Search

Additional file 4: Table S1. Patient demographics, genotype, and functional outcome measures. Table S2. STIR-positive muscles by baseline STIR hyperintensity levels. Table S3. LMEM results, 6MWT and Go-30-Feet regressed on visit and individual muscle fat fractions. Fat fraction coefficients correspond to IQR change increases.

BETi block DUX4 target gene expression in FSHD myoblasts. (A) The expression level of ZSCAN4 mRNA after 72 h of treatment with the BETi (+)-JQ1, I-BET762, or I-BET151 in 54-2 FSHD1 myoblasts. (B) The BETi RVX-208 inhibits ZSCAN4 and TRIM43 expression in FSHD2 MB200 myoblasts treated for 72 h with EC50s of 350 nM and 280 nM, respectively. Error bars indicate the standard deviation from the mean of three biological replicates. (PDF 53 kb)

The DUX4-responsive ZSCAN4 reporter system used for screening epigenetic modifier compounds. (A) Schematic depicting the ZSCAN4 promoter luciferase reporter vector (top) that includes four tandem DUX4 binding sites (D4BS), and a control reporter in which three of the four D4BS have been mutated (bottom). (B) Activity of the ZSCAN4 reporter vectors in 54-2 FSHD1 and 54-6 control (non-FSHD) myoblasts and 6Â day differentiated myotubes. Error bars indicate the standard deviation from the mean of three biological replicates. (PDF 66Â kb)

DUX4 binding at HS16. UCSC genome track of aligned DUX4 ChIP-sequencing reads from a previously published study (GEO GSE33838 [43]) showing a peak of DUX4 binding at HS16 within the cluster control region of the PCDH cluster. Sequence in the peak region is listed highlighting the DUX4 consensus binding sequence in red. (PDF 151Â kb)

DUX4 binding at HS16. UCSC genome track of aligned DUX4 ChIP-sequencing reads from a previously published study (GEO GSE33838 [43]) showing a peak of DUX4 binding at HS16 within the cluster control region of the PCDH cluster. Sequence in the peak region is listed highlighting the DUX4 consensus binding sequence in red. (PDF 151Â kb)

BETi block DUX4 target gene expression in FSHD myoblasts. (A) The expression level of ZSCAN4 mRNA after 72 h of treatment with the BETi (+)-JQ1, I-BET762, or I-BET151 in 54-2 FSHD1 myoblasts. (B) The BETi RVX-208 inhibits ZSCAN4 and TRIM43 expression in FSHD2 MB200 myoblasts treated for 72 h with EC50s of 350 nM and 280 nM, respectively. Error bars indicate the standard deviation from the mean of three biological replicates. (PDF 53 kb)

Characterization of BET gene expression. (A-C) 54-2 FSHD1 and 54-6 control (non-FSHD, Normal) myoblasts were induced to differentiate into myotubes and gene expression measured. Differentiation efficiency was determined by examining the early differentiation marker MYOG and the late differentiation marker MYH2 (A). As expected, DUX4 targets were strongly induced upon differentiation of FSHD1 but not control myoblasts (B). The levels of BET family member (BRD2, BRD3, BRD4, BRDT) mRNAs are shown in (C). (D) Tracks showing ChIP-seq and RNA-seq reads mapped to the BRDT locus in DUX4-expressing muscle cells. Error bars indicate the standard deviation from the mean of three biological replicates. (PDF 117Â kb)

Primary screening data for the Pharmakon 1600 library. Quantitative PCR detection of DUX4 target gene MBD3L2 is plotted for each library plate (88 compounds per plate). Data for each plate is normalized to no drug controls (n = 8), which were set to 1, as described in Methods. Plate means are indicated with horizontal lines and positive control data from each plate (I-BET762, n = 8) are collectively plotted in the rightmost column (Pos Controls). (PDF 126 kb)

Transcriptional expression of the PCDHγ cluster in control, FSHD1, and FSHD2 individuals. RNA expression analysis of PCDHγ cluster isoform members in primary (A) myoblast cells and (B) myotube cells. Results represent log10 relative expression by qRT-PCR analysis of the indicated gene after normalization to the internal control gene GUS1. For each gene, the value of expression in control individuals was then arbitrarily set to 1. *Bonferroni adjusted P

