA-D5, SC-71, and BF-F3 staining were pseudo colored white, red, green, and blue, respectively. Scale bars 100 μm.
Identification of type IIA and IIX myofibers in bovine skeletal muscle
To confirm the presence or absence of type IIX myofibers in these bovine longissimus dorsi samples, we stained serial bovine muscle cross-sections with a primary antibody against dystrophin in combination with a cocktail of mouse monoclonal antibodies against MyHC IIX (6H1) and all myosin isoforms except for type IIX (BF-35) (Figure 2). This staining revealed many IIX myofibers as present based on both negative immunoreactivity with BF-35 and positive immunoreactivity for 6H1 (Figure 2). Furthermore, type IIA and type IIX myofibers stained equally brightly with the SC-71 antibody, suggesting that these bovine myofiber types cannot be reliably distinguished based on relative SC-71 staining intensity (Figure 2).
Distinguishing type IIA and IIX myofibers in bovine skeletal muscle: A: Cross-sections (10 μm) of bovine longissimus dorsi muscle biopsy samples were stained with a primary antibody against dystrophin [MANDYS1(3B7)s, MIgG2a, 1:10] to label the myofiber boundaries, together with a cocktail of mouse monoclonal primary antibodies against MyHC type I (BA-D5c, MIgG2b, 1:100) and MyHC type IIA (SC-71c, MIgG1, 1:100). Primary antibody staining was visualized using Alexa Fluor-conjugated secondary antibodies (1: 500) including Goat Anti-Mouse IgG2a Alexa Fluor 647 (to detect dystrophin), Goat Anti-Mouse IgG1 Alexa Fluor 488 (to detect SC-71), and Goat Anti-Mouse IgG2b Alexa Fluor 555 (to detect BA-D5). Dystrophin, BA-D5, and SC-71 staining was pseudo colored white, red, and green, respectively. B: Serial cross-sections from the same bovine longissimus dorsi muscle biopsy sample shown in panel A were stained with a primary antibody against dystrophin [MANDYS1(3B7)s, MIgG2a, 1:10] to label the myofiber boundaries in combination with a cocktail of primary antibodies against MyHC type IIX (6H1s, MIgM, 1:10) and all MyHC isoforms except for type IIX (BF-35c, MIgG1, 1:100). Primary antibody staining was visualized using Alexa Fluor-conjugated secondary antibodies (1:500) including Goat Anti-Mouse IgG2a Alexa Fluor 647 (to detect dystrophin), Goat Anti-Mouse IgG1 Alexa Fluor 488 (to detect BF-35), and Goat Anti-Mouse IgM Alexa Fluor 555 (to detect 6H1). Dystrophin, BF-35, and 6H1 staining were pseudo colored white, red, and green, respectively. Scale bars are 200 μm.
Simultaneous identification of type I, IIA, and IIX fibers in bovine muscle
To establish a reliable method of bovine muscle myofiber type identification, we next tested different MyHC-specific primary antibody combinations. We used a primary antibody against laminin together with a combination of monoclonal primary antibodies against MyHC type I (BA-D5), type IIX (6H1), and all but type IIX (BF-35) (Figure 3). This primary antibody cocktail was able to identify type I myofibers (as BA-D5posBF-35pos cells), type IIA myofibers (as BA-D5negBF-35pos cells), and type IIX myofibers (as BF-35neg 6H1pos cells) on a single tissue section (Figure 3).
Simultaneous identification of type I, IIA, and IIX myofibers in bovine skeletal muscle: Cross-sections (10 μm) of bovine longissimus dorsi muscle biopsy samples were stained with a primary antibody against laminin (ab7463, Rabbit, 1:100) to label the myofiber boundaries, together with a cocktail of mouse monoclonal primary antibodies against MyHC type I (BA-D5, MIgG2b), MyHC type IIX (6H1, MIgM), and all MyHC isoforms except type IIX (BF-35, MIgG1). Primary antibody binding was visualized with Alexa Fluor-conjugated secondary antibodies (1:500), including Goat Anti-Rabbit Alexa Fluor 350 (to detect laminin), Goat Anti-Mouse IgG1 Alexa Fluor 488 (to detect BF-35), Goat Anti-Mouse IgG2b Alexa Fluor 555 (to detect BA-D5), and Goat Anti-Mouse IgM Alexa Fluor 647 (to detect 6H1). Laminin, BA-D5, BF-35, and 6H1 staining were pseudo colored white, red, green, and blue, respectively. Scale bars are 100 μm.
Similar reactivity of MyHC antibodies when applied individually or in a cocktail
To confirm whether the BA-D5, BF-35, and 6H1 primary antibodies act similarly when applied as a cocktail, serial bovine muscle cross-sections were stained with these primary antibodies either individually or in combination (Figure 4). The reactivity of these primary antibodies was found to be highly similar when applied in a cocktail, when compared to incubation with each individual primary antibody on separate serial sections of the same tissue sample (Figure 4). Notably, however, the 6H1 primary antibody, while clearly staining type IIX (BF-35neg) myofibers, did cross-react with type I myofibers in bovine muscle both when applied in isolation or in a cocktail format (Figure 4).
Reactivity of MyHC primary antibodies with bovine skeletal muscle when applied individually or as a cocktail: Serial muscle cross-sections (10 μm) of bovine longissimus dorsi muscle biopsy samples were stained with a primary antibody against laminin (ab7463, Rabbit, 1:100) to label the myofiber boundaries together with mouse monoclonal primary antibodies against MyHC type I (BA-D5c, MIgG2b, 1:100), MyHC type IIX (6H1s, MIgM, 1:10), and all MyHC isoforms except type IIX (BF-35c, MIgG1, 1:100), applied either individually or in combination as a cocktail. Primary antibody staining was visualized with a cocktail of Alexa Fluor-conjugated secondary antibodies (1:500), including Goat Anti-Rabbit 350 (to detect laminin), Goat Anti-Mouse IgG2b Alexa Fluor 555 (to detect BA-D5), Goat Anti-Mouse IgM Alexa Fluor 647 (to detect 6H1), and Goat Anti-Mouse IgG1 Alexa Fluor 488 (to detect BF-35). Laminin, BA-D5, BF-35, and 6H1 were pseudo colored white, red, green, and blue, respectively. Scale bars are 100 μm.
High-throughput automated image analysis of bovine muscle cross-sections
We next questioned whether existing published high-content image analysis methods designed and validated in mice might be successfully applied to the analysis of bovine muscle samples. We focused our initial efforts on the MuscleJ 1.0.2 plugin for FIJI/ImageJ (Mayeuf-Louchart et al., 2018). Based on our initial qualitative observations, MuscleJ could accurately segment myofibers on cross-sections of bovine muscle biopsies when primary antibodies against either laminin or dystrophin were used to stain the myofiber boundaries (Figure 5). MuscleJ could also identify type I (BA-D5pos) myofibers in bovine skeletal muscle cross-sections, although in our study, an adjustment in the default sensitivity threshold for myofiber type detection from 3 × standard deviations (Std) from the mean to 1 × Std was required to achieve type I myofiber counts that were consistent with manual identification (Figure 6). Therefore, the relative proportions of type I myofibers and type I myofiber type–specific CSA (e.g., BA-D5pos cells) vs. type II myofibers (e.g., BA-D5neg cells) can be quantified with the default MuscleJ 1.0.2 plugin with some relatively minor modifications of the default sensitivity threshold (3 × Std) for myofiber type detection to the source code. Unfortunately, however, IIA vs. IIX myofibers could not be distinguished using MuscleJ due to the differences in the reactivity of the SC-71 antibody between rodent and bovine muscle tissue.
Automated myofiber segmentation in bovine skeletal muscle samples using MuscleJ software: Cross-sections (10 μm) of bovine longissimus dorsi muscle biopsy samples were stained with primary antibodies against either laminin (ab7463, Rabbit 1:100) or dystrophin [MANDYS1(3B7)s, MIgG2a, 1:10] to delineate the myofiber boundaries. Primary antibody binding was visualized using Alexa Fluor-conjugated secondary antibodies (1:500), including either Goat Anti-Rabbit Alexa Fluor 350 (to detect laminin) or Goat Anti-Mouse IgG2a Alexa Fluor 647 (to detect dystrophin). Laminin and dystrophin staining were pseudo-colored white. Stitched panoramic images of the entire muscle biopsy cross-section were obtained using an automated fluorescent microscope (Echo Revolution) and analyzed using the MuscleJ plugin for FIJI/ImageJ. Segmented myofiber regions of interest (ROIs) were overlaid on the original images, and corresponding cartography maps of segmented myofibers colored according to their measured cross-sectional areas (CSA) are shown. Scale bars are 400 μm on stitched mosaic images and 100 μm on representative fields of view.
