Skip to main content
PMC home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2007 Nov 5;104(46):18157–18162. doi: 10.1073/pnas.0708659104

DUX4, a candidate gene of facioscapulohumeral muscular dystrophy, encodes a transcriptional activator of PITX1

Manjusha Dixit *, Eugénie Ansseau †, Alexandra Tassin †, Sara Winokur ‡, Rongye Shi *, Hong Qian *, Sébastien Sauvage †, Christel Mattéotti †, Anne M van Acker §, Oberdan Leo §, Denise Figlewicz , Marietta Barro ‖, Dalila Laoudj-Chenivesse ‖, Alexandra Belayew †, Frédérique Coppée †,††, Yi-Wen Chen *,**,††
PMCID: PMC2084313  PMID: 17984056

Abstract

Facioscapulohumeral muscular dystrophy (FSHD) is an autosomal dominant disorder linked to contractions of the D4Z4 repeat array in the subtelomeric region of chromosome 4q. By comparing genome-wide gene expression data from muscle biopsies of patients with FSHD to those of 11 other neuromuscular disorders, paired-like homeodomain transcription factor 1 (PITX1) was found specifically up-regulated in patients with FSHD. In addition, we showed that the double homeobox 4 gene (DUX4) that maps within the D4Z4 repeat unit was up-regulated in patient myoblasts at both mRNA and protein level. We further showed that the DUX4 protein could activate transient expression of a luciferase reporter gene fused to the Pitx1 promoter as well as the endogenous Pitx1 gene in transfected C2C12 cells. In EMSAs, DUX4 specifically interacted with a 30-bp sequence 5′-CGGATGCTGTCTTCTAATTAGTTTGGACCC-3′ in the Pitx1 promoter. Mutations of the TAAT core affected Pitx1-LUC activation in C2C12 cells and DUX4 binding in vitro. Our results suggest that up-regulation of both DUX4 and PITX1 in FSHD muscles may play critical roles in the molecular mechanisms of the disease.

Keywords: D4Z4, expression profiling, homeodomain, atrophy

Facioscapulohumeral muscular dystrophy (FSHD) is the third most common inherited muscular dystrophy. Patients show progressive weakness and atrophy of the muscles in the face, upper arms, and shoulder girdle to lower limbs, and a right/left asymmetry of onset is common (1, 2). The disease is autosomal dominant and is associated with shortening of the D4Z4 repeat array from 11–100 to 1–10 copies in the 4q35 subtelomeric region (1, 3, 4). Several molecular mechanisms have been proposed to explain how the deletion might activate transcription of genes in the region (5–8, 13). Candidate genes that might contribute to the FSHD phenotype were reported in several studies. FSHD region gene 1 (FRG1), FSHD region gene 2 (FRG2), and adenine nucleotide translocator 1 (ANT1) were reported up-regulated in FSHD muscles (6), although other studies showed controversial results (7, 9, 10). Transgenic animals with Frg1 overexpression show a muscular dystrophy phenotype (11). However, it is not clear whether the mouse is a FSHD model, and it is generally considered that other genes in the region contribute to the disease (1).

Although there is consensus that FSHD is a disorder of transcription regulation, the molecular pathways leading to muscular dystrophy and other unique clinical features of the disease are far from clear, including the asymmetric distribution of muscle weakness. Each D4Z4 repeat unit contains an ORF with a double homeobox putatively encoding the DUX4 protein (12, 40). Although initially considered as “junk” DNA, the DUX4 ORF was recently shown to be conserved in evolution for >100 million years (14). Because there are hundreds of homologous 3.3-kb elements with highly similar DUX genes dispersed in the human genome (15), it has been very difficult to specifically amplify DUX4 mRNA against the background of other DUX mRNAs unlinked to FSHD (16). Therefore, it is not known whether DUX4 is expressed and activated by the array contraction, and whether it is involved in FSHD.

In this study, we tested the hypothesis that genes which show disease-specific changes in patients with FSHD are more likely to be involved in the early stages of disease progression. By comparing the genome-wide expression profiles of FSHD to those of 11 other neuromuscular disorders, we identified that paired-like homeodomain transcription factor 1 (PITX1) and four other genes were specifically up-regulated in muscle biopsies of FSHD patients. We further established the regulatory relationship between the DUX4 gene within the D4Z4 repeat unit and the PITX1 gene.

Results

PITX1 Is Specifically Up-Regulated in Muscles of Patients with FSHD.

In this study, we hypothesized that genes involved in the early steps of FSHD were likely to be disease-specific. By comparing expression profiles of FSHD to those of 11 other neuromuscular diseases, we can filter out expression changes reflecting secondary pathological processes, such as degeneration, regeneration, inflammation, and fibrosis formation. After expression profiling both clinically affected and less-affected muscles of patients with FSHD, we first identified genes differentially expressed in the FSHD muscles comparing with the normal controls (P < 0.05). We then compared the FSHD profiles vs. the other 11 neuromuscular disorders as a group and identified five genes specifically changed in FSHD [supporting information (SI) Table 1].

Among the five genes specifically changed in FSHD, PITX1 showed the most dramatic up-regulation in both the unaffected (17-fold, P = 2 × 10−4) and affected (12-fold, P = 7.7 × 10−6) muscles of patients with FSHD (Fig. 1). We validated this result by real-time quantitative RT-PCR in FSHD muscle biopsies. This analysis showed PITX1 was 11-fold up-regulated in FSHD unaffected muscles (P = 2 × 10−5) and 24-fold in affected muscles (P = 2 × 10−3) compared with muscles from healthy volunteers.

Fig. 1.

Fig. 1.

