Skip to main content
PMC home page
eLife logoLink to eLife
. 2019 Dec 12;8:e51114. doi: 10.7554/eLife.51114

Straightjacket/α2δ3 deregulation is associated with cardiac conduction defects in myotonic dystrophy type 1

Emilie Auxerre-Plantié 1, Masayuki Nakamori 2, Yoan Renaud 1, Aline Huguet 3,4, Caroline Choquet 5, Cristiana Dondi 1, Lucile Miquerol 5, Masanori P Takahashi 6, Geneviève Gourdon 3,4, Guillaume Junion 1, Teresa Jagla 1, Monika Zmojdzian 1,✉, Krzysztof Jagla 1,✉
Editors: Margaret Buckingham7, Utpal Banerjee8
PMCID: PMC6908436  PMID: 31829940

Abstract

Cardiac conduction defects decrease life expectancy in myotonic dystrophy type 1 (DM1), a CTG repeat disorder involving misbalance between two RNA binding factors, MBNL1 and CELF1. However, how DM1 condition translates into conduction disorders remains poorly understood. Here we simulated MBNL1 and CELF1 misbalance in the Drosophila heart and performed TU-tagging-based RNAseq of cardiac cells. We detected deregulations of several genes controlling cellular calcium levels, including increased expression of straightjacket/α2δ3, which encodes a regulatory subunit of a voltage-gated calcium channel. Straightjacket overexpression in the fly heart leads to asynchronous heartbeat, a hallmark of abnormal conduction, whereas cardiac straightjacket knockdown improves these symptoms in DM1 fly models. We also show that ventricular α2δ3 expression is low in healthy mice and humans, but significantly elevated in ventricular muscles from DM1 patients with conduction defects. These findings suggest that reducing ventricular straightjacket/α2δ3 levels could offer a strategy to prevent conduction defects in DM1.

Research organism: D. melanogaster

Introduction

Myotonic dystrophy type 1 (DM1), the most prevalent muscular dystrophy in adults (Theadom et al., 2014), is caused by a CTG triplet repeat expansion in the 3’ untranslated region of the dystrophia myotonica protein kinase (dmpk) gene. Transcripts of the mutated dmpk, containing expanded CUGs, form hairpin-like secondary structures in the nuclei, and sequester RNA-binding proteins with affinity to CUG-rich sequences. Among them, Muscleblind-like 1 (MBNL1) protein is trapped within the repeats, forming nuclear foci aggregates that hallmark the disease (Davis et al., 1997; Taneja et al., 1995). In parallel, the CUGBP- and ELAV-like family member 1 (CELF1) is stabilized (Kuyumcu-Martinez et al., 2007), creating misbalance between MBNL1 and CELF1. This leads to missplicing of several transcripts and a general shift from adult to fetal isoforms (Freyermuth et al., 2016; Kino et al., 2009; Savkur et al., 2001). In addition, repeat toxicity induces a range of splice-independent alterations including impaired transcript stability (Sicot et al., 2011). A combination of splice-dependent and splice-independent events thus underlies DM1 pathogenesis, with the latter remaining largely unexplored. DM1 affects mainly skeletal muscles and the heart, with about 80% of DM1 patients showing impaired heart function with arrhythmia and conduction disturbance, which can sometimes end in heart block and sudden death (de Die-Smulders et al., 1998; Groh et al., 2008; Mathieu et al., 1999). Cardiac symptoms, and particularly conduction defects, thus decrease life expectancy in DM1 (Wang et al., 2009). Data suggest that cardiac phenotypes, including conduction defects, are due to MBNL1/CELF1 misbalance. It was shown in a DM1 mouse model that PKC phosphorylates CELF1 leading to increased CELF1 levels, whereas PKC inhibition caused CELF1 reduction and amelioration of cardiac dysfunction (Wang et al., 2009). This suggests that increased CELF1 levels could cause heart phenotypes in DM1, a possibility supported by findings that heart-specific upregulation of CELF1 reproduces functional and electrophysiological cardiac changes observed in DM1 patients and mouse model (Koshelev et al., 2010). In parallel, analyses of Mbnl1 mutant mice (Dixon et al., 2015) and evidence that misregulation of MBNL1-splice target gene SCN5A encoding a cardiac sodium channel leads to cardiac arrhythmia and conduction delay (Freyermuth et al., 2016), indicate that Mbnl1 contributes to DM1 heart phenotypes. However, in spite of aberrant SCN5A splicing (Freyermuth et al., 2016) and downregulation of a large set of miRNAs (Kalsotra et al., 2014), gene deregulations causing cardiac dysfunctions in DM1 remain to be characterized.