Automated detection of type I myofibers in bovine skeletal muscle using MuscleJ software: Cross-sections (10 μm) of bovine longissimus dorsi muscle biopsy samples were stained with a primary antibody against laminin (ab7463, Rabbit, 1:100) to label the myofiber boundaries in combination with a mouse monoclonal primary antibody against MyHC type I (BA-D5c, MIgG2b, 1:100). Primary antibody binding was visualized using Alexa Fluor-conjugated secondary antibodies (1:500) including Goat Anti-Rabbit Alexa Fluor 647 (to detect laminin) and Goat Anti-Mouse IgG2b Alexa Fluor 555 (to detect BA-D5). Laminin and BA-D5 staining were pseudo-colored white and red, respectively. Stitched panoramic images of entire muscle biopsy cross-sections were obtained using an automated fluorescent microscope (Echo Revolution) and analyzed using the MuscleJ plugin for FIJI/ImageJ. Segmented myofiber regions of interest (ROIs) were overlaid on the original images, and corresponding cartography maps of type I myofibers (red) versus non-type I (e.g., type II) myofibers (gray) are shown. Scale bars are 400 μm on stitched mosaic images and 100 μm on representative fields of view.
Customization of MuscleJ for analysis of bovine myofiber type
To attempt to automate the detection and measurement of IIA and IIX myofibers in bovine muscle, we next assessed whether MuscleJ could detect staining with the BF-35 primary antibody. While MuscleJ was not originally designed to identify BF-35 staining, we found that the function originally designed to detect type IIB myofibers on mouse muscle samples worked well to identify BF-35pos myofibers on bovine muscle cross-sections (Figure 7). Moreover, when combined with the detection of type I myofibers (BA-D5pos cells) using a staining intensity threshold of 1 × Std from the mean, MuscleJ could also identify type I (BA-D5pos cells), type IIA (BA-D5negBF-35pos cells), and type IIX myofibers (BA-D5negBF-35neg cells) on a single tissue section (Figure 7). According to these new designations for bovine myofiber type classification, we made extensive custom modifications to the MuscleJ 1.0.2 source code, user interface (UI), and results outputs. Since an adjustment in the default sensitivity threshold of type I myofiber identification was required in our hands, and this may potentially vary between laboratories, we incorporated a new user interface (UI) option to enable users to easily set custom intensity thresholds for positive myofiber type identification. We also added new UI options to generate cartography maps with custom color schemes. Finally, we added the option to generate cartography results maps of myofiber area and myofiber type profile simultaneously from a single run. We named this new custom bovine-specific MuscleJ 1.0.2 version MuscleBos after the scientific bovine genus Bos (from Latin bōs: cow, ox, bull).
Simultaneous detection of type I, IIA, and IIX myofibers in bovine skeletal muscle using customized MuscleJ software: Bovine longissimus dorsi muscle biopsy cross-sections were stained with a primary antibody against laminin (ab7463, Rabbit, 1:100) to label the myofiber boundaries together with a cocktail of mouse monoclonal primary antibodies against MyHC type I (BA-D5c, MIgG2b, 1:100), all but IIX (BF-35c, MIgG1, 1:100), and type IIX (6H1s, IgM, 1:10). Primary antibody binding was detected using Alexa Fluor-conjugated secondary antibodies (1:500) including Goat Anti-Rabbit Alexa Fluor 350 (to detect laminin), Goat Anti-Mouse IgG2b Alexa Fluor 555 (to detect BA-D5), Goat Anti-Mouse IgG1 Alexa Fluor 488 (to detect BF-35), and Goat Anti-Mouse IgM Alexa Fluor 647 (to detect 6H1). Laminin, BA-D5, SC-71, and 6H1 staining was pseudo colored white, red, green, and blue, respectively. Stitched panoramic images of the entire muscle biopsy cross-section were obtained using an automated fluorescent microscope (Echo Revolution) and analyzed using the MuscleJ plugin for FIJI/ImageJ with custom modifications. Original images and myofiber type cartography maps of type I (red), type IIA (green), and type IIX (gray) are shown. Scale bars are 400 μm on stitched mosaic images and 100 μm on representative fields of view.
Validation of MuscleBos software
We next sought to validate our initial prototype version of MuscleBos via side-by-side comparisons with manual image analysis. A trained expert manually analyzed a random selection of images of longissimus dorsi biopsy cross-sections obtained from multiparous Holstein dairy cows (Gouveia et al., 2024). The same images were analyzed in parallel using MuscleBos software (Figure 8A). Manual analysis identified a total of 2,755 myofibers, while MuscleBos counted a total of 2661 myofibers. The average number of myofibers per image identified by MuscleBos was 211.92, which did not differ significantly from manual myofiber counts of 204.69 (Figure 8B). MuscleBos completed this task in ∼13 min, achieving an average speed of ∼200 myofibers/min. In contrast, it took the trained expert approximately 13 h working at a pace of ∼3.5 myofibers/min to complete this task. MuscleBos was therefore approximately 60 times faster than manual image analysis.
Validation of MuscleBos vs. manual quantification: A: Bovine longissimus dorsi muscle cross-section (10 μm) were stained with a primary antibody against laminin (ab7463, Rabbit, 1:100) to label the myofiber boundaries together with a cocktail of mouse monoclonal antibodies against MyHC type I (BA-D5c, MIgG2b, 1:100), MyHC type IIX (6H1s, MIgM, 1:10), and all MyHC types except IIX (BF-35c, MIgG1, 1:100). Primary antibody staining was visualized with Alexa Fluor-conjugated secondary antibodies (1:500) including Goat Anti-Rabbit Alexa Fluor 350 (to detect laminin), Goat Anti-Mouse IgG2b Alexa Fluor 555 (to detect BA-D5), Goat Anti-Mouse IgG1 Alexa Fluor 488 (to detect BF-35), and Goat Anti-Mouse IgM Alexa Fluor 647 (to detect 6H1). Laminin, BA-D5, BF-35, and 6H1 were pseudo colored white, red, green, and blue, respectively. Stitched panoramic images of the entire muscle biopsy cross-section were obtained using an automated fluorescent microscope (Echo Revolution) and analyzed using the MuscleBos plugin for FIJI/ImageJ. An original image and corresponding manually or MuscleBos-generated myofiber type cartography maps are shown. Scale bars on representative fields of view are 100 μm. B–E: Comparison of manual and MuscleBos analysis of the same bovine muscle cross-sections for total myofiber count (B), the percentage of type I myofibers (C), the percentage of type IIA myofibers (D), and the percentage of type IIX myofibers (E). F–H: Comparison of manual and MuscleBos analysis of the same bovine muscle cross-sections for mean myofiber cross-sectional area (CSA) measurements for all myofibers irrespective of type (F), type I myofibers (G), type IIA myofibers (H), and type IIX myofibers (I). Dot plots represent data from each individual animal (biological replicates). NS denotes no significant difference between manual analysis and MuscleBos. *Denotes p < 0.05 for manual analysis vs. MuscleBos.
The percentage of type I (Figure 8C), IIA (Figure 8D), and IIX (Figure 8E) myofibers did not differ between MuscleBos and manual image analysis. Total myofiber cross-sectional area (CSA) (Figure 8F), type I myofiber CSA (Figure 8G), and type IIA myofiber CSA (Figure 8H) measurements were also similar between MuscleBos and manual analysis. Our initial version of MuscleBos did measure the CSA of the type IIX myofibers to be significantly smaller when compared with manual image analysis (Figure 8I). This was found to be due to MuscleBos identifying some small unstained spaces between adjacent myofibers as type IIX myofibers (BA-D5negBF-35neg cells) (Figure 8A). To address this potential limitation, we incorporated a new option into the MuscleBos UI to set user-defined myofiber type–specific CSA exclusion thresholds. For type IIX myofibers, we set a default exclusion threshold of ≤1,000 μm2 since very few, if any, type IIX myofibers below this size were identified in manual image analysis. Excluding these very small unstained tissue areas did not influence the average myofiber count (Figure 8B), the percentage of type I (Figure 8C), type IIA (Figure 8D), and IIX (Figure 8E) myofibers, or the total mean myofiber CSA (Figure 8F). Nevertheless, applying this option improved the consistency between MuscleBos and manual analysis for measurement of the mean type IIX myofiber CSA (Figure 8I).
