Skip to main content
NCBI home page
Journal of Animal Science logoLink to Journal of Animal Science
. 2023 Jun 20;101:skad205. doi: 10.1093/jas/skad205

Semi-automated technique for bovine skeletal muscle fiber cross-sectional area and myosin heavy chain determination

Luke K Fuerniss 1, Bradley J Johnson 2,✉
PMCID: PMC10294558  PMID: 37338173

Abstract

Myosin heavy chain (MyHC) type and muscle fiber size are informative but time-consuming variables of interest for livestock growth, muscle biology, and meat science. The objective of this study was to validate a semi-automated protocol for determining MyHC type and size of muscle fibers. Muscle fibers obtained from the longissimus and semitendinosus of fed beef carcasses were embedded and frozen within 45 min of harvest. Immunohistochemistry was used to distinguish MyHC type I, IIA, and IIX proteins, dystrophin, and nuclei in transverse sections of frozen muscle samples. Stained muscle cross sections were imaged and analyzed using two workflows: 1) Nikon workflow which used Nikon Eclipse inverted microscope and NIS Elements software and 2) Cytation5 workflow consisting of Agilent BioTek Cytation5 imaging reader and Gen5 software. With the Cytation5 workflow, approximately six times more muscle fibers were evaluated compared to the Nikon workflow within both the longissimus (P < 0.01; 768 vs. 129 fibers evaluated) and semitendinosus (P < 0.01; 593 vs. 96 fibers evaluated). Combined imaging and analysis took approximately 1 h per sample with the Nikon workflow and 10 min with the Cytation5 workflow. When muscle fibers were evaluated by the objective thresholds of the Cytation5 workflow, a greater proportion of fibers were classified as glycolytic MyHC types, regardless of muscle (P < 0.01). Overall mean myofiber cross-sectional area was 14% smaller (P < 0.01; 3,248 vs. 3,780) when determined by Cytation5 workflow than when determined by Nikon workflow. Regardless, Pearson correlation of mean muscle fiber cross-sectional areas determined by Nikon and Cytation5 workflows was 0.73 (P < 0.01). In both workflows cross-sectional area of MyHC type I fibers was the smallest and area of MyHC type IIX fibers was the largest. These results validated the Cytation5 workflow as an efficient and biologically relevant tool to expedite data capture of muscle fiber characteristics while using objective thresholds for muscle fiber classification.

Keywords: immunohistochemistry, muscle fiber, MyHC, skeletal muscle

A semi-automated technique for evaluating muscle fiber size and myosin heavy chain type can image and analyze six times more myofibers in one-sixth of the time compared to the manual workflow with similar biological results.

Introduction

Skeletal muscle tissue is the most relevant tissue of the beef carcass for protein production and its antemortem growth has implications for efficiency of production (Thornton, 2019). Divergence in muscle fiber properties like contractile speed, metabolic substrates, oxygen availability, and size of muscle fibers are related to myosin heavy chain isoforms (MyHC) and affect meat quality and growth efficiency (Picard et al., 2020). Early investigation of muscle fiber properties focused on color (red or pale white), contractile speed, and sarcoplasmic volume (Denny-Brown, 1929). When Engel and Irwin (1967) identified that lesser myofibrillar ATPase activity was associated with tonic, slow-contracting muscle fibers of transverse fresh-frozen muscle samples. Muscle fibers were then grouped as slow or fast twitch (Guth and Samaha, 1969). Myosin heavy chain type I fibers were considered the slow-twitch fibers, and MyHC type II fibers were considered fast twitch fibers, being further separated into MyHC types IIA and IIB based on sensitivity to pH changes (Brooke and Kaiser, 1970). With the development of immunohistochemical techniques, an additional MyHC—MyHC type IIX—was identified as an intermediate to the more glycolytic MyHC type IIB and more oxidative MyHC type IIA (Schiaffino et al., 1989; Gorza, 1990). Most muscle fibers express only one MyHC isoform, but muscle fibers expressing more than one MyHC exist and are referred to as “hybrid” muscle fibers (Pette and Staron, 2000; Song et al., 2020). Hybrid muscle fibers could represent fibers transitioning from expression of one MyHC type to another (Pierobon-Bormioli et al., 1981). Differences in MyHC types and myofiber cross-sectional areas are observed between cattle of differing ages and breed types (Albrecht et al., 2006) and as growth technologies are applied (Gonzalez et al., 2007; Ebarb et al., 2017; Hergenreder et al., 2017). Myofiber MyHC type and size remain variables of interest for livestock growth, muscle biology, and meat science (Wellmann et al., 2021; Foraker et al., 2022). While immunohistochemistry is widely used for determination of myofiber MyHC type and size, manually evaluating enough muscle fibers is time consuming and prone to the subjectivity of human judgement. Gel electrophoresis has been validated to rapidly evaluate relative abundance of MyHC proteins (Scheffler et al., 2018), but the method does not enable evaluation of individual muscle fiber frequency and size. The objective of this study was to validate a semi-automated, objective immunohistochemical protocol for determining MyHC type and size of muscle fibers.

