Abstract
Isolating individual muscle fibers and characterizing their myosin heavy chain (MHC) content using SDS-PAGE has become an increasingly common method for describing skeletal muscle fiber type proportions. In this study, we aimed to assess how the number of muscle fibers analyzed, and whether they are characterized in the order of isolation or randomly selected from a larger pool of muscle fibers, affects the precision of fiber type proportion estimates. A total of 170 individual muscle fibers were isolated from vastus lateralis biopsies from each of eight human subjects, and their MHC isoform content was analyzed using SDS-PAGE. To evaluate the precision of fiber type proportion estimates, we employed a resampling approach, varying both the muscle fiber sample size (25, 50, or 100 fibers) and the selection method (ordered vs. random selection). Our results indicate that when analyzing a small number of muscle fibers, precision improves if the fibers are randomly selected from a larger pool rather than characterized in the order they were isolated. These findings have important implications for designing experiments to assess skeletal muscle fiber heterogeneity and its role in health and disease.
Keywords: Muscle fibers, Muscle phenotype, Myosin heavy chain, Accuracy, Number of fibers, Selection of fibers
Subject terms: Biological techniques, Biological models
Introduction
Skeletal muscle is a heterogeneous tissue composed of various fiber types, and their proportion within a muscle plays a crucial role in determining its overall metabolic and functional properties. These properties have significant implications for physical performance and metabolic health1–4. Precise determination of the muscle fiber type proportions is essential for understanding muscle function in both physiological and pathological contexts, including the effects of sex differences5aging6obesity7inactivity8and microgravity exposure9. Muscle fibers are broadly classified based on their myosin heavy chain (MHC) isoform content into type I, type IIa, and type IIx fibers, with hybrid fibers co-expressing more than one MHC isoform (e.g., type I/IIa or IIa/IIx). Since fiber type composition dictates a muscle’s metabolic and functional characteristics4,10,11precisely determining its fiber type proportions is essential for muscle research.
Various techniques are available to determine muscle fiber type composition, as previously discussed in detail12. While antibody-based techniques, such as immunofluorescence on cryosectioned muscle samples, are commonly used for muscle fiber type classification, they have inherent limitations, as, for example, when precise discrimination of MHC IIx is required13. Isolating single muscle fibers and analyzing their MHC isoform content using silver-stained sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) is considered the gold standard. This method enables accurate separation of MHC I, IIa, and IIx bands, also allowing for the precise identification of hybrid fiber types that are not readily distinguishable using other techniques12,13. Describing the proportion of hybrid fibers is important in muscle physiology research, as these fibers function as intermediates during fiber type transitions and reflect the muscle’s dynamic equilibrium between fiber types14,15. Their prevalence varies with aging16obesity17and exercise training18,19and neglecting to characterize them risks oversimplified fiber-type classifications and obscures the muscle’s adaptive responses to (patho)physiological stimuli.
Analyzing too few fibers from a muscle biopsy sample may lead to imprecise estimates of muscle fiber proportions, undermining the validity of research findings and potentially masking statistically significant differences (i.e., whether the analysis meets formal statistical significance criteria). This, in turn, can affect result interpretation and limit the ability to draw meaningful conclusions. Theoretically, precision in determining fiber type proportions improves as more fibers are phenotyped for their MHC isoform content. However, practical constraints such as time and cost often limit the analysis of large numbers of muscle fibers, and analyzing 25 fibers per biopsy has been proposed as a representative measure of muscle fiber type distribution20. Moreover, muscle fibers are typically isolated in an orderly fashion from a specific region of a biopsy sample. The impact of muscle fiber sample size and isolation order on the precision of determining muscle fiber type proportions has not been described. This information is crucial for guiding the design of experiments involving single muscle fiber typing, particularly when using the widely employed and recommended SDS-PAGE technique12,21.
Our objective was to assess how the number of muscle fibers phenotyped, as well as whether they were phenotyped in the order of isolation or randomly selected from a larger pool of isolated fibers, affects the precision of determining fiber type proportions in muscle. We hypothesized that the precision of muscle fiber phenotyping improves either by increasing the number of phenotyped fibers or, when phenotyping a smaller number, by randomly selecting them from a larger pool of fibers isolated from the biopsy sample.
Methods
Overall study design
The study was carried out in accordance with the Declaration of Helsinki principles and performed after obtaining approval from the Institutional Review Board at Mayo Clinic (IRB # 20-003294), and informed consent was obtained from all subjects. We studied eight subjects (4 females/4 males), determined to be apparently healthy, at the Ambulatory Infusion Center (AIC) at Mayo Clinic in Scottsdale, Arizona. Study subjects reported initially for a screening visit in the morning following an overnight fast (i.e., 12 h). Before obtaining written consent, the study’s purpose, the specific experimental procedures, and the associated risks were explained to each participant. Screening included medical history, standard physical examination, electrocardiogram, and routine blood tests. Subjects were excluded from the study if screening showed that it is not safe for them to engage in the study procedures. For the collection of the muscle tissue, the subjects arrived at the AIC on a separate day after an overnight fast.
Muscle biopsy
Muscle tissue was collected from the vastus lateralis muscle using a Bergstrom needle and the suction technique. A small area in the vastus lateralis muscle was numbed with lidocaine, and an incision was made to allow the passage of the biopsy needle. All procedures were conducted under a sterile environment. A vastus lateralis muscle sample (~ 20 mg) was collected and examined under a microscope to ensure it is free from connective and fat tissue and blood clots. The sample was placed in a cold skinning solution (125 mM K propionate, 2 mM EGTA, 1 mM MgCl2, 20 mM Imidazole, [pH7.0], and 50% glycerol), and stored at −20 °C until the isolation of single muscle fibers.