Open in a new tab

PITX1 is specifically up-regulated in FSHD compared with 11 other neuromuscular diseases. The expression level of the gene with arbitrary unit was determined by either dCHIP or Affymetrix MAS5.0. The sample size of each disease was normal healthy muscle (NHM), n = 15; juvenile dermatomyositis (JDM), n = 25; human spastic paraplegia (HSP), n = 4; FSHD, unaffected, n = 5, affected, n = 9; fukutin-related protein deficiency (FKRD), n = 7; Emery–Dreifuss muscular dystrophy, lamin A/C deficiency) (ED-L), n = 4; Emery–Dreifuss muscular dystrophy, emerin deficiency (ED-E), n = 4; dysferlinopathy (DYSF), n = 10; Duchenne muscular dystrophy (DMD), n = 10; Calpain-3 deficiency (CALP), n = 10; Becker muscular dystrophy (BMD), n = 5; acute quadriplegic myopathy (AQM), n = 5; and amyotrophic lateral sclerosis (ALS), n = 9.

To determine whether the genes at the 4q35 region were differentially expressed in muscles of patients with FSHD, we examined the expression data of genes with probe sets designed on the U133A microarrays. The profile data showed that the mRNA expression levels of DUX4, FRG1, tubulin, β polypeptide 4, member Q (TUBB4Q), and ANT1 were not different among clinically affected FSHD, clinically unaffected FSHD, and control samples at the mRNA level. No expression of DUX4 and TUBB4Q was detected by the microarrays, whereas the expression of FRG1 and ANT1 was detected (Affymetrix present calls). It is important to note that because of the existence of numerous homologues in the human genome, the design of probe sets specifically targeting the DUX4 and FRG1 genes on chromosome 4q35 is unlikely. To verify the specificity of the Affymetrix probe sets representing DUX4 and FRG1 genes, we BLAST searched the Affymetrix target sequences and individual probe sequences against the nonredundant (nr) database at National Center for Biotechnology Information (www.ncbi.nlm.nih.gov/BLAST) and confirmed that the probe sets were not specific to 4q35. The target sequence of DUX4 aligned with homologues at chromosomes 2, 4, 10, 18, 22, and Y, whereas the target sequence of FRG1 aligned with homologues on chromosomes 4, 9, 20, and 22. A BLAST search of individual probe sequences gave similar results.

DUX4 Is a Transcription Activator of PITX1.

PITX1 is a member of the paired family of homeodomain transcription factors (17). That the PITX1 gene is located on chromosome 5q31, not 4q35, suggested that the change of PITX1 and other genes with FSHD-specific changes may occur downstream of one or few genes adjacent to the D4Z4 arrays, namely FRG1, FRG2, and DUX4. Homeobox-containing genes are often part of regulatory networks involved in embryonic development and express transcription factors that can regulate other homeobox genes in the network (18). Because both PITX1 and DUX4 are homeodomain transcription factors, we hypothesized that DUX4 was an upstream regulator of PITX1. To test this hypothesis, we first searched the PITX1 promoter region and looked for putative DUX4-binding site. Three guidelines were used: (i) the site might have a core motif (TAAT) often found in homeodomain-binding site, (ii) the sequence would be in an evolutionary conserved region, and (iii) the sequence might share similarity with the homologous DUX1-binding site (19). We identified a putative binding site ≈500 bp upstream from the transcription start site (SI Fig. 6). The “Vertebrate Multiz Alignment and Conservation” function of the University of California, Santa Cruz, genome browser (http://genome.ucsc.edu) was used to determine that the region was highly conserved among six species (mouse, dog, rabbit, rat, human, and opossum). In addition to the TAAT core motif, the site shares an adjacent 2 bp of exact match with the DUX1-binding sequence (SI Fig. 6).

To test whether the putative DUX4-binding site was functional, we first amplified a 369-bp mouse Pitx1 promoter fragment containing the site and cloned it into the pGL3-basic luciferase reporter vector. Cotransfection of C2C12 cells with the pGL3-Pitx1 promoter construct (Pitx1-LUC) and a pCIneo-DUX4 expression vector resulted in 7.4-fold increase in luciferase activity at 24 h (Fig. 2a) compared with samples cotransfected with the insertless pCIneo vector (n = 4, P = 1.6 × 10−21). Cotransfection of Pitx1-LUC and pClneo-DUX4c expressing a 4q35 homologue of DUX4 led to a 4.3-fold (P = 0.002) up-regulation of the luciferase activity, but no significant change was observed with pCIneo-DUX1 that expressed a non4q35 homologue.

Fig. 2.

Fig. 2.

Open in a new tab

DUX4 activates the Pitx1 promoter in transient expression and specifically interacts with a cis element in the Pitx1 promoter region. (a) C2C12 cells were cotransfected with the pCIneo expression vector encoding DUX4, DUX4c, or DUX1, as indicated, and a Pitx1-LUC reporter vector where the Pitx1 promoter is either wild type (WT) or mutated (M; TAAT to TACC in the putative homeodomain-binding site). Luciferase activity was assayed on cells lysed 24 h after transfection. Data are provided as fold induction compared with luciferase activities obtained in cotransfection of the Pitx1-LUC vectors with the insertless pCIneo vector. (b) EMSA was performed with nuclear extract prepared from C2C12 cells at 16 h after transfection. Wild-type probe (lanes 1–3, 5, 6, 8, and 9) and mutated probes (lanes 4, 7, and 10) were end-labeled and incubated with nuclear extracts from cells transfected with pCIneo-DUX4 (lane 2–4), -DUX4c (lanes 5–7), or -DUX1 (lanes 8–10). Supershifts were determined by incubating probe and protein complexes with mAb 9A12 that recognizes DUX4 and DUX4c (lanes 3, 6, and 9). Small arrows in lanes 2 and 6 and arrowheads in lanes 3 and 6 indicate shifts and supershifts, respectively.

To determine the specificity of the interaction between DUX4 and the putative binding site, we mutated the TAAT sequence to TACC in the Pitx1-LUC. Cotransfection of this mutated vector with pClneo-DUX4 reduced luciferase activity 4-fold (P = 1.7 × 10−17). Cotransfection with pClneo-DUX4c also led to an ≈2-fold decrease in luciferase activity (P = 0.047), whereas DUX1 showed no effect as expected.