Materials and Methods

Animal sampling

Preharvest cattle management was in accordance with “The Guide for Care and Use of Agricultural Animals in Research and Teaching”. Since all cattle were harvested in a commercial processing facility and muscle samples were collected postmortem, no IACUC approval was sought for this research.

Muscle sample embedding

Samples of longissimus (N = 49) and semitendinosus (N = 46) muscles were collected from carcasses described by Waller et al. (2023). All cattle were the progeny of crossbred beef bulls (Limousin and Angus) and crossbred dairy cows (Holstein and Jersey). Steers were fed at a commercial feedlot and harvested at the University of Arizona Food Products and Safety Laboratory. Within 45 min of harvest, muscle samples approximately 1 cm in diameter and 2 cm long were collected. Longissimus samples were collected at the third lumbar vertebrae 5 cm lateral to the spinous process, and semitendinosus samples were collected 5 cm distal to the ischiatic tuberosity. Muscle samples were embedded in optimum cutting temperature media (VWR International, Radnor, PA), frozen in supercooled 2-methylbutane, packed in dry ice for transport, and eventually stored at −80 °C.

Immunohistochemistry

Embedded samples were equilibrated to −20 °C overnight before 10 μm-thick cross sections were cut using a Leica CM1950 cryostat (Leica Biosystems, Buffalo Grove, IL) with the chamber temperature at −21 °C. Five cross sections of muscle were placed on positively charged glass slides (Superfrost Plus, VWR International). Muscle cross sections were immunohistochemically stained similar to methods described by (Gonzalez et al., 2007; Paulk et al., 2014; Hergenreder et al., 2016). Cross sections were fixed in 4% paraformaldehyde phosphate buffered saline (PBS, pH 7.4) for 10 min at room temperature. After fixation, slides were washed three times for 5 min each in PBS. Blocking solution consisting of 0.2% Triton-100X, 2% bovine serum albumin, and 5% horse serum in PBS was applied to slides, incubated for 30 min at room temperature, and then removed. Primary antibodies (Table 1) were diluted in blocking solution, applied to cross sections, and incubated for 1 h at room temperature. Since myosin heavy chain (MyHC) type IIB fibers are not present in bovine skeletal muscle (Maccatrozzo et al., 2009), only MyHC type I, IIA, and IIX were considered. Primary antibody solution was removed, and slides were washed three times for 5 min each in PBS. Without exposure to light, secondary antibodies (Table 2) were diluted in blocking solution, applied to cross sections, and incubated for 30 min at room temperature. Secondary antibody solution was removed, and slides were washed three times for 5 min each in PBS. ProLong Gold Antifade Mountant with DAPI (Life Technologies, Carlsbad, CA) was applied to cross sections before covering with thin glass coverslips (VWR International) and incubating for 24 h at 4 °C in the dark. Coverslips were sealed after 24 h of incubation in preparation for imaging.

Table 1.

Primary antibodies

Antibody Target antigen Isotype Host Source Dilution
BA-D5 MyHC type I
(MYH7 gene)		 IgG2b Mouse DSHB1 1:100
BF-35 All MyHC except type IIX
(MYH2 gene)		 IgG1 Mouse DSHB1 1:75
PA137587 Dystrophin
(DMD gene)		 IgG Rabbit Invitrogen2 1:100
Open in a new tab

1Developmental Studies Hybridoma Bank, Iowa Ave, Iowa City.

2Invitrogen, Carlsbad, CA.

Table 2.

Secondary antibodies

Antibody Reactivity Target isotype Conjugate Host Source1 Dilution
A-21143 Mouse IgG2B Alexa-Fluor 546 Goat Invitrogen 1:1000
A-21126 Mouse IgG1 Alexa-Fluor 633 Goat Invitrogen 1:1000
A-11008 Rabbit IgG Alexa-Fluor 488 Goat Invitrogen 1:1000
Open in a new tab

1Invitrogen, Carlsbad, CA.