Isolation of muscle fibers and identification of their MHC isoforms
Muscle bundles were trimmed at both ends, perpendicular to the long axis, to ensure uniform fiber length throughout the bundle. Single muscle fibers of approximately 2.5 mm in apparent length were extracted from the muscle bundle using fine tweezers under a light microscope (Leica S9i) in the same skinning solution. Single muscle fibers (170 from each subject) were progressively isolated from different areas of each muscle biopsy sample22,23placed in separate microtubes containing ~ 20 ul of a sample buffer (1% SDS, 23 mM EDTA, 0.008% bromophenyl blue, 15% glycerol, 715 mM b-mercaptoethanol, [pH 6.8]), and labeled according to the order that they were isolated from the tissue biopsy. Isolated muscle fibers were then heated at 95 °C for 5 min and stored at −20 °C until further analysis.
MHC isoforms in the isolated muscle fibers were identified using SDS-PAGE, by preparing 3% stacking and 5% resolving gels, and run at 150 volts for ~ 16 h at 5 °C (SE 600 Series; Hoefer, San Francisco, CA, USA). Gels were silver stained using a series of chemical washes to fix and prepare the gel for silver stain (fixing solution contained 10% acetic acid and 50% ethanol; for 10 min; crosslinking solution contained 10% glutaraldehyde). The silver stain was prepared using 2% ammonium hydroxide, 1.5% sodium hydroxide, and 0.7% silver nitrate. Gels were then washed in the silver stain one at a time for no more than 10 min, before 3 separate washes with miliQ water for at least 3 min, and placed in a developing solution containing 0.25% citric acid and 0.01% formaldehyde until bands became visible. The reaction was stopped using approximately 2 ml of 10% acetic acid before imaging of the gels on a light box and a 12 mega pixel camera (Pixel 5/7, Google Inc). MHC isoform bands on the gels were identified based on their distinct, well-established migration pattern on SDS-PAGE gels determined by their molecular weight24,25. Muscle fibers were categorized based on the MHC isoforms present in each lane as pure muscle fibers, if only one MHC isoform was present (i.e., Type I or Type IIa or Type IIx), or hybrid muscle fibers if more than one MHC isoforms were present (i.e., Type I/IIa, Type Type IIa/IIx). Example of a stained gel showing the different MHC isoforms in single muscle fibers is shown in Fig. 1.
Fig. 1.
Example of a stained gel showing the different human myosin heavy chain (MHC) isoforms in single muscle fibers. The muscle fibers were isolated from a muscle biopsy sample under a microscope and analyzed (along with muscle homogenate as a reference) for MHC isoform content using sodium dodecyl sulfate-polyacrylamide gel electrophoresis and silver staining. Pure and hybrid muscle fibers are depicted.
Calculations
The ratio of each type of muscle fiber to the total 170 muscle fibers analyzed per subject was calculated to determine a true proportion of each muscle fiber type in the muscle biopsy sample. Resampling allows for estimating the sampling distribution of the underlying sample by repeatedly resampling from the observed data. The process generates multiple simulated samples that can provide insights into the variability and distribution of a statistic and enable inference about the precision by which population parameters are estimated. This process involved resampling a specified number of observations from the sample (i.e., subsampling) with replacement. We applied resampling 1,000 times to the first 25, 50, and 100 muscle fibers phenotyped (referred to as Ordered Selection Fiber Typing) and also randomly to 25, 50, and 100 muscle fibers phenotyped (referred to as Random Selection Fiber Typing) from the total of 170 muscle fibers isolated from each subject.
Statistical analyses
The normality of the data was evaluated using the Kolmogorov-Smirnov test. Histograms and violin plots were used for graphical representation of the distribution of the median of the muscle fiber type proportions generated from the resampling processes. Data are presented as quartiles (median, 25th and 75th percentile) and differences between data sets were evaluated using the Kruskal-Wallis test. Post-hoc pairwise comparisons were performed using the Dunn’s test. Bland-Altman plots26 were used to assess the agreement between proportion of type of muscle fibers in muscle estimated from the resampling process and that calculated based on all 170 muscle fibers phenotyped from each subject. All statistical analyses were performed using GraphPad Prism Software (version 8.4.3).
Results
Subjects’ age and body mass index were (mean ± SD, range) (31 ± 8, 19–44 years) and (28 ± 8, 20–40, kg/m2). The proportion of the different types of muscle fibers in the total of 170 muscle fibers phenotyped for each subject, ranked from most to least dominant were as follows (mean ± SD, range): Type I (0.37 ± 0.20, 0.12–0.63), Type IIa (0.34 ± 0.14, 0.11–0.50), Type IIa/IIx (0.17 ± 0.10, 0.03–0.35), Type I/IIa (0.06 ± 0.08, 0.00–0.24), and Type IIx (0.03 ± 0.04, 0.00–0.12).
The datasets generated from the resampling process, which described the proportion of muscle fiber types using 25, 50, or 100 fibers, were not normally distributed for any subject in either the Ordered Selection Fiber Typing or the Random Selection Fiber Typing data. Consequently, all data from the resampling process were analyzed using quartiles (median, 25th, and 75th percentiles).
