Optimization of the fixation, cryoprotection, and sectioning conditions for Fluorescence Imaging of Myofibrils with Image Deconvolution (FIM-ID).

Plantaris muscles from mice were collected and either (A-B) immediately flash-frozen and sectioned at -20° C, or (C-D) subjected to optimized fixation, cryoprotection, and sectioning conditions. Cross-sections of the muscles were stained with phalloidin CF680R to identify f-actin and SERCA 1 was immunolabeled with Alexa 594 to identify the periphery of the myofibrils. Images of SERCA 1 (top) and f-actin (bottom) were captured from identical regions with a widefield fluorescence microscope. (E-F) The images in C-D were subjected to deconvolution. Scale bar in all images = 10 μm.

The effect of sectioning temperature on surface artifacts.

An EDL muscle from a mouse was fixed with 4% paraformaldehyde, cryoprotected with 45% sucrose, and longitudinally sectioned at -15, -20, -25, or -30° C. The sections were stained for total protein and then five randomly selected regions of interest (ROI) were imaged. Each fiber within the ROls was scored for the presence of artifacts on the surface (a score of 1 indicates no artifacts and a score of 10 indicates extensive distortions). (A) Representative images from the surface and mid-depth of the sections. Scale bars = 10 μm. (B) Violin plot of the surface scores for the fibers at each of the different sectioning temperatures. Thick bars represent the median and thin bars represent the quartiles, n = 28-39 fibers / group. Data was analyzed with one-way ANOVA. Significantly different from, * -15°, t -20° C. Scale bar in all images = 10 μm.

Validation of an automated pipeline for measuring myofibril size with FIM-ID.

A cross-section from a mouse plantaris muscle was subjected to FIM-10, and then eight regions of interest (ROI) from a fiber (A) were subjected to automated and manual measurements of myofibril size. (B) Representation of an ROI from the fiber in A, scale bar = 5 μm. (C) Example of how the automated CellProfiler (CP) pipeline identifies the myofibrils (green). (D) Illustration of the myofibrils in CP that met the morphological criteria for subsequent measurements of cross-sectional area (CSA). (E) Example of the same ROI in B-0 that was manually assessed for myofibrils that met the morphological criteria for subsequent measurements of CSA (cyan). (F) Scatter plot of the CSA of all myofibrils that were automatically measured by CP, or manually measured by independent investigators (n = 6 investigators). The black bars represent the mean for each group. (G) For each ROI, the mean myofibril CSA as determined by CP was compared with the mean myofibril CSA from all of the manual measurements. The solid line represents the best fit from linear regression, the dashed lines represent the 95% confidence intervals, R indicates Pearson’s correlation coefficient, and P indicates the likelihood that the relationship is significantly different from zero.

Automated measurements of myofibril size and number with FIM-ID versus manual measurements with electron microscopy.

For each mouse, one plantaris muscle was (A) processed for imaging with electron microscopy (EM), and the contralateral plantaris was (B) processed for FIM-ID, n = 4 mice. In the EM images, glycolytic fibers (Gly) and oxidative fibers (Ox) fibers were distinguished by the presence of extensive intermyofibrillar mitochondria (arrow). In FIM-ID, the autofluorescence signal (Auto) was used to distinguish the Gly vs. Ox fibers. The ‘Auto’ signal was also merged with the signal from SERCA1 to provide a comprehensive decoration of the periphery of the myofibrils. (C-D) Scatter plots that illustrate the relationships between (C) myofibril cross-sectional area (CSA) vs. fiber CSA, and (D) the number of myofibrils per fiber vs. fiber CSA. The automated measurements from the FIM-ID samples are shown in green and the manual measurements from the EM samples are shown in gray. Individual values are presented as triangles for the Gly fibers and circles for the Ox fibers, n = 144-160 fibers / condition (36-40 fibers per muscle). The solid lines represent the best fit from linear regression for each condition, R indicates Pearson’s correlation coefficient, and P indicates the likelihood of a significant difference between the automated FIM-ID and the manual EM measurements. (E-F) The data in C-D was used to compare ratios of (E) the myofibril CSA to fiber CSA, and (F) the myofibril number to fiber CSA. The data is provided for all of the analyzed fibers and also separated according to fiber type. The large dots indicate the mean for each muscle. The data in C and D were analyzed with extra-sums-of-squares F-tests and the data in E and F were analyzed with paired t-tests or a Wilcoxon-matched paired t-test when the test for normality failed (i.e., the EM measurements of oxidative fibers in F).

The radial growth of fibers that occurs in response to chronic mechanical overload is largely mediated by myofibrillogenesis.

(A) The plantaris muscles of mice were subjected to a chronic mechanical overload (MOV) or sham surgery, allowed to recover for 16 days, and then mid-belly cross sections were subjected to FIM-I0. (B) lmmunolabeling for dystrophin was used to identify the periphery of muscle fibers, autofluorescence (Auto) was used to distinguish glycolytic fibers (Gly) from highly oxidative fibers (Ox), and the signal from SERCA1 plus Auto was merged and used to identify the periphery of the myofibrils, scale bar= 25 μm. (C-G) Graphs contain the results from all of the fibers that were analyzed, as well as the same data after it was separated according to fiber type. (C) fiber CSA, (D) the area per fiber occupied by myofibrils, (E) the area per fiber occupied by intermyofibrillar components, (F) myofibril CSA, and (G) the number of myofibrils per fiber. The data are presented as the mean ± SEM, n = 11-15 muscles/ group (All = 24-54 fibers / muscle, Ox and Gly = 10-24 fibers / muscle, Gly = 10-24 fibers / muscle, and an average of 356 ± 17 myofibrils / fiber). Student’s t-tests were used to analyze the data in the ‘All’ graphs and two-way ANOVA was used to analyze the data in the fiber type graphs. Insets show the P values for the main effects of MOV, fiber type (FT), and the interaction (Int).* Significant effect of MOV, P < 0.05.

The radial growth of fibers that occurs in response to resistance exercise is largely mediated by myofibrillogenesis.

(A) Biopsies of the vastus lateralis were collected before (PRE) and after (POST) participants performed 7 weeks of progressive resistance exercise (RE). (B) Cross-sections were immunolabeled for dystrophin (to identify the periphery of muscle fibers), SERCA1 (to identify the periphery of the myofibrils in Type II fibers), or SERCA2 (to identify the periphery of the myofibrils in Type I fibers), and subjected to FIM-ID, scale bar = 50 μm. (C-G) Graphs contain the values for each subject expressed relative to their PRE sample. (C) fiber CSA, (D) the area per fiber occupied by myofibrils, (E) the area per fiber occupied by intermyofibrillar components, (F) myofibril CSA, and (G) the number of myofibrils per fiber. The data are presented as the mean ± SEM, n = 7 participants (SERCA1 15-33 fibers/ participant, SERCA2 8-38 fibers / participant, and an average of 1101 ± 60 myofibrils / fiber). Significance was determined by repeated measures two-way ANOVA. Insets show the P values for the main effects of RE, fiber type (FT), and the interaction (Int). * Significant effect of RE, P < 0.05.