Manual imaging and image analysis: Nikon workflow

Slides were imaged using 200 × magnification with an inverted fluorescence microscope (Nikon Eclipse, Ti-E; Nikon Instruments, Inc., Mellville, NY) with a UV light source (Intensilight C-HGFIE; Nikon Instruments, Inc.) and a CoolSnapES2 monochrome camera (Photometrics, Tucson, AZ) as previously described (Hosford et al., 2015; Smith et al., 2019; Baggerman et al., 2021). Five images (approximately 440 μm × 330 μm) were taken per sample, one from each of the five muscle cross sections. Location of images was manually selected and focused by hand. Images of each selected area were captured using brightfield and fluorescent channels with filters optimized for DAPI, FITC, TRITC, and Texas Red. NIS Elements Imaging software (Nikon Instruments, Inc.) was used to recolor images (­Figure 1A); blue was used for nuclei (DAPI), green was used for dystrophin (Alexa-Fluor 488), red was used for MyHC type I (Alexa-Fluor 546), and yellow was used for MyHC type IIA (Alexa-Fluor 633 in absence of Alexa-Fluor 546). The intensity range and gamma parameters (brightness and contrast) were manually adjusted to visually identify MyHC types. The image was annotated to indicate the MyHC type for each individual muscle fiber. Muscle fibers with membrane proteins not completely visible (cut off from field of view) were not considered. Each fiber was counted, and its area in square μm was recorded after manually tracing the dystrophin surrounding the muscle fiber. Data from all five images were aggregated per sample.

Figure 1.

Figure 1.

Open in a new tab

Example muscle fiber cross sections visualized with Nikon Eclipse with NIS Elements software (A) or with Cytation5 with Gen5 software (B).

Automated imaging and image analysis: Agilent/BioTek workflow

The same slides were imaged using BioTek Cytation5 with Gen5 software (Agilent, Santa Clara, CA). The automated protocol is summarized in Supplementary File S1. Generally, the center region of the slide (approximately 1.18 cm by 1.96 cm) was imaged using brightfield microscopy and 40 × magnification. From the image, two regions of interest were defined on a single muscle cross section. Each region of interest was approximately 1,000 μm × 1,000 μm. Because 40 × magnification and 200 × magnification yielded biologically similar results with the Cytation5 workflow (Supplementary File S2), fluorescent images were captured with 40 × magnification to improve efficiency of image capture. The image was autofocused using the GFP channel; images were captured with filters optimized for GFP, Texas Red, TRITC, and DAPI and recolored as described above (Figure 1B). Then, a data reduction step was used to recognize muscle fibers based on GFP fluorescence of dystrophin staining. Muscle fibers were defined as the absence of GFP fluorescence. Within each fiber, average fluorescent intensities of TRITC and Texas Red were used to classify each fiber’s MyHC type (Figure 2) by the objective standards in Table 3. Fiber counts and areas were aggregated across both regions of interest for each sample.

Figure 2.

Figure 2.

Open in a new tab

Muscle fibers identified by MyHC type by Cytation5 with Gen5 software. Layers include raw image (A), all recognized muscle fibers (B), muscle fibers classified as MyHC type I (C), muscle fibers classified as MyHC type IIA (D), muscle fibers classified as MyHC type IIX (E), and all recognized muscle fibers colored by MyHC type (F).

Table 3.

Cytation5 classification standards for each MyHC type

MyHC type Average fiber TRITC intensity Average fiber Texas red intensity
I ≥7,500 ≥22,800
IIA <7,500 ≥22,800
IIX <7,500 <22,800
Open in a new tab

Statistical analysis

The number of muscle fibers counted per sample were tested with linear models were fit using the lme4 package of R (Bates et al., 2015). The linear model included main effects of muscle (longissimus or semitendinosus) and method (Nikon or Cytation5 workflow), interaction of muscle and method, and carcass as a random effect to account for lack of independence. Mean fiber area was compared using linear models with fiber area determined by the Nikon workflow as the independent variable and fiber area determined by the Cytation5 workflow as the dependent variable; relationship between the two variables was summarized by Pearson correlation. Proportion of muscle fibers classified as each MyHC type was analyzed within muscle by ordinal logistic regression using the ordinal package (Christensen, 2022) and RVAideMemoire packages of R (Hervé, 2022). When a significant effect of method on the MyHC type distribution was detected, the proportions of fibers identified by each method was compared within MyHC type by logistic regression. Testing was completed on the log odds scale and back transformed to be presented as proportions. Estimated marginal means were calculated using the emmeans package of R (Lenth, 2022). All statistical analysis was performed in R version 4.2.1 (R Core Team, 2022). Statistical significance was evaluated compared to α of 0.05. Visualizations were also built in R using the ggplot2 package (Wickham, 2016), cowplot package (Wilke, 2020), and ggpubr package (Kassambara, 2020).

Results

Muscle fiber counts

An interaction for the total number of muscle fibers observed per sample was observed between the method of imaging (Nikon or Cytation5 workflows) and the muscle (longissimus or semitendinosus) from which the sample originated (P < 0.01; Figure 3). Total cross-sectional muscle area imaged with the Nikon workflow was approximately 0.73 mm2; total cross-sectional muscle area imaged with the Cytation5 workflow was approximately 2.00 mm2. Within both muscles, the Nikon workflow (compared to the Cytation5 workflow) observed and recorded data on a greater number of muscle fibers (P < 0.05; Figure 3). Among longissimus samples, the Cyation5 workflow identified 5.95 times as many muscle fibers as the Nikon workflow (768 vs. 129 fibers counted). Among semitendinosus samples, the Cyation5 workflow identified 6.18 times as many muscle fibers as the Nikon workflow (593 vs. 96 fibers counted).