Fiber type determination based on ordered and random selection of a given number of phenotyped muscle fibers
Table 1 presents the median, 25th, and 75th percentile values describing the proportions of the different types of muscle fibers for each study participant. These values were derived from the Ordered and Random selection of phenotyped muscle fibers, and when resampling 25, 50, and 100 muscle fibers. The values describing the proportion of muscle fiber types based on the Ordered selection fiber typing analysis were statistically significantly different across the counts of muscle fibers analyzed (i.e., 25, 50 and 100) for all muscle fiber types and in all subjects (P < 0.05). However, except for the least represented fiber types in muscle (i.e., Type I/IIa, Type IIx), these values were not statistically significantly different across the counts of muscle fibers analyzed (i.e., 25, 50 and 100) based on the Random selection fiber typing analysis (P > 0.05). This is further illustrated in Fig. 2, which shows violin plots for the most abundant (i.e., Type I; top) and least abundant (i.e., Type I/IIa; bottom) fiber types in muscle, based on the Ordered and Random selection analyses for fiber typing.
Table 1.
Estimates of the proportion of fiber types in muscle obtained by resampling 25, 50, or 100 of the phenotyped muscle fibers using ordered (Ordered) and random (Random) selection of phenotyped muscle fibers for each subject.
| Fiber Type | Ordered | Random | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Fiber Count | Fiber Count | |||||||||
| 25 | 50 | 100 | P | 25 | 50 | 100 | P | |||
| S1 | I |
0.48 (0.40, 0.56) |
0.46 (0.42, 0.68) |
0.59 (0.56, 0.63) |
< 0.0001 |
0.56 (0.52, 0.64) |
0.56 (0.52, 0.62) |
0.57 (0.54, 0.60) |
0.67 | |
| I/IIa | N/A |
0.06 (0.04, 0.08) |
0.05 (0.04, 0.06) |
< 0.0001 |
0.04 (0.04, 0.08) |
0.06 (0.04, 0.08) |
0.06 (0.05, 0.08) |
< 0.05 | ||
| IIa |
0.52 (0.44, 0.60) |
0.48 (0.44, 0.52) |
0.34 (0.30. 0.37) |
< 0.0001 |
0.36 (0.28, 0.40) |
0.34 (0.29, 0.38) |
0.34 (0.31, 0.37) |
0.42 | ||
| IIa/IIx | N/A | N/A |
0.02 (0.01, 0.03) |
< 0.0001 |
0.04 (0.00, 0.04) |
0.02 (0.02, 0.04) |
0.03 (0.02, 0.04) |
< 0.0001 | ||
| IIx | N/A | N/A | N/A | - | N/A | N/A | N/A | - | ||
| S2 | I |
0.32 (0.24, 40) |
0.28 (0.24, 0.32) |
0.28 (0.25, 0.31) |
< 0.0001 |
0.24 (0.16, 0.28) |
0.22 (0.18, 0.28) |
0.23 (0.20, 0.26) |
0.74 | |
| I/IIa |
0.28 (0.20, 0.36) |
0.20 (0.16, 0.24) |
0.20 (0.17, 0.22) |
< 0.0001 |
0.24 (0.20, 0.28) |
0.24 (0.20, 0.28) |
0.24 (0.21, 0.27) |
0.15 | ||
| IIa |
0.04 (0.00, 0.04) |
0.04 (0.02, 0.06) |
0.04 (0.03, 0.05) |
< 0.01 |
0.12 (0.08, 0.16) |
0.12 (0.08, 0.14) |
0.11 (0.09. 0.13) |
0.14 | ||
| IIa/IIx |
0.04 (0.00, 0.08) |
0.18 (0.14, 0.22) |
0.18 (0.15,0.20) |
< 0.0001 |
0.16 (0.12, 0.20) |
0.14 (0.12, 0.18) |
0.15 (0.12, 0.17) |
0.95 | ||
| IIx | N/A | N/A | N/A | - | N/A | N/A | N/A | - | ||
| S3 | I |
0.20 (0.16, 0.24) |
0.32 (0.28, 0.36) |
0.34 (0.31, 0.37) |
< 0.0001 |
0.32 (0.24, 0.36) |
0.32 (0.28, 0.36) |
0.32 (0.29, 0.35) |
0.62 | |
| I/IIa | N/A | N/A | N/A | - |
0.00 (0.00, 0.04) |
0.00 (0.00, 0.02) |
0.01 (0.00, 0.02) |
< 0.0001 | ||
| IIa |
0.32 (0.24, 0.40) |
0.26 (0.22, 0.30) |
0.30 (0.27, 0.33) |
< 0.0001 |
0.36 (0.32, 0.44) |
0.38 (0.32, 0.42) |
0.37 (0.34, 0.41) |
0.74 | ||
| IIa/IIx |
0.48 (0.40, 0.56) |
0.42 (0.38, 0.46) |
0.36 (0.33, 0.39) |
< 0.0001 |
0.28 (0.20, 0.32) |
0.26 (0.22, 0.32) |
0.27 (0.24, 0.30) |
0.89 | ||
| IIx | N/A | N/A | N/A | - |
0.00 (0.00, 0.04) |
0.02 (0.00, 0.02) |
0.02 (0.01, 0.03) |
< 0.0001 | ||
| S4 | I |
0.12 (0.08, 0.16) |
0.10 (0.08, 0.12) |
0.11 (0.09, 0.13) |
< 0.0001 |