To evaluate direct and specific binding between the DUX4 protein and its putative binding site in the Pitx1 promoter region, a 30-bp oligonucleotide probe 5′-CGGATGCTGTCTTCTAATTAGTTTGGACCC-3′ containing the putative binding site was tested in EMSA. When the probe was incubated with nuclear extracts of C2C12 cells transfected with pClneo-DUX4 or DUX4c, complexes were formed (Fig. 2b). Addition of the 9A12 mAb directed against DUX4 and DUX4c caused further retardation of the complexes and confirmed the presence of either protein. A probe with the two mismatches could still form the specific complexes (data not shown). However, a probe with six mutated base pairs abolished the binding of both DUX4 and DUX4c. No binding or supershift was observed with extracts containing DUX1.

Many homeodomain proteins bind to the TAAT core motif in vitro with similar affinities, whereas their specificity is regulated in vivo through interaction with different protein partners that help discriminate TAAT sites in different sequence contexts (20–24). To show that DUX4 and DUX4c did not interact with just any TAAT motif, we used another 30-bp oligonucleotide probe containing the other TAAT site in the same evolutionary conserved region (SI Fig. 6) and indeed observed no complex formation with DUX4, DUX4c, or DUX1 (data not shown).

To evaluate whether DUX4 could also activate the endogenous Pitx1 gene, we transfected C2C12 cells with pClneo-DUX4 and performed a coimmunofluorescence staining of DUX4 and Pitx1. Both proteins were detected in the same nuclei 24 h after transfection, whereas only background signals were observed in cells transfected with insertless pClneo (Fig. 3). These experiments were performed in the presence of a proteasome inhibitor, because we found that Pitx1 contained a PEST motif, which is a sequence enriched in proline (P), glutamic acid (E), serine (S), and threonine (T). Such a motif targets proteins for rapid degradation by the proteasome (25). Indeed, in the absence of a proteasome inhibitor, Pitx1 was detected at a later time (36 h) (data not shown).

Fig. 3.

Fig. 3.

Open in a new tab

DUX4 induces expression of endogenous Pitx1 gene in C2C12 cells. C2C12 cells were transfected with the pCIneo-DUX4 expression vector (Left), the insertless pCIneo as a negative control (Center), or a Pitx1 expression vector as a positive control (Right). Double immunofluorescence was performed 24 h after transfection with mAb 9A12 and a secondary antibody coupled to FITC for DUX4 (green, Upper) and with a rabbit serum raised against a Pitx1-specific peptide and secondary antibodies coupled to Texas red (Lower). Arrows point to three nuclei coexpressing DUX4 and Pitx1. (Scale bar, 20 μm.)

Characterization of the DUX4 Gene.

To characterize the 5′ end of the transcript, we performed 5′-RACE by using nested PCR with primers nos. 68 and 73 (Fig. 4a; SI Table 2), which yielded a 400-bp fragment from both control and FSHD samples. After cloning the amplicons into the pCR4 vector, 9 FSHD and 17 control inserts were sequenced and found to be identical to the DUX4 but not the DUX4c genomic sequence. All 5′-RACE products corresponded to a single start site mapped 48 bp downstream from the TACAA box (position 10,732 in GenBank no. AF117653) within a typical initiator sequence (GCACCTG) overlapping with a putative E box upstream of the ORF start codon (Fig. 4a). This site was previously detected in C2C12 cells transfected with a genomic fragment containing one D4Z4 unit (26).

Fig. 4.

Fig. 4.

Open in a new tab

Determination of the DUX4 mRNA 5′ and 3′ends. (a) Schematic representation of the DUX4 promoter with the putative CATT, GC, TACAA, and E boxes, the ATG translation start codon, and the primers used in 5′-RACE (arrows, nos. 68 and 73). The two 5′ ends found previously in total RNAs of C2C12 cells transfected with pGEM42 are indicated. One of these is identical to the single transcription start site determined here on total RNAs of FSHD and control primary myoblasts (*). (b Upper) Scheme of the EcoRI genomic fragment end as cloned in pGEM42 with the stop codon of the DUX4 ORF and the pLAM region. The KpnI and EcoRI restriction sites, the polyA addition signal (ATTAAA), and the primers used in 3′-RACE (arrows, nos. 94 and 95) are indicated. (Lower) Mapping of the ends obtained by 3′-RACE, with introns A and B.

To characterize the 3′ end of the DUX4 mRNA, we first set up 3′-RACE conditions on total RNA extracted from C2C12 cells transfected with pGEM42 that contains the 13.5-kb genomic fragment of a patient with two D4Z4 repeats (12). This approach avoided the transcription background of hundreds of human DUX genes not linked to FSHD. We used an oligo dT containing primer for the RT step and obtained two 3′-RACE products of ≈490 and 350 bp that were cloned in pCR4. Their sequences demonstrated alternative splicing of a 136-bp intron between the stop codon and the end of the D4Z4 unit (Fig. 4b). In addition, all of the products had spliced out a 345-bp intron located in the pLAM region that flanks the telomeric side of the D4Z4 array. Both splicing variants have heterogeneous 3′ ends between positions 12873 and 12883 (in GenBank accession no. AF117653), that most likely resulted from the use of a polyadenylation signal in the pLAM sequence (ATTAAA, 12852–12858 in GenBank accession no. AF117653). This signal was predicted by the FGENESH gene-finding program (27). By using the 3′-RACE, both transcripts were then detected in total RNAs extracted from FSHD myoblast primary cultures but not controls.

In conclusion, the DUX4 gene showed alternatively spliced introns in its 3′ untranslated region. Its heterogeneous mRNA 3′ ends mapped downstream of the ORF, outside of the D4Z4 unit, in the pLAM region that contributed an additional intron to the DUX4 gene in the most distal D4Z4 unit. In addition, our data suggest the DUX4 gene in the most distal D4Z4 element is likely to be the only one that could be transcribed into a polyadenylated RNA in vivo and yield a protein product.

Expression of DUX4 mRNAs in Primary Myoblasts of Patients with FSHD.