Figure 3.

Figure 3.

Open in a new tab

Muscle fiber counts per sample for cross sections of the longissimus (LD) and semitendinosus (ST) as imaged and analyzed by Nikon Eclipse with NIS Elements software (Nikon) or by Cytation5 with Gen5 software (Cytation5).

Classification of myosin heavy chain type

Relative abundance of each MyHC type was analyzed by ordinal logistic regression.

A significant interaction between method and muscle was observed for the distribution of MyHC types (P < 0.01). Within both the longissimus and semitendinosus muscles, method of imaging affected the distribution of MyHC types (P < 0.01; Figure 4). Imaging with the Cytation5 workflow caused the distribution of MyHC types to be observed as more glycolytic. Within the longissimus, fewer muscle fibers were classified as MyHC type I (P < 0.05), and more fibers were classified as MyHC type IIX (P < 0.05) when the Cytation5 workflow was used. Within the semitendinosus, fewer muscle fibers were classified as MyHC type I (P < 0.05), fewer muscle fibers were classified as MyHC type IIA (P < 0.05), and more fibers were classified as MyHC type IIX (P < 0.05) when the Cytation5 workflow was used.

Figure 4.

Figure 4.

Open in a new tab

Percentage of muscle fibers classified as each MyHC type. P-value presented represents the test of the main effect of analysis method (Nikon Eclipse with NIS Elements software, Nikon; or by Cytation5 with Gen5 software, Cytation5) on the distribution of muscle fiber types when tested by ordinal logistic regression. Analysis was completed separately for longissimus (LD) and semitendinosus (ST) muscles.

Muscle fiber cross-sectional area

No interaction was observed between method of analysis and muscle for overall mean muscle fiber cross-sectional area (P = 0.46). On average, samples evaluated by the Cyation5 workflow had 14% smaller (P < 0.01) mean muscle fiber cross-sectional area than when evaluated by the Nikon workflow (3,248 vs. 3,780; standard error = 89.2 µm2). Longissimus samples had smaller (P < 0.01) mean muscle fiber cross-sectional area than semitendinosus samples (3,119 vs. 3,909; standard error = 82.6 µm2). Pearson correlation of mean muscle fiber cross-sectional areas determined by Nikon and Cytation5 workflows was 0.73 (P < 0.01; Figure 5). Within each MyHC type, Pearson correlation of mean muscle fiber cross-sectional areas determined by Nikon and Cytation5 workflows ranged between 0.62 and 0.73 (P < 0.01; Table 4).

Figure 5.

Figure 5.

Open in a new tab

Mean muscle fiber cross-sectional area as imaged and analyzed by Nikon Eclipse with NIS Elements software (Nikon) or by Cytation5 with Gen5 software (Cytation5).

Table 4.

Summary of relationship between Nikon and Cytation5 results for mean fiber area of each MyHC type

MyHC type Nikon mean fiber area ± SD, µm2 Cytation5 mean fiber area ± SD, µm2 Pearson correlation Regression slope Regression intercept
I 2,776 ± 686 2,487 ± 728 0.62
(P < 0.01)		 0.66 ± 0.086
(P < 0.01)		 655 ± 247
(P < 0.01)
IIA 3,812 ± 938 3,050 ± 935 0.73
(P < 0.01)		 0.73 ± 0.071
(P < 0.01)		 283 ± 278
(P = 0.31)
IIX 4,476 ± 1,106 3,703 ± 904 0.73
(P < 0.01)		 0.60 ± 0.060
(P < 0.01)		 1,028 ± 267
(P < 0.01)
Open in a new tab

Discussion

Image capture

Nikon and Cytation5 workflows performed similarly for capturing fluorescent images. Qualitatively, images obtained from the Nikon and Cytation5 workflow were similar as visualized. All targeted elements (nuclei, membrane-associated protein, MyHC type I, MyHC type IIA, and MyHC type IIX) are present in images from both workflows. A minor difference in the images is the clarity of focus favoring the Nikon workflow. However, this is likely related to the objective lens used to capture the images. For the Nikon workflow, 200 × magnification was used whereas 40 × magnification was used for the Cytation5 workflow. Regardless, a clear, easily annotated image was created from both imaging platforms.