0.12 (0.08, 0.16) |
0.08 (0.08, 0.14) |
0.12 (0.09, 0.14) |
0.87 | |
| I/IIa | N/A | N/A | N/A | - | N/A | N/A | N/A | - | ||
| IIa |
0.56 (0.48, 0.64) |
0.48 (0.44, 0.52) |
0.54 (0.50, 0.57) |
< 0.0001 |
0.48 (0.44, 0.56) |
0.50 (0.44, 0.56) |
0.50 (0.47, 0.53) |
0.54 | ||
| IIa/IIx |
0.32 (0.25, 0.40) |
0.40 (0.36, 0.44) |
0.31 (0.8, 0.34) |
< 0.0001 |
0.36 (0.28, 0.40) |
0.34 (0.30, 0.38) |
0.34 (0.31, 0.38) |
0.58 | ||
| IIx | N/A |
0.02 (0.00, 0.04) |
0.04 (0.03, 0.05) |
< 0.0001 |
0.04 (0.00, 0.04) |
0.04 (0.02, 0.06) |
0.03 (0.02, 0.05) |
0.16 | ||
| S5 | I |
0.24 (0.20, 0.28) |
0.18 (0.14, 0.22) |
0.24 (0.21, 0.27) |
< 0.0001 |
0.28 (0.24, 0.36) |
0.28 (0.24, 0.34) |
0.29 (0.26, 0.31) |
0.57 | |
| I/IIa | N/A |
0.02 (0.00, 0.04) |
0.03 (0.02, 0.04) |
< 0.0001 |
0.04 (0.00, 0.08) |
0.04 (0.02, 0.06) |
0.05 (0.03, 0.06) |
< 0.01 | ||
| IIa |
0.64 (0.60, 0.72) |
0.58 (0.54, 0.62) |
0.55 (0.52, 0.58) |
< 0.0001 |
0.48 (0.40, 0.56) |
0.48 (0.44, 0.54) |
0.48 (0.44, 0.51) |
0.26 | ||
| IIa/IIx |
0.04 (0.00, 0.08) |
0.18 (0.14, 0.22) |
0.16 (0.14, 0.18) |
< 0.0001 |
0.16 (0.12, 0.20) |
0.16 (0.14, 0.20) |
0.16 (0.14, 0,19) |
0.78 | ||
| IIx |
0.08 (0.04, 0.12) |
0.04 (0.02, 0.06) |
0.02 (0.10, 0.03) |
< 0.0001 |
0.00 (0.00, 0.04) |
0.02 (0.00, 0.04) |
0.02 (0.01, 0.03) |
< 0.001 | ||
| S6 | I |
0.60 (0.52, 0.68) |
0.68 (0.64, 0.72) |
0.61 (0.58, 0.64) |
< 0.0001 |
0.64 (0.56, 0.68) |
0.62 (0.58, 0.66) |
0.62 (0.59, 0.65) |
0.45 | |
| I/IIa |
0.04 (0.00, 0.08) |
0.04 (0.02, 0.06) |
0.07 (0.05, 0.08) |
< 0.0001 |
0.04 (0.04, 0.08) |
0.06 (0.04, 0.08) |
0.05 (0.04, 0.07) |
< 0.01 | ||
| IIa |
0.20 (0.12, 0.24) |
0.16 (0.12, 0.20) |
0.21 (0.18, 0.24) |
< 0.0001 |
0.24 (0.16, 0.28) |
0.22 (0.18, 0.26) |
0.22 (0.19, 0.25) |
0.48 | ||
| IIa/IIx |
0.12 (0.08, 0.16) |
0.10 (0.08, 0.12) |
0.10 (0.08, 0.12) |
< 0.0001 |
0.08 (0.04, 0.12) |
0.08 (0.06, 0.12) |
0.09 (0.07, 0.11) |
0.15 | ||
| IIx |
0.04 (0.00, 0.08) |
0.02 (0.00, 0.04) |
0.01 (0.00, 0.20) |
< 0.0001 | N/A |
0.00 (0.00, 0.02) |
0.00 (0.00, 0.01) |
< 0.0001 | ||
| S7 | I |
0.24 (0.20, 0.32) |
0.24 (0.20, 0.28) |
0.21 (0.18, 0.24) |
< 0.0001 |
0.21 (0.16, 0.28) |
0.22 (0.18, 0.26) |
0.22 (0.19, 0.24) |
0.20 | |
| I/IIa | N/A |
0.02 (0.00, 0.04) |
0.02 (0.01, 0.03) |
< 0.0001 |
0.00 (0.00, 0.04) |
0.00 (0.00, 0.02) |
0.01 (0.00, 0.02) |
< 0.0001 | ||
| IIa |
0.32 (0.24, 0.40) |
0.32 (0.28, 0.36) |
0.40 (0.37, 0.43) |
< 0.0001 |
0.42 (0.36, 0.48) |
0.42 (0.38, 0.47) |
0.42 (0.39, 0.46) |
0.88 | ||
| IIa/IIx |
0.20 (0.12, 0.24) |
0.24 (0.20, 0.28) |
0.22 (0.19, 0.25) |
< 0.0001 |
0.21 (0.16, 0.28) |
0.22 (0.18, 0.26) |
0.22 (0.20, 0.25) |
0.83 | ||
| IIx | N/A | N/A | N/A | - | N/A | N/A | N/A | - | ||
| S8 | I |
0.64 (0.57, 0.71) |
0.61 (0.56, 0.66) |
0.60 (0.56, 0.63) |
< 0.0001 |
0.64 (0.56,0.68) |
0.62 (0.58, 0.68) |
0.63 (0.60, 0.66) |
0.61 | |
| I/IIa |
0.04 (0.00, 0.08) |
0.04 (0.02, 0.06) |
0.02 (0.01, 0.03) |
< 0.0001 |
0.00 (0.00, 0.04) |
0.02 (0.00, 0.04) |
0.02 (0.01, 0.03) |
< 0.0001 | ||
| IIa |
0.16 (0.12, 0.21) |
0.16 (0.12, 0.20) |
0.25 (0.22, 0.28) |
< 0.0001 |
0.24 (0.20, 0.29) |
0.24 (0.20, 0.28) |
0.24 (0.21, 0.27) |
0.34 | ||
| IIa/IIx |
0.17 (0.12, 0.21) |
0.18 (0.14, 0.22) |
0.13 (0.11, 0.15) |
< 0.0001 |
0.12 (0.08, 0.16) |
0.10 (0.08, 0.14) |
0.10 (0.08, 0.13) |
0.48 | ||
| IIx | N/A | N/A | N/A | - | N/A | N/A | N/A | - | ||
Data are median (25th and 75th percentiles). S1 through S8 are subjects 1 through 8. P values are from Kurskal-Wallis tests across the number of fibers (i.e., 25, 50, 100) within the Ordered and Random selection of phenotyped muscle fibers. N/A, no fibers of this type were detected.
