Abstract

Defects in mRNA 3′end formation have been described to alter transcription termination, transport of the mRNA from the nucleus to the cytoplasm, stability of the mRNA and translation efficiency. Therefore, inhibition of polyadenylation may lead to gene silencing. Here, we choose facioscapulohumeral dystrophy (FSHD) as a model to determine whether or not targeting key 3′ end elements involved in mRNA processing using antisense oligonucleotide drugs can be used as a strategy for gene silencing within a potentially therapeutic context. FSHD is a gain-of-function disease characterized by the aberrant expression of the Double homeobox 4 (DUX4) transcription factor leading to altered pathogenic deregulation of multiple genes in muscles. Here, we demonstrate that targeting either the mRNA polyadenylation signal and/or cleavage site is an efficient strategy to down-regulate DUX4 expression and to decrease the abnormally high-pathological expression of genes downstream of DUX4. We conclude that targeting key functional 3′ end elements involved in pre-mRNA to mRNA maturation with antisense drugs can lead to efficient gene silencing and is thus a potentially effective therapeutic strategy for at least FSHD. Moreover, polyadenylation is a crucial step in the maturation of almost all eukaryotic mRNAs, and thus all mRNAs are virtually eligible for this antisense-mediated knockdown strategy.

Introduction

The cleavage and polyadenylation of the 3′end of pre-mRNAs are fundamental processing steps for the maturation of the vast majority of eukaryotic mRNAs. In human cells, these reactions are governed by >80 RNA-binding proteins and by regulatory cis-acting RNA sequence elements (1–3). The key element dictating the cleavage is a six-nucleotide (nt) motif called the poly(A) signal (PAS). Most of the mammalian mRNAs contain the PAS consensus AAUAAA or AUUAAA hexamer or close variants (4–6), which is recognized by the multimeric set of cleavage and polyadenylation factors. This RNA-protein interaction determines the site of cleavage which usually occurs 10–30 nt downstream of the PAS. The second important element is a Uracil/Guanine and uracil (U/GU)-rich sequence [downstream sequence element (DSE)] contacted by the cleavage stimulation factor and located 30–60 nt downstream the PAS motif (7). In most cases, these co-transcriptional maturations are required for nuclear export and stability of mRNAs, and their efficient translation (8), and thus consequently represent attractive and interesting drug and/or genetic targets for suppression of gene expression. Indeed, the functional importance of the 3′ end mRNA processing has been highlighted by the discovery of mutations in the PAS cis-element causing or contributing to human diseases including haemotological diseases, neurogenerative diseases or cancer (9).

The aim of our study was to determine whether or not targeting elements involved in pre-mRNA maturation (such as PAS or DSE) using antisense oligonucleotide (AO) drugs can be a strategy for gene silencing within a potentially therapeutic context. We focused on faciocapulohumeral dystrophy (FSHD), which is a rare autosomal dominant neuromuscular disorder with an incidence of 1:14 000–1:20 000 characterized by the atrophy of specific groups of muscles (10). This pathology is caused by a loss of epigenetic marks within the D4Z4 macrosatellite located in the sub-telomeric region of chromosome 4 leading to chromatin relaxation (11). In 95% of the FSHD patients (named FSHD1), this chromatin relaxation is associated with a contraction of the D4Z4 array (12,13). The remaining 5% of the FSHD patients (named FSHD2) do not present a contraction of D4Z4, but the vast majority of them carry a mutation in the epigenetic modifier gene SMCHD1 (14,15) . This loss of epigenetic marks, when associated with a permissive chromosome 4 [i.e. carrying the 4qA haplotype containing the ATTAAA PAS commonly found in human (16)], leads to the aberrant transcription of a double-homeobox transcription factor named Double homeobox 4 (DUX4) whose ORF is present in each D4Z4 repeat (17). DUX4 protein and mRNA have been robustly detected in adult and fetal FSHD1 and FSHD2 cells and biopsies whereas they can be found at very low levels in control (18–21). DUX4 mRNA is only expressed in 1/1000 nuclei, but the DUX4 protein can be found in 0.5–9% of nuclei (18,22), suggesting that although DUX4 is not transcribed in most nuclei, the resulting protein may spread further (23). DUX4 is a transcription factor and its overexpression is described to disturb several cellular pathways (23–30). Moreover, it was shown that even if DUX4 expression has not been directly linked to patient's phenotype, DUX4 may play a major role in the pathophysiology of FSHD because: (i) it has been shown that at least one D4Z4 repeat is needed for FSHD onset (31), (ii) only alleles with the 4qA type (containing the AUUAAA PAS for DUX4 mRNA) are associated with FSHD (32,33), (iii) contraction of the D4Z4 array on chromosome 10, which carries a mutated PAS (AUCAAA), does not lead to FSHD (16), (iv) DUX4-induced gene expression is the major molecular signature in FSHD skeletal muscles (34) and (v) DUX4 expression is the common point between FSHD1 and FSHD2 patients (16).

Several therapeutic strategies targeting DUX4 expression have been proposed in the literature: RNA interference, 2′-O-methyl AOs targeting intron–exon junctions or over-expression of truncated DUX4 (27,35–37). Here, we describe a new therapeutic AO-based therapeutic approach for FSHD using phosphorodiamidate morpholino oligomers (PMOs) targeting the key elements of 3′ end processing of the DUX4 transcript and which represent the key disease-permissive feature of the FSHD1 and 2 locus. Indeed, during the past 10 years, synthetic AOs have emerged as a promising strategy for drug therapy of genetic disorders affecting skeletal muscles such as the muscular dystrophies and motor-neuron diseases. PMOs have been mainly described to bind with high affinity the targeted pre-mRNAs and mRNAs and to form a stable hybrid that efficiently blocks the access of splicing or translation machineries (so-called steric blocking AOs). PMOs can also form a structure mimicking a tRNA precursor, which is recognized by RNaseP leading to mRNA cleavage (38,39). Numerous studies investigating the therapeutic potential of antisense technology have been already performed and several clinical trials are now in progress for muscular dystrophies (40–43).

In this article, we choose to target the 3′ key elements of DUX4 pre-mRNA, focusing on the disease-permissive PAS, the cleavage site and the U/GU-rich DSE sequence. We demonstrated in vitro that targeting the 3′end elements of mRNA with antisense drugs can be an efficient therapeutic strategy for a genetic disease. We observed that targeting DUX4 3′key elements leads to an efficient extinction of DUX4, may redirect the mRNA cleavage site and prevents aberrant expression of genes downstream of the DUX4 transcription factor.

Results

Determination of 3′ end key elements of DUX4 mRNA and PMO design

For muscle tissue, one PAS (AUUAAA) has been described for DUX4 mRNA, located 766 bp downstream of the stop codon in genomic DNA (16). Because the DUX4 gene gives rise to two DUX4-FL isoforms (DUX4-FL1 with spliced intron 1 and DUX4-FL2 with non-spliced intron 1), the cleavage site of DUX4 was precisely determined by RT-3′ rapid amplification of cDNA-ends by polymerase chain reaction (3′RACE-PCR) using primers allowing the detection of each isoform. Total RNAs were extracted from FSHD myotubes at Day 4 of differentiation when DUX4 expression is the highest (21). The sequences of the PCR products revealed the presence of three different cleavage sites located 16–22 bp after the PAS for both DUX4-FL1 and -FL2 DUX4 isoforms (Fig. 1A and B). However, these cleavage sites are not used at the same frequencies. The cleavage site located 22 bases downstream of the PAS represents 78% of the DUX4-FL1 isoforms, but only 41% of DUX4-FL2 isoforms thus showing the interplay between mRNA cleavage site and splicing (Fig. 1A and B). In order to investigate the therapeutic potential of AONs targeting 3′ key elements of DUX4 mRNA, we designed AONs covering either the PAS (PMO-PAS), the cleavage sites and the polyA (PMO-CS1–3) or the U/GU-rich (PMO-DSE) sequence (Fig. 1C). Briefly, PMOs were designed to (i) contain little or no self-complementarity, (ii) form no >16 contiguous intra-strand hydrogen bonds, (iii) contain no >9 total guanines (>36% G) or >3 contiguous guanines to ensure good water solubility and (iv) where possible to target >1 of the 3′ key elements with 1 PMO. For these reasons, the cleavage site PMOs contains either a single mismatch to target (PMO-CS2 and PMO-CS3) or do not target contiguous bases (PMO-CS1). These PMOs were of varying lengths (25–30 mers) (Table 1).

Table 1.