To gain further insight into mechanisms underlying cardiac DM1 phenotypes, we used previously described Drosophila DM1 models (Picchio et al., 2013). The heart of the fruit fly is simple in structure, but like the human heart, it displays pacemaker-regulated rhythmic beating, involving functions of conserved ion channels (Ocorr et al., 2007; Taghli-Lamallem et al., 2016). We simulated pathogenic MBNL1/CELF1 misbalance specifically in the fly heart by attenuating the Drosophila MBNL1 ortholog Muscleblind (Mbl) and by overexpressing the CELF1 counterpart Bruno-3 (Bru-3) (Picchio et al., 2018). This caused asynchronous heartbeat (anterior and posterior heart segments beating at different rates), which in Drosophila results from partial conduction block (Birse et al., 2010). Using these two fly DM1 models, we hoped to identify molecular players involved in DM1-associated conduction defects. We did not observe asynchronous heartbeats in flies expressing in the heart 960CTG repeats. This DM1 model (Picchio et al., 2013) developed other cardiac phenotypes such as arrhythmia. To identify deregulated genes underlying conduction defects, we applied a heart-targeted TU-tagging approach (Miller et al., 2009) followed by RNA sequencing. This cardiac cell-specific genome-wide approach yielded a discrete number of evolutionarily conserved candidate genes with altered cardiac expression in both DM1 models used, including regulators of cellular calcium. Among them, we found increased transcript levels of straightjacket (stj)/CACNA2D3 (α2δ3), which encodes a major regulatory subunit of Ca-α1D/Cav1.2 voltage-gated calcium channel, and is a key regulator of calcium current density, which triggers cardiac contraction (Bodi, 2005; Dolphin, 2013; Hoppa et al., 2012; Mesirca et al., 2015). The role of stj transcript level in proper conduction is supported by cardiac-specific overexpression of stj, which leads to asynchronous heartbeat. Conversely, attenuating its expression by cardiac-specific knockdown in both DM1 fly heart models reverses cardiac asynchrony. Our hypothesis that stj contributes to the cardiac DM1-associated pathology is supported by our finding that ventricular α2δ3 expression level is low in healthy mouse and human hearts, but is significantly increased in DM1 patients with cardiac conduction defects. Hence lowering α2δ3 in ventricular cardiomyocytes could offer a potential treatment strategy for DM1-associated conduction defects and in particular intraventricular conduction delay (IVCD).

Results

Imbalance of MBNL1 and CELF1 counterparts in the Drosophila heart results in asynchronous heartbeat

The Drosophila counterparts of MBNL1 and CELF1, the Mbl and Bru-3 proteins, are both present in the fly heart cells, including contractile cardiomyocytes, and localize predominantly but not exclusively to the nuclei (Figure 1—figure supplement 1). To mimic the DM1-associated imbalance between these two RNA binding factors, we either knocked down Mbl, or overexpressed Bru-3. We used the GAL4/UAS system with a cardiac-specific GAL4 driver line, Hand-GAL4 to target UAS-driven transgene expression in the adult fly heart (Figure 1A). We found that the UAS-MblRNAi line efficiently attenuated Mbl expression, whereas UAS-Bru-3 generated Bru-3 overexpression in GAL4-expressing cardiac cells (Figure 1—figure supplement 1). The Hand-GAL4 driven overexpression of Bru-3 and attenuation of Mbl, both appeared homogenous along the heart tube.

Figure 1. Cardiac-specific knockdown of MBNL1 ortholog and overexpression of CELF1 counterpart in Drosophila lead to asynchronous heartbeats.

(A) The adult Drosophila heart expressing Hand-Gal4-driven GFP (Hand>GFP). Note that this Gal4 line drives expression exclusively in the heart. Arrows indicate cardiomyocytes (CM) and pericardial cells (PC) and A2–A5 denote abdominal segments. Scale bar, 150 μm. (B) Movie and M-modes views illustrating asynchronous heartbeats in Hand>MblRNAi flies registered in two adjacent heart segments (A3 and A4). Two-sided arrows indicate heart diameter in diastolic state. (C) Barplot graph showing percentage of flies with conduction defects in the different genetic contexts tested. Note the higher impact of attenuation of Mbl compared to overexpression of Bru-3. Number of fly hearts tested (n) is indicated and statistical significance (Fisher’s exact test) denoted by * (p<0.05) and *** (p<0.001). Below are examples of M-modes illustrating ‘partial’ and ‘total’ heart blocks observed in Hand>MblRNAi and Hand>Bru-3 flies developing conduction defects.

Figure 1.

Figure 1—figure supplement 1. Bru-3 and Mbl are expressed in the adult fly heart and their transcript and protein levels are affected in Hand>Bru-3 and Hand>MblRNAi flies.

Figure 1—figure supplement 1.