Application of MuscleBos software
We next applied MuscleBos to the analysis of muscle biopsy samples obtained from a previously published study in which longissimus dorsi muscle depth was measured in multiparous Holstein dairy cows via ultrasonography at 42 d before expected calving, and cows were assigned based on baseline muscle depth to either the high muscle group (HM, ≥4.6 cm, n = 26) or low muscle group (LM, ≤4.6 cm, n = 22). (Gouveia et al., 2024). Cows were further randomized to receive dietary supplementation with control (CON) or branched-chain volatile fatty acid (BCVFA) supplementation from 42 d before expected calving until parturition. Dietary supplementation with BCVFAs did not influence any indices of myofiber size or type. Therefore, data are presented here pooled over dietary supplementation treatment.
MuscleBos completed myofiber type analysis of a total of 27,508 myofibers from available immunofluorescent images of a total of 90 biopsy samples in a duration of 2 h and 17 min, achieving an average rate of ∼200 myofibers/min. Representative myofiber type staining of cross-sections of longissimus dorsi muscle biopsy samples from HM and LM cows obtained pre- and postpartum are shown in Figure 9A. The overall myofiber type profile of these samples was found to be 27.27% type I, 47.33% type IIA, and 25.41% type IIX (Figure 9B). The mean CSA of type I myofibers was lower than both type IIA and type IIX myofibers (Figure 9C). There was a significant effect of both group and time, but no group × time interaction effect for overall myofiber CSA (Figure 9D). HM cows showed a larger myofiber CSA than LM cows, and myofiber CSA decreased from pre to postpartum (Figure 9D). There was no significant effect of group, time, or group × time interaction effect for type I myofiber CSA (Figure 9E). Type IIA myofiber CSA showed a significant main effect of both group and time, but no group × time interaction (Figure 9F). Type IIA myofiber CSA was larger in HM vs. LM cows and decreased from pre- to post-parturition (Figure 9F). Type IIX myofiber CSA showed a trend towards a significant main effect of group, but no effect of time, or group × time interaction effect (Figure 9G). Type IIX myofiber CSA overall tended to be higher in HM vs LM cows (Figure 9G). There was no significant effect of time, group, or time by group interaction for the percentage of type I myofibers (Figure 9H). The percentage of type IIA myofibers showed a significant main effect of group and a statistical trend for a group-by-time interaction effect (Figure 9I). HM cows had a greater proportion of type IIA myofibers than LM cows, and this was especially true in the postpartum biopsy samples (Figure 9I). The proportion of type IIX myofibers showed a significant main effect of group and a group × time interaction effect (Figure 9J). HM cows had a lower percentage of type IIX myofibers specifically in the postpartum period when compared to LM cows (Figure 9J).
Application of MuscleBos for the characterization of bovine muscle phenotype: (A) Muscle biopsy samples were obtained from the longissimus dorsi of high muscle (HM) or low muscle (LM) multiparous dairy cows at 21 d before expected calving (PRE) and at 21-d postpartum (POST). Biopsy cross-sections (10 μm) were stained with a primary antibody against laminin (ab7463, Rabbit, 1:100) to label the myofiber boundaries together with a cocktail of mouse monoclonal antibodies against MyHC type I (BA-D5c, MIgG2b, 1:100), MyHC type IIX (6H1s, MIgM, 1:10), and all MyHC types except IIX (BF-35c, MIgG1, 1:10). Primary antibody staining was visualized with Alexa Fluor conjugated secondary antibodies (1:500) including Goat Anti-Rabbit Alexa Fluor 350 (to detect laminin), Goat Anti-Mouse IgG2b Alexa Fluor 555 (to detect BA-D5), Goat Anti-Mouse IgG1 Alexa Fluor 488 (to detect BF-35), and Goat Anti-Mouse IgM Alexa Fluor 647 (to detect 6H1). Laminin, BA-D5, BF-35, and 6H1 were pseudo colored white, red, green, and blue, respectively. Scale bars are 100 μm. B-J: Stitched panoramic images of the entire muscle biopsy cross-section were obtained using an automated fluorescent microscope (Echo Revolution) and analyzed using the MuscleBos plugin for FIJI/ImageJ. B: The overall myofiber type profile of pooled bovine longissimus dorsi muscle biopsy samples showing the percentage composition of type I, type IIA, and type IIX myofibers as determined by MuscleBos. C: The mean cross-sectional area (CSA) type I, type IIA, and type IIX myofibers in pooled bovine longissimus dorsi muscle biopsy samples as determined by MuscleBos. D: Comparison of mean myofiber CSA irrespective of myofiber type between bovine longissimus dorsi muscle biopsy samples obtained from HM or LM cows at PRE and POST parturition. E-G: Comparison of the mean CSA of type I (E), type IIA (F), and type IIX (G) myofibers in bovine longissimus dorsi muscle biopsy samples obtained from HM or LM cows at PRE and POST parturition. H-J: Comparison of percentage myofiber type composition of type I (H), type IIA (I), and type IIX (J) myofibers in bovine longissimus dorsi muscle biopsy samples obtained from HM or LM cows PRE and POST parturition. Values are mean ± SEM with dot plots representing data obtained from each individual animal (biological replicates). Symbols denote *P < .05, **P < .01, ***P < .001, and ****P < .0001 between the indicated groups.
MuscleBos analysis of hybrid type IIA/IIX myofibers
The most commonly observed hybrid myofiber type present in bovine muscle has been reported to be IIA/IIX myofibers (Song et al., 2020). Most currently available automated software options for myofiber type analysis do not directly account for the presence of such hybrid IIA/IIX (or IIX/IIB in rodents) subtypes and rather use a purely binary approach to classify myofibers as either IIX (unstained) or not IIX (stained). Manual counting of these types can potentially be subjective and prone to investigator bias. Nevertheless, there may be special cases in which investigators are particularly interested in including such putative hybrid fiber types in their analysis. On this basis, we have incorporated into our latest version of MuscleBos (v1.1) the optional ability to classify non-type I (BA-D5neg) myofibers that stain relatively dim for the BF-35 primary antibody, as type IIA/IIX hybrid myofibers. Based on qualitative observations of images analyzed here, we classified BA-D5neg myofibers with a BF-35 staining intensity with ≥ 0.5 SD from the mean as pure type IIA (BF-35pos), those ≥ 0 SD and ≤ 0.5 SD from the mean fiber intensity as hybrid IIA/IIX (BF-35dim), and those with a staining intensity below the mean as pure type IIX (BF-35neg). Co-staining with the 6H1 (type IIX) primary antibody confirmed that the BA-D5negBF-35dim myofiber population expressed relatively greater amounts of type IIX MyHC when compared with the BF-35pos population, thereby confirming their tentative identity as bona fide hybrid type IIA/IIX myofibers (Figure 10A).
Optional detection of hybrid type IIA/IIX myofibers with MuscleBos: (A) Muscle biopsy samples were obtained from the longissimus dorsi of high muscle (HM) or low muscle (LM) multiparous dairy cows at 21 d before expected calving (PRE) and at 21-d postpartum (POST). Biopsy cross-sections (10 μm) were stained with a primary antibody against laminin (ab7463, Rabbit, 1:100) to label the myofiber boundaries together with a cocktail of mouse monoclonal antibodies against MyHC type I (BA-D5c, MIgG2b, 1:100), MyHC type IIX (6H1s, MIgM, 1:10), and all MyHC types except IIX (BF-35c, MIgG1, 1:10). Primary antibody staining was visualized with Alexa Fluor conjugated secondary antibodies (1:500) including Goat Anti-Rabbit Alexa Fluor 350 (to detect laminin), Goat Anti-Mouse IgG2b Alexa Fluor 555 (to detect BA-D5), Goat Anti-Mouse IgG1 Alexa Fluor 488 (to detect BF-35), and Goat Anti-Mouse IgM Alexa Fluor 647 (to detect 6H1). Laminin, BA-D5, BF-35, and 6H1 were pseudo colored white, red, green, and blue, respectively. Scale bars are 100 μm. An original image and corresponding MuscleBos-generated myofiber type cartography maps with or without hybrid type IIA/IIX fiber detection are shown. B: The overall myofiber type profile of pooled bovine longissimus dorsi muscle biopsy samples showing the percentage composition of type I, type IIA, type IIX, and type IIA/IIX myofibers as determined by MuscleBos. C-D: Comparison of percentage myofiber type composition of pure type IIA (C) and hybrid type IIA/IIX (D) myofibers in bovine longissimus dorsi muscle biopsy samples obtained from HM or LM cows PRE and POST parturition.