Summary of the characteristics of the PMOs

PMO name PMO sequence 5′-3′ Leash sequence 5′-3′ Length (bp) %GC 
PMO-PAS (−13 + 12) GGGCATTTTAATATATCTCTGAACT gattgGATATATTAAAATGCCCgtgat 25 0.32 
PMO-CS1 (−5 + 32) CTATAGGATCCACAGGGCATTTTAATATC gttacCCCTGTGGATCCTATAGgtgat 29 0.38 
PMO-CS2 (−1 + 27) GGATCCACAGGGAGGAGGCATTTTAATA gattgTCCTCCCTGTGGATCCgtgat 28 0.46 
PMO-CS3 (+2 + 31) TATAGGATCCACAGGGAGGAGGCATTTTAA gaatgTCCCTGTGGATCCTATAgtgat 30 0.43 
PMO-DSE (+32 + 56) CATCACACAAAAGATGCAAATCTTC gattgGCATCTTTTGTGTGATGgtgat 25 0.36 
PMO-control CCTCTTACCTCAGTTACAATTTATA gattgTAACTGAGGTAAGAGGgtgat 25 0.32 
PMO name PMO sequence 5′-3′ Leash sequence 5′-3′ Length (bp) %GC 
PMO-PAS (−13 + 12) GGGCATTTTAATATATCTCTGAACT gattgGATATATTAAAATGCCCgtgat 25 0.32 
PMO-CS1 (−5 + 32) CTATAGGATCCACAGGGCATTTTAATATC gttacCCCTGTGGATCCTATAGgtgat 29 0.38 
PMO-CS2 (−1 + 27) GGATCCACAGGGAGGAGGCATTTTAATA gattgTCCTCCCTGTGGATCCgtgat 28 0.46 
PMO-CS3 (+2 + 31) TATAGGATCCACAGGGAGGAGGCATTTTAA gaatgTCCCTGTGGATCCTATAgtgat 30 0.43 
PMO-DSE (+32 + 56) CATCACACAAAAGATGCAAATCTTC gattgGCATCTTTTGTGTGATGgtgat 25 0.36 
PMO-control CCTCTTACCTCAGTTACAATTTATA gattgTAACTGAGGTAAGAGGgtgat 25 0.32 

The locations, the PMO sequences and the leash sequences (uppercases indicate the complementary region between leash and PMO, lowercases indicates the free ends), the percentage of GC and the length are shown.

Figure 1.

3′ end key elements of DUX4 mRNA and PMO design. (A and B) To determine the cleavage sites of DUX4 mRNA, a RT using an oligo-dT adapter primer was realized on total mRNAs isolated from FSHD myotubes at Day 4 of differentiation. The 3′RACE-PCR was performed using forward primer matching Exon 1 and reverse primer recognizing the adapter. The PCR products were cloned into a TOPO-TA plasmid and 27 and 17 clones were sequenced for DUX4-FL1 (A) and -FL2 (B), respectively. The frequencies of the cleavage site usages are indicated. Arrows indicate the location of the primers used. The poly(A) site is bolded and underlined and the poly(A) tail is in bold. (C) Sequences recognized by the different PMOs targeting DUX4. The position +1 is defined as the first base of the poly(A) site. The sequences recognized by the PMOs are indicated as well as their position on the mRNA. Deletions are represented by hyphen (-). Nucleotidic substitutions are indicated.

Figure 1.

3′ end key elements of DUX4 mRNA and PMO design. (A and B) To determine the cleavage sites of DUX4 mRNA, a RT using an oligo-dT adapter primer was realized on total mRNAs isolated from FSHD myotubes at Day 4 of differentiation. The 3′RACE-PCR was performed using forward primer matching Exon 1 and reverse primer recognizing the adapter. The PCR products were cloned into a TOPO-TA plasmid and 27 and 17 clones were sequenced for DUX4-FL1 (A) and -FL2 (B), respectively. The frequencies of the cleavage site usages are indicated. Arrows indicate the location of the primers used. The poly(A) site is bolded and underlined and the poly(A) tail is in bold. (C) Sequences recognized by the different PMOs targeting DUX4. The position +1 is defined as the first base of the poly(A) site. The sequences recognized by the PMOs are indicated as well as their position on the mRNA. Deletions are represented by hyphen (-). Nucleotidic substitutions are indicated.

PMOs induce a down expression of DUX4 mRNA

The efficacy of each PMO was evaluated in a dose-dependent manner after transfection into differentiated immortalized FSHD clones. Total RNAs were extracted from myotubes and RT-PCR allowing the detection of either DUX4-all (Fig. 2A), or DUX4-FL1 and -FL2 (Fig. 2B) was performed. No modification of DUX4-all mRNA was observed with PMO-control compared to unlipofected cells thus showing that introduction of PMO-control does not modify DUX4 expression. With PMO-CS1, -CS2 and -DSE, 27% ± 5, 25% ± 17 and 36% ± 10 reduction of DUX4-all mRNA was observed, respectively, at the highest PMO concentration used compared with PMO-control (Fig. 2A). The highest efficacies were obtained with the PMO-PAS and -CS3 for which a dose-dependent reduction of DUX4-all levels was observed with 40% ± 6 and 52% ± 7 inhibition at 50 nm, respectively (Fig. 2A). These two AOs were investigated further to ascertain if both DUX4 isoforms were identically targeted by them. A decreased expression of both isoforms was observed at 50 nm when primers allowing the detection of both DUX4-FL1 and FL-2 [DUX4-3′ untranslated region (3′UTR) oligonucleotides] were used (Fig. 2B).

Figure 2.

DUX4 expression in the presence of the PMOs. Immortalized FSHD myotubes were transfected with PMOs in a dose-dependent manner. (A) The percentage of residual DUX4-all mRNA was measured by semi-quantitative PCR using the ImageJ software on gel scans. Independent transfections for PMO-control, -PAS, -CS1, -CS2, CS3 and -DSE (n ≥ 6) were performed. Values are compared with unlipofected conditions. All data represent mean ± standard error of mean (SEM). A MANOVA and a Newman–Keuls post hoc test were performed (*P < 0.05; **P < 0.01 and ***P < 0.001). (B) Representative RT-PCR analysis using DUX4-3′UTR primers in the presence of PMO-control, -PAS and -CS3 at 50 nm. As a negative control, a PMO targeting a human β-globin mutation that causes β-thalassemia was be used. B2M was used as the reference (housekeeping) gene.

Figure 2.

DUX4 expression in the presence of the PMOs. Immortalized FSHD myotubes were transfected with PMOs in a dose-dependent manner. (A) The percentage of residual DUX4-all mRNA was measured by semi-quantitative PCR using the ImageJ software on gel scans. Independent transfections for PMO-control, -PAS, -CS1, -CS2, CS3 and -DSE (n ≥ 6) were performed. Values are compared with unlipofected conditions. All data represent mean ± standard error of mean (SEM). A MANOVA and a Newman–Keuls post hoc test were performed (*P < 0.05; **P < 0.01 and ***P < 0.001). (B) Representative RT-PCR analysis using DUX4-3′UTR primers in the presence of PMO-control, -PAS and -CS3 at 50 nm. As a negative control, a PMO targeting a human β-globin mutation that causes β-thalassemia was be used. B2M was used as the reference (housekeeping) gene.

PMO-CS3 induces a redirection of cleavage region

Because 22–54% of human mRNA have more than one poly(A) sites (4,5), and because redirections of the PAS have been described in the presence of AOs targeting the poly(A) of other mRNAs (44,45), in silico analyses were performed which revealed the presence of putative alternative PAS downstream of Exon 3 (Fig. 3A). A redirection of the poly(A) and/or cleavage sites was thus investigated in the presence of the different PMOs at the highest concentration by 3′RACE nested PCR using forward primers located in Exon 3. A switch in cleavage site or poly(A) usage was not observed with any of the PMOs except PMO-CS3 (Fig. 3B). The sequence of this supplemental band revealed that the cleavage site of the residual DUX4 mRNA in the presence of PMO-CS3 was ∼40 nt upstream of the canonical cleavage site (Fig. 3C), thus suggesting that an alternative PAS was used to generate this new cleavage site. However, in silico analysis of the genomic region upstream of this alternative cleavage site (up to DUX4 Exon 2) did not reveal the presence of any of the 13 PAS previously described (4). This new cleavage site is used in ∼20% of the residual DUX4 mRNAs.

Figure 3.