To evaluate expression of full-length DUX4 mRNA, total RNA was extracted from FSHD (5 and 7 D4Z4 units) and control myoblast cultures, reverse-transcribed with a primer (no. 407) located in pLAM, 3′ of both introns, and amplified by PCR with an additional primer (no. 222; SI Table 2) located upstream of the ORF start codon (Fig. 5a). A strong band of 2,000 and a weaker one of 1,700 bp were observed in cells transfected with pGEM42 (Fig. 5b). No fragment was obtained either upon omission of reverse transcriptase or in RNA extracted from cells transfected with the insertless vector. The 1,700-bp fragment and a weak signal at ≈2 kb were observed in differentiating FSHD myoblasts but not controls. The products were cloned in pCR4, and only the sequences corresponding to the 1,700-bp fragment were found to be identical to DUX4; they had all spliced out the intron located in the pLAM region. Identical sequences were also obtained on clones derived from two additional FSHD myoblast lines. No sequence identical to DUX4 was found in any control myoblast sample. The observed intron splicing clearly showed our amplified products derived from bona fide mRNAs and not from contaminating DNA.

Fig. 5.

Fig. 5.

Open in a new tab

Detection of the DUX4 mRNAs in FSHD primary myoblasts. (a) Schematic representation of the DUX4 RNAs with alternative 3′ ends and the primers used for RT (no. 407) and PCR (nos. 222 and 407). (b) For controls, RT-PCR was performed on 3–4 μg of total RNA extracted from C2C12 cells transfected with the pGEM7Z vector containing either no insert (lanes 2 and 3) or the 13.5-kb genomic fragment of a patient with 2 D4Z4 units (pGEM42, lanes 4 and 5). RT-PCR was similarly performed on total RNA of control (N036, 9719, and C20) and FSHD primary myoblasts (F22, 5 D4Z4 units; M038, 7 units) either in proliferation (lanes 6–9) or in differentiation (diff) (lanes 10–13). RNA samples were incubated (+) or not (−) with DNase I and RT, as indicated. As a positive control (lane 14), PCR was performed on the pGEM42 vector present in a control sample (as in lanes 4 and 5) not treated with DNase I.

Expression of the DUX4 Protein in Primary Myoblasts of Patients with FSHD.

Conceptual translation of the DUX4 ORF (GenBank accession no. AF117653) provided a 424-residue protein, a size larger than initially reported following correction of sequencing errors in very GC-rich D4Z4 (12). We raised a mAb (9A12 mAb) against the 253 last residues of DUX4 expressed as a recombinant protein in E. coli and found that it cross reacted with DUX4c but no other DUX protein.

The antibody specificity was shown on extracts of human rhabdomyosarcoma TE671 cells transfected with pCIneo-DUX4 or -DUX4c (SI Fig. 7a). The 52-kDa DUX4 and 47-kDa DUX4c bands disappeared upon competition with a 10-fold excess of the recombinant protein. A band with the same apparent molecular weight as DUX4 but not DUX4c could be observed in the FSHD but not in the control sample (SI Fig. 7b). To evaluate mAb 9A12 sensitivity, we performed a similar Western blot with decreasing amounts of FSHD myoblast extract (SI Fig. 7c Top). DUX4 was noticed in all lanes but not in 30-μg control myoblast extract. The blot was stripped and incubated with a rabbit serum raised against a DUX4-specific peptide. As found in a previous study, this serum detected DUX4 only in extracts of cells transfected with pCIneo-DUX4 but not in myoblasts (SI Fig. 7c Middle) (26). We performed additional Western blots with mAb 9A12 and again detected DUX4 in 40-μg extracts of four FSHD but not one additional control myoblast lines (SI Fig. 7d). We detected DUX4 protein expression in four additional FSHD lines but not in two other control myoblast lines (data not shown). Densitometric quantitation of DUX4 expression levels failed to show a correlation with D4Z4 copy number in these samples.

Discussion

FSHD is believed to be a disorder of gene regulation. Although the genetic defect linked to the disease has been known for 15 years, the molecular pathways leading to muscular dystrophy and other unique clinical features are far from clear (3). By comparing with other muscular dystrophies, one may filter out shared secondary molecular pathways such as degeneration/regeneration, inflammatory infiltration, and fibrosis. Therefore, the genes that show disease-specific changes in patients with FSHD are more likely to be involved in the early stages of the disease progression. Indeed, we identified PITX1 specifically up-regulated in the patients with FSHD. PITX1 is a member of a paired family of homeoproteins and is involved in specification of hind limb identity. Pitx1-deleted mice exhibit striking abnormalities in morphogenesis and growth of the hind limb, resulting in structural changes in both muscle and bone to more closely resemble the corresponding forelimb structures (28, 29). The gene shows a similar expression pattern in chick, and the targeted ectopic expression of chick Pitx1 in the developing wing bud causes several morphological changes that affect the appearance, digital patterning, and muscle organization in a way that more resembles the hind limb (30). In addition to affecting anterior/posterior limb development, Pitx1 has also been shown to affect left–right symmetry (28, 31–33). FSHD differs from all other types of muscular dystrophy in both the pattern of muscle involvement (head and shoulder girdle) and the marked right/left asymmetry seen in many patients. Our finding of aberrant regulation of the PITX1 gene provides a potential molecular basis for the enigmatic presentation and progression of FSHD. Recent studies reporting that Pitx1 regulates expression of the IFNα gene in inflammatory cells and is a suppressor of RAS and tumorigenicity suggest the gene could be contributing to other clinical aspects of FSHD (34–36), including the involvement of atypical inflammatory responses and defects in myoblasts reported previously (37–39).