Reanalysis of the above LM vs. HM cow dataset revealed that approximately one quarter of the total type IIA population (47.33% of total myofibers) could be classified as hybrid IIA/IIX myofibers (12.37% of total myofibers) using this approach (Figure 10B). Interestingly, this subclass analysis revealed a significant main effect of time, but no effect of group, or group × time interaction for the percentage of pure type IIA myofibers. Both HM and LM cows showed a reduced proportion of pure type IIA myofibers in the postpartum period (Figure 10C). Furthermore, there was a main effect of group, but no effect of time, or time × group interaction effect for the proportion of hybrid IIA/IIX myofibers. HM cows possess a relatively greater percentage of hybrid IIA/IIX myofibers when compared to LM cows, irrespective of time-point. (Figure 10D).
Discussion
The present study established and validated a high-throughput immunofluorescence-based method for characterizing bovine skeletal myofiber type composition and myofiber-type–specific CSA. By combining optimized antibody cocktails with automated fluorescent microscopy and high-content image analysis using the new MuscleBos plugin for FIJI/ImageJ, we demonstrated that reliable, rapid, and automatic identification/quantification of type I, IIA, and IIX myofibers can be achieved on bovine muscle cross-sections. This approach substantially reduces the time and effort required for image analysis while maintaining accuracy comparable to manual annotation, enabling its application in large-scale studies of bovine muscle biology and meat characteristics.
Traditional histochemical classification based on myosin ATPase staining techniques (Brooke and Kaiser 1970) originally identified 3 myofiber types (I, IIA, and IIB) in cattle (Holmes and Ashmore 1972), as well as in humans (Brooke and Kaiser 1970), and horses (Rome et al., 1990; Sosnicki et al., 1989). Some more recent studies employing this pioneering methodology have also continued to report the myosin ATPase acid-labile/alkali-stable population as “type IIB” fibers in cattle, consistent with this classical nomenclature (Bureš et al., 2015). Nevertheless, consistent with our findings, other more recent studies using myofiber type–specific monoclonal primary antibodies have reported that MyHC IIX, in fact, appears to be the only fast-twitch glycolytic isoform that is present in trunk and limb muscles of human (Smerdu et al., 1994), equine (Chikuni et al., 2004a), and bovine species (Chikuni et al., 2004b; Maccatrozzo et al., 2004; Song et al., 2020; Toniolo et al., 2005). The MyHC IIB protein isoform is indeed present in some large mammals, most notably swine (Kim et al., 2013). While functionally absent in bovine limb and trunk muscles, true type IIB myofibers have also been detected in specialized eye muscles of cattle (Maccatrozzo et al., 2004; Toniolo et al., 2005). Some studies have further suggested that the mRNA encoding MyHC type IIB (MYH4) is indeed expressed together with variably detectable protein levels in limb muscles of some specialized breeds of cattle, such as the French beef breed, Blonde d’Aquitaine (Gagaoua et al., 2015; Oury et al., 2010; Picard and Cassar-Malek 2009). On this basis, our method may not be applicable to all bovine muscle types or cattle breeds. Nevertheless, the methods described here should be generally applicable for the analysis of limb and trunk muscles of the most common breeds of dairy and beef cattle. Finally, although not validated here, the MuscleBos workflow should, in theory, also be applicable to the analysis of both human and equine skeletal muscle cross-sections, which, like cattle, express MyHC I, IIA, and IIX (but not IIB) at the protein level, at least in limb and trunk muscles (Chikuni et al., 2004a; Smerdu et al., 1994)
Another key finding of our study is that the widely used SC-71 primary antibody, which in rodents reacts selectively with type IIA myofibers, exhibited broader reactivity in bovine muscle. Specifically, in the bovine longissimus dorsi muscle biopsy samples analyzed here, the SC-71 antibody clearly labeled all type II myofibers, including both IIA and IIX myofibers. Although initially unexpected based on our prior experience, this finding is indeed consistent with several previously published reports in cattle (Cheng et al., 2020; Duris et al., 2000; Kim et al., 2016; Maccatrozzo et al., 2004; Song et al., 2020). In human skeletal muscle, the SC-71 antibody also cross-reacts with IIX myofibers, albeit with notably dimmer staining than type IIA myofibers, thereby potentially allowing for type IIA and IIX myofibers to be visually distinguished based on relative staining intensity (D’Souza et al., 2018). In contrast, we observed no such difference in the staining intensity between type IIA and IIX myofiber populations in bovine muscle. This observation precludes the use of the SC-71 primary antibody to distinguish type IIA and IIX myofibers in bovine skeletal muscle. This species-specific limitation underscores the necessity of robust primary antibody validation in non-rodent samples and highlights the importance of establishing alternative unique antibody combinations to accurately resolve myofiber types in cattle.
We demonstrated that a cocktail of BA-D5 (type I-specific), BF-35 (all but IIX-specific), and 6H1 (IIX-specific) antibodies provides robust differentiation of the 3 major adult bovine myofiber types on a single tissue section. This was expected based on early reports, which already comprehensively characterized the reactivity of each of these primary antibodies in isolation on bovine muscle (Duris et al., 2000; Kim et al., 2016; Maccatrozzo et al., 2004; Toniolo et al., 2005). Indeed, similar primary antibody cocktails as described here have been successfully used by several other groups to enable simultaneous detection of the 3 major adult bovine MyHC isoforms (I, IIA, and IIX) (Cheng et al., 2020; Ebarb et al., 2016; Ebarb et al., 2017; Fuerniss and Johnson, 2023; Fuerniss et al., 2023; Matney et al., 2021; Phelps et al., 2014; Song et al., 2020; Wesley et al., 2025; Zeitz et al., 2018). Importantly, when tested both individually and in combination, we found that BA-D5, BF-35, and 6H1 primary antibodies displayed consistent reactivity, supporting their utility in multiplexed staining protocols. Nevertheless, some cross-reactivity of the type IIX-specific 6H1 primary antibody with type I myofibers was noted in our work. This finding is consistent with the results from a prior study, which also reported cross-reactivity of the 6H1 antibody with type I myofibers in beef cattle (Song et al., 2020). Therefore, it appears to be preferable to characterize bovine type IIX myofibers based on their relative lack of BF-35 signal, either alone or in combination with their positive reactivity with 6H1, rather than based on 6H1 staining alone.
We found that the default version of MuscleJ 1.0.2, which was originally designed and validated for use on mouse muscle samples, could accurately segment myofiber boundaries on bovine muscle biopsy cross-sections stained with either laminin or dystrophin. Moreover, type I myofibers could readily be distinguished from non-type I (e.g., pooled type IIA + IIX) myofibers. However, on the samples analyzed in our work, adjustment of the sensitivity threshold was required to accurately capture all BA-D5+ myofibers that were counted in manual image analysis, emphasizing the potential need for optimization of software parameters for individual laboratory-, species-, and/or antibody-specific contexts. When BF-35 and 6H1 primary antibodies were incorporated into a cocktail together with BA-D5, our new customized bovine-specific version of MuscleJ 1.0.2 that we termed MuscleBos could also successfully automate the classification of type I, IIA, and IIX myofibers, providing powerful quantitative measures of bovine myofiber type distribution and myofiber type–specific CSA.