PMO-CS3 induces a switch in cleavage site usage. A redirection of poly(A) usage was investigated in the presence of the different PMOs. (A) In silico analyses and literature (18) reveal the presence of APA site in the subtelomeric region of chromosome 4. (B) 3′RACE nested PCR using forward primers located in Exon 3 shows a switch in cleavage site only in the presence of PMO-CS3. The bands with (double asterisks) or without (single asterisk) a redirection of the cleavage site are indicated. (C) The sequence of the most abundant mRNA carrying the redirected cleavage site (DUX4 pre-mRNA). The sequence of poly(A) site is underlined and bolded. The poly(A) tail is in bold. The frequencies of each variant showing alternative cleavage site usage are indicated (14 analyzed sequences).

Figure 3.

PMO-CS3 induces a switch in cleavage site usage. A redirection of poly(A) usage was investigated in the presence of the different PMOs. (A) In silico analyses and literature (18) reveal the presence of APA site in the subtelomeric region of chromosome 4. (B) 3′RACE nested PCR using forward primers located in Exon 3 shows a switch in cleavage site only in the presence of PMO-CS3. The bands with (double asterisks) or without (single asterisk) a redirection of the cleavage site are indicated. (C) The sequence of the most abundant mRNA carrying the redirected cleavage site (DUX4 pre-mRNA). The sequence of poly(A) site is underlined and bolded. The poly(A) tail is in bold. The frequencies of each variant showing alternative cleavage site usage are indicated (14 analyzed sequences).

PMO-PAS and -CS3 induce a knockdown in expression of genes downstream of DUX4 in FSHD cells

Because the DUX4 protein is very hard to detect, expression levels of DUX4 biomarkers were investigated. Indeed, DUX4 being a transcription factor, the expression levels of several genes downstream of DUX4 have been described to be modulated after DUX4 expression among them TRIM43, MBD3L2 and ZSCAN4 (21,27,46). The expression of these three genes was determined by RT-qPCR in the presence of the PMOs in a dose-dependent manner. No significant modulation of their expressions was observed in the presence of PMO-control compared with unlipofected condition. PMO-CS1, -CS2 and -DSE moderately modulate expression of genes downstream of DUX4 at the lowest concentrations used, but a down expression of TRIM43 is observed at the highest concentration. The best modulation of the genes downstream of DUX4 was observed in the presence of PMO-PAS and -CS3 in a dose-dependent manner (Fig. 4).

Figure 4.

Expression of the genes downstream of DUX4 in the presence of the PMOs. Expression levels of TRIM43 (A), ZSCAN4 (B) and MBD3L2 (C) were measured by RT-qPCR in FSHD myotubes (4 days of differentiation) in the presence of the PMOs at different concentrations and compared with the PMO control at the same concentration. Values are compared with unlipofected conditions. B2M was used as a normalizer. Data represent mean ± SEM from independent transfections for PMO-control, -PAS, -CS1, -CS2, CS3 and -DSE (n ≥ 6). A MANOVA and a Newman–Keuls post hoc test were performed (*P < 0.05; **P < 0.01 and ***P < 0.001).

Figure 4.

Expression of the genes downstream of DUX4 in the presence of the PMOs. Expression levels of TRIM43 (A), ZSCAN4 (B) and MBD3L2 (C) were measured by RT-qPCR in FSHD myotubes (4 days of differentiation) in the presence of the PMOs at different concentrations and compared with the PMO control at the same concentration. Values are compared with unlipofected conditions. B2M was used as a normalizer. Data represent mean ± SEM from independent transfections for PMO-control, -PAS, -CS1, -CS2, CS3 and -DSE (n ≥ 6). A MANOVA and a Newman–Keuls post hoc test were performed (*P < 0.05; **P < 0.01 and ***P < 0.001).

Since we have previously published that DUX4 is able to spread from one FSHD nucleus to neighboring nuclei thus inducing an aberrant activation of genes downstream of DUX4 in many nuclei (23), one possible explanation for the lower expression of TRIM43, MBD3L2 and ZSCAN4 observed in the presence of the PMO-PAS and -CS3 could have been an inhibition of myotube formation mediated by the PMOs. PMO-transfected cells were immunostained with the MF20 antibody recognizing all the myosin heavy chains (Fig. 5A) and fusion indexes were calculated by counting the number of nuclei in MF20-positive myotubes containing three or more nuclei as the percentage of the total number of nuclei (Fig. 5B). No statistically significant differences were observed with PMO treatment thus showing that the down-regulation of the genes downstream of DUX4 in the presence of the PMOs was not due to myotube formation impairment, but rather to lowered DUX4 expression.

Figure 5.

Fusion index of immortalized FSHD myotubes in the presence of the PMOs. Immortalized FSHD myotubes were transfected at Day 2 of differentiation with the different PMOs at 50 nm. At Day 4 of differentiation, cells were stained with MF20 antibody recognizing all the myosin heavy chain and counterstained with 4′,6-diamidino-2-phenylindole (A). Fusion indexes were calculated by counting the number of nuclei in myotubes containing more than two nuclei as the percentage of the total number of nuclei (B). The transfections were performed six times for the unlipofected, four times for PMO-control, -PAS -DSE and two times for PMO-CS1, -CS2 and -CS3. In each experiment, 1000 nuclei were counted. All data represent mean ± standard deviation.

Figure 5.

Fusion index of immortalized FSHD myotubes in the presence of the PMOs. Immortalized FSHD myotubes were transfected at Day 2 of differentiation with the different PMOs at 50 nm. At Day 4 of differentiation, cells were stained with MF20 antibody recognizing all the myosin heavy chain and counterstained with 4′,6-diamidino-2-phenylindole (A). Fusion indexes were calculated by counting the number of nuclei in myotubes containing more than two nuclei as the percentage of the total number of nuclei (B). The transfections were performed six times for the unlipofected, four times for PMO-control, -PAS -DSE and two times for PMO-CS1, -CS2 and -CS3. In each experiment, 1000 nuclei were counted. All data represent mean ± standard deviation.

No significant off-targets are predicted

The potential off-target effects were also evaluated, PMOs targeting the DUX4 transcript were submitted to short-input-sequence basic alignment search tool (BLAST) searches against coding and non-coding human genome-nucleotide collection databases (nr/nr and nr/nt: see Table 2). According to BLAST a very strong match is considered to have E-value < 1 × 10−4, whereas a poor match has E-value > 1 × 10−3. For each PMO, the two BLAST returns showing highest alignment scores (lowest E-values) were selected for further analysis. With the exception of PMO-CS1, which contains specifically selected deletions, all designed PMOs returned a very strong match to the desired DUX4 transcript (E-value ≤ 2 × 10−5), whereas all of the off-target gene candidates show high E-values > 0.5. Predicted off-target sequence matches that showed homology on their negative or forward strand, corresponding to RNA transcript, were selected for the calculation of total free energy of binding using the RNAup Server (Table 2; Supplementary Material, Fig. S1). The only potential off-target interactions on the RNA transcript (i.e. negative strand) was GYS2 for PMO-PAS and PTPRT for PMO-CS1, but overall free energies of binding were low (−16.62 and −6.60, respectively), and with major mismatches (8/25 and 12/29, respectively). These data suggest a low probability of interference of any of the designed PMOs targeting DUX4 with off target RNAs.

Table 2.

Prediction of the anti-DUX4 off target candidates

PMO name Target name E-value % identity No. of base overlap Predicted homology with (+/−) strand Coding/non-coding Overall free binding energy (ΔGi
PMO-PAS DUX4 4.00E-05 100 25/25 − Non-coding −32.04 
IL1RAPL2 0.58 72 18/25 N.A. N.A. 
GYS2 2.3 68 17/25 − Non-coding −16.62 
PMO-CS1 C1ORF101 0.91 72 21/29 N.A. N.A. 
DUX4 3.6 59 17/29 − Non-coding −21.82 
PTPRT 3.6 59 17/29 − Non-coding −6.60 
PMO-CS2 DUX4 2.00E-05 96 27/28 − Non-coding −38.02 
OTOF 0.82 75 21/28 N.A. N.A. 
POR 0.82 64 18/28 N.A. N.A. 
PMO-CS3 DUX4 2.00E-05 97 29/30 − Non-coding −37.93 
OTOF 0.99 70 21/30 N.A. N.A. 
POR 0.99 60 18/30 N.A. N.A. 
PMO-DSE DUX4 4.00E-05 100 25/25 − Non-coding −27.41 
TCRA/TC 0.58 72 18/25 N.A. N.A. 
KCNJ6 2.3 68 17/25 N.A. N.A. 
PMO name Target name E-value % identity No. of base overlap Predicted homology with (+/−) strand Coding/non-coding Overall free binding energy (ΔGi
PMO-PAS DUX4 4.00E-05 100 25/25 − Non-coding −32.04 
IL1RAPL2 0.58 72 18/25 N.A. N.A. 
GYS2 2.3 68 17/25 − Non-coding −16.62 
PMO-CS1 C1ORF101 0.91 72 21/29 N.A. N.A. 
DUX4 3.6 59 17/29 − Non-coding −21.82 
PTPRT 3.6 59 17/29 − Non-coding −6.60 
PMO-CS2 DUX4 2.00E-05 96 27/28 − Non-coding −38.02 
OTOF 0.82 75 21/28 N.A. N.A. 
POR 0.82 64 18/28 N.A. N.A. 
PMO-CS3 DUX4 2.00E-05 97 29/30 − Non-coding −37.93 
OTOF 0.99 70 21/30 N.A. N.A. 
POR 0.99 60 18/30 N.A. N.A. 
PMO-DSE DUX4 4.00E-05 100 25/25 − Non-coding −27.41 
TCRA/TC 0.58 72 18/25 N.A. N.A. 
KCNJ6 2.3 68 17/25 N.A. N.A. 