This study reports introns in the DUX4 gene and identifies a direct transcription target of the DUX4 protein. Detection of DUX4 expression has been unsuccessful in the past (10, 15, 40–43). SI Table 3 presents a step-by-step comparison of the unsuccessful protocols with ours. In our study, a single initiation site at the 5′ end of the gene and alternative spliced isoforms of DUX4 transcripts were identified in FSHD myoblasts. These transcripts likely use a polyadenylation site in the pLAM region. Although we detected the 5′ end of DUX4 mRNAs in both FSHD and control myoblasts, the 3′ end sequence specific to the most distal unit of DUX4 was detected only in patients' samples. This observation suggested that, whereas other DUX4 copies might be transcribed, only the gene in the most distal D4Z4 unit could extend to a polyadenylation signal present in the flanking pLAM region. Therefore, these polyA+ DUX4 mRNAs are likely to be the only ones with an extended half life to be translated into a protein product in vivo. This could explain why at least one or part of a D4Z4 unit with one full copy of the DUX4 gene is required on the FSHD allele to cause the disease. Interestingly, some of the differences between the 4qA and 4qB alleles are found in the region just distal to the D4Z4 array, which may affect both the splicing and polyadenylation signals (44, 45). Contractions of the D4Z4 repeat array are associated with the FSHD phenotype only when they occur on the 4qA allele (45, 46). It will be interesting to investigate whether the distal D4Z4 unit on a contracted 4qB allele is able to produce a protein product. The 10q26 region presents high sequence similarity to 4q35, including a D4Z4-like repeat array with DUX10 genes encoding a protein identical to DUX4 (47). Contractions of this repeat array are not associated with FSHD (44). Interestingly, none of our 3′ RACE or RT-PCR products contained the BlnI restriction site characteristic of the D4Z4-like array in 10q26, suggesting no stable DUX transcript could derive from chromosome 10.

We showed by transient cotransfection with a Pitx1-LUC reporter and DUX expression vectors that DUX4 could activate transcription about twice as much as DUX4c, although both proteins could similarly bind to the Pitx1 promoter element in EMSA. Similar DNA binding is expected, because these proteins are identical over their first 342 residues, including the double homeodomain. However, in contrast to DUX4c, DUX4 presented a strong transcriptional activity in the yeast one-hybrid system that was mediated by its unique C-terminal (tail) region (ref. 47 and unpublished data). This difference in transcriptional activity was much stronger in the yeast system than was observed here in mammalian muscle cells and probably reflects the nature of cofactors within the transcriptional complex (20–24). Low expression of a transcription factor is an event frequently occurring as an initial trigger leading to differentiation during embryonic development. Similarly, low levels of a potent trans-factor such as DUX4 could activate expression of a gene encoding another trans-factor such as PITX1 that would be produced at higher levels and could itself activate other genes. Such transcription factor cascades are well known for homeoproteins and could contribute to the mechanism of FSHD.

The lack of Pitx1-LUC activation by DUX1 is correlated with its lack of interaction with the Pitx1 promoter site as confirmed by EMSA. DUX1 is a transcription factor with unknown function that binds to a palindromic TAAT/ATTA site instead of a single TAATTA as found here for DUX4/DUX4c. This difference in binding specificity could be explained by sequence divergence in their DNA-binding domains. The first and second homeodomains present 11 and 2 mismatches in DUX4/DUX4c vs. DUX1, respectively.

In summary, we identified PITX1 as showing aberrant transcriptional regulation in FSHD compared with 11 other neuromuscular disorders. We further showed that the DUX4 protein encoded by a gene mapped within the D4Z4 repeat, interacted in vitro with a cis-element in the Pitx1 promoter region and activated transient expression of both a Pitx1-LUC vector and the endogenous Pitx1 gene in C2C12 cells. Our characterization of the DUX4 gene structure suggested that polyadenylated DUX4 mRNAs could be produced only from the most distal D4Z4 unit at the FSHD locus. Our data provide a direct link between the genetic defect in 4q35 and the specific up-regulation of a gene in FSHD muscles. Moreover the transcription factor properties of both DUX4 and PITX1 would explain the large transcriptome dysregulation observed in FSHD myoblasts and muscles and the unique clinical characteristics in patients (9, 10, 48). In conclusion, our data identified a candidate mechanism involving interplay between DUX4 and PITX1, which may help to explain the long-elusive biology of FSHD.

Materials and Methods

Muscle Biopsies and Expression Profiling.

The muscle biopsies of nine patients with FSHD described in a previous study were used here for expression profiling and quantitative RT-PCR (10). Nine affected muscles from patients with FSHD and five clinically unaffected muscles from five of those patients were expression-profiled by using the Affymetrix Human Genome U133A microarrays. Normal sex- and age-matched adult controls were obtained from six healthy adult volunteers. Expression profiles of 11 neuromuscular disorders were used as disease controls, as described (49). The procedures and quality control were done as described in our previous publications (50–52). For details, see SI Text.

Primary Myoblasts.

Muscle biopsies were performed according to a procedure approved either by the University of Rochester Research Subjects Review Board or by current ethical and legislative rules of France, as described (53). Primary myoblast cultures were derived from biopsies as described (39). Myoblasts were grown in DMEM with 10% FCS and 1% Ultroser (Cyphergen). Differentiation to myotubes was induced by changing confluent myoblasts to DMEM containing 2% horse serum (differentiation medium). Cells were harvested either during the exponential proliferation phase or 4–6 days after differentiation was induced.

Luciferase Reporter Assay.

C2C12 cells were cotransfected with the Pitx1-LUC wild-type or mutant promoter construct and either pCIneo vector by using FuGENE 6 (Roche Applied Sciences) and 0.5 μg of each plasmid (for details, see SI Text). Quadruplicates were performed. After 24 h, transfected cells were lysed directly in the dish with 100 μl of reporter lysis buffer, and luciferase activity was evaluated by using Luciferase Assay System (Promega) in Centro LB 960 Luminometer (Berthold Technologies).

EMSA.

EMSA was performed with a 30-bp oligonucleotide containing the putative DUX4-binding site as a probe (for details, see SI Text). Nuclear protein extract was prepared from C2C12 cells 16 h posttransfected with pCIneo-DUX4, DUX4c, and DUX1 expression vectors by using Cell Lytic NuCLEAR Extraction Kit (Sigma). To obtain supershift of protein–DNA complexes, 1 μg of the anti-DUX4/DUX4c mAb was added to the mixtures. DNA–protein complexes were resolved on a 6% Novex DNA retardation polyacrylamide gel (Invitrogen), then transferred onto positively charged nylon membranes by electroblotting. For signal detection, the chemiluminescent substrate CSPD (10 mg/ml) was used in 1:100 dilutions, and the emitted light was recorded on an x-ray film.