Transcriptional expression of the PCDHγ cluster in control, FSHD1, and FSHD2 individuals. RNA expression analysis of PCDHγ cluster isoform members in primary (A) myoblast cells and (B) myotube cells. Results represent log10 relative expression by qRT-PCR analysis of the indicated gene after normalization to the internal control gene GUS1. For each gene, the value of expression in control individuals was then arbitrarily set to 1. *Bonferroni adjusted P

Gene-specific primers used for ChIP and qPCR analysis. (PDF 25Â kb)

BETi and beta-2 adrenergic agonists do not suppress the transcriptional activation function of DUX4. 54-6 control (non-FSHD) myoblasts that lack DUX4 were transfected with a DUX4 expression vector and DUX4 activity monitored by measuring mRNA levels of the endogenous DUX4 target gene MBD3L2. Compounds were added 5 h after transfection, a time at which there is little detectable MBD3L2 expression, and mRNA levels assessed at 24 h after transfection. The level of MBD3L2 at 24 h is expressed as ‘Fold Activation’ over the level present at 5 h, which was set to 1. Compounds were added in a 12 point, 3-fold dilution series to cover concentrations well above those required to inhibit DUX4 expression. The data point farthest to the left for each of the compounds represents a DMSO (no drug) control. The approximate EC50s of the beta-2 adrenergic agonist clenbuterol (X) and the BETi I-BET762 (Y) for blocking DUX4 expression in FSHD myoblasts are indicated on the x-axis. (PDF 31 kb)

Characterization of BET gene expression. (A-C) 54-2 FSHD1 and 54-6 control (non-FSHD, Normal) myoblasts were induced to differentiate into myotubes and gene expression measured. Differentiation efficiency was determined by examining the early differentiation marker MYOG and the late differentiation marker MYH2 (A). As expected, DUX4 targets were strongly induced upon differentiation of FSHD1 but not control myoblasts (B). The levels of BET family member (BRD2, BRD3, BRD4, BRDT) mRNAs are shown in (C). (D) Tracks showing ChIP-seq and RNA-seq reads mapped to the BRDT locus in DUX4-expressing muscle cells. Error bars indicate the standard deviation from the mean of three biological replicates. (PDF 117Â kb)

Transcriptional expression of DUX4 target genes and the PCDHγ cluster in SMCHD1 KD 4qB cell lines. RNA expression analysis of (A) the DUX4 target genes ZSCAN4 and LEUTX in primary myotubes and the PCDHγ cluster isoform members in primary (B) myoblast and (C) myotube cells after SMCHD1 shRNA KD or control shRNA KD targeting luciferase and GFP. Results represent qRT-PCR analysis of the indicated gene after normalization to the internal control gene GUS1. For each gene, the value of expression in control shRNA samples was then arbitrarily set to 1. *Bonferroni adjusted P

Differentially methylated regions identified by 450k. (PDF 48Â kb)

Differentially methylated regions identified by RRBS. (PDF 170Â kb)

Differentially methylated regions identified by 450k. (PDF 48Â kb)

Methylation of the TCEB3C cluster and ZAV macrosatellite in FSHD2 and control individuals. Relative methylation levels found in FSHD2 and control individuals by RRBS, displayed as a methylation heatmap with color scale from yellow, no/low number of methylated CpGs at the locus to dark blue, high number of methylated CpGs at (A) the TCEB3C cluster and (B) the ZAV macrosatellite in primary myoblast cells. Bisulfite sequencing of (C) the TCEB3C cluster displayed as a heatmap depicting the average methylation fraction at a specific CpG from ESME analysis, with color scale from yellow to dark blue indicating a low to high percent of methylation, and (D) the ZAV macrosatellite depicted as single molecule clones for individuals, where black circles represent a methylated cytosine and open circles an unmethylated cytocine; gender of the individual is indicated by a red line for females and blue line for males in peripheral blood mononuclear cells. (PDF 109Â kb)