N.A., non-applicable; IL1RAPL2, interleukin 1 receptor accessory protein-like 2; GYS2, glycogen synthase 2; C1ORF101, chromosome 1 open reading frame 101; PTPRT, protein tyrosine phosphatase receptor type T; OTOF, otoferlin; POR, P450 cytochrome oxidoreductase; TCRA/TCRD, T-cell receptor alpha–delta locus; KCNJ6, potassium channel inwardly rectifying subfamily J member.

Each PMO sequence was run thought the BLAST software, optimized for short-input sequences and limited to human and nt/nr database. Overall the binding energy was calculated only for the predicted targets that show sequence homology with the PMO on their negative strand (−). The units of ΔGi are expressed in kcal/mol.

Discussion

Several strategies using AOs have been described during the last decade. Upon binding, AOs can alter the original function of the targeted mRNA through an array of different mechanisms. In the literature, AOs have been described to (i) prevent 5′cap formation, (ii) bind the pre-mRNA to modulate splicing by masking the splicing sequences to force exon inclusion or exclusion, (iii) induce mRNA degradation after the recognition of the duplex AO-mRNA (depending on the chemical modification used), (iv) inhibit ribosome access to the mRNA leading to the suppression of protein translation and (v) modulate polyadenylation selection in transcripts with more than one poly(A) in their 3′UTR (39). However, to our knowledge, the possibility to target the 3′UTR key elements of the mammalian pre-mRNA to inhibit the polyadenylation in genetic diseases has not been studied.

As previously described, defects in mRNA 3′end formation can result in disruption of gene expression processes in a number of ways, including: transcription termination; transport of the mRNA from the nucleus to the cytoplasm; stability of the mRNA; translation efficiency and the inhibition of polyadenylation, which in turn might lead to gene silencing (47). In this study, we investigate the possibility of interference in the 3′ end mRNA processing as a therapeutic approach in the FSHD model. Since FSHD is a gain-of-function disease, characterized by the aberrant over-expression of the DUX4 gene, the therapeutic strategies should therefore be based on gene silencing rather than gene rescue. Here, we show that targeting the 3′end elements involved in mRNA processing is an effective way to decrease DUX4 expression and prevent aberrant expression of genes downstream of DUX4.

Among the five AONs tested, only two of them lead to a significant decrease of DUX4 mRNA levels under the conditions evaluated. PMOs length could play an important role in their efficacies, since PMO-CS3 is the only 30-mer used in this study and is the AO showing the best efficiency. The number of mismatches could also play a role affecting their relative activity when compared with a completely homologous sequence of equal length. Secondary structures may also have affected effectiveness of the PMOs and secondary structures of Exon 3 of DUX4 or the AON were analyzed in silico using thermodynamic methods. Indeed, the probability of interaction between the AON and the targeted mRNA can be impacted by the presence of closed or opened conformations at the site of hybridization (47). When calculating the minimum free energy of a folded mRNA molecule using mfold (http://mfold.rna.albany.edu/?q=mfold/RNA-Folding-Form2.3, date last accessed, January 13, 2016) (48,49), the Exon 3 of DUX4 was predicted to form highly stable secondary structure with free energies of −56.3 kcal/mol (Fig. 6). The resulting outputs show that DUX4 poly(A) is located within the region of an open conformation, whereas the cleavage site and the DSE lay within a closed region. The PAS, thus, appears more accessible than the DSE and this could explain the low efficacy of the PMO-DSE which totally overlap a closed region.

Figure 6.

Secondary structure predictions of DUX4 mRNA. Mfold analysis was performed on the 3′end of DUX4 (starting first nucleotide of Exon 3 and ending 230 nucleotides downstream). The PAS and the cleavage region are boxed in red. The positions of the sequences recognized by the PMOs are in yellow for the PMOs inducing a strong down-regulation of DUX4, whereas the others are in blue. Examples of opened and closed RNA secondary structures are indicated.

Figure 6.

Secondary structure predictions of DUX4 mRNA. Mfold analysis was performed on the 3′end of DUX4 (starting first nucleotide of Exon 3 and ending 230 nucleotides downstream). The PAS and the cleavage region are boxed in red. The positions of the sequences recognized by the PMOs are in yellow for the PMOs inducing a strong down-regulation of DUX4, whereas the others are in blue. Examples of opened and closed RNA secondary structures are indicated.

Different mechanisms may be involved in the down-regulation of the targeted mRNA in the presence of the PMOs. The first one is that the duplex formation of the PMO with the mRNA 3′UTR could prevent poly(A) tail elongation. The second one could be the removal of the entire tail mediated by a deadenylase-dependent cutting of the poly(A) tail as previously described in zebrafish (50), thus affecting mRNA stability. Another possibility is that spatial rearrangements of the 3′UTR in the presence of the PMO could have modulated the nonsense mediated decay (NMD) pathway (51). On the one hand, it was previously shown that NMD is an endogenous suppressor of DUX4 mRNA that contributes to the low expression of DUX4 in FSHD cells (30). But on the other hand, DUX4 expression was also described to inhibit NMD by decreasing Up-frameshift protein 1 (UPF-1) levels (30). The PMO-DUX4 duplex formation, causing a destabilization of DUX4 mRNA by at least one of the mechanisms described earlier, could have thus increased NMD activity, enhancing DUX4 repression.

Recent discoveries have revealed that many human genes use more than one poly(A) and cleavage sites, indicating that alternative polyadenylation (APA) is a widespread phenomenon. The use of APA has been described to play important roles in several human diseases such as (i) in oculopharyngeal muscular dystrophy, in which mutations in PolyA binding protein nuclear 1 (PABPN1) result in a global enhancement of the usage of proximal cleavage sites, leading to an unbalanced formation of alternative mRNA 3′ ends and to a modification of gene expression through aberrant microRNA regulation (52); (ii) in β-thalassemia, in which a loss of function of globin 3′end processing linked to an alteration of the poly(A) site is one of the causes of this disease, resulting in the generation of an unstable transcript that is ∼900 nt longer (1,7); (iii) a single-nucleotide polymorphism in the U/GU rich region downstream of an alternative PAS in the fibrinogen gamma haplotype 2 gene has been shown to affect the usage of this poly(A) and has been associated with an increased risk of deep-venous thrombosis (53). Considering that the use of AOs targeting the poly(A) site can lead to the use of APA and cleavage site (44,45), one important point raised in our experiments was to determine if such a redirection was observed in the presence of the PMOs. Although in the presence of PMO-CS3 the cleavage site was detected to be redirected by ∼40 nt upstream of its normal position, it was still considerably distal to any of the previously described alternative PASs (4). Interestingly, this finding does not fit with the current consensus describing that there are four general classes of APA, including: tandem 3′UTR APA; alternative terminal exon APA; intronic APA and internal exon APA (1). This could be explained by the steric congestion at the PAS and cleavage site provided by the PMO-CS3 and the cellular proteins involved in the polyadenylation process, respectively. The cleavage factors would be then forced to cleave the pre-mRNA upstream of the PAS, resulting in a new cleavage site, but still utilizing the original poly(A) site. Importantly, in the 14 sequences analyzes, the same cleavage region was observed, strongly suggesting a specificity to the target sequence. Alternatively, a novel APA was used that has not been previously described (4).

A redirection of the cleavage site associated to the alternative poly(A) site located in Exon 7 of DUX4 was not observed either. It was shown that chromosomes 4 and 10 produce DUX4 mRNA in human testes (18). These transcripts can use alternative 3′ exons with a PAS in Exon 7, leading to a DUX4 mRNA carrying Exons 1, 2, 6 and 7 or Exons 1, 2, 4–7. Exons 3 and 7, thus, seem to be mutually exclusive. It is, therefore, not surprising to not observe a redirection of the cleavage site to Exon 7 when Exon 3 is targeted.