(A–B) RT-qPCR analyses of Bru-3 overexpression and Mbl attenuation in the adult hearts from Hand>Bru-3 (A) and Hand>MblRNAi (B) flies. Three biological replicates (n = 10–12 hearts per replicate) were analysed for each condition. Statistical significance (Mann Whitney U test) is denoted by * (p<0.05) (C–F) Optical projection views of A3 cardiac segment showing expression of Bru-3 (C–D) and Mbl (E–H) proteins in the hearts of wild-type (Wt) (C, E), Hand>Bru-3 (D) and Hand>MblRNAi (F) flies. (G, H) show confocal optical sections at the level of cardiomyocytes (CM) to document attenuation of Mbl in CMs nuclei. Bru-3 protein is detected in the nuclei of CMs, pericardial cells (PC) and ventral longitudinal muscle (VLM). A faint granular cytoplasmic expression could also be detected in PCs (C). Notice an increase of Bru-3 levels in Hand>Bru-3 context (D) with more cytoplasmic Bru-3 signal detected in PCs and CMs. Similarly, Mbl accumulation is observed in CMs, PCs and the VLM nuclei (E). Apparent sarcomeric localization of Mbl is also detected in circular fibers of CMs and in the VLM (E). Notice Bru-3 and Mbl expression in nuclei of somatic muscles (sm). Scale bars (C–F) 100 μm; (G,H) 50 μm.
Figure 1—figure supplement 2. Effect of simultaneous cardiac attenuation of Mbl and overexpression of Bru-3 on conduction defects.

Figure 1—figure supplement 2.

Barplot graph showing percentage of flies with conduction defects. Note a moderate additive effect of simultaneous attenuation of Mbl and overexpression of Bru-3. Number of fly hearts tested (n) is indicated and statistical significance (Fisher’s exact test) denoted by ns (p>0.05) and *** (p<0.001).
Figure 1—video 1. An example of asynchronous heartbeat phenotype observed in Hand>MblRNAi context.
Download video file (11.1MB, mp4)
High-speed 20 s time-lapse movie.
Open in a new tab

To assess the cardiac function in these two DM1 models we used the semi-automated optical heartbeat analysis (SOHA) method, which consists in filming the adult beating heart from hemi-dissected flies and allows analysis of the contractile and rhythmic heart parameters (Fink et al., 2009; Ocorr et al., 2007). Besides frequently observed arrhythmia, about 58% of Hand>MblRNAi and 28% of Hand>Bru-3 flies displayed asynchronous heartbeat (Figure 1B,C and Figure 1—video 1), a hallmark of conduction disturbance (Birse et al., 2010). In some cases, asynchronous heartbeat can result in total heart block when at least one heart segment stops contracting. About 3% of Hand>MblRNAi and 4% of Hand>Bru-3 flies did develop a total heart block (Figure 1C). We also tested heartbeat in flies in which Mbl was attenuated in parallel to Bru-3 overexpression (Figure 1—figure supplement 2) and found a moderate additive effect on conduction defects of these two deregulations present in DM1. Taken together, these results show that the imbalance of Mbl/MBNL1 and Bru-3/CELF1 in the fly heart increases the occurrence of asynchronous heartbeat. This prompted us to perform transcriptional profiling of Hand>MblRNAi and Hand>Bru-3 in cardiac cells to identify misregulated gene expression underlying this phenotype.

Transcriptional profiling of cardiac DM1 Drosophila models using heart-specific TU-tagging approach

Transcriptional profiling of cardiac cells from Hand>MblRNAi and Hand>Bru-3 flies was performed using a cell-specific approach called TU-tagging (Miller et al., 2009), followed by RNA-seq. This method allows tissue-specific RNA isolation by expressing the uracil phosphoribosyltransferase (UPRT) enzyme cell-specifically and incorporation into neo-synthesized transcripts of a modified uracil, 4TU. Here, Hand-GAL4-driven UPRT expression in cardiac cells enabled us to isolate transcripts from adult heart cells only and we combined UAS-UPRT with UAS-MblRNAi and UAS-Bru-3 lines to perform TU-tagging in these two DM1 contexts (Figure 2A). Comparison of RNA-seq data from input RNA fraction and heart-specific TU-fraction from control Hand>LacZ;UPRT flies was used to assess the suitability of TU-tagging in low-abundance heart cells. We found that transcripts of cardiac-specific genes such as Hand and Tin were highly enriched in the cardiac-tagged TU fractions, whereas the transcripts of Npc1b and Obp19b genes expressed in midgut (Voght et al., 2007) and gustatory cells (Galindo and Smith, 2001), respectively, were depleted (Figure 2B). Thus the TU-tagging method proved well-suited to the transcriptional profiling of adult Drosophila cardiac cells, allowing the identification of both global gene deregulations (analyzed here) and isoform-specific differential gene expression (in a parallel study).

Figure 2. Cardiac-specific transcriptional profiling using TU-tagging method.