FSHD myoblasts grown continuously in low dose BETi maintain suppressed DUX4 target gene expression but do not re-establish D4Z4 repeat DNA methylation. (A) DUX4 target gene expression in MB200 FSHD2 myoblasts grown for three weeks in media containing DMSO or 100 nM (+)-JQ1 compared to gene expression in 54-6 control (non-FSHD) myoblasts. (B) Structure of the D4Z4 repeat unit showing regions analyzed by bisulfate sequencing in this study (ADS3747 and ADS1454) relative to the DUX4 open reading frame (ORF). ADS3747 spans positions 665-708 and ADS1454 spans positions 2230-2361 with respect to the start of the KpnI site. The locations of previously published methylation-sensitive restriction sites and regions analyzed by bisulfite sequencing (DR1, DR2 and DR3) are indicated. (C-D) Average percent methylation across 9 CpG sites within ADS3747 (C) or 10 CpG sites within ADS1454 (D) in 54-6 control (non-FSHD, Normal), 54-2 FSHD1, and MB200 FSHD2 myoblasts, as well as MB200 FSHD2 myoblasts grown for three weeks in media containing DMSO or 100 nM (+)-JQ1. (E) Slow recovery of DUX4 target gene expression after BETi withdrawal. MB200 FSHD2 myoblasts grown for 3 weeks in 100 nM (+)-JQ1 were split and seeded onto culture plates in the absence of drug at ~10% confluence to allow for continued growth. ZSCAN4 mRNA levels in untreated MB200 FSHD2 myoblasts (Control), MB200 FSHD2 myoblasts maintained continuously in drug (JQ1) and MB200 FSHD2 myoblasts grown for the indicated times after compound withdrawal are shown. Error bars in (A) and (E) indicate the standard deviation from the mean of three biological replicates. (PDF 122 kb)

Gene-specific primers used for ChIP and qPCR analysis. (PDF 25Â kb)

The DUX4-responsive ZSCAN4 reporter system used for screening epigenetic modifier compounds. (A) Schematic depicting the ZSCAN4 promoter luciferase reporter vector (top) that includes four tandem DUX4 binding sites (D4BS), and a control reporter in which three of the four D4BS have been mutated (bottom). (B) Activity of the ZSCAN4 reporter vectors in 54-2 FSHD1 and 54-6 control (non-FSHD) myoblasts and 6Â day differentiated myotubes. Error bars indicate the standard deviation from the mean of three biological replicates. (PDF 66Â kb)

FSHD myoblasts grown continuously in low dose BETi maintain suppressed DUX4 target gene expression but do not re-establish D4Z4 repeat DNA methylation. (A) DUX4 target gene expression in MB200 FSHD2 myoblasts grown for three weeks in media containing DMSO or 100 nM (+)-JQ1 compared to gene expression in 54-6 control (non-FSHD) myoblasts. (B) Structure of the D4Z4 repeat unit showing regions analyzed by bisulfate sequencing in this study (ADS3747 and ADS1454) relative to the DUX4 open reading frame (ORF). ADS3747 spans positions 665-708 and ADS1454 spans positions 2230-2361 with respect to the start of the KpnI site. The locations of previously published methylation-sensitive restriction sites and regions analyzed by bisulfite sequencing (DR1, DR2 and DR3) are indicated. (C-D) Average percent methylation across 9 CpG sites within ADS3747 (C) or 10 CpG sites within ADS1454 (D) in 54-6 control (non-FSHD, Normal), 54-2 FSHD1, and MB200 FSHD2 myoblasts, as well as MB200 FSHD2 myoblasts grown for three weeks in media containing DMSO or 100 nM (+)-JQ1. (E) Slow recovery of DUX4 target gene expression after BETi withdrawal. MB200 FSHD2 myoblasts grown for 3 weeks in 100 nM (+)-JQ1 were split and seeded onto culture plates in the absence of drug at ~10% confluence to allow for continued growth. ZSCAN4 mRNA levels in untreated MB200 FSHD2 myoblasts (Control), MB200 FSHD2 myoblasts maintained continuously in drug (JQ1) and MB200 FSHD2 myoblasts grown for the indicated times after compound withdrawal are shown. Error bars in (A) and (E) indicate the standard deviation from the mean of three biological replicates. (PDF 122 kb)

Percent input ChIP-qPCR for representative individuals. ChIP-qPCR for all ChIP examined loci displayed as a percent of the input for one control, FSHD1, and FSHD2 individual. Please note that it is difficult to compare single copy loci, HS17-17â ˛and HS19-20, to multi-copy loci tRNA and D4Z4. (PDF 831Â kb)

Methylation of the TCEB3C cluster and ZAV macrosatellite in FSHD2 and control individuals. Relative methylation levels found in FSHD2 and control individuals by RRBS, displayed as a methylation heatmap with color scale from yellow, no/low number of methylated CpGs at the locus to dark blue, high number of methylated CpGs at (A) the TCEB3C cluster and (B) the ZAV macrosatellite in primary myoblast cells. Bisulfite sequencing of (C) the TCEB3C cluster displayed as a heatmap depicting the average methylation fraction at a specific CpG from ESME analysis, with color scale from yellow to dark blue indicating a low to high percent of methylation, and (D) the ZAV macrosatellite depicted as single molecule clones for individuals, where black circles represent a methylated cytosine and open circles an unmethylated cytocine; gender of the individual is indicated by a red line for females and blue line for males in peripheral blood mononuclear cells. (PDF 109Â kb)