Targeting the 3′ end elements of mRNA could thus represent an alternative approach for gene silencing compared with classic AO approaches leading to exon skipping, translation inhibition or mRNA destruction into the cytoplasm. The strategy to interfere with the mRNA 3′end processing provides a number of advantages. First, polyadenylation is a crucial step required for the maturation of almost all eukaryotic mRNAs (except for histones), meaning that virtually any mRNA can be subjected to this strategy. Secondly, genes with only 1 exon can be targeted using AON hybridizing the 3′ end key elements. Thirdly, AON targeting the PASs may be investigated in a therapeutic context for some disease caused by utilization of inadequate APAs. For example, some mantle cell lymphoma patients express isoforms of cyclinD1 mRNA with short 3′UTR compared with healthy individuals. Shortening of the 3′UTR is caused by the mutations in the CCND1 gene leading to the creation of a premature PAS. This mutated mRNA lacks most of the 3′UTR region that contains mRNA destabilizing elements, making the CCDN1 mRNA abnormally stable (54). Therefore, targeting a premature poly(A) with AON to promote the use of the distal poly(A) may also serve as an promising approach to treat disease.

Materials and Methods

PMO were manufactured and supplied by Gene Tools (LLC, Philomath, OR, USA). The DNA leashes for PMO transfection were synthesized by Eurogentec and their sequences are indicated in Table 1. Before transfection, the PMOs (2.5 µl at 1 mm) were annealed with the leash (25 µl at100 µm) in the final volume of 50 µl (12.5 µl of PBS10× and 10 µl of RNAse-free water) (final concentration of leashed PMO 50 µm) using the following protocol: 95°C for 5 min, 85°C for 1 min, 75°C for 1 min, 65°C for 5 min, 55°C for 1 min, 45°C for 1 min, 35°C for 5 min, 25°C for 1 min and then held at 15°C. Efficient annealing was verified using gel electrophoresis band shift as previously described (55). Leashed PMOs were stored at 4°C.

Immortalized FSHD clones have been characterized previously (56). Cells were cultivated in the proliferation medium [4 vols of Dulbecco's modified Eagle medium (DMEM), 1 vol of 199 medium, fetal bovine serum 20%, gentamycin 50 µg/ml (Life Technologies, Saint Aubin, France)] supplemented with insulin 5 µg/ml, dexamethasone 0.2 µg/ml, β-fibroblast growth factor (β-FGF) 0.5 ng/ml, human epidermal growth factor (hEGF) 5 ng/ml and fetuine 25 µg/ml. The differentiation was induced by replacing the proliferation medium by DMEM supplemented with insulin (10 µg/ml). Two days after the induction of differentiation, the transfections were realized using Lipofectamine® RNAiMax Reagent (Life Technologies) according to the manufacturer's instructions. Cells were harvested at Day 4 of differentiation.

RNA extraction from cultured cells was performed using Trizol according the manufacturer's protocol (Life Technologies, Saint Aubin, France). RNA concentration was determined using a nanodrop ND-1000 spectrophotometer (Thermo Scientific, Wilmington, DE, USA). Reverse transcription was performed on 1 µg of total RNA in a 10 µL final volume (Roche Transcriptor First Strand cDNA Synthesis kit, Roche, Meylan, France) with a NVT primer (according to the International Union of Pure and Applied Chemistry, N = any base; V = adenine or cytosine or guanine and T = thymine) (GCGAGCTCCGCGGCCGCGTTTTTTTTTTTVN) (10 μM) according to the manufacturer instructions. The PCR for DUX4-all was performed on 1 µl of cDNA using 2X Reddy Mix PCR Master Mix (Thermo Scientific) according the manufacturer's instructions. The PCR program was the following: 94°C for 5 min, followed by 35 cycles at 94°C for 20 s and 60°C for 20 s and 72°C for 20 s, finished with 72°C for 7 min. Quantitative PCR (qPCR) was designed according to the minimum of information for publication of quantitative (MIQE) standards (57). The primers used for both PCR and qPCR are indicated in Table 3. Quantitative PCRs were performed in a final volume of 9 µl with 4 µl of RT product, 0.18 µl of each forward and reverse primers at 20 µm, and 4.5 µl of SYBR® Green mastermix 2× (Roche, Meylan, France) in a lightcycler LC480 (Roche). The qPCR cycling conditions were 94°C for 5 min, followed by 50 cycles at 94°C for 10 s and 60°C for 5 s and 72°C for 5 s. All the amplicons were validated using sequencing. B2M was the most appropriate gene for normalization: a multiparametric analysis of variance (MANOVA) and a Newman–Keuls post hoc test did not reveal any correlation between B2M expression and PMO addition (at different concentrations).

Table 3.

Primers used in this study

 Targeted gene Primer Accession number Sequence Location Size (bp) qPCR efficiency (calculated from slope) 
RT-oligonucleotides  Oligo_dT_3′RACE N.A. GCTGTCAACGATACGCTACGTAACGGCATGACAGTGTTTTTTTTTTTTTTTTTTTTTTTT N.A. N.A. N.A. 
Oligo_NVT N.A. GCGAGCTCCGCGGCCGCGTTTTTTTTTTTVN N.A. N.A. N.A. 
PCR oligonucleotides DUX4-all DUX4-all_fw ENSG00000260596 CCCAGGTACCAGCAGACC Spanning Exon 2–Exon 3 164 bp N.A. 
DUX4-all_rev TCCAGGAGATGTAACTCTAATCCA Exon 3 
DUX4-3′UTR DUX4-3′UTR_fw HQ266760 (DUX4-fl1) and HQ266761 (DUX4-fl2AGGCGCAACCTCTCCTAGAAAC Exon 1 368 bp and 504 bp N.A. 
DUX4-3′UTR_rev TCCAGGAGATGTAACTCTAATCCA Exon 3 
qPCR oligonucleotides B2M B2M_fw NM_004048.2 CTCTCTTTCTGGCCTGGAGG Exon 1 67 bp 100% 
B2M_rev TGCTGGATGACGTGAGTAAACC Exon 2 
MBD3L2 MBD3L2_fw NM_144614.3 CGTTCACCTCTTTTCCAAGC Exon 1 142 bp 106% 
MBD3L2_rev AGTCTCATGGGGAGAGCAGA Exon 2 
TRIM 43 TRIM43_fw NM_138800.1 ACCCATCACTGGACTGGTGT Exon 6 100 bp 109% 
TRIM43_rev CACATCCTCAAAGAGCCTGA Exon 7 
ZSCAN4 ZSCAN4_962U20_fw NM_152677.2 CTGGAGCAGTTTATGATTGG Exon 3 162 bp 98% 
ZSCAN4_rev AGCTTCCTGTCCCTGCATGT Exon 4 
3′RACE-PCR oligonucleotides  3RACE_3′_primer_rev N.A. GCTGTCAACGATACGCTACGTAACG N.A. N.A. N.A. 
3RACE_3′_nested_rev N.A. CGCTACGTAACGGCATGACAGTG N.A. N.A. N.A. 
DUX4 3RACE_Exon1_fw ENSG00000260596 CTCTCCTAGAAACGGAGGCCCCG Exon 1 N.A. N.A. 
3RACE_Exon1_nested_fw CTCGCTGGAAGCACCCCTCAGC Exon 1 N.A. N.A. 
3RACE_3′_Exon3_fw ENSG00000260596 CGCACCCCGGCTGACGTGCAAG Exon 3 N.A. N.A. 
3RACE_3′_Exon3_nested_fw CGCTGGCCTCTCTGTGCCCTTG Exon 3 N.A. N.A. 
 Targeted gene Primer Accession number Sequence Location Size (bp) qPCR efficiency (calculated from slope) 
RT-oligonucleotides  Oligo_dT_3′RACE N.A. GCTGTCAACGATACGCTACGTAACGGCATGACAGTGTTTTTTTTTTTTTTTTTTTTTTTT N.A. N.A. N.A. 
Oligo_NVT N.A. GCGAGCTCCGCGGCCGCGTTTTTTTTTTTVN N.A. N.A. N.A. 
PCR oligonucleotides DUX4-all DUX4-all_fw ENSG00000260596 CCCAGGTACCAGCAGACC Spanning Exon 2–Exon 3 164 bp N.A. 
DUX4-all_rev TCCAGGAGATGTAACTCTAATCCA Exon 3 
DUX4-3′UTR DUX4-3′UTR_fw HQ266760 (DUX4-fl1) and HQ266761 (DUX4-fl2AGGCGCAACCTCTCCTAGAAAC Exon 1 368 bp and 504 bp N.A. 
DUX4-3′UTR_rev TCCAGGAGATGTAACTCTAATCCA Exon 3 
qPCR oligonucleotides B2M B2M_fw NM_004048.2 CTCTCTTTCTGGCCTGGAGG Exon 1 67 bp 100% 
B2M_rev TGCTGGATGACGTGAGTAAACC Exon 2 
MBD3L2 MBD3L2_fw NM_144614.3 CGTTCACCTCTTTTCCAAGC Exon 1 142 bp 106% 
MBD3L2_rev AGTCTCATGGGGAGAGCAGA Exon 2 
TRIM 43 TRIM43_fw NM_138800.1 ACCCATCACTGGACTGGTGT Exon 6 100 bp 109% 
TRIM43_rev CACATCCTCAAAGAGCCTGA Exon 7 
ZSCAN4 ZSCAN4_962U20_fw NM_152677.2 CTGGAGCAGTTTATGATTGG Exon 3 162 bp 98% 
ZSCAN4_rev AGCTTCCTGTCCCTGCATGT Exon 4 
3′RACE-PCR oligonucleotides  3RACE_3′_primer_rev N.A. GCTGTCAACGATACGCTACGTAACG N.A. N.A. N.A. 
3RACE_3′_nested_rev N.A. CGCTACGTAACGGCATGACAGTG N.A. N.A. N.A. 
DUX4 3RACE_Exon1_fw ENSG00000260596 CTCTCCTAGAAACGGAGGCCCCG Exon 1 N.A. N.A. 
3RACE_Exon1_nested_fw CTCGCTGGAAGCACCCCTCAGC Exon 1 N.A. N.A. 
3RACE_3′_Exon3_fw ENSG00000260596 CGCACCCCGGCTGACGTGCAAG Exon 3 N.A. N.A. 
3RACE_3′_Exon3_nested_fw CGCTGGCCTCTCTGTGCCCTTG Exon 3 N.A. N.A. 