Muscle fiber identification

The greatest advancement of the Cytation5 workflow is the ability of the software to identify features of the image and use threshold-based systems to classify those features. The Cytation5 workflow was able to identify muscle fibers based on staining of membrane proteins. Muscle fibers were defined as the absence of staining for membrane proteins. Thus, accuracy of results produced with the Cytation5 workflow could be sensitive to embedding and freezing techniques that cause ice crystals to form in the extracellular space and push muscle fibers apart; gaps between muscle fibers could be identified as muscle fibers. Rapid rate of freezing and minimizing the time interval between in vivo tissue and frozen sample should alleviate this potential inaccuracy. With the sample collection and preservation techniques used for this study, this potential challenge was not observed with great frequency. When the Nikon workflow is used, low-quality samples could be less impactful on accuracy because the perimeter of each cell membrane is traced manually. However, human judgement is required to identify muscle fibers; small muscle fibers could be excluded from results because of subjective decision that the fiber was a freezing artifact. The Cytation5 workflow functioned as an objective method to identify muscle fibers.

Number, MyHC type, and size of muscle fibers evaluated

Although the Cytation5 workflow imaged about 3 × the muscle cross-sectional area of the Nikon workflow, the Cytation5 identified and evaluated almost 6 × more muscle fibers. This effect was caused by the size of the field of view. Since the Cytation5 imaged a larger field of view (1,000 μm × 1,000 μm compared to 440 μm × 330 μm), a greater area relative to the perimeter was enabled. Since muscle fibers with part of their perimeter out of the field of view (“primary edge objects” in Cytation5 protocol) are removed from analysis to prevent distortion of area measurement, minimizing the perimeter relative to the area imaged maximizes the number of muscle fibers counted per area imaged. This effect could have been negated by using the 4 × objective instead of the 20 × objective with the Nikon workflow. However, the 4 × objective of the Nikon workflow makes image analysis difficult because of marginal image quality. Additionally, each individual fiber takes approximately 30 s to trace and measure with the Nikon workflow. If 100 fibers were evaluated per sample using both methods, the Cytation5 workflow would save approximately 50 min per sample. If 600 fibers were evaluated, the Cytation5 workflow would save approximately 5 h per sample.

Myosin heavy chain type was determined manually with the Nikon workflow and by objective standards presented in Table 3. While proportion of fibers classified as each MyHC was within previously published ranges (Hergenreder et al., 2016; Smith et al., 2019; Wellmann et al., 2021), differences were detected in proportions of MyHC types between workflows. Regardless of muscle, the Cytation5 workflow identified the muscle fiber profile to be more glycolytic than the Nikon workflow. Subjectively, each imaging workflow appears to classify the visual colors of muscle fibers correctly. In cases of poor sample quality, the Cytation5 workflow could identify gaps between muscle fibers as MyHC type IIX fiber; ­however, this was not visually observed with frequency to suggest that this was a driving factor of differences observed in this study. One factor that could have affected differences in proportions of MyHC types observed was primary edge objects. Since glycolytic fibers are generally larger in size, glycolytic fibers would be more likely to have part of the perimeter out of the field of view and, therefore, be omitted from the data. This effect could have caused results from the Nikon workflow to underestimate the proportion of glycolytic fibers. Another factor that could have led to differences in oxidative and glycolytic fiber determination is that the exact same region of the muscle sample was not imaged for each analysis. In the Nikon workflow, location of the captured image was determined by hand. Each of the five images were captured from unique muscle cross sections; however, the location of the image within the cross section could have overlapped the region imaged by another capture from a separate cross section, failing to capture a large enough unique area to describe the true biological variation of the muscle. This uncertainty is avoided in the Cytation5 workflow by imaging two unique locations within the same muscle cross section.

Muscle fiber cross-sectional area was similar to previously published results (Hergenreder et al., 2017; Smith et al., 2019; Foraker et al., 2022). However, mean muscle fiber cross-sectional areas evaluated by the Cytation5 workflow across all fiber types were only 80% to 90% of the mean areas determined by the Nikon workflow. One factor that could have contributed to this as a technical effect was how the muscle fiber boundary was determined. With the Nikon workflow, the muscle fiber boundary was manually traced, with the perimeter overlayed on the fluorescence of dystrophin. In the Cytation5 workflow, the muscle fiber boundary was defined as the boundary where dystrophin was detected and where it was not. This is visualized in Figure 2B where the muscle fiber area is filled, but the green stain of the dystrophin is still visible, seeming to justify the smaller reported mean muscle fiber cross-sectional areas with the Cytation5 workflow. This phenomenon is consistent with fiber areas of the larger MyHC types having greater disparity between the Nikon and Cytation5 workflows since the radius of the fiber would have a quadratic effect on its area. Additionally, detection of spaces between fibers as MyHC type IIX fibers could have negatively biased mean fiber area of Cytation5 results. However, even if 5% of the MyHC type IIX fibers identified were not biologically MyHC type IIX fibers and had only 10% of the cross-sectional area of true MyHC type IIX fibers, overall mean MyHC type IIX fiber cross-sectional area would only be decreased 4.5%. This magnitude compared to the magnitude of difference indicated that this effect was not the only cause of differences between methods, and visual observation finds little evidence of detection of spaces between fibers as MyHC type IIX fibers.