Fig. 2.
Estimates of muscle fiber type proportions based on resampling different numbers of phenotyped fibers isolated in either an ordered or random fashion from the muscle biopsy sample. Violin plots depicting the distributions of the median, 25th, and 75th percentiles of the proportions of Type I (top) and Type I/IIa (bottom) muscle fibers for a single subject when resampling (i.e., 1000 times) 25, 50, and 100 phenotyped fibers isolated in Ordered or Random fashion from the muscle biopsy. Dunn’s test was used for pairwise comparisons. P values between the different muscle fiber counts are as shown. Note: N/A, no muscle fibers of this type were detected in the first 25 fibers isolated from the muscle biopsy sample.
Frequency distribution of fiber type estimates based on ordered and random selection of a given number of phenotyped muscle fibers
We plotted the frequency of the distribution of the median describing the proportion of muscle fiber types for a single subject generated from the Ordered and Random selection of the muscle fibers phenotyped and for the most abundant (i.e., Type I) and least abundant (i.e., Type I/IIa) fiber types in muscle (Fig. 3). These histograms show that the width of the distribution of the muscle fiber type estimates is becoming narrower as the count of muscle fibers analyzed increases for both the Ordered and Random selection fiber typing approaches. Also, the central tendencies for the Ordered and Random selection fiber typing approaches are closest at the highest number of muscle fiber count (i.e., 100 muscle fibers) for both the most and least abundant fibers in muscle. Moreover, the estimates describing the proportion of fiber types in muscle exhibit a greater clustering away from the central tendency (i.e., true value) at the lowest counts of muscle fibers (i.e., 25 muscle fibers compared to 100 muscle fibers), and this is especially evident for the least abundant muscle fiber types (Fig. 3B).
Fig. 3.
Frequency of the distribution of the median describing the proportion of different types of fibers in skeletal muscle when using ordered and random selection for phenotyping isolated muscle fibers. Histograms showing the frequency of the distribution of the median describing the proportion of Type I (A) and Type I/IIa (B) muscle fibers in skeletal muscle for a single subject when resampling (i.e., 1000 times) 25, 50, and 100 muscle fibers using Ordered and Random selection for phenotyping fibers isolated from a muscle biopsy sample.
Agreement of ordered and random muscle fiber phenotyped with observed muscle fiber phenotype
We constructed Bland-Altman plots to evaluate the agreement of the estimated proportions of the various muscle fiber types generated from the Ordered and Random selection fiber typing analyses with the corresponding observed proportions based on all 170 muscle fibers analyzed. Representative plots for the most abundant (i.e., Type I; Fig. 4A) and least abundant (i.e., Type I/IIa; Fig. 4B) fiber types in muscle are shown in Fig. 4.
Fig. 4.
Agreement between estimated and observed proportions of different types of fibers in skeletal muscle when using ordered and random selection for phenotyping isolated muscle fibers. Bland-Altman scatter plots illustrating the agreement between the estimated and observed proportions of Type I (A) and Type I/IIa (B) muscle fibers across all subjects. Estimated proportions are based on either Ordered (left) or Random (right) selection for phenotyping isolated muscle fibers, while observed proportions are determined using all 170 isolated muscle fibers.
The mean difference line in each plot, which when compared to the x-axis, describes the bias of the Ordered or Random approaches when selecting fibers for phenotyping, and the lines for the upper and lower 95% limits of agreement are positioned closer to zero (i.e., the x-axis) for most abundant types of fibers (i.e., Type I) when the Random selection fiber typing is employed across all the number (i.e., 25 vs. 50 vs. 100) of muscle fibers phenotyped (Fig. 4A). However, these lines are closer to zero for most abundant types of muscle fibers when only a large number of muscle fibers are phenotyped in the Ordered selection fiber typing approach. The same observations generally apply to the least abundant types of muscle fibers as well (Fig. 4B). Additionally, when only a few muscle fibers can be phenotyped from a muscle biopsy sample, the systematic bias is considerably reduced with Random selection fiber typing compared to Ordered Selection fiber typing when characterizing the proportion of fiber types in muscle.
