Figure 5

Ratiometric fura-2 imaging with 10 mM caffeine induction after treatment with p38 inhibitors (A) SB203580 (# of myotubes from left-right: n = 56, 30, 28, 43, 45, 57, 46, 79) and (B) SB202190 (n = 78, 44, 25, 45, 50, 58, 50, 142) or after (C) siRNA targeting p38α (Mapk14), p38β (Mapk11), and p38γ (Mapk12) (n = 17, 27,18, 23, 11, 11, 33, 13). Panels on the left show the peak change in calcium concentration within 30 s of adding caffeine normalized to the average of 10 s of baseline resting free Ca2+ concentration ([Ca2+]i), that is normalized Δ ratio, while plots on the right show calcium levels normalized to [Ca2+]i after addition of caffeine. This data shows that p38 inhibition or knockdown impairs Ca2+ release in wild type myotubes while promoting some Ca2+ release in Ryr1 knockout cells. Note: labels directly above each group indicate the statistical significance compared to DMSO control within the same genotype. Data are presented as mean ± SEM derived from Ca2+ measurements of individual myotubes (Figure 5—figure supplement 3A–C). Corresponding plots showing [Ca2+]i levels without normalization are also provided (Figure 5—figure supplement 4D–F). Statistical analysis by two-way ANOVA followed by Tukey’s multiple comparisons post-test where *p<0.05, **p<0.01, ***p<0.001. (D) Overnight treatment of myotubes from Y522S mutant mice with 10 μM SB203580 significantly reduced a temperature-dependent increase in resting Ca2+ concentration.

Expression Data

Expression Detail
Antibody Labeling
Phenotype Data

Phenotype Detail
Acknowledgments
This image is the copyrighted work of the attributed author or publisher, and ZFIN has permission only to display this image to its users. Additional permissions should be obtained from the applicable author or publisher of the image. Full text @ Elife