Regardless of differences in absolute muscle fiber area, both workflows identified MyHC type I fibers as the smallest MyHC type IIX fibers as the largest. The positive relationship between the mean areas evaluated by each method identified that the methods are consistent with each other and ultimately rank samples from a given population similarly. Additionally, similar variability was observed in both methods. With the observed variation and sample size, an approximately 250 µm2 difference between treatments could be detected by either method with 95% confidence and 80% power.

Potential protocol improvements

The Cytation5 workflow could be improved by identifying MyHC type IIX fibers with an antibody such as SC-71 based on the work of Song et al. (2020). This technique would enable fiber recognition by fluorescent staining of myosin independent of cell membrane structure, a protocol change that would be possible using the Cytation5 workflow. However, SC-71 is generally produced as a mouse IgG1 antibody, an isotype and subclass already used in this workflow. Adding SC-71 to this protocol would require an additional step in the staining procedure to apply another primary antibody solution followed by a secondary antibody solution with an anti-IgG1 conjugate different than Alexa-Fluor 633 to be imaged by a unique fluorescent channel or to replace an existing channel. After applying and washing the primary and secondary antibodies described by the current protocol, SC-71 (Developmental Studies Hybridoma Bank) could be diluted in blocking solution and applied to cross sections for 1 h. After three washes in PBS, Goat antiMouse IgG1 Cross-Adsorbed Secondary Antibody, Alexa Fluor 350 (Invitrogen) could be diluted in blocking solution and applied to cross sections for 30 min before washing. Then, mounting media without DAPI could be applied before the coverslip.

This protocol modification would add about 2 h to the length of the immunohistochemical staining protocol but would enable fiber recognition by fluorescent staining of myosin independent of cell membrane structure and allow for the dystrophin antibody and DAPI stain to be removed. This method would inhibit simultaneous investigation of muscle fiber and nuclei characteristics but would add certainty to fiber identification, lessen cost, and allow identification of hybrid MyHC type IIA + IIX fibers described by Song et al. (2020). Although the frequency of these transition fibers has been reported to be less than 5% of all fibers, the faintly yellow-stained muscle fibers in Figure 1A could be an instance of these hybrid fibers. In previous workflows, these fibers were classified as either MyHC type IIA or MyHC type IIX and could contribute to the variation in proportion of fibers classified as each MyHC type. For adapting this protocol for other species that express MyHC type IIB in skeletal muscle, the BF-F3 antibody could be added as previously described by Song et al. (2020).

Summary

Overall, the Nikon and Cytation5 workflows had similar findings. When evaluated by Cytation5, muscle fiber area was less, and muscle fiber MyHC type was determined to be more glycolytic than when evaluated by the Nikon workflow. However, the Cytation5 workflow evaluated a greater number of muscle fibers and added certainty that muscle fibers were only evaluated once per sample. The Cytation5 workflow was a marked improvement in workflow efficiency, counting approximately 6 × the number of muscle fiber in about 15% of the time. Additional accuracy could be added to the outlined protocol by adding a fluorescent stain for MyHC type IIX fibers.

Supplementary Material

skad205_suppl_Supplementary_File_S1
Click here for additional data file. (144.5KB, pdf)
skad205_suppl_Supplementary_File_S2
Click here for additional data file. (304.3KB, pdf)

Acknowledgments

We express appreciation to Kinsie Arnst of Agilent Technologies, Inc., and Oscar J. Benitez of Texas Tech University for assistance with protocol development and Clarissa Strieder-Barboza of Texas Tech University for partnership with technology acquisition.

Glossary

Abbreviations

DAPI

4ʹ,6-diamidino-2-phenylindole

FITC

fluorescein isothiocyanate

GFP

green fluorescent protein

MyHC

myosin heavy chain

PBS

phosphate buffered saline

TRITC

tetramethylrhodamine

Contributor Information

Luke K Fuerniss, Department of Animal and Food Sciences, Texas Tech University, Lubbock, TX 79409, USA.

Bradley J Johnson, Department of Animal and Food Sciences, Texas Tech University, Lubbock, TX 79409, USA.

Conflict of interest statement. The authors declare no real or perceived conflicts of interest.