5′- and 3′-RACE.

Primary myoblast total RNA was extracted with the Aurum Total RNA Mini kit (Bio-Rad) or NucleoSpin RNAII (Macherey-Nagel). DNase-treated total RNA were submitted to 5′- and 3′-RACE by using the RLM-RACE kit (Ambion). The products were cloned and sequenced (for details, see SI Text.)

RT-PCR.

RT was done on 2–4 μg of DNase-treated RNA with primer no. 407 and 200 units of SuperScript III reverse transcriptase in a 20-μl final volume (Invitrogen; see SI Table 3). Eight or 12 μl of cDNA was used for PCR in a 50-μl final volume containing 2.5 units of platinum Pfx, 3× PCRX enhancer solution (Invitrogen), and 15 pmol of each primer (nos. 222 and 407). A step-by-step procedure is provided in SI Text.

Immunofluorescence.

Cells were transfected with 1 μg of plasmid DNA. The proteasome inhibitor MG132 (25 mM; Sigma) was added 24 or 36 h after transfection, and immunostaining was performed by standard procedures (SI Text) 5 h later by using the 9A12 mAb and the anti-Pitx1 rabbit serum (1:50). The pCMV-Pitx1 expression vector (MGC-13954) was purchased from American Type Culture Collection containing full-length Pitx1 cDNA.

Western Blotting.

Transfected control TE671 cells or myoblast primary cultures were lysed on ice in 150 μl of 10 mM Tris·HCl, pH 7.4/150 mM NaCl/0.1% Triton X-100, with protease inhibitors.

Proteins were separated by electrophoresis (10% SDS/PAGE) and transferred onto a PVDF membrane (Amersham Biosciences). After blocking, the Western blot was incubated with the 9A12 mAb (dilution 1:1,000 in PBS/0.2% Tween/2% BSA) followed by secondary antibodies coupled to HRP (Amersham Biosciences). We also used a rabbit serum raised against a DUX4-specific peptide (26). HRP was detected either with LiteAblot (Euroclone) or Super Signal West Femto Maximum Sensitivity Substrate (Pierce). Chemiluminescence was detected on ECL films (Amersham Biosciences). Detection of loading controls α-tubulin (mAb, Sigma–Aldrich) or actin (rabbit polyclonal; Sigma–Aldrich) was similarly performed after membrane stripping (SI Text).

Supplementary Material

Supporting Information
pnas_0708659104_index.html (18.8KB, html)

Acknowledgments

We thank Dr. Eric P. Hoffman for constructive review of the manuscript. The study was partially supported by National Institutes of Health Grants NIAMS 5R21AR048318 (to Y.-W.C. and S.W.) and R21-NS046242 (to H. J. Worman and A.B.) and Facioscapulohumeral Society Grants DR-01 (to A.B.) and MGBF-011 (to Y.-W.C.). A.B.'s group was also funded by the Association Belge Contre les Maladies Neuromusculaires (Belgium), the Asssociation Française Contre les Myopathies (AFM, France), and the Muscular Dystrophy Association (Grant 3209). Additional support was generously provided by the Fischer–Shaw Foundation (to S.W. and A.B.) and by Mr. D. Frenzel (A.B.). E.A. and A.T. held predoctoral fellowships from the Fonds pour la Formation à la Recherche dans l'Industrie et dans l'Agriculture (Belgium), and M.B. held a predoctoral fellowship from the Ministry of Research and Technology (France). The AFM provided pre- and postdoctoral fellowships to S.S. and D.L.-C., respectively.

Abbreviation

FSHD

facioscapulohumeral muscular dystrophy.

Footnotes

The authors declare no conflict of interest.

Data deposition: The sequences reported in this paper have been deposited in the Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov) (accession no. GSE9397) and Children's National Medical Center Public Expression Profiling Resources (http://pepr.cnmcresearch.org) databases.

This article contains supporting information online at www.pnas.org/cgi/content/full/0708659104/DC1.