N.A., non-applicable.

For the 3′RACE, the RT was realized with an oligo-dT adapter primer: (GCTGTCAACGATACGCTACGTAACGGCATGACAGTGTTTTTTTTTTTTTTTTTTTTTTTT) as described in Table 3. The primary PCRs were performed in a final volume of 15 µl with 1 µl of RT product, 3 µl of reverse primer at 20 µm (Table 3) and 1 µl of primer at 20 µm, 7.5 µl 2X Reddy Mix PCR Master Mix (Thermo Scientific). The PCR cycling conditions were 94°C for 5 min, followed by 5 cycles at 94°C for 20 s and 72°C for 50 s, then 5 cycles at 94°C for 20 s and 70°C for 30 s and 72°C for 20 s and 25 cycles at 94°C for 20 s and 68°C for 20 s and 72°C for 30 s, finished with 72°C for 7 min. The nested PCRs were realized in a final volume of 15 µl with 1 µl of primary PCR product, 1 µl of reverse and/or forward primers 20 µm, 7.5 µl 2X Reddy Mix PCR Master Mix (Thermo Scientific). The PCR cycling conditions were 94°C for 5 min, followed by 35 cycles at 94°C for 20 s and 60°C for 20 s and 72°C for 30 s, finished with 72°C for 7 min. Nested PCR products were purified according to NucleoSpin Gel and PCR Clean-Up manufacturer's instructions (Macherey–Nagel). Fragments were cloned into TOPO-TA vector with TOPO-TA Cloning kit (Life Technologies). Cleavage sites were determined by sequencing at least 10 colony-forming units.

Supplementary Material

Supplementary Material is available at HMG online.

Conflict of Interest statement. None declared.

Funding

This study was financially supported by the French Association against Myopathies (AFM-Téléthon, France) and the National Research Agency (FSHDecipher, ANR-13-BSV1-0004) and The Rosetrees Trust, UK.

References

1
Elkon
R.
,
Ugalde
A.P.
,
Agami
R.
(
2013
)
Alternative cleavage and polyadenylation: extent, regulation and function
.
Nat. Rev.
 ,
14
,
496
506
.
2
Nunes
N.M.
,
Li
W.
,
Tian
B.
,
Furger
A.
(
2010
)
A functional human Poly(A) site requires only a potent DSE and an A-rich upstream sequence
.
EMBO J.
 ,
29
,
1523
1536
.
3
Proudfoot
N.J.
(
2011
)
Ending the message: poly(A) signals then and now
.
Genes Dev.
 ,
25
,
1770
1782
.
4
Tian
B.
,
Hu
J.
,
Zhang
H.
,
Lutz
C.S.
(
2005
)
A large-scale analysis of mRNA polyadenylation of human and mouse genes
.
Nucleic Acids Res.
 ,
33
,
201
212
.
5
Beaudoing
E.
,
Freier
S.
,
Wyatt
J.R.
,
Claverie
J.M.
,
Gautheret
D.
(
2000
)
Patterns of variant polyadenylation signal usage in human genes
.
Genome Res.
 ,
10
,
1001
1010
.
6
Gruber
A.R.
,
Martin
G.
,
Keller
W.
,
Zavolan
M.
(
2014
)
Means to an end: mechanisms of alternative polyadenylation of messenger RNA precursors
.
Wiley Interdiscip. Rev.
 ,
5
,
183
196
.
7
Danckwardt
S.
,
Hentze
M.W.
,
Kulozik
A.E.
(
2008
)
3’ end mRNA processing: molecular mechanisms and implications for health and disease
.
EMBO J.
 ,
27
,
482
498
.
8
Sachs
A.
(
1990
)
The role of poly(A) in the translation and stability of mRNA
.
Curr. Opin. Cell Biol.
 ,
2
,
1092
1098
.
9
Hollerer
I.
,
Grund
K.
,
Hentze
M.W.
,
Kulozik
A.E.
(
2014
)
mRNA 3'end processing: A tale of the tail reaches the clinic
.
EMBO Mol. Med.
 ,
6
,
16
26
.
10
Tawil
R.
,
van der Maarel
S.M.
,
Tapscott
S.J.
(
2014
)
Facioscapulohumeral dystrophy: the path to consensus on pathophysiology
.
Skeletal Muscle
 ,
4
,
12
.
11
van der Maarel
S.M.
,
Miller
D.G.
,
Tawil
R.
,
Filippova
G.N.
,
Tapscott
S.J.
(
2012
)
Facioscapulohumeral muscular dystrophy: consequences of chromatin relaxation
.
Curr. Opin. Neurol.
 ,
25
,
614
620
.
12
van Deutekom
J.C.
,
Wijmenga
C.
,
van Tienhoven
E.A.
,
Gruter
A.M.
,
Hewitt
J.E.
,
Padberg
G.W.
,
van Ommen
G.J.
,
Hofker
M.H.
,
Frants
R.R.
(
1993
)
FSHD associated DNA rearrangements are due to deletions of integral copies of a 3.2 kb tandemly repeated unit
.
Hum. Mol. Genet.
 ,
2
,
2037
2042
.
13
Wijmenga
C.
,
Hewitt
J.E.
,
Sandkuijl
L.A.
,
Clark
L.N.
,
Wright
T.J.
,
Dauwerse
H.G.
,
Gruter
A.M.
,
Hofker
M.H.
,
Moerer
P.
,
Williamson
R.
et al
. (
1992
)
Chromosome 4q DNA rearrangements associated with facioscapulohumeral muscular dystrophy
.
Nat. Genet.
 ,
2
,
26
30
.
14
Lemmers
R.J.
,
Tawil
R.
,
Petek
L.M.
,
Balog
J.
,
Block
G.J.
,
Santen
G.W.
,
Amell
A.M.
,
van der Vliet
P.J.
,
Almomani
R.
,
Straasheijm
K.R.
et al
. (
2012
)
Digenic inheritance of an SMCHD1 mutation and an FSHD-permissive D4Z4 allele causes facioscapulohumeral muscular dystrophy type 2
.
Nat. Genet.
 ,
44
,
1370
1374
.
15
Lemmers
R.J.
,
van den Boogaard
M.L.
,
van der Vliet
P.J.
,
Donlin-Smith
C.M.
,
Nations
S.P.
,
Ruivenkamp
C.A.
,
Heard
P.
,
Bakker
B.
,
Tapscott
S.
,
Cody
J.D.
et al
. (
2015
)
Hemizygosity for SMCHD1 in facioscapulohumeral muscular dystrophy Type 2: consequences for 18p deletion syndrome
.
Hum. Mutat.
 ,
36
,
679
683
.
16
Lemmers
R.J.
,
van der Vliet
P.J.
,
Klooster
R.
,
Sacconi
S.
,
Camano
P.
,
Dauwerse
J.G.
,
Snider
L.
,
Straasheijm
K.R.
,
van Ommen
G.J.
,
Padberg
G.W.
et al
. (
2010
)
A unifying genetic model for facioscapulohumeral muscular dystrophy
.
Science
 