Literature Cited

  1. Albrecht, E., Teuscher F., Ender K., and Wegner J.. . 2006. Growth- and breed-related changes of muscle bundle structure in cattle. J. Anim. Sci. 84:2959–2964. doi: 10.2527/jas.2006-345 [DOI] [PubMed] [Google Scholar]
  2. Baggerman, J. O., Thompson A. J., Jennings M. A., Hergenreder J. E., Rounds W., Smith Z. K., and Johnson B. J.. . 2021. Effects of encapsulated methionine on skeletal muscle growth and development and subsequent feedlot performance and carcass characteristics in beef steers. Animals. 11:1627–1643. doi: 10.3390/ani11061627 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bates, D., Mächler M., Bolker B., and Walker S.. . 2015. Fitting linear mixed-effects models using lme4. J. Stat. Softw. 67:1–48. doi: 10.18637/jss.v067.i01 [DOI] [Google Scholar]
  4. Brooke, M. H., and Kaiser K. K.. . 1970. Three “myosin adenosine triphosphatase” systems: the nature of their pH lability and sulfhydryl dependence. J. Histochem. Cytochem. 18:670–672. doi: 10.1177/18.9.670 [DOI] [PubMed] [Google Scholar]
  5. Christensen, R. H. B. 2022. Ordinal: regression models for ordinal data. https://CRAN.R-project.org/package=ordinal
  6. Denny-Brown, D. E. 1929. The histological features of striped muscle in relation to its functional activity. Proc. R. Soc. Lond. B: Biol. Sci. 104:371–411. doi: 10.1098/RSPB.1929.0014 [DOI] [Google Scholar]
  7. Ebarb, S. M., Phelps K. J., Drouillard J. S., Maddock-Carlin K. R., Vaughn M. A., Burnett D. D., Noel J. A., Van Bibber- Krueger C. L., Paulk C. B., Grieger D. M., . et al. 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: 10.2527/jas2016.1263 [DOI] [PubMed] [Google Scholar]
  8. Engel, W. K., and Irwin R. L.. . 1967. A histochemical-physiological correlation of frog skeletal muscle fibers. Am. J. Physiol. 213:511–518. doi: 10.1152/AJPLEGACY.1967.213.2.511 [DOI] [PubMed] [Google Scholar]
  9. Foraker, B. A., Johnson B. J., Rathmann R. J., Legako J. F., Brooks J. C., Miller M. F., Woerner D. R., Foraker B. A., Johnson B. J., Rathmann R. J., . et al. 2022. Expression of beef- versus dairy-type in crossbred beef × dairy cattle does not impact shape, eating quality, or color of strip loin steaks. Meat Muscle Biol. 6:13926–13927. doi: 10.22175/MMB.13926 [DOI] [Google Scholar]
  10. Gonzalez, J. M., Carter J. N., Johnson D. D., Ouellette S. E., and Johnson S. E.. . 2007. Effect of ractopamine-hydrochloride and trenbolone acetate on longissimus muscle fiber area, diameter, and satellite cell numbers in cull beef cows. J. Anim. Sci. 85:1893–1901. doi: 10.2527/jas.2006-624 [DOI] [PubMed] [Google Scholar]
  11. Gorza, L. 1990. Identification of a novel type 2 fiber population in mammalian skeletal muscle by combined use of histochemical myosin ATPase and anti-myosin monoclonal antibodies. J. Histochem. Cytochem. 38:257–265. doi: 10.1177/38.2.2137154 [DOI] [PubMed] [Google Scholar]
  12. Guth, L., and Samaha F. J.. . 1969. Qualitative differences between actomyosin ATPase of slow and fast mammalian muscle. Exp. Neurol. 25:138–152. doi: 10.1016/0014-4886(69)90077-6 [DOI] [PubMed] [Google Scholar]
  13. Hergenreder, J. E., Legako J. F., Dinh T. T. N., Broadway P. R., Spivey K. S., Baggerman J. O., Hutcheson J. P., Corrigan M. E., Johnson B. J., Hergenreder J. E., . et al. 2017. Zilpaterol hydrochloride affects cellular muscle metabolism and lipid components of 10 different muscles in feedlot heifers. Meat Muscle Biol. 1:192–206. doi: 10.22175/MMB2017.02.0013 [DOI] [Google Scholar]
  14. Hergenreder, J. E., Legako J. F., Dinh T. T. N., Spivey K. S., Baggerman J. O., Broadway P. R., Beckett J. L., Branine M. E., and Johnson B. J.. . 2016. Zinc methionine supplementation impacts gene and protein expression in calf-fed Holstein steers with minimal impact on feedlot performance. Biol. Trace Elem. Res. 171:315–327. doi: 10.1007/s12011-015-0521-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hervé, M. 2022. RVAideMemoire: testing and plotting procedures for biostatistics. https://CRAN.R-project.org/package=RVAideMemoire
  16. Hosford, A. D., Hergenreder J. E., Kim J. K., Baggerman J. O., Ribeiro F. R. B., Anderson M. J., Spivey K. S., Rounds W., and Johnson B. J.. . 2015. Effects of supplemental lysine and methionine with zilpaterol hydrochloride on feedlot performance, carcass merit, and skeletal muscle fiber characteristics in finishing feedlot cattle. J. Anim. Sci. 93:4532–4544. doi: 10.2527/jas.2015-9047 [DOI] [PubMed] [Google Scholar]