Validation against expert manual annotation confirmed that MuscleBos delivered highly comparable results. Our initial prototype version of MuscleBos produced similar total myofiber counts, myofiber type proportions, and myofiber CSA values when compared to manual image analysis. However, our initial version, which was based on the default MuscleJ myofiber segmentation code, consistently underestimated type IIX myofiber CSA on our bovine muscle samples. MuscleJ was originally designed to identify type IIX myofibers based on a lack of staining with other primary antibodies. Because of this, it sometimes incorrectly identified spaces between adjacent myofibers as small type IIX myofibers on the samples tested in our work. This limitation may reflect the larger average size of bovine vs. rodent myofibers and/or the looser packing of myofibers in muscle biopsy samples from larger species when compared to complete cross-sections of entire mouse muscle samples in which myofibers are more densely organized. To address this potential limitation, we incorporated into MuscleBos the ability to add user-defined myofiber type–specific CSA exclusion thresholds. Excluding unstained objects (e.g., IIX myofibers) with CSA measurements of ≤ 1,000 μm2 further improved the consistency between MuscleBos and manual image analysis.
When applied to biopsy samples obtained from dairy cows differing in prepartum longissimus dorsi muscle depth, MuscleBos revealed statistically significant differences in both percentage myofiber type composition and myofiber type–specific CSA. Specifically, the HM cows displayed larger CSA of type IIA and IIX (but not type I) myofibers and had a relatively higher proportion of type IIA myofibers during the post-partum period. In contrast, LM cows showed lower type IIA and IIX (but not type I) myofiber CSA and a greater proportion of type IIX myofibers during the postpartum period. We were further able to demonstrate myofiber type–specific atrophy of type IIA myofibers from pre- to post-partum. These data show for the first time that fast oxidative-glycolytic myofibers are preferentially mobilized during the transition period, during which dairy cows lose large amounts of skeletal muscle mass. Overall, these findings demonstrate that automated image analysis can capture physiologically relevant changes in myofiber morphology and can serve as a powerful tool to study the underlying cellular basis of skeletal muscle remodeling in response to adaptation to lactation and/or nutritional changes in dairy cattle.
We based MuscleBos on the MuscleJ 1.0.2 plugin for FIJI/ImageJ (Mayeuf-Louchart et al., 2018). This decision was primarily due to our prior familiarity with MuscleJ and recent success in utilizing this plugin for the automated quantitative analysis of myofiber type in rodent skeletal muscle (Belbis et al., 2025; Markworth et al., 2021; Markworth et al., 2020; Sharma et al., 2025). Nevertheless, the immunofluorescent staining and imaging protocol described here should, in theory, also be compatible with several other freely available muscle-specific image analysis software options (see below for details). The first prerequisite would be the ability to identify staining of various MyHC-specific primary antibodies following segmentation of the myofiber boundaries. This is a feature that is offered by most (Babcock et al., 2020; Bergmeister et al., 2016; Bergmeister et al., 2017; Danckaert et al., 2023; Encarnacion-Rivera et al., 2020; Kastenschmidt et al., 2019; Mayeuf-Louchart et al., 2018; Rahmati and Rashno, 2021; Smith and Barton, 2014; Waisman et al., 2021; Wen et al., 2018), but not all software options (Desgeorges et al., 2019; Gu et al.,2024; Zhang et al.,2025). The second requirement would be the ability to identify myofibers that are double positive for 2 different primary antibodies, e.g., type IIA myofibers as BA-D5pos/BF-35pos cells. Several other software originally designed and validated on mouse samples do allow for the potential detection of hybrid muscle myofiber types expressing 2 MyHC types, most commonly I-IIA myofibers (e.g., BA-D5pos/SC-71pos cells) (Babcock et al., 2020; Bergmeister et al., 2016; Bergmeister et al., 2017; Danckaert et al., 2023; Encarnacion-Rivera et al., 2020; Kastenschmidt et al., 2019; Smith and Barton, 2014; Waisman et al., 2021). Therefore, the staining and imaging methods established here may also be compatible with QuantiMus (Kastenschmidt et al., 2019), SMASH (Smith and Barton, 2014), MyoSight (Babcock et al., 2020), Myosoft (Encarnacion-Rivera et al., 2020), the unnamed ImageJ plugin of Bergmeister et al., 2016 (Bergmeister et al., 2016; Bergmeister et al., 2017), Cellpose + LabelsToRois (Waisman et al., 2021), and MuscleJ2 (Danckaert et al., 2023). However, some degree of customization, optimization, and validation would most likely be required, especially for those that only allow for fully automated analysis without any user input or customization (e.g., Danckaert et al., 2023; Rahmati and Rashno, 2021). On the other hand, the bovine-specific primary antibody cocktails used here would most likely be incompatible with MyoVision (Wen et al., 2018), Open-CSAM (Desgeorges et al., 2019), MyoView (Rahmati and Rashno, 2021), the unnamed FIJI/ImageJ plugin by Reyes-Fernandez (Reyes-Fernandez et al., 2019), MyoV (Gu et al., 2024), and MyoAnalyst (Zhang et al., 2025).
The methodology described here has several limitations. Firstly, the combination of BA-D5, BF-35, and 6H1 primary antibodies did not allow for the potential detection of hybrid type I/IIA myofibers. Hybrid type I-IIA myofibers are rarely observed in bovine muscle (Song et al., 2020). Nevertheless, they could potentially be identified via manual analysis of serial sections stained with a cocktail of BA-D5 and SC-71 if deemed to be of particular interest. Finally, the accuracy of both myofiber type and CSA determination with MuscleBos, like MuscleJ 1.0.2 on which it was based, depends on the quality of antibody staining of the myofiber boundaries, e.g., with laminin, dystrophin, or wheat germ agglutinin. Therefore, dim or uneven staining of the myofiber boundaries may lead to inaccurate myofiber segmentation results, which would compromise the downstream accuracy of both myofiber type proportion and myofiber CSA measurements.
Conclusions
In conclusion, this study provides a validated framework for rapid, high-throughput analysis of bovine muscle myofiber type and morphology by incorporating automated fluorescence microscopy and automated image analysis with MuscleBos that is accurate, efficient, and scalable. The methodological advances described here should facilitate more comprehensive investigations into bovine muscle biology, with implications for animal physiology, growth performance, and meat quality.
Acknowledgments
This work was supported by the U.S. Department of Agriculture (USDA) National Institute of Food and Agriculture (NIFA) (Research Capacity Fund [HATCH Multistate], project award no. 7004451 [NC1184]); the USDA NIFA (grant number 2022-67015-36317) (JPB); Purdue University AgSEED Crossroads funding (JFM); and Purdue University College of Agriculture laboratory startup funding (JFM). We would like to thank S. Haag, M. Simonds, and the Purdue Dairy Farm staff for their assistance with this project. Monoclonal antibodies, including MANDYS1(3B7)s, BA-D5c, SC-71c, 6H1s, and BF-35c, were obtained from the Developmental Studies Hybridoma Bank (DSHB), created by the NICHD of the NIH and maintained at The University of Iowa, Department of Biology, Iowa City, IA 52242.
Data and Code Availability
The MuscleBos software is open-source and freely available. The source code, installation instructions, and user tutorial documentation can be accessed on the GitHub repository: https://github.com/jmarkwor/MuscleBos. The underlying microscopy datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request.
Conflict of Interest
The authors declare that they have no competing interests.
Author Contributions
Hamood Rehman: Investigation, Formal analysis, Methodology, Software, Visualization, Writing - Original draft. Kyrstin M. Gouveia: Investigation, Resources, Writing - Review & Editing. Rebecca K. Coombe: Investigation, Formal analysis, Validation, Writing - Review & Editing, Jacquelyn P. Boerman: Funding acquisition, Project administration, Resources, Supervision, Writing - Review & Editing. J. Alex Pasternak: Software, Methodology, Validation, Supervision, Project administration, Writing - Review & Editing. James F. Markworth: Funding acquisition, Project administration, Supervision, Formal Analysis, Visualization, Validation, Writing – Review & Editing.