Discussion
While it is widely recognized that skeletal muscle exhibits remarkable heterogeneity in fiber type composition, a universally accepted single parameter for broadly classifying these fibers remains elusive and the subject of ongoing research27. Nonetheless, the MHC isoforms remain the most widely used markers for classifying muscle fiber types in research. The single-fiber SDS-PAGE method for MHC determination is considered the gold standard and has gained popularity for its ability to precisely characterize MHC content in individual muscle fibers, significantly advancing our understanding of muscle metabolism and function12,21. Using this method, we applied a resampling approach to analyze the distribution of the median representing the proportion of fiber types in muscle when 25, 50, or 100 fibers were phenotyped. The resampling approach helps assess the variability of specific statistics across different subsets of data, providing insights into the robustness and generalizability of findings when working with a limited dataset. Additionally, it highlights potential biases that may emerge when analyzing a limited number of muscle fibers. Our research focused on how the number of muscle fibers phenotyped and the randomness of their selection from a larger pool of muscle fibers influence the precision of estimates describing the proportion of fiber types in muscle. This information can inform the design of experiments aimed at characterizing muscle fiber phenotypes in human biopsy samples.
Our analysis included a range across the continuum of fiber types in muscle, such as pure (e.g., Type I) and hybrid (e.g., Type I/IIa) muscle fibers. The single-fiber SDS-PAGE approach is particularly well-suited for assessing fiber types present in low proportions in muscle (e.g., Type I/IIa, IIx)12,13. We used histograms to plot the distributions of the resampled medians, allowing us to visually compare the differences in data between the Ordered and Random selection approaches for muscle fiber typing. The frequency distribution plots clearly demonstrate that the precision of muscle fiber type estimates improves (i.e., variability decreases) as more muscle fibers are phenotyped. This trend holds true regardless of whether the muscle fibers were phenotyped in the order they were isolated or randomly selected from a larger pool. These findings confirm that phenotyping a progressively larger number of fibers improves the accuracy of muscle fiber type proportion estimates. It is worth noting, however, that in the case of more abundant muscle fiber types (e.g., pure Type I fibers), fiber type proportions can still be estimated with a relatively high degree of precision by analyzing a smaller number of fibers. Nevertheless, the precision of this estimate improves further when these muscle fibers are randomly selected from a larger pool of fibers isolated from muscle. In contrast, estimates for less abundant muscle fiber types are inherently less precise, regardless of the selection method.
Our data support previous findings that the precision of determining muscle fiber type proportions progressively improves as the number of phenotyped muscle fibers increases28. Our research also builds on previous studies showing that the proportion of fiber types in muscle can be estimated by phenotyping as few as 25 fibers20. We demonstrate that the precision of muscle fiber phenotyping improves when fibers are randomly selected from a larger pool of isolated muscle fibers. This is particularly evident when only a small number of fibers can be phenotyped. Estimating the distribution of muscle fiber types becomes especially challenging when studying populations in which certain fiber types, such as hybrid fibers or Type IIx fibers, are rare. This may include older individuals, those with obesity or Type 2 diabetes, individuals with spinal cord injuries, or those exposed to experimental conditions associated with muscle unloading7,9,18,29,30. When the fiber types of interest are rare, our findings caution against characterizing muscle fiber type proportions using only 25 fibers20especially if they are merely the first 25 isolated from a muscle biopsy. In such cases, it is important to consider the expected fiber type distribution of the target study population when designing the experiments. This consideration will guide the number of muscle fibers that need to be isolated before phenotyping. By ensuring a sufficiently large pool of muscle fibers for random selection and phenotyping, researchers can achieve more precise estimates of muscle fiber type proportions. This approach reduces variability and enhances statistical power to detect significant differences31 in muscle fiber type proportions under the given experimental conditions.
A substantial body of evidence highlights the importance of using standardized protocols to obtain precise estimates when characterizing muscle fibers28,32–34. Improved experimental approaches can enhance our understanding of muscle fiber type distributions and improve the interpretation and generalizability of the findings. Fiber length may influence the likelihood of selection during the fiber isolation process, potentially introducing bias toward certain fiber types35. However, adhering to standardized procedures, such as trimming the ends of muscle bundles to ensure uniform fiber length, as described in this paper and related work35and avoiding preferential selection based on fiber length can help minimize such bias. In the present study, we aimed to isolate fibers of similar apparent length but did not precisely assess fiber length post-dissection. Therefore, we cannot rule out the possibility of a slight bias toward longer fibers, such as Type IIa fibers, which may appear marginally longer35. This represents a potential limitation of our Ordered selection approach. Nonetheless, we consider such bias unlikely in the present study, as the fiber length differences associated with specific fiber type are relatively small (~ 10% of the bundle length) and not discernible during our isolation process. Consequently, lower precision observed in fiber type proportion estimates when fibers were evaluated in the orderly fashion is more likely due to the natural tendency of muscle fibers of the same type to cluster together36,37combined with an increased likelihood of consecutively isolating fibers from a single portion of the bundle due to the ease of peeling them off in sequence.
Randomly selecting and phenotyping muscle fibers from a larger pool yields more consistent and precise estimates of muscle fiber types compared to phenotyping the same number of fibers in the order they were isolated from muscle. Our Bland-Altman plots further support these conclusion, showing that true and estimated muscle fiber type proportions across subjects are more tightly clustered around the mean difference line and with narrower limits of agreement for the Random selection approach compared to the Ordered selection approach. These findings indicate that randomly selecting muscle fibers from a larger pool of isolated fibers results in more consistent and precise muscle fiber phenotyping across subject populations compared to phenotyping fewer fibers in the order they were isolated. When fibers are analyzed in the order they are isolated, obtaining estimates closer to the true proportion of muscle fiber types can be improved solely by increasing the number of phenotyped fibers.