References

  • 1.van der Maarel SM, Frants RR, Padberg GW. Biochim Biophys Acta. 2006;1772:186–194. doi: 10.1016/j.bbadis.2006.05.009. [DOI] [PubMed] [Google Scholar]
  • 2.Tawil R, Figlewicz DA, Griggs RC, Weiffenbach B. Ann Neurol. 1998;43:279–282. doi: 10.1002/ana.410430303. [DOI] [PubMed] [Google Scholar]
  • 3.Wijmenga C, Hewitt JE, Sandkuijl LA, Clark LN, Wright TJ, Dauwerse HG, Gruter AM, Hofker MH, Moerer P, Williamson R, et al. Nat Genet. 1992;2:26–30. doi: 10.1038/ng0992-26. [DOI] [PubMed] [Google Scholar]
  • 4.Tupler R, Berardinelli A, Barbierato L, Frants R, Hewitt JE, Lanzi G, Maraschio P, Tiepolo L. J Med Genet. 1996;33:366–370. doi: 10.1136/jmg.33.5.366. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Winokur ST, Bengtsson U, Feddersen J, Mathews KD, Weiffenbach B, Bailey H, Markovich RP, Murray JC, Wasmuth JJ, Altherr MR, et al. Chromosome Res. 1994;2:225–234. doi: 10.1007/BF01553323. [DOI] [PubMed] [Google Scholar]
  • 6.Gabellini D, Green MR, Tupler R. Cell. 2002;110:339–348. doi: 10.1016/s0092-8674(02)00826-7. [DOI] [PubMed] [Google Scholar]
  • 7.Jiang G, Yang F, van Overveld PG, Vedanarayanan V, van der Maarel S, Ehrlich M. Hum Mol Genet. 2003;12:2909–2921. doi: 10.1093/hmg/ddg323. [DOI] [PubMed] [Google Scholar]
  • 8.Masny PS, Bengtsson U, Chung SA, Martin JH, van Engelen B, van der Maarel SM, Winokur ST. Hum Mol Genet. 2004;13:1857–1871. doi: 10.1093/hmg/ddh205. [DOI] [PubMed] [Google Scholar]
  • 9.van Overveld PG, Lemmers RJ, Sandkuijl LA, Enthoven L, Winokur ST, Bakels F, Padberg GW, van Ommen GJ, Frants RR, van der Maarel SM. Nat Genet. 2003;35:315–317. doi: 10.1038/ng1262. [DOI] [PubMed] [Google Scholar]
  • 10.Winokur ST, Chen YW, Masny PS, Martin JH, Ehmsen JT, Tapscott SJ, van der Maarel SM, Hayashi Y, Flanigan KM. Hum Mol Genet. 2003;12:2895–2907. doi: 10.1093/hmg/ddg327. [DOI] [PubMed] [Google Scholar]
  • 11.Gabellini D, D'Antona G, Moggio M, Prelle A, Zecca C, Adami R, Angeletti B, Ciscato P, Pellegrino MA, Bottinelli R, et al. Nature. 2006;439:973–977. doi: 10.1038/nature04422. [DOI] [PubMed] [Google Scholar]
  • 12.Gabriels J, Beckers MC, Ding H, De Vriese A, Plaisance S, van der Maarel SM, Padberg GW, Frants RR, Hewitt JE, Collen D, et al. Gene. 1999;236:25–32. doi: 10.1016/s0378-1119(99)00267-x. [DOI] [PubMed] [Google Scholar]
  • 13.Winokur ST, Bengtsson U, Vargas JC, Wasmuth JJ, Altherr MR, Weiffenbach B, Jacobsen SJ. Hum Mol Genet. 1996;5:1567–1575. doi: 10.1093/hmg/5.10.1567. [DOI] [PubMed] [Google Scholar]
  • 14.Clapp J, Mitchell LM, Bolland DJ, Fantes J, Corcoran AE, Scotting PJ, Armour JA, Hewitt JE. Am J Hum Genet. 2007;81:264–279. doi: 10.1086/519311. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Lyle R, Wright TJ, Clark LN, Hewitt JE. Genomics. 1995;28:389–397. doi: 10.1006/geno.1995.1166. [DOI] [PubMed] [Google Scholar]
  • 16.Beckers M, Gabriëls J, van der Maarel S, De Vriese A, Frants RR, Collen D, Belayew A. Gene. 2001;264:51–57. doi: 10.1016/s0378-1119(00)00602-8. [DOI] [PubMed] [Google Scholar]
  • 17.Tremblay JJ, Goodyer CG, Drouin J. Neuroendocrinology. 2000;71:277–286. doi: 10.1159/000054547. [DOI] [PubMed] [Google Scholar]
  • 18.Tumpel S, Cambronero F, Ferretti E, Blasi F, Wiedemann L, Krumlauf R. Dev Biol. 2007;302:646–660. doi: 10.1016/j.ydbio.2006.10.029. [DOI] [PubMed] [Google Scholar]
  • 19.Ding H, Beckers MC, Plaisance S, Marynen P, Collen D, Belayew A. Hum Mol Genet. 1998;7:1681–1694. doi: 10.1093/hmg/7.11.1681. [DOI] [PubMed] [Google Scholar]
  • 20.Kalionis B, O'Farrell PH. Mech Dev. 1993;43:57–70. doi: 10.1016/0925-4773(93)90023-q. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Kumar P, Nazarali AJ. Cell Mol Neurobiol. 2001;21:369–388. doi: 10.1023/a:1012654122046. [DOI] [PubMed] [Google Scholar]
  • 22.Gehring WJ, Qian YQ, Billeter M, Furukubo-Tokunaga K, Schier AF, Resendez-Perez D, Affolter M, Otting G, Wuthrich K. Cell. 1994;78:211–223. doi: 10.1016/0092-8674(94)90292-5. [DOI] [PubMed] [Google Scholar]
  • 23.Chan SK, Ryoo HD, Gould A, Krumlauf R, Mann RS. Development (Cambridge, UK) 1997;124:2007–2014. doi: 10.1242/dev.124.10.2007. [DOI] [PubMed] [Google Scholar]
  • 24.Ryoo HD, Mann RS. Genes Dev. 1999;13:1704–1716. doi: 10.1101/gad.13.13.1704. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Rechsteiner M, Rogers SW. Trends Biochem Sci. 1996;21:267–271. [PubMed] [Google Scholar]
  • 26.Kowaljow V, Marcowycz A, Ansseau E, Conde CB, Sauvage S, Matteotti C, Arias C, Corona ED, Nunez NG, Leo O, et al. Neuromuscul Disord. 2007;17:611–623. doi: 10.1016/j.nmd.2007.04.002. [DOI] [PubMed] [Google Scholar]
  • 27.Salamov AA, Solovyev VV. Genome Res. 2000;10:516–522. doi: 10.1101/gr.10.4.516. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Lanctot C, Moreau A, Chamberland M, Tremblay ML, Drouin J. Development (Cambridge, UK) 1999;126:1805–1810. doi: 10.1242/dev.126.9.1805. [DOI] [PubMed] [Google Scholar]