329
,
1650
1653
.
17
Snider
L.
,
Asawachaicharn
A.
,
Tyler
A.E.
,
Geng
L.N.
,
Petek
L.M.
,
Maves
L.
,
Miller
D.G.
,
Lemmers
R.J.
,
Winokur
S.T.
,
Tawil
R.
et al
. (
2009
)
RNA transcripts, miRNA-sized fragments and proteins produced from D4Z4 units: new candidates for the pathophysiology of facioscapulohumeral dystrophy
.
Hum. Mol. Genet.
 ,
18
,
2414
2430
.
18
Snider
L.
,
Geng
L.N.
,
Lemmers
R.J.
,
Kyba
M.
,
Ware
C.B.
,
Nelson
A.M.
,
Tawil
R.
,
Filippova
G.N.
,
van der Maarel
S.M.
,
Tapscott
S.J.
et al
. (
2010
)
Facioscapulohumeral dystrophy: incomplete suppression of a retrotransposed gene
.
PLoS Genet.
 ,
6
,
e1001181
.
19
Jones
T.I.
,
Chen
J.C.
,
Rahimov
F.
,
Homma
S.
,
Arashiro
P.
,
Beermann
M.L.
,
King
O.D.
,
Miller
J.B.
,
Kunkel
L.M.
,
Emerson
C.P.
Jr.
et al
. (
2012
)
Facioscapulohumeral muscular dystrophy family studies of DUX4 expression: evidence for disease modifiers and a quantitative model of pathogenesis
.
Hum. Mol. Genet.
 ,
21
,
4419
4430
.
20
Broucqsault
N.
,
Morere
J.
,
Gaillard
M.C.
,
Dumonceaux
J.
,
Torrents
J.
,
Salort-Campana
E.
,
Maues de Paula
A.
,
Bartoli
M.
,
Fernandez
C.
,
Chesnais
A.L.
et al
. (
2013
)
Dysregulation of 4q35- and muscle-specific genes in fetuses with a short D4Z4 array linked to Facio-Scapulo-Humeral Dystrophy
.
Hum. Mol. Genet.
 ,
22
,
4206
4214
.
21
Ferreboeuf
M.
,
Mariot
V.
,
Bessieres
B.
,
Vasiljevic
A.
,
Attie-Bitach
T.
,
Collardeau
S.
,
Morere
J.
,
Roche
S.
,
Magdinier
F.
,
Robin-Ducellier
J.
et al
. (
2014
)
DUX4 and DUX4 downstream target genes are expressed in fetal FSHD muscles
.
Hum. Mol. Genet.
 ,
23
,
171
181
.
22
Block
G.J.
,
Narayanan
D.
,
Amell
A.M.
,
Petek
L.M.
,
Davidson
K.C.
,
Bird
T.D.
,
Tawil
R.
,
Moon
R.T.
,
Miller
D.G.
(
2013
)
Wnt/beta-catenin signaling suppresses DUX4 expression and prevents apoptosis of FSHD muscle cells
.
Hum. Mol. Genet.
 ,
22
,
390
396
.
23
Ferreboeuf
M.
,
Mariot
V.
,
Furling
D.
,
Butler-Browne
G.
,
Mouly
V.
,
Dumonceaux
J.
(
2014
)
Nuclear protein spreading: implication for pathophysiology of neuromuscular diseases
.
Hum. Mol. Genet.
 ,
23
,
4125
4133
.
24
Kowaljow
V.
,
Marcowycz
A.
,
Ansseau
E.
,
Conde
C.B.
,
Sauvage
S.
,
Matteotti
C.
,
Arias
C.
,
Corona
E.D.
,
Nunez
N.G.
,
Leo
O.
et al
. (
2007
)
The DUX4 gene at the FSHD1A locus encodes a pro-apoptotic protein
.
Neuromuscul. Disord.
 ,
17
,
611
623
.
25
Dixit
M.
,
Ansseau
E.
,
Tassin
A.
,
Winokur
S.
,
Shi
R.
,
Qian
H.
,
Sauvage
S.
,
Matteotti
C.
,
van Acker
A.M.
,
Leo
O.
et al
. (
2007
)
DUX4, a candidate gene of facioscapulohumeral muscular dystrophy, encodes a transcriptional activator of PITX1
.
Proc. Natl Acad. Sci. USA
 ,
104
,
18157
18162
.
26
Wallace
L.M.
,
Garwick
S.E.
,
Mei
W.
,
Belayew
A.
,
Coppee
F.
,
Ladner
K.J.
,
Guttridge
D.
,
Yang
J.
,
Harper
S.Q.
(
2011
)
DUX4, a candidate gene for facioscapulohumeral muscular dystrophy, causes p53-dependent myopathy in vivo
.
Ann. Neurol.
 ,
69
,
540
552
.
27
Geng
L.N.
,
Yao
Z.
,
Snider
L.
,
Fong
A.P.
,
Cech
J.N.
,
Young
J.M.
,
van der Maarel
S.M.
,
Ruzzo
W.L.
,
Gentleman
R.C.
,
Tawil
R.
et al
. (
2012
)
DUX4 activates germline genes, retroelements, and immune mediators: implications for facioscapulohumeral dystrophy
.
Dev. Cell
 ,
22
,
38
51
.
28
Xu
H.
,
Wang
Z.
,
Jin
S.
,
Hao
H.
,
Zheng
L.
,
Zhou
B.
,
Zhang
W.
,
Lv
H.
,
Yuan
Y.
(
2014
)
Dux4 induces cell cycle arrest at G1 phase through upregulation of p21 expression
.
Biochem. Biophys. Res. Commun.
 ,
446
,
235
240
.
29
Rickard
A.M.
,
Petek
L.M.
,
Miller
D.G.
(
2015
)
Endogenous DUX4 expression in FSHD myotubes is sufficient to cause cell death and disrupts RNA splicing and cell migration pathways
.
Hum. Mol. Genet.
 ,
24
,
5901
5914
.
30
Feng
Q.
,
Snider
L.
,
Jagannathan
S.
,
Tawil
R.
,
van der Maarel
S.M.
,
Tapscott
S.J.
,
Bradley
R.K.
(
2015
)
A feedback loop between nonsense-mediated decay and the retrogene DUX4 in facioscapulohumeral muscular dystrophy
.
eLife
 ,
4
, .
31
Tupler
R.
,
Berardinelli
A.
,
Barbierato
L.
,
Frants
R.
,
Hewitt
J.E.
,
Lanzi
G.
,
Maraschio
P.
,
Tiepolo
L.
(
1996
)
Monosomy of distal 4q does not cause facioscapulohumeral muscular dystrophy
.
J. Med. Genet.
 ,
33
,
366
370
.
32
Lemmers
R.J.
,
de Kievit
P.
,
Sandkuijl
L.
,
Padberg
G.W.
,
van Ommen
G.J.
,
Frants
R.R.
,
van der Maarel
S.M.
(
2002
)
Facioscapulohumeral muscular dystrophy is uniquely associated with one of the two variants of the 4q subtelomere
.
Nat. Genet.
 ,
32
,
235
236
.
33
Thomas
N.S.
,
Wiseman
K.
,
Spurlock
G.
,
MacDonald
M.
,
Ustek
D.
,
Upadhyaya
M.
(
2007
)
A large patient study confirming that facioscapulohumeral muscular dystrophy (FSHD) disease expression is almost exclusively associated with an FSHD locus located on a 4qA-defined 4qter subtelomere
.
J. Med. Genet.
 ,
44
,
215
218
.
34
Yao
Z.
,
Snider
L.
,
Balog
J.
,
Lemmers
R.J.
,
Van Der Maarel
S.M.
,
Tawil
R.
,
Tapscott
S.J.
(
2014
)
DUX4-induced gene expression is the major molecular signature in FSHD skeletal muscle
.
Hum. Mol. Genet.
 ,
23
,
5342
5352
.
35
Wallace
L.M.
,
Liu
J.
,
Domire
J.S.
,
Garwick-Coppens
S.E.
,
Guckes
S.M.
,
Mendell
J.R.
,
Flanigan
K.M.
,
Harper
S.Q.
(
2012
)
RNA interference inhibits DUX4-induced muscle toxicity in vivo: implications for a targeted FSHD Therapy
.
Mol. Ther.
 ,
20
,
1417
1423
.
36
Mitsuhashi
H.
,
Mitsuhashi
S.
,
Lynn-Jones
T.
,
Kawahara
G.
,
Kunkel
L.M.
(
2013
)
Expression of DUX4 in zebrafish development recapitulates facioscapulohumeral muscular dystrophy