Literature Cited
Babcock, L. W., A. D. Hanna, N. H. Agha, and S. L. Hamilton. 2020. Myosight-semi-automated image analysis of skeletal muscle cross sections. Skelet. Muscle 10:33. doi: https://doi.org/10.1186/s13395-020-00250-5
Belbis, M. D., B. Ro, L. E. Schepers, K. H. Kim, Z. Yap, T. A. Schultz, M. D. Tallman, M. F. Killam, S. E. Hobart, O. J. Fafowora, J. N. Bray, M. E. Fister, V. Silva, H. Ur Rehman, B. Sharma, X. Lu, G. P. T. Areas, J. F. Markworth, K. Kim, T. E. Ryan, S. Kuang, C. J. Goergen, T. P. Gavin, B. T. Roseguini, and D. M. Hirai. 2025. Effects of heat therapy on exercise tolerance in a rat model of heart failure with preserved ejection fraction. J. Appl. Physiol. 139:185–205. doi: https://doi.org/10.1152/japplphysiol.00247.2025
Bergmeister, K. D., M. Groger, M. Aman, A. Willensdorfer, K. Manzano-Szalai, S. Salminger, and O. C. Aszmann. 2016. Automated muscle fiber type population analysis with ImageJ of whole rat muscles using rapid myosin heavy chain immunohistochemistry. Muscle Nerve 54:292–299. doi: https://doi.org/10.1002/mus.25033
Bergmeister, K. D., M. Groger, M. Aman, A. Willensdorfer, K. Manzano-Szalai, S. Salminger, and O. C. Aszmann. 2017. A rapid automated protocol for muscle fiber population analysis in rat muscle cross sections using myosin heavy chain immunohistochemistry. J. Vis. Exp. doi: https://doi.org/10.3791/55441
Bloemberg, D. and J. Quadrilatero. 2012. Rapid determination of myosin heavy chain expression in rat, mouse, and human skeletal muscle using multicolor immunofluorescence analysis. PLoS One 7:e35273. doi: https://doi.org/10.1371/journal.pone.0035273
Brooke, M. H. and K. K. Kaiser. 1970. Muscle fiber types: How many and what kind? Arch. Neurol. 23:369–79. doi: https://doi.org/10.1001/archneur.1970.00480280083010
Bureš, D., L. Bartoň, R. Kotrba, and T. Lesiów. 2015. Muscle fiber characteristics of four muscles from different cattle breeds and their relation to meat instrumental toughness. 61st International Congress of Meat Science and Technology (ICoMST), Clermont-Ferrand, France August 23–28 2015.
Calkins, C. R., T. R. Dutson, G. C. Smith, Z. L. Carpenter, and G. W. Davis. 1981. Relationship of fiber type composition to marbling and tenderness of bovine muscle. J. Food. Sci. 46:708–710. doi: http://doi.org/10.1111/j.1365-2621.1981.tb15331.x
Cheng, H., S. Song, E. Y. Jung, J. Y. Jeong, S. T. Joo, and G. D. Kim. 2020. Comparison of beef quality influenced by freeze-thawing among different beef cuts having different muscle fiber characteristics. Meat Sci. 169:108206. doi: https://doi.org/10.1016/j.meatsci.2020.108206
Chikuni, K., S. Muroya, and I. Nakajima. 2004a. Absence of the functional myosin heavy chain 2b isoform in equine skeletal muscles. Zoolog. Sci. 21:589–596. doi: http//doi.org/10.2108/zsj.21.589
Chikuni, K., S. Muroya, and I. Nakajima. 2004b. Myosin heavy chain isoforms expressed in bovine skeletal muscles. Meat Sci. 67:87–94. doi: https://doi.org/10.1016/j.meatsci.2003.09.011
Clancy, M. J., J. M. Lester, and J. F. Roche 1986. The effects of anabolic agents and breed on the fibers of the longissimus muscle of male cattle. J. Anim. Sci. 63:83–91. doi: https://doi.org/10.2527/jas1986.63183x
D’Souza, R. F., N. Zeng, J. F. Markworth, V. C. Figueiredo, L. A. Roberts, T. Raastad, J. S. Coombes, J. M. Peake, D. Cameron-Smith, and C. J. Mitchell. 2018. Divergent effects of cold water immersion versus active recovery on skeletal muscle fiber type and angiogenesis in young men. Am. J. Physiol. Regul. Integr. Comp. Physiol. 314:R824–R833. doi: https//doi.org/10.1152/ajpregu.00421.2017
Danckaert, A., A. Trignol, G. Le Loher, S. Loubens, B. Staels, H. Duez, S. L. Shorte, and A. Mayeuf-Louchart. 2023. Musclej2: A rebuilding of musclej with new features for high-content analysis of skeletal muscle immunofluorescence slides. Skelet. Muscle 13:14. doi: https://doi.org/10.1186/s13395-023-00323-1
Desgeorges, T., S. Liot, S. Lyon, J. Bouviere, A. Kemmel, A. Trignol, D. Rousseau, B. Chapuis, J. Gondin, R. Mounier, B. Chazaud, and G. Juban. 2019. Open-csam, a new tool for semi-automated analysis of myofiber cross-sectional area in regenerating adult skeletal muscle. Skelet. Muscle 9:2. doi: https://doi.org/10.1186/s13395-018-0186-6
Duris, M. P., B. Picard, and Y. Geay. 2000. Specificity of different anti-myosin heavy chain antibodies in bovine muscle. Meat Sci. 55:67–78. doi: https://doi.org/10.1016/s0309-1740(99)00127-8
Ebarb, S. M., J. S. Drouillard, K. R. Maddock-Carlin, K. J. Phelps, M. A. Vaughn, D. D. Burnett, C. L. Van Bibber-Krueger, C. B. Paulk, D. M. Grieger, and J. M. Gonzalez. 2016. Effect of growth-promoting technologies on longissimus lumborum muscle fiber morphometrics, collagen solubility, and cooked meat tenderness. J. Anim. Sci. 94:869–881. doi: https://doi.org/10.2527/jas.2015-9888
Ebarb, S. M., K. J. Phelps, J. S. Drouillard, K. R. Maddock-Carlin, M. A. Vaughn, D. D. Burnett, J. A. Noel, C. L. Van Bibber-Krueger, C. B. Paulk, D. M. Grieger, and J. M. Gonzalez. 2017. Effects of anabolic implants and ractopamine-HCl on muscle fiber morphometrics, collagen solubility, and tenderness of beef longissimus lumborum steaks. J. Anim. Sci. 95:1219–1231. doi: https://doi.org/10.2527/jas.2016.1263
Encarnacion-Rivera, L., S. Foltz, H. C. Hartzell, and H. Choo. 2020. Myosoft: An automated muscle histology analysis tool using machine learning algorithm utilizing Fiji/ImageJ software. PLoS One 15:e0229041. doi: https://doi.org/10.1371/journal.pone.0229041
Fuerniss, L. K. and B. J. Johnson. 2023. Semi-automated technique for bovine skeletal muscle fiber cross-sectional area and myosin heavy chain determination. J. Anim. Sci. 101. https://do.org/10.1093/jas/skad205https://do.org/10.1093/jas/skad205
Fuerniss, L. K., J. D. Young, J. R. Hall, K. R. Wesley, S. M. Bowman, L. D. Felizari, D. R. Woerner, R. J. Rathmann, and B. J. Johnson. 2023. Body, carcass, and steak dimensions of straightbred Holstein calves and Angus-sired calves from Holstein, Jersey, and crossbred beef dams. J. Anim. Sci. 101. https://org/10.1093/jas/skad358https://org/10.1093/jas/skad358
Gagaoua, M., E. M. Terlouw, A. Boudjellal, and B. Picard. 2015. Coherent correlation networks among protein biomarkers of beef tenderness: What they reveal. J. Proteomics 128:365–74. doi: https://doi.org/10.1016/j.jprot.2015.08.022
Gonzalez, J. M., S. E. Johnson, T. A. Thrift, J. D. Savell, S. E. Ouellette, and D. D. Johnson. 2009. Effect of ractopamine-hydrochloride on the fiber type distribution and shelf-life of six muscles of steers. J. Anim. Sci. 87:1764–1771. doi: https://doi.org/10.2527/jas.2008-1469
Gouveia, K. M., L. M. Beckett, T. M. Casey, and J. P. Boerman 2024. Production responses of multiparous dairy cattle with differing prepartum muscle reserves and supplementation of branched-chain volatile fatty acids. J. Dairy Sci. 107:11655–11668. doi: https://doi.org/10.3168/jds.2024-24915
Greenwood, P. L., N. W. Tomkins, R. A. Hunter, P. G. Allingham, S. Harden, and G. S. Harper. 2009. Bovine myofiber characteristics are influenced by postweaning nutrition. J. Anim. Sci. 87:3114–23. doi: https://doi.org/10.2527/jas.2009-1936
Gu, S., C. Wen, Z. Xiao, Q. Huang, Z. Jiang, H. Liu, J. Gao, J. Li, C. Sun, and N. Yang. 2024. Myov: A deep learning-based tool for the automated quantification of muscle fibers. Brief Bioinform. 25. doi: https://doi.org/10.1093/bib/bbad528
Hoh, J. F. Y. 2023. Developmental, physiologic and phylogenetic perspectives on the expression and regulation of myosin heavy chains in mammalian skeletal muscles. J. Comp. Physiol. B. 193:355–382 doi: https://doi.org/10.1007/s00360-023-01499-0
Holmes, J. H. and C. R. Ashmore. 1972. A histochemical study of development of muscle fiber type and size in normal and “double muscled” cattle. Growth 36:351–72 https://www.ncbi.nlm.nih.gov/pubmed/4653615https://www.ncbi.nlm.nih.gov/pubmed/4653615
Hwang, Y. H., G. D. Kim, J. Y. Jeong, S. J. Hur, and S. T. Joo. 2010. The relationship between muscle fiber characteristics and meat quality traits of highly marbled hanwoo (Korean native cattle) steers. Meat Sci. 86:456–61. doi: https://doi.org/10.1016/j.meatsci.2010.05.034
Johnston, D. M., W. G. Moody, J. A. Boling, and N. W. Bradley. 1981. Influence of breed type, sex, feeding systems, and muscle bundle size on bovine fiber type characteristics. J. Food Sci. 46:1760–1765. doi: https://doi.org/10.1111/j.1365-2621.1981.tb04480.x.