  17. Kassambara, A. 2020. ggpubr: “ggplot2” based publication ready plots. https://CRAN.R-project.org/package=ggpubr
  18. Lenth, R. v. 2022. emmeans: estimated marginal means, aka least-squares means.
  19. Maccatrozzo, L., Patruno M., Toniolo L., Reggiani C., and Mascarello F.. . 2009. 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. doi: 10.4081/908 [DOI] [PubMed] [Google Scholar]
  20. Paulk, C. B., Tokach M. D., Nelssen J. L., Burnett D. D., Vaughn M. A., Phelps K. J., Dritz S. S., DeRouchey J. M., Goodband R. D., Woodworth J. C., . et al. 2014. Effect of dietary zinc and ractopamine hydrochloride on pork chop muscle fiber type distribution, tenderness, and color characteristics. J. Anim. Sci. 92:2325–2335. doi: 10.2527/jas.2013-7318 [DOI] [PubMed] [Google Scholar]
  21. Pette, D., and Staron R. S.. . 2000. Myosin isoforms, muscle fiber types, and transitions. Microsc. Res. Tech. 50:500–509. doi: [DOI] [PubMed] [Google Scholar]
  22. Picard, B., Gagaoua M., and Gagaoua M.. . 2020. Muscle fiber properties in cattle and their relationships with meat qualities: an overview. J. Agric. Food Chem. 68:6021–6039. doi: 10.1021/acs.jafc.0c02086 [DOI] [PubMed] [Google Scholar]
  23. Pierobon-Bormioli, S., Sartore S., Dalla Libera L., Vitadello M., and Schiaffino S.. . 1981. “Fast” isomyosins and fiber types in mammalian skeletal muscle. J. Histochem. Cytochem. 29:1179–1188. doi: 10.1177/29.10.7028858 [DOI] [PubMed] [Google Scholar]
  24. R Core Team. 2022. R: a language and environment for statistical computing. R Foundation for Statistical Computing. https://www.r-project.org/ [Google Scholar]
  25. Scheffler, T. L., Leitner M. B., and Wright S. A.. . 2018. Technical note: protocol for electrophoretic separation of bovine myosin heavy chain isoforms and comparison to immunohistochemistry analysis. J. Anim. Sci. 96:4306–4312. doi: 10.1093/jas/sky283 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Schiaffino, S., Gorza L., Sartore S., Saggin L., Ausoni S., Vianello M., Gundersen K., and Lømo T.. . 1989. Three myosin heavy chain isoforms in type 2 skeletal muscle fibres. J. Muscle Res. Cell Motil. 10:197–205. doi: 10.1007/BF01739810 [DOI] [PubMed] [Google Scholar]
  27. Smith, Z. K., Kim J., and Johnson B. J.. . 2019. Feedlot performance and biological responses to coated and non-coated steroidal implants containing trenbolone acetate and estradiol benzoate in finishing beef steers. J. Anim. Sci. 97:4371–4385. doi: 10.1093/jas/skz298 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Song, S., Ahn C. H., and Kim G. D.. . 2020. Muscle fiber typing in bovine and porcine skeletal muscles using immunofluorescence with monoclonal antibodies specific to myosin heavy chain isoforms. Food Sci. Anim. Resour. 40:132–144. doi: 10.5851/kosfa.2019.e97 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Thornton, K. J. 2019. Triennial growth symposium: the nutrition of muscle growth: impacts of nutrition on the proliferation and differentiation of satellite cells in livestock species. J. Anim. Sci. 97:2258–2269. doi: 10.1093/jas/skz081 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Waller, B. E., Garcia S. R., Woerner D. R., Johnson B. J., and Wulf D. M.. . 2023. Abstracts from the 2022 AMSA Reciprocal Meat Conference: utilizing the F94L myostatin gene mutation to improve carcass characteristics and cutout in beef × dairy crossbred cattle. Meat Muscle Biol. 6:9–10. doi: 10.22175/mmb.15450 [DOI] [Google Scholar]
  31. Wellmann, K. B., Kim J., Urso P. M., Smith Z. K., and Johnson B. J.. . 2021. Evaluation of vitamin A status on myogenic gene expression and muscle fiber characteristics. J. Anim. Sci. 99:1–10. doi: 10.1093/JAS/SKAB075 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Wickham, H. 2016. ggplot2: elegant graphics for data analysis. Springer. [Google Scholar]
  33. Wilke, C. O. 2020. cowplot: streamlined Plot Theme and plot annotations for “ggplot2.”https://CRAN.R-project.org/package=cowplot

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

skad205_suppl_Supplementary_File_S1
Click here for additional data file. (144.5KB, pdf)
skad205_suppl_Supplementary_File_S2
Click here for additional data file. (304.3KB, pdf)

Articles from Journal of Animal Science are provided here courtesy of Oxford University Press

RESOURCES