Fig 1.
Sample images of ultrasonography diaphragm acquisition.
The figure shows representative frames of the diaphragm acquired in M-mode (A) and B-mode (B), by ultrasonography. Diaphragm amplitude was measured in M-mode as the distance in mm, directly provided by the Vevo2100 software, between the baseline and the peak of contraction. The B-mode acquisition was used to perform echodensity analysis performed by using the Image J software. Mean pixel echodensity was obtained from the analysis of the outlined diaphragm section of constant dimensions of 4514.0 ± 17.6 pixels.
Fig 2.
Analysis of hyperechoic tissue.
In A, is shown a schematic representation of the analysis performed to evaluate the possible contribution of hyperechoic upper layer of tissue (abdominal wall) on signal’s attenuation of deeper layers. In B, is shown a sample image of the algorithm that scans the upper portion of the images, one column at a time, retrieving information from these pixels. The image on the left shows an ultrasonography acquisition where the red cursor indicates the column of pixel analyzed at that moment. The line charts on the right show Normalized Pixel Intensity sampled by the red cursor (up) and 10-Moving Average of Pixel Intensity of pixels sampled by the red cursor (down).
Fig 3.
Measurement of diaphragm amplitude by ultrasonography.
The scatter plot graph illustrates individual mouse values, mean and standard error of the mean (SEM) obtained for diaphragm amplitude in wt (●) and mdx (○) mice at 3 (M3) and 6 (M6) months of age. A statistically significant difference among groups was found by ANOVA (F = 35.7, p < 2.5 x 10−10). Bonferroni post hoc correction for individual differences between groups is as follows: significantly different vs * wt M3 (p < 7.9 x 10−9), # wt M6 (p < 9.9 x 10−8).
Fig 4.
Diaphragm echodensity evaluation.
The scatter plot graph in A shows individual mouse values, mean and standard error of the mean (SEM) obtained for pixel echodensity in wt (●) and mdx (○) mice at 3 (M3) and 6 (M6) months of age. A statistically significant difference among groups was found by ANOVA at M6 (F = 5.08, p < 0.007). Bonferroni post hoc correction for individual differences between groups is as follows: significantly different vs # wt M6 (p < 0.003). In B and C are shown the histograms of pixel distribution (right panel) obtained by measuring echodensity in wt (B) and mdx (C) mice at M6, with sample images used for analysis (left panel). Each histogram represents the number of repetitions (count; reported on the Y axis) of each pixel value (pixel intensity; indicated on the X axis).
Fig 5.
Evaluation of abdominal wall thickness during ultrasound measurements.
Correlation between AW (abdominal wall) Thickness detected with Canny Edge Detection and the Below-AW P.I. (pixel intensity) / AW P.I. ratio (mean). On the top, on the right of the chart, is showed the R2 (r-squared).
Table 1.
Echocardiography.
Table 2.
In vivo force and exercise performance.
Fig 6.
Correlation of diaphragm amplitude with in vivo fatigability and ex vivo force.
In A is shown the linear correlation obtained by plotting individual mouse values from wt (●) and mdx (○) mice for total distance run (m) during the e-test at 3 and 6 months of age, expressed as a function of diaphragm amplitude absolute values (mm) at the same time points. In B and C is shown the linear correlation obtained by plotting individual mouse values from wt (●) and mdx (○) mice at 6 months of age for diaphragm specific isometric tetanic force (sP0, kN/m2), expressed as a function of absolute (B, mm) or body weight normalized (C, mm/g) diaphragm amplitude.
Fig 7.
Diaphragm ex vivo isometric contraction.
The scatter plot graphs in A and B show individual normalized values, mean, and standard error of the mean (SEM) for maximal isometric twitch (A, sPtw measured in kN/m2) and tetanic (B, sP0 measured in kN/m2) tension of diaphragm strips isolated from wt (●) and mdx (○) mice at 3 (M3) and 6 (M6) months of age. In detail, values at M3 were obtained from one cohort of wt and one of mdx mice (n = 9 animals per group, with one wt excluded from data analysis due to experimental issues), dedicated to ex vivo assessments at this time point. Values at M6 were obtained from wt and mdx mice of the main experimental groups (n = 9 for each genotype). Statistically significant differences among groups were found by ANOVA for sPtw (F > 16.6, p < 1.5 x 10−6) and sP0 (F > 27.6, p < 9.3 x 10−9). Bonferroni post hoc correction for individual differences between groups is as follows: significantly different vs * wt M3 (9.9 x 10−6 < p < 0.001), # wt M6 (3.5 x 10−8 < p < 1.5 x 10−6).
Fig 8.
Diaphragm levels of pro-fibrotic marker TGF-β1 and histopathology.
The scatter plot graph in A illustrates individual mouse values, mean and standard error of the mean (SEM) for TGF-β1 levels normalized to total protein content (pg/μg), measured by ELISA in diaphragm samples from wt (●) and mdx (○) mice at 3 (M3) and 6 (M6) months of age. No statistically significant differences among groups were found by ANOVA followed by Bonferroni post hoc correction. The percentage increase of TGF-β1 levels calculated in M3 and M6 mdx mice with respect to age-matched wt is indicated above the bars. In B are shown representative sections of hematoxylin & eosin stained DIA muscles from mdx mice at M3 and M6 (10× magnification). Images are characterized by a disorganized tissue architecture with the presence of centronucleated fibers, inflammatory infiltrates and areas of non-muscle tissue, typical signs of dystrophic histopathology. In C are shown representative sections of DIA muscles from dystrophic mice at M3 and M6 (10× magnification) stained with Masson’s trichrome. The staining allows to discriminate nuclei (dark brown), muscle fibers cytoplasm (pink-red), collagen and bone (blue).