  • 29.Szeto DP, Rodriguez-Esteban C, Ryan AK, O'Connell SM, Liu F, Kioussi C, Gleiberman AS, Izpisua-Belmonte JC, Rosenfeld MG. Genes Dev. 1999;13:484–494. doi: 10.1101/gad.13.4.484. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Logan M, Tabin CJ. Science. 1999;283:1736–1739. doi: 10.1126/science.283.5408.1736. [DOI] [PubMed] [Google Scholar]
  • 31.Cole NJ, Tanaka M, Prescott A, Tickle C. Curr Biol. 2003;13:R951–R952. doi: 10.1016/j.cub.2003.11.039. [DOI] [PubMed] [Google Scholar]
  • 32.Shapiro MD, Marks ME, Peichel CL, Blackman BK, Nereng KS, Jonsson B, Schluter D, Kingsley DM. Nature. 2004;428:717–723. doi: 10.1038/nature02415. [DOI] [PubMed] [Google Scholar]
  • 33.Tanaka M, Hale LA, Amores A, Yan YL, Cresko WA, Suzuki T, Postlethwait JH. Dev Biol. 2005;281:227–239. doi: 10.1016/j.ydbio.2005.02.016. [DOI] [PubMed] [Google Scholar]
  • 34.Civas A, Island ML, Genin P, Morin P, Navarro S. Biochimie. 2002;84:643–654. doi: 10.1016/s0300-9084(02)01431-1. [DOI] [PubMed] [Google Scholar]
  • 35.Island ML, Mesplede T, Darracq N, Bandu MT, Christeff N, Djian P, Drouin J, Navarro S. Mol Cell Biol. 2002;22:7120–7133. doi: 10.1128/MCB.22.20.7120-7133.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Kolfschoten IG, van Leeuwen B, Berns K, Mullenders J, Beijersbergen RL, Bernards R, Voorhoeve PM, Agami R. Cell. 2005;121:849–858. doi: 10.1016/j.cell.2005.04.017. [DOI] [PubMed] [Google Scholar]
  • 37.Padberg G, Adams C. In: Neurogenetics. Pulst S-M, editor. Oxford: Oxford Univ Press; 2000. pp. 105–116. [Google Scholar]
  • 38.Arahata K, Ishihara T, Fukunaga H, Orimo S, Lee JH, Goto K, Nonaka I. Muscle Nerve. 1995;2:S56–S66. [PubMed] [Google Scholar]
  • 39.Winokur ST, Barrett K, Martin JH, Forrester JR, Simon M, Tawil R, Chung SA, Masny PS, Figlewicz DA. Neuromuscul Disord. 2003;13:322–333. doi: 10.1016/s0960-8966(02)00284-5. [DOI] [PubMed] [Google Scholar]
  • 40.Hewitt JE, Lyle R, Clark LN, Valleley EM, Wright TJ, Wijmenga C., van Deutekom JC, Francis F, Sharpe PT, Hofker M, et al. Hum Mol Genet. 1994;3:1287–1295. doi: 10.1093/hmg/3.8.1287. [DOI] [PubMed] [Google Scholar]
  • 41.Yip DJ, Picketts DJ. FEBS Lett. 2003;537:133–138. doi: 10.1016/s0014-5793(03)00110-8. [DOI] [PubMed] [Google Scholar]
  • 42.Osborne RJ, Welle S, Venance SL, Thornton CA, Tawil R. Neurology. 2007;68:569–577. doi: 10.1212/01.wnl.0000251269.31442.d9. [DOI] [PubMed] [Google Scholar]
  • 43.Alexiadis V, Ballestas ME, Sanchez C, Winokur S, Vedanarayanan V, Warren M, Ehrlich M. Biochim Biophys Acta. 2007;1769:29–40. doi: 10.1016/j.bbaexp.2006.11.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.van Geel M, Dickson MC, Beck AF, Bolland DJ, Frants RR, van der Maarel SM, de Jong PJ, Hewitt JE. Genomics. 2002;79:210–217. doi: 10.1006/geno.2002.6690. [DOI] [PubMed] [Google Scholar]
  • 45.Lemmers RJ, de Kievit P, Sandkuijl L, Padberg GW, van Ommen GJ, Frants RR, van der Maarel SM. Nat Genet. 2002;32:235–236. doi: 10.1038/ng999. [DOI] [PubMed] [Google Scholar]
  • 46.Thomas NS, Wiseman K, Spurlock G, Macdonald M, Ustek D, Upadhyaya M. J Med Genet. 2006;44:215–218. doi: 10.1136/jmg.2006.042804. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Coppée F, Mattéotti C, Ansseau E, Sauvage S, Leclercq I, Leroy A, Marcowycz A, Figlewicz D, Ding H, Belayew A. In: Facioscapulohumeral Muscular Dystrophy: Clinical Medicine and Mol Cell Biol. Cooper D, Upadhyahya M, editors. Oxford: BIOS; 2004. pp. 117–134. chap 9. [Google Scholar]
  • 48.Tupler R, Perini G, Pellegrino MA, Green MR. Proc Natl Acad Sci USA. 1999;96:12650–12654. doi: 10.1073/pnas.96.22.12650. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Bakay M, Wang Z, Melcon G, Schiltz L, Xuan J, Zhao P, Sartorelli V, Seo J, Pegoraro E, Angelini C, et al. Brain. 2006;129:996–1013. doi: 10.1093/brain/awl023. [DOI] [PubMed] [Google Scholar]
  • 50.Chen YW, Hubal MJ, Hoffman EP, Thompson PD, Clarkson PM. J Appl Physiol. 2003;95:2485–2494. doi: 10.1152/japplphysiol.01161.2002. [DOI] [PubMed] [Google Scholar]
  • 51.Chen YW, Zhao P, Borup R, Hoffman EP. J Cell Biol. 2000;151:1321–1336. doi: 10.1083/jcb.151.6.1321. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Tumor Analysis Best Practices Working Group. Nat Rev Genet. 2004;5:229–237. doi: 10.1038/nrg1297. [DOI] [PubMed] [Google Scholar]
  • 53.Laoudj-Chenivesse D, Carnac G, Bisbal C, Hugon G, Bouillot S, Desnuelle C, Vassetzky Y, Fernandez A. J Mol Med. 2005;83:216–224. doi: 10.1007/s00109-004-0583-7. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information
pnas_0708659104_index.html (18.8KB, html)
pnas_0708659104_3.pdf (15.1KB, pdf)
pnas_0708659104_4.pdf (11.5KB, pdf)
pnas_0708659104_5.pdf (29KB, pdf)
pnas_0708659104_1.pdf (267.1KB, pdf)
pnas_0708659104_2.pdf (165.5KB, pdf)

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES