DUX4 differentially regulates transcriptomes of human rhabdomyosarcoma and mouse C2C12 cells

Abstract

Facioscapulohumeral muscular dystrophy (FSHD) is linked to the deletion of the D4Z4 arrays at chromosome 4q35. Recent studies suggested that aberrant expression of double homeobox 4 (DUX4) from the last D4Z4 repeat causes FSHD. The aim of this study is to determine transcriptomic responses to ectopically expressed DUX4 in human and mouse cells of muscle lineage. We expression profiled human rhabdomyosarcoma (RD) cells and mouse C2C12 cells transfected with expression vectors of DUX4 using the Affymetrix Human Genome U133 Plus 2.0 Arrays and Mouse Genome 430 2.0 Arrays, respectively. A total of 2267 and 150 transcripts were identified to be differentially expressed in the RD and C2C12 cells, respectively. Amongst the transcripts differentially expressed in the RD cells, MYOD and MYOG (2 fold, p<0.05), and six MYOD downstream targets were up-regulated in RD but not C2C12 cells. Furthermore, 13 transcripts involved in germline function were dramatically induced only in the RD cells expressing DUX4. The top 3 IPA canonical pathways affected by DUX4 were different between the RD (inflammation, BMP signaling and NRF-2 mediated oxidative stress) and the C2C12 cells (p53 signaling, cell cycle regulation and cellular energy metabolism). Amongst the 40 transcripts shared by the RD and C2C12 cells, UTS2 was significantly induced by 76 fold and 224 fold in the RD and C2C12 cells, respectively. The differential expression of MYOD, MYOG and UTS2 were validated using real-time quantitative RT-PCR. We further validated the differentially expressed genes in immortalized FSHD myoblasts and showed up-regulation of MYOD, MYOG, ZSCAN4 and UTS2. The results suggest that DUX4 regulates overlapped and distinct groups of genes and pathways in human and mouse cells as evident by the selective up-regulation of genes involved in myogenesis and gametogenesis in human RD and immortalized cells as well as the different molecular pathways identified in the cells.

Figure 1. Up-regulation of MYOD and MYOG in response to ectopic DUX4 expression in RD cells.

Expression levels of myogenic markers MYOD (A) and MYOG (B) were determined by expression profiling human RD cells transfected with an expression vector either encoding DUX4 or insertless (control), respectively. The differential expression of MYOD (C) and MYOG (D) were validated using real-time qRT-PCR (n = 4). Normalized expression levels of the transcripts were calculated using GAPDH as a reference. ** p<0.01, * p <0.05.

Figure 2. Scatter plot analysis of transcripts regulated by ectopically expressed DUX4 in RD and C2C12 cells.

To clearly visualize transcripts highly induced by DUX4, only transcripts changed >2-fold were used for analysis. Log transformed expression levels of transcripts in cells transfected with the insertless vector was plotted against expression levels of transcripts in cells transfected with the DUX4 expression vector. A cluster of transcripts highly induced by DUX4 (circled) in RD (A.) but not C2C12 cells (B.) was observed.

Figure 3. Up-regulation of human UTS2 and mouse Uts2 was observed in RD and C2C12 cells ectopically expressing DUX4, respectively.

Expression levels of human UTS2 and mouse Uts2 were determined by expression profiling RD (A) and C2C12 (B) cells transfected with an expression vector either encoding DUX4 or insertless (control), respectively. The expression changes in RD (C) and C2C12 (D) were validated using real-time qRT-PCR (n = 4). Normalized expression levels of the transcripts were calculated using GAPDH and Gapdh as a reference in both cell lines. ** p<0.01, * p <0.05.

Figure 4. Up-regulation of MYOD, MYOG, ZSCAN4, and UTS2 in FSHD immortalized cells.

MYOD (A.), MYOG (B.), ZSCAN4 (C.), and UTS2 (D.) levels were quantified in FSHD immortalized cells and control cells using real-time qRT-PCR (n = 4). Values representing expression levels of transcripts were calculated using GAPDH as a reference. ** p<0.01.