TSS and DEGs

To aggregate the CAGE tag start sites (CTSS) into peaks of TSS we employed the DPI1 clustering tool. After applying a permissive threshold (see Methods section and Supplementary data) there were 170 875 TSS, associated with 13337 genes in total in slow muscle, and 126 885 TSS associated with 11915 genes in fast muscle (for details see ST2). Processes of atrophy and recovery in both muscles were associated with strong changes in RNA transcription profile. The most dramatic changes in gene expression were observed in the slow muscles on the first day of recovery, both in phase vs. control (see ST3 A and B) and time-course (see ST3 C and D) comparisons. In contrast, the fast muscle showed comparatively weaker changes in expression during the whole recovery process. To note, the differentially expressed gene (DEG) composition signatures of three different recovery phases of the fast muscle had only a few overlapping features. Both phase-control and time course signatures performed well as phase classifiers, except for one animal measured on the last day of recovery. The same applied to the statistics of the DEGs. Most of the peak signatures were common to both phase-control and time-course comparison (Figure 2B) and the common set formed three clusters of expression across all experimental phases (Figure 2A).

A B

Figure 2. A. Clusterization of differentially expressed peaks (both genic and intergenic) common to both phase-control and time-course comparisons. FDR threshold 5e-04. Clusters of the metabolism regulation through MAP kinase activity, actin biosynthesis, nucleosome assembly, and lipid metabolism are shown in black, green, blue, and red, respectively (See SF). B. Common and unique differentially expressed peaks between phase-control (Ph-Ct) and time course (TC) comparisons. Soleus muscle. FDR threshold 5e-04

The differentially expressed genes in the slow muscle on the first day of recovery also outnumbered any other phase of the experiment by several-fold. Interestingly, in the fast muscle, the number of DEGs on the first day of recovery was dramatically lower than on the other phases of the experiment, except for the last, seventh day of recovery (Figure 3A)

Figure 3. Differentially expressed genes (DEGs) and differentially expressed enhancers in soleus and EDL muscles. A. Number of DEGs. B. Number of differentially expressed enhancers. C. DEGs in the fast and slow muscles (as compared to control) on days 1, 3 and 7 of disuse (shown as D1, D3, D7, respectively) and on days 1, 3, and 7 of recovery (shown as R1, R3, R7, respectively). D. Differentially expressed enhancers in the fast and slow muscles (as compared to control) on days 1, 3 and 7 of disuse (shown as D1, D3, D7, respectively) and on days 1, 3, and 7 of recovery (shown as R1, R3, R7, respectively).

Remarkably, most DEGs, as in recovering fast muscles, were specific to a certain phase of recovery (Figure 3C) which was not the case in disused fast muscles or both disused and recovering slow muscles. Such divergence suggests phase specificity of pathways involved in the recovery of fast muscle. In contrast, most time-course DEGs in slow muscles on the third day of recovery compared to the first day of recovery were the same as on the first day of recovery compared to the last day of disuse (see ST4).