.
Hum. Mol. Genet.
 ,
22
,
568
577
.
37
Vanderplanck
C.
,
Ansseau
E.
,
Charron
S.
,
Stricwant
N.
,
Tassin
A.
,
Laoudj-Chenivesse
D.
,
Wilton
S.D.
,
Coppee
F.
,
Belayew
A.
(
2011
)
The FSHD atrophic myotube phenotype is caused by DUX4 expression
.
PLoS One
 ,
6
,
e26820
.
38
Kole
R.
,
Krainer
A.R.
,
Altman
S.
(
2012
)
RNA therapeutics: beyond RNA interference and antisense oligonucleotides
.
Nat. Rev. Drug Discov.
 ,
11
,
125
140
.
39
DeVos
S.L.
,
Miller
T.M.
(
2013
)
Antisense oligonucleotides: treating neurodegeneration at the level of RNA
.
Neurotherapeutics
 ,
10
,
486
497
.
40
Cirak
S.
,
Arechavala-Gomeza
V.
,
Guglieri
M.
,
Feng
L.
,
Torelli
S.
,
Anthony
K.
,
Abbs
S.
,
Garralda
M.E.
,
Bourke
J.
,
Wells
D.J.
et al
. (
2011
)
Exon skipping and dystrophin restoration in patients with Duchenne muscular dystrophy after systemic phosphorodiamidate morpholino oligomer treatment: an open-label, phase 2, dose-escalation study
.
Lancet
 ,
378
,
595
605
.
41
Mendell
J.R.
,
Rodino-Klapac
L.R.
,
Sahenk
Z.
,
Roush
K.
,
Bird
L.
,
Lowes
L.P.
,
Alfano
L.
,
Gomez
A.M.
,
Lewis
S.
,
Kota
J.
et al
. (
2013
)
Eteplirsen for the treatment of Duchenne muscular dystrophy
.
Ann. Neurol.
 ,
74
,
637
647
.
42
Jarmin
S.
,
Kymalainen
H.
,
Popplewell
L.
,
Dickson
G.
(
2013
)
New developments in the use of gene therapy to treat Duchenne muscular dystrophy
.
Expert Opin. Biol. Ther.
 ,
14
,
209
230
.
43
Voit
T.
,
Topaloglu
H.
,
Straub
V.
,
Muntoni
F.
,
Deconinck
N.
,
Campion
G.
,
De Kimpe
S.J.
,
Eagle
M.
,
Guglieri
M.
,
Hood
S.
et al
. (
2014
)
Safety and efficacy of drisapersen for the treatment of Duchenne muscular dystrophy (DEMAND II): an exploratory, randomised, placebo-controlled phase 2 study
.
Lancet
 ,
13
,
987
996
.
44
Raz
V.
,
Buijze
H.
,
Raz
Y.
,
Verwey
N.
,
Anvar
S.Y.
,
Aartsma-Rus
A.
,
van der Maarel
S.M.
(
2014
)
A novel feed-forward loop between ARIH2 E3-ligase and PABPN1 regulates aging-associated muscle degeneration
.
Am. J. Pathol.
 ,
184
,
1119
1131
.
45
Vickers
T.A.
,
Wyatt
J.R.
,
Burckin
T.
,
Bennett
C.F.
,
Freier
S.M.
(
2001
)
Fully modified 2’ MOE oligonucleotides redirect polyadenylation
.
Nucleic Acids Res.
 ,
29
,
1293
1299
.
46
Mariot
V.
,
Roche
S.
,
Hourde
C.
,
Portilho
D.
,
Sacconi
S.
,
Puppo
F.
,
Duguez
S.
,
Rameau
P.
,
Caruso
N.
,
Delezoide
A.L.
et al
. (
2015
)
Correlation between low FAT1 expression and early affected muscle in facioscapulohumeral muscular dystrophy
.
Ann. Neurol.
 ,
78
,
387
400
.
47
Millevoi
S.
,
Vagner
S.
(
2010
)
Molecular mechanisms of eukaryotic pre-mRNA 3’ end processing regulation
.
Nucleic Acids Res.
 ,
38
,
2757
2774
.
48
Zuker
M.
(
2003
)
Mfold web server for nucleic acid folding and hybridization prediction
.
Nucleic Acids Res.
 ,
31
,
3406
3415
.
49
Wu
J.C.
,
Gardner
D.P.
,
Ozer
S.
,
Gutell
R.R.
,
Ren
P.
(
2009
)
Correlation of RNA secondary structure statistics with thermodynamic stability and applications to folding
.
J. Mol. Biol.
 ,
391
,
769
783
.
50
Wada
T.
,
Hara
M.
,
Taneda
T.
,
Qingfu
C.
,
Takata
R.
,
Moro
K.
,
Takeda
K.
,
Kishimoto
T.
,
Handa
H.
(
2012
)
Antisense morpholino targeting just upstream from a poly(A) tail junction of maternal mRNA removes the tail and inhibits translation
.
Nucleic Acids Res.
 ,
40
,
e173
.
51
Eberle
A.B.
,
Stalder
L.
,
Mathys
H.
,
Orozco
R.Z.
,
Muhlemann
O.
(
2008
)
Posttranscriptional gene regulation by spatial rearrangement of the 3’ untranslated region
.
PLoS Biol.
 ,
6
,
e92
.
52
Jenal
M.
,
Elkon
R.
,
Loayza-Puch
F.
,
van Haaften
G.
,
Kuhn
U.
,
Menzies
F.M.
,
Oude Vrielink
J.A.
,
Bos
A.J.
,
Drost
J.
,
Rooijers
K.
et al
. (
2012
)
The poly(A)-binding protein nuclear 1 suppresses alternative cleavage and polyadenylation sites
.
Cell
 ,
149
,
538
553
.
53
Uitte de Willige
S.
,
Rietveld
I.M.
,
De Visser
M.C.
,
Vos
H.L.
,
Bertina
R.M.
(
2007
)
Polymorphism 10034C>T is located in a region regulating polyadenylation of FGG transcripts and influences the fibrinogen gamma'/gammaA mRNA ratio
.
J. Thromb. Haemost.
 ,
5
,
1243
1249
.
54
Wiestner
A.
,
Tehrani
M.
,
Chiorazzi
M.
,
Wright
G.
,
Gibellini
F.
,
Nakayama
K.
,
Liu
H.
,
Rosenwald
A.
,
Muller-Hermelink
H.K.
,
Ott
G.
et al
. (
2007
)
Point mutations and genomic deletions in CCND1 create stable truncated cyclin D1 mRNAs that are associated with increased proliferation rate and shorter survival
.
Blood
 ,
109
,
4599
4606
.
55
Popplewell
L.J.
,
Graham
I.R.
,
Malerba
A.
,
Dickson
G.
(
2011
)
Bioinformatic and functional optimization of antisense phosphorodiamidate morpholino oligomers (PMOs) for therapeutic modulation of RNA splicing in muscle
.
Methods Mol. Biol.
 ,
709
,
153
178
.
56
Krom
Y.D.
,
Dumonceaux
J.
,
Mamchaoui
K.
,
den Hamer
B.
,
Mariot
V.
,
Negroni
E.
,
Geng
L.N.
,
Martin
N.
,
Tawil
R.
,
Tapscott
S.J.
et al
. (
2012
)
Generation of isogenic D4Z4 contracted and noncontracted immortal muscle cell clones from a mosaic patient: a cellular model for FSHD
.
Am. J. Pathol.
 ,
181
,
1387
1401
.
57
Bustin
S.A.
,
Benes
V.
,
Garson
J.A.
,
Hellemans
J.
,
Huggett
J.
,
Kubista
M.
,
Mueller
R.
,
Nolan
T.
,
Pfaffl
M.W.
,
Shipley
G.L.
et al
. (
2009
)
The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments
.
Clin. Chem.
 ,
55
,
611
622
.

Author notes

L.C. and V.M. contributed equally.
Present address: NIHR Biomedical Research Centre, Institute of Child Health, University College London, 30 Guilford Street, London WC1N1EH, UK.