Our results highlight the importance of adopting well-informed methodologies when seeking to examine the proportions of muscle fiber types in human biopsy samples, particularly when focusing on less common fiber types such as Type IIx or hybrid muscle fiber types. Precise estimation of muscle fiber types is crucial given the growing interest in muscle physiology and the role of the different fiber types in various disease states7. Integrating considerations outlined here into experimental designs will help ensure reproducible results and robust conclusions about the implications of muscle fiber types in health and disease.
Conclusions
The analysis of single fibers isolated from skeletal muscle biopsies is a cornerstone of modern muscle physiology research and is expected to become even more prevalent in future studies. To ensure that the collected data accurately reflects the true proportions of muscle fiber types, researchers should phenotype a sufficient number of fibers randomly selected from a larger pool of fibers isolated from the biopsy sample. The number of fibers required for this purpose increases as the expected proportion of certain fiber types in the muscle decreases under the specific experimental circumstances.
Acknowledgements
We acknowledge the support of our Study Coordinator, Brooke Brown, and the nursing staff at the Ambulatory Infusion Center at Mayo Clinic in Arizona. We thank the subjects for their participation in the study.
Author contributions
N.S. and C.S.K. designed the experiments and wrote the manuscript. N.S., L.R.R., E.DF. and C.S.K. performed the experiments and analyzed data. M.R.R. and C.S.K.performed statistical analyses. All authors reviewed the manuscript.
Funding
The study was supported by National Institutes of Health/National Institute of Diabetes and Digestive and Kidney Diseases grant R01DK123441 (CSK).
Data availability
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Declarations
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Pette, D. Metabolic heterogeneity of muscle fibres. J. Exp. Biol.115, 179–189. 10.1242/jeb.115.1.179 (1985). [DOI] [PubMed] [Google Scholar]
- 2.Liu, G., Mac Gabhann, F. & Popel, A. S. Effects of fiber type and size on the heterogeneity of oxygen distribution in exercising skeletal muscle. PLoS One. 7, e44375. 10.1371/journal.pone.0044375 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bottinelli, R. & Reggiani, C. Human skeletal muscle fibres: molecular and functional diversity. Prog Biophys. Mol. Biol.73, 195–262. 10.1016/s0079-6107(00)00006-7 (2000). [DOI] [PubMed] [Google Scholar]
- 4.Schiaffino, S. & Reggiani, C. Fiber types in mammalian skeletal muscles. Physiol. Rev.91, 1447–1531. 10.1152/physrev.00031.2010 (2011). [DOI] [PubMed] [Google Scholar]
- 5.Nuzzo, J. L. Sex differences in skeletal muscle fiber types: A meta-analysis. Clin. Anat.37, 81–91. 10.1002/ca.24091 (2024). [DOI] [PubMed] [Google Scholar]
- 6.Korhonen, M. T. et al. Aging, muscle fiber type, and contractile function in sprint-trained athletes. J. Appl. Physiol. (1985). 101, 906–917. 10.1152/japplphysiol.00299.2006 (2006). [DOI] [PubMed] [Google Scholar]
- 7.Serrano, N., Hyatt, J. K., Houmard, J. A., Murgia, M. & Katsanos, C. S. Muscle fiber phenotype: a culprit of abnormal metabolism and function in skeletal muscle of humans with obesity. Am. J. Physiol. Endocrinol. Metab.325, E723–E733. 10.1152/ajpendo.00190.2023 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Trappe, S. et al. Human single muscle fibre function with 84 day bed-rest and resistance exercise. J. Physiol.557, 501–513. 10.1113/jphysiol.2004.062166 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Bagley, J. R., Murach, K. A. & Trappe, S. W. Microgravity-Induced fiber type shift in human skeletal muscle. Gravitational Space Biology. 26, 34–40 (2012). [Google Scholar]
- 10.Zierath, J. R. & Hawley, J. A. Skeletal muscle fiber type: influence on contractile and metabolic properties. PLoS Biol.2, e348. 10.1371/journal.pbio.0020348 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Vanhatalo, A. et al. The mechanistic bases of the power-time relationship: muscle metabolic responses and relationships to muscle fibre type. J. Physiol.594, 4407–4423. 10.1113/JP271879 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Tobias, I. S. & Galpin, A. J. Moving human muscle physiology research forward: an evaluation of fiber type-specific protein research methodologies. Am. J. Physiol. Cell. Physiol.319, C858–C876. 10.1152/ajpcell.00107.2020 (2020). [DOI] [PubMed] [Google Scholar]
- 13.Murach, K. A. et al. Fiber typing human skeletal muscle with fluorescent immunohistochemistry. J. Appl. Physiol. (1985). 127, 1632–1639. 10.1152/japplphysiol.00624.2019 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Medler, S. Mixing it up: the biological significance of hybrid skeletal muscle fibers. J. Exp. Biol.10.1242/jeb.200832 (2019). [DOI] [PubMed] [Google Scholar]
- 15.Pette, D. & Staron, R. S. Mammalian skeletal muscle fiber type transitions. Int. Rev. Cytol.170, 143–223. 10.1016/s0074-7696(08)61622-8 (1997). [DOI] [PubMed] [Google Scholar]
- 16.D’Antona, G. et al. The effect of ageing and immobilization on structure and function of human skeletal muscle fibres. J. Physiol.552, 499–511. 10.1113/jphysiol.2003.046276 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kemp, J. G., Blazev, R., Stephenson, D. G. & Stephenson, G. M. Morphological and biochemical alterations of skeletal muscles from the genetically obese (ob/ob) mouse. Int J Obes (Lond) 33, 831–841 (2009). https://doi.org/ijo2009100 [pii] 10.1038/ijo.2009.100. [DOI] [PubMed]
- 18.Williamson, D. L., Gallagher, P. M., Carroll, C. C., Raue, U. & Trappe, S. W. Reduction in hybrid single muscle fiber proportions with resistance training in humans. J. Appl. Physiol. (1985). 91, 1955–1961. 10.1152/jappl.2001.91.5.1955 (2001). [DOI] [PubMed] [Google Scholar]
- 19.Moreillon, M. et al. Hybrid fiber alterations in exercising seniors suggest contribution to fast-to-slow muscle fiber shift. J. Cachexia Sarcopenia Muscle. 10, 687–695. 10.1002/jcsm.12410 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Murach, K. A. et al. Improving human skeletal muscle myosin heavy chain fiber typing efficiency. J. Muscle Res. Cell. Motil.37, 1–5. 10.1007/s10974-016-9441-9 (2016). [DOI] [PubMed] [Google Scholar]
- 21.Pandorf, C. E., Caiozzo, V. J., Haddad, F. & Baldwin, K. M. A rationale for SDS-PAGE of MHC isoforms as a gold standard for determining contractile phenotype. J. Appl. Physiol. (1985). 108, 222–222. 10.1152/japplphysiol.01233.2009 (2010). author reply 226. [DOI] [PubMed] [Google Scholar]
- 22.Krakova, D. et al. Muscle fiber type grouping does not change in response to prolonged resistance exercise training in healthy older men. Exp. Gerontol.173, 112083. 10.1016/j.exger.2023.112083 (2023). [DOI] [PubMed] [Google Scholar]
- 23.Toien, T. et al. The impact of life-long strength versus endurance training on muscle fiber morphology and phenotype composition in older men. J. Appl. Physiol. (1985). 135, 1360–1371. 10.1152/japplphysiol.00208.2023 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Broos, S. et al. Role of alpha-actinin-3 in contractile properties of human single muscle fibers: a case series study in paraplegics. PLoS One. 7, e49281. 10.1371/journal.pone.0049281 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Williamson, D. L., Godard, M. P., Porter, D. A., Costill, D. L. & Trappe, S. W. Progressive resistance training reduces myosin heavy chain coexpression in single muscle fibers from older men. J. Appl. Physiol. (1985). 88, 627–633. 10.1152/jappl.2000.88.2.627 (2000). [DOI] [PubMed] [Google Scholar]
- 26.Giavarina, D. Understanding Bland Altman analysis. Biochem. Med. (Zagreb). 25, 141–151. 10.11613/BM.2015.015 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Moreno-Justicia, R. et al. Human skeletal muscle fiber heterogeneity beyond myosin heavy chains. Nat. Commun.16, 1764. 10.1038/s41467-025-56896-6 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Nederveen, J. P. et al. Variability in skeletal muscle fibre characteristics during repeated muscle biopsy sampling in human Vastus lateralis. Appl. Physiol. Nutr. Metab.45, 368–375. 10.1139/apnm-2019-0263 (2020). [DOI] [PubMed] [Google Scholar]
- 29.Mogensen, M. et al. Mitochondrial respiration is decreased in skeletal muscle of patients with type 2 diabetes. Diabetes56, 1592–1599. 10.2337/db06-0981 (2007). [DOI] [PubMed] [Google Scholar]
- 30.Andersen, J. L., Mohr, T., Biering-Sorensen, F., Galbo, H. & Kjaer, M. Myosin heavy chain isoform transformation in single fibres from m. vastus lateralis in spinal cord injured individuals: effects of long-term functional electrical stimulation (FES). Pflugers Arch.431, 513–518. 10.1007/BF02191897 (1996). [DOI] [PubMed] [Google Scholar]
- 31.Norton, B. J. & Strube, M. J. Understanding statistical power. J. Orthop. Sports Phys. Ther.31, 307–315. 10.2519/jospt.2001.31.6.307 (2001). [DOI] [PubMed] [Google Scholar]
- 32.Long, D. E. et al. Short-term repeated human biopsy sampling contributes to changes in muscle morphology and higher outcome variability. J. Appl. Physiol. (1985). 135, 1403–1414. 10.1152/japplphysiol.00441.2023 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Hughes, M. C. et al. Mitochondrial bioenergetics and fiber type assessments in microbiopsy vs. Bergstrom percutaneous sampling of human skeletal muscle. Front. Physiol.6, 360. 10.3389/fphys.2015.00360 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Horwath, O. et al. Variability in Vastus lateralis fiber type distribution, fiber size, and myonuclear content along and between the legs. J. Appl. Physiol. (1985). 131, 158–173. 10.1152/japplphysiol.00053.2021 (2021). [DOI] [PubMed] [Google Scholar]
- 35.Privett, G. E., Ricci, A. W., Ortiz-Delatorre, J. & Callahan, D. M. Predicting myosin heavy chain isoform from postdissection fiber length in human skeletal muscle fibers. Am. J. Physiol. Cell. Physiol.326, C749–C755. 10.1152/ajpcell.00700.2023 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Greising, S. M., Medina-Martinez, J. S., Vasdev, A. K., Sieck, G. C. & Mantilla, C. B. Analysis of muscle fiber clustering in the diaphragm muscle of sarcopenic mice. Muscle Nerve. 52, 76–82. 10.1002/mus.24641 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Lexell, J., Henriksson-Larsen, K. & Sjostrom, M. Distribution of different fibre types in human skeletal muscles. 2. A study of cross-sections of whole m. vastus lateralis. Acta Physiol. Scand.117, 115–122. 10.1111/j.1748-1716.1983.tb07185.x (1983). [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.