Joo, S. H., K. W. Lee, Y. H. Hwang, and S. T. Joo. 2017. Histochemical characteristics in relation to meat quality traits of eight major muscles from Hanwoo steers. Korean J. Food Sci. Anim. Resour. 37:716–725. doi: https://doi.org/10.5851/kosfa.2017.37.5.716
Jurie, C., B. Picard, and Y. Geay. 1998. Influences of the method of housing bulls on their body composition and muscle fibre types. Meat Sci. 50:457–469. doi: https://doi.org/10.1016/s0309-1740(98)00059-x
Jurie, C., B. Picard, and Y. Geay. 1999. Changes in the metabolic and contractile characteristics of muscle in male cattle between 10 and 16 months of age. Histochem. J. 31:117–122 doi: https://doi.org/10.1023/a:1003589320910
Kastenschmidt, J. M., K. L. Ellefsen, A. H. Mannaa, J. J. Giebel, R. Yahia, R. E. Ayer, P. Pham, R. Rios, S. A. Vetrone, T. Mozaffar, and S. A. Villalta. 2019. Quantimus: A machine learning-based approach for high precision analysis of skeletal muscle morphology. Front. Physiol. 10:1416. doi: https://doi.org/10.3389/fphys.2019.01416
Kellermeier, J. D., A. W. Tittor, J. C. Brooks, M. L. Galyean, D. A. Yates, J. P. Hutcheson, W. T. Nichols, M. N. Streeter, B. J. Johnson, and M. F. Miller. 2009. Effects of zilpaterol hydrochloride with or without an estrogen-trenbolone acetate terminal implant on carcass traits, retail cutout, tenderness, and muscle fiber diameter in finishing steers. J. Anim. Sci. 87:3702–3711. doi: https://doi.org/10.2527/jas.2009-1823
Kim, G. D., J. Y. Jeong, H. S. Yang, and S. T. Joo. 2013. Identification of myosin heavy chain isoforms in porcine longissimus dorsi muscle by electrophoresis and mass spectrometry. Electrophoresis 34:1255–1261. doi: https://doi.org/10.1002/elps.201300049
Kim, G. D., H. S. Yang, and J. Y. Jeong. 2016. Comparison of characteristics of myosin heavy chain-based fiber and meat quality among four bovine skeletal muscles. Korean J. Food Sci. Anim. Resour. 36:819–828. doi: https://doi.org/10.5851/kosfa.2016.36.6.819
Kirchofer, K. S., C. B. Calkins, and B. L. Gwartney. 2002. Fiber-type composition of muscles of the beef chuck and round. J. Anim. Sci. 80:2872–2878. doi: https://doi.org/10.2527/2002.80112872x
Komiya, Y., W. Mizunoya, K. Kajiwara, I. Yokoyama, H. Ogasawara, and K. Arihara. 2020. Correlation between skeletal muscle fiber type and responses of a taste sensing system in various beef samples. Anim. Sci. J. 91:e13425. doi: https://doi.org/10.1111/asj.13425
Lang, Y., S. Zhang, P. Xie, X. Yang, B. Sun, and H. Yang. 2020. Muscle fiber characteristics and postmortem quality of longissimus thoracis, psoas major and semitendinosus from Chinese Simmental bulls. Food. Sci. Nutr. 8:6083–6094. doi: https://doi.org/10.1002/fsn3.1898
Li, Z., M. Ha, D. Frank, M. Hastie, and R. D. Warner. 2023. Muscle fibre type composition influences the formation of odour-active volatiles in beef. Food Res. Int. 165:112468. doi: https://doi.org/10.1016/j.foodres.2023.112468
Maccatrozzo, L., M. Patruno, L. Toniolo, C. Reggiani, and F. Mascarello. 2004. Myosin heavy chain 2b isoform is expressed in specialized eye muscles but not in trunk and limb muscles of cattle. Eur. J. Histochem. 48:357–366. https://www.ncbi.nlm.nih.gov/pubmed/15718201https://www.ncbi.nlm.nih.gov/pubmed/15718201
Markworth, J. F., L. A. Brown, E. Lim, J. A. Castor-Macias, J. Larouche, P. C. D. Macpherson, C. Davis, C. A. Aguilar, K. R. Maddipati, and S. V. Brooks 2021. Metabolipidomic profiling reveals an age-related deficiency of skeletal muscle pro-resolving mediators that contributes to maladaptive tissue remodeling. Aging Cell. 20:e13393. doi: https://doi.org/10.1111/acel.13393
Markworth, J. F., L. A. Brown, E. Lim, C. Floyd, J. Larouche, J. A. Castor-Macias, K. B. Sugg, D. C. Sarver, P. C. Macpherson, C. Davis, C. A. Aguilar, K. R. Maddipati, and S. V. Brooks. 2020. Resolvin d1 supports skeletal myofiber regeneration via actions on myeloid and muscle stem cells. JCI Insight 5. doi: https://doi.org/10.1172/jci.insight.137713
Markworth, J. F., B. Durainayagam, V. C. Figueiredo, K. Liu, J. Guan, A. K. H. MacGibbon, B. Y. Fong, A. C. Fanning, A. Rowan, P. McJarrow, and D. Cameron-Smith 2017. Dietary supplementation with bovine-derived milk fat globule membrane lipids promotes neuromuscular development in growing rats. Nutr Metab (Lond) 14:9 https://doi.or/10.1186/s12986-017-0161-y.https://doi.or/10.1186/s12986-017-0161-y
Matney, M. J., M. E. Gravely, T. G. O’Quinn, J. S. Drouillard, K. J. Phelps-Ronningen, T. A. Houser, A. W. Hobson, H. M. Alcocer, and J. M. Gonzalez. 2021. Effects of extended postmortem aging and intramuscular location on protein degradation, muscle fiber morphometrics, and tenderness of beef longissimus lumborum and semitendinosus steaks. J. Anim. Sci. 99. doi: https://doi.org/10.1093/jas/skab263
Mayeuf-Louchart, A., D. Hardy, Q. Thorel, P. Roux, L. Gueniot, D. Briand, A. Mazeraud, A. Bougle, S. L. Shorte, B. Staels, F. Chretien, H. Duez, and A. Danckaert 2018. Musclej: A high-content analysis method to study skeletal muscle with a new Fiji tool. Skelet. Muscle 8:25. doi: https://doi.org/10.1186/s13395-018-0171-0
Meunier, B., B. Picard, T. Astruc, and R. Labas. 2010. Development of image analysis tool for the classification of muscle fibre type using immunohistochemical staining. Histochem. Cell Biol. 134:307–317. doi: https://doi.org/10.1007/s00418-010-0733-7
Oury, M. P., R. Dumont, C. Jurie, J. F. Hocquette, and B. Picard. 2010. Specific fibre composition and metabolism of the rectus abdominis muscle of bovine Charolais cattle. BMC Biochem. 11:12. doi: https://doi.org/10.1186/1471-2091-11-12
Ozawa, S., T. Mitsuhashi, M. Mitsumoto, S. Matsumoto, N. Itoh, K. Itagaki, Y. Kohno, and T. Dohgo. 2000. The characteristics of muscle fiber types of longissimus thoracis muscle and their influences on the quantity and quality of meat from japanese black steers. Meat Sci. 54:65–70. doi: https://doi.org/10.1016/s0309-1740(99)00072-8
Phelps, K. J., J. S. Drouillard, J. S. Jennings, B. E. Depenbusch, C. L. Van Bibber-Krueger, K. A. Miller, M. A. Vaughn, D. D. Burnett, S. M. Ebarb, T. A. Houser, S. E. Johnson, and J. M. Gonzalez. 2014. Effects of the programmed nutrition beef program on meat quality characteristics. J. Anim. Sci. 92:1780–1791. doi: https://doi.org/10.2527/jas.2013-7231