Summary of material used in the studies. (PDF 43Â kb)

Transcriptional expression of DUX4 target genes and the PCDHγ cluster in SMCHD1 KD 4qB cell lines. RNA expression analysis of (A) the DUX4 target genes ZSCAN4 and LEUTX in primary myotubes and the PCDHγ cluster isoform members in primary (B) myoblast and (C) myotube cells after SMCHD1 shRNA KD or control shRNA KD targeting luciferase and GFP. Results represent qRT-PCR analysis of the indicated gene after normalization to the internal control gene GUS1. For each gene, the value of expression in control shRNA samples was then arbitrarily set to 1. *Bonferroni adjusted P

Percent input ChIP-qPCR for representative individuals. ChIP-qPCR for all ChIP examined loci displayed as a percent of the input for one control, FSHD1, and FSHD2 individual. Please note that it is difficult to compare single copy loci, HS17-17â ˛and HS19-20, to multi-copy loci tRNA and D4Z4. (PDF 831Â kb)

PKA inhibition does not prevent formoterol-mediated inhibition of DUX4 expression. (A-B) Expression of DUX4, DUX4 target genes, and the myogenic marker CKM in MB073 FSHD1 (A) or MB200 FSHD2 (B) myotubes differentiated for 48 h and treated with 1 nM formoterol, 10 uM H89, or both compounds during the final 8 h of culturing. Error bars indicate the standard deviation from the mean of three biological replicates. P-values were calculated using a two-tailed, two-sample Student’s t-test assuming unequal variance. *, p

PKA inhibition does not prevent formoterol-mediated inhibition of DUX4 expression. (A-B) Expression of DUX4, DUX4 target genes, and the myogenic marker CKM in MB073 FSHD1 (A) or MB200 FSHD2 (B) myotubes differentiated for 48 h and treated with 1 nM formoterol, 10 uM H89, or both compounds during the final 8 h of culturing. Error bars indicate the standard deviation from the mean of three biological replicates. P-values were calculated using a two-tailed, two-sample Student’s t-test assuming unequal variance. *, p

Table S1 FSHD-CTRN ReSolve Investigators (DOCX 21 kb)

The effect of formoterol on DUX4-mediated cell death. MB135 control (non-FSHD) myoblasts that stably express a doxycycline-inducible DUX4 transgene [37] were used to test the effect of the beta-2 agonist formoterol on DUX4-mediated cell death. (A) Bright field images showing cell morphology after DUX4 expression at 24 h post doxycycline induction, with DMSO or 1 nM formoterol added during the last 16 h. (B) Cell counts from wells imaged in (A). (C) Expression of the DUX4 transgene and endogenous DUX4 target gene MBD3L2 from cells treated as in (A). (D) Western blot showing expression of exogenous DUX4, endogenous MBD3L2, and endogenous Histone H3 as a loading control from cells treated as in (A). Error bars indicate the standard deviation from the mean of three biological replicates. (PDF 1068 kb)

Checklist supplementary information. Supplementary information to implementing an international, multi-centre, academic-led clinical trial: checklist. (DOCX 115 kb)

Checklist supplementary information. Supplementary information to implementing an international, multi-centre, academic-led clinical trial: checklist. (DOCX 115 kb)

The effect of formoterol on DUX4-mediated cell death. MB135 control (non-FSHD) myoblasts that stably express a doxycycline-inducible DUX4 transgene [37] were used to test the effect of the beta-2 agonist formoterol on DUX4-mediated cell death. (A) Bright field images showing cell morphology after DUX4 expression at 24 h post doxycycline induction, with DMSO or 1 nM formoterol added during the last 16 h. (B) Cell counts from wells imaged in (A). (C) Expression of the DUX4 transgene and endogenous DUX4 target gene MBD3L2 from cells treated as in (A). (D) Western blot showing expression of exogenous DUX4, endogenous MBD3L2, and endogenous Histone H3 as a loading control from cells treated as in (A). Error bars indicate the standard deviation from the mean of three biological replicates. (PDF 1068 kb)

Table S1 FSHD-CTRN ReSolve Investigators (DOCX 21 kb)