Hrough the peak finding algorithm of ChIPseeqer [15], which identifies peaks with increased Bcl-3 binding compared to weight bearing muscle. By using a low stringency peak height cutoff, 49,000 Bcl-3 peaks were found. These peaks were evenly distributed across the mouse genome (Figure S1). Using a web based tool called Nebula [20,21] a component of the Galaxy suite of programs, the distribution of Bcl-3 peaks from unloaded musclewere compared to random peaks found from the input fraction of chromatin (Figure 1). This showed that the Bcl-3 ChIP had succeeded in enriching many Bcl-3 binding sites (i.e., peaks) near the activation sites of transcription across the entire genome. We then took the sequence alignments from unloaded muscle Bcl-3 ChIP-seq and compared them to weight bearing muscle sequences in order to find peaks that were at least 2-fold increased using the peak finder in ChIPseeqer. By using this level of stringency for peak finding we obtained 2,817 Bcl-3 peaks in unloaded compared to weight bearing muscle. Phastcon analysis using the Cistrome/Galaxy program [21] was used to show that the peaks were located at phylogenetically conserved sites (Figure 2). Annotation of the parts of genes associated with unloading-induced peaks showed that they were mainly inA Bcl-3 Network Controls Muscle AtrophyTable 2. qPCR of selected proteolysis genes with increased Bcl-3 promoter binding.Gene Arih2 Ate1 Fbxo6 Itch Rlim Rnf13 Psmb7 Sod1 Trim63 Ubb Ubr1 Fold change, control vs. unloaded. doi:10.1371/journal.pone.0051478.tFold activation 2.1 1.5 1.8 1.4 1.6 1.5 1.9 1.8 2.0 1.8 1.Also found in the GO pathways and shown in Table 1 are genes that function in the reduction of reactive oxygen species, including SOD1, and several phosphatases. The other GO terms having genes represented are those involved in regulating myogenesis, particularly in the Wnt pathway, and those in glucose metabolism, including glycogen phosphorylase and 7-phosphofructokinase, genes that liberate glucose and control its glycolytic metabolism JI 101 site respectively. Several of the genes, especially the E3 ligases found as Bcl-3 targets by ChIP-seq were subject to qPCR to verify gene activation during unloading and these data are shown in Table 2. The advantage of iPAGE is that it can find the most important functions of the overrepresented genes having peaks with unloading in an unbiased fashion. However, there are other genes with Bcl-3 peaks in the promoter region that are likely to be important to atrophy. For example, several proteolytic pathway genes not identified by iPAGE also show Bcl-3 peaks with unloading (Psmc1, Psmb7, Ube2b, Ubb, Cul4a, Rnf135, Rnf13, Atg3). For transcription factors, Foxo1, Foxo3, and Cebpa show peaks as well as several translation initiating genes including Eif4b and Eif3f. All of the genes with unloading-induced increased Bcl-3 binding in their promoters are listed in Table S1.promoters (Figure 3). We then focused on the peaks in the promoters of the genes found, from 24 to +2 kb relative to the TSS (n = 845).Direct and Indirect Targets of Bcl-Since we were interested in further describing direct and indirect targets of the Bcl-3 transactivator at the genome-wide level, we used the algorithms of ChIPArray [25] to bring together our ChIPseq data on Bcl-3 binding to promoters with the genes whose mRNA was (-)-Calyculin A upregulated as determined by global gene expression array (28,853 transcripts) of control vs. unloaded muscle (Figure 6). ChIPArray found 241 direc.Hrough the peak finding algorithm of ChIPseeqer [15], which identifies peaks with increased Bcl-3 binding compared to weight bearing muscle. By using a low stringency peak height cutoff, 49,000 Bcl-3 peaks were found. These peaks were evenly distributed across the mouse genome (Figure S1). Using a web based tool called Nebula [20,21] a component of the Galaxy suite of programs, the distribution of Bcl-3 peaks from unloaded musclewere compared to random peaks found from the input fraction of chromatin (Figure 1). This showed that the Bcl-3 ChIP had succeeded in enriching many Bcl-3 binding sites (i.e., peaks) near the activation sites of transcription across the entire genome. We then took the sequence alignments from unloaded muscle Bcl-3 ChIP-seq and compared them to weight bearing muscle sequences in order to find peaks that were at least 2-fold increased using the peak finder in ChIPseeqer. By using this level of stringency for peak finding we obtained 2,817 Bcl-3 peaks in unloaded compared to weight bearing muscle. Phastcon analysis using the Cistrome/Galaxy program [21] was used to show that the peaks were located at phylogenetically conserved sites (Figure 2). Annotation of the parts of genes associated with unloading-induced peaks showed that they were mainly inA Bcl-3 Network Controls Muscle AtrophyTable 2. qPCR of selected proteolysis genes with increased Bcl-3 promoter binding.Gene Arih2 Ate1 Fbxo6 Itch Rlim Rnf13 Psmb7 Sod1 Trim63 Ubb Ubr1 Fold change, control vs. unloaded. doi:10.1371/journal.pone.0051478.tFold activation 2.1 1.5 1.8 1.4 1.6 1.5 1.9 1.8 2.0 1.8 1.Also found in the GO pathways and shown in Table 1 are genes that function in the reduction of reactive oxygen species, including SOD1, and several phosphatases. The other GO terms having genes represented are those involved in regulating myogenesis, particularly in the Wnt pathway, and those in glucose metabolism, including glycogen phosphorylase and 7-phosphofructokinase, genes that liberate glucose and control its glycolytic metabolism respectively. Several of the genes, especially the E3 ligases found as Bcl-3 targets by ChIP-seq were subject to qPCR to verify gene activation during unloading and these data are shown in Table 2. The advantage of iPAGE is that it can find the most important functions of the overrepresented genes having peaks with unloading in an unbiased fashion. However, there are other genes with Bcl-3 peaks in the promoter region that are likely to be important to atrophy. For example, several proteolytic pathway genes not identified by iPAGE also show Bcl-3 peaks with unloading (Psmc1, Psmb7, Ube2b, Ubb, Cul4a, Rnf135, Rnf13, Atg3). For transcription factors, Foxo1, Foxo3, and Cebpa show peaks as well as several translation initiating genes including Eif4b and Eif3f. All of the genes with unloading-induced increased Bcl-3 binding in their promoters are listed in Table S1.promoters (Figure 3). We then focused on the peaks in the promoters of the genes found, from 24 to +2 kb relative to the TSS (n = 845).Direct and Indirect Targets of Bcl-Since we were interested in further describing direct and indirect targets of the Bcl-3 transactivator at the genome-wide level, we used the algorithms of ChIPArray [25] to bring together our ChIPseq data on Bcl-3 binding to promoters with the genes whose mRNA was upregulated as determined by global gene expression array (28,853 transcripts) of control vs. unloaded muscle (Figure 6). ChIPArray found 241 direc.