Picard, B. and I. Cassar-Malek. 2009. Evidence for expression of IIB myosin heavy chain isoform in some skeletal muscles of blonde d’Aquitaine bulls. Meat Sci. 82:30–36. doi: https://doi.org/10.1016/j.meatsci.2008.11.022
Picard, B. and M. Gagaoua. 2020. Muscle fiber properties in cattle and their relationships with meat qualities: An overview. J. Agr. Food Chem. 68:6021–6039. doi: https://doi.org/10.1021/acs.jafc.0c02086
Picard, B., C. Jurie, M. P. Duris, and G. Renand. 2006. Consequences of selection for higher growth rate on muscle fibre development in cattle. Livest. Sci. 102:107–120. doi: https://doi.org/10.1016/j.livsci.2005.12.001
Picard, B., J. Robelin, and Y. Geay. 1995. Influence of castration and postnatal energy restriction on the contractile and metabolic characteristics of bovine muscle. Annales De Zootechnie 44:347–357. doi: https://doi.org/10.1051/animres:19950402
Rahmati, M. and A. Rashno. 2021. Automated image segmentation method to analyse skeletal muscle cross section in exercise-induced regenerating myofibers. Sci. Rep. 11:21327. doi: https://doi.org/10.1038/s41598-021-00886-3
Reyes-Fernandez, P. C., B. Periou, X. Decrouy, F. Relaix, and F. J. Authier. 2019. Automated image-analysis method for the quantification of fiber morphometry and fiber type population in human skeletal muscle. Skelet. Muscle 9:15. doi: https://doi.org/10.1186/s13395-019-0200-7
Rome, L. C., A. A. Sosnicki, and D. O. Goble. 1990. Maximum velocity of shortening of three fibre types from horse soleus muscle: Implications for scaling with body size. J. Physiol. 431:173–185. doi: https://doi.org/10.1113/jphysiol.1990.sp018325
Sawano, S. and W. Mizunoya. 2022. History and development of staining methods for skeletal muscle fiber types. Histol. Histopathol. 37:493–503. doi: https://doi.org/10.14670/HH-18-422
Schiaffino, S. 2018. Muscle fiber type diversity revealed by anti-myosin heavy chain antibodies. FEBS J. 285:3688–3694. doi: https://doi.org/10.1111/febs.14502
Schiaffino, S. and C. Reggiani. 1994. Myosin isoforms in mammalian skeletal muscle. J. Appl. Physiol. (1985). 77:493–501. doi: https://doi.org/10.1152/jappl.1994.77.2.493
Seideman, S. C. and J. D. Crouse. 1986. The effects of sex condition, genotype and diet on bovine muscle fiber characteristics. Meat Sci. 17:55–72. doi: https://doi.org/10.1016/0309-1740(86)90083-5
Sharma, B., X. Lu, H. Rehman, V. C. Figueiredo, C. Davis, H. Van Remmen, S. Kuang, S. V. Brooks, K. R. Maddipati, and J. F. Markworth. 2025. Leukocyte-type 12/15-lipoxygenase is essential for timely inflammation-resolution and effective tissue regeneration following skeletal muscle injury. bioRxiv. doi: https://doi.org/10.1101/2025.05.13.653766
Smerdu, V., I. Karsch-Mizrachi, M. Campione, L. Leinwand, and S. Schiaffino. 1994. Type IIX myosin heavy chain transcripts are expressed in type IIB fibers of human skeletal muscle. Am. J. Physiol. 267:C1723–C1728. doi: https://doi.org/10.1152/ajpcell.1994.267.6.C1723
Smith, L. R. and E. R. Barton. 2014. Smash - semi-automatic muscle analysis using segmentation of histology: A MATLAB application. Skelet. Muscle 4:21. doi: https://doi.org/10.1186/2044-5040-4-21
Song, S., C. H. Ahn, and G. D. Kim. 2020. Muscle fiber typing in bovine and porcine skeletal muscles using immunofluorescence with monoclonal antibodies specific to myosin heavy chain isoforms. Food Sci. Anim. Res. 40:132–144. doi: https://doi.org/10.5851/kosfa.2019.e97
Sosnicki, A. A., G. J. Lutz, L. C. Rome, and D. O. Goble. 1989. Histochemical and molecular determination of fiber types in chemically skinned single equine skeletal muscle fibers. J. Histochem. Cytochem. 37:1731–1738. doi: https://doi.org/10.1177/37.11.2530270
Toniolo, L., L. Maccatrozzo, M. Patruno, F. Caliaro, F. Mascarello, and C. Reggiani. 2005. Expression of eight distinct MHC isoforms in bovine striated muscles: Evidence for MHC-2 b presence only in extraocular muscles. J. Exp. Biol. 208:4243–4253. doi: https://doi.org/10.1242/jeb.01904.
Vestergaard, M., P. Henckel, N. Oksbjerg, and K. Sejrsen. 1994. The effect of cimaterol on muscle fiber characteristics, capillary supply, and metabolic potentials of longissimus and semitendinosus muscles from young friesian bulls. J. Anim. Sci. 72:2298–2306. doi: https://doi.org/10.2527/1994.7292298x
Waisman, A., A. M. Norris, M. Elias Costa, and D. Kopinke. 2021. Automatic and unbiased segmentation and quantification of myofibers in skeletal muscle. Sci. Rep. 11:11793. doi: https://doi.org/10.1038/s41598-021-91191-6
Wegner, J., E. Albrecht, I. Fiedler, F. Teuscher, H. J. Papstein, and K. Ender. 2000. Growth- and breed-related changes of muscle fiber characteristics in cattle. J. Anim. Sci. 78:1485–1496. doi: https://doi.org/10.2527/2000.7861485x
Wen, Y., K. A. Murach, I. J. Vechetti, Jr., C. S. Fry, C. Vickery, C. A. Peterson, J. J. McCarthy, and K. S. Campbell. 2018. Myovision: Software for automated high-content analysis of skeletal muscle immunohistochemistry. J. Appl. Physiol. (1985). 124:40–51. doi: https://doi.org/10.1152/japplphysiol.00762.2017
Wesley, K. R., L. K. Fuerniss, J. R. Hall, J. D. Young, F. B. Green, P. N. Smith, J. P. Hutcheson, and B. J. Johnson. 2025. Effects of implant status and breed type on feedlot performance, carcass characteristics, sera metabolites, and immunohistochemical responses in finishing steers. J. Anim. Sci. 103. doi: https://doi.org/10.1093/jas/skaf003
Zeitz, J. O., A. Weber, E. Most, W. Windisch, C. Bolduan, J. Geyer, F. J. Romberg, C. Koch, and K. Eder. 2018. Effects of supplementing rumen-protected niacin on fiber composition and metabolism of skeletal muscle in dairy cows during early lactation. J. Dairy. Sci. 101:8004–8020. doi: https://doi.org/10.3168/jds.2018-14490
Zhang, B., S. Wang, Y. Wang, C. Liang, and H. Zhang. 2025. Myoanalyst: An ImageJ plugin for accurate and automatic myofiber segmentation and analysis in skeletal muscle cross sections. Histochem. Cell. Biol. 163:89. doi: https://doi.org/10.1007/s00418-025-02414-0
Zhang, M., Y. L. Liu, C. Y. Fu, J. Wang, S. Y. Chen, J. Yao, and S. J. Lai. 2014. Expression of myhc genes, composition of muscle fiber type and their association with intramuscular fat, tenderness in skeletal muscle of Simmental hybrids. Mol. Biol. Rep. 41:833–840. doi: https://doi.org/10.1007/s11033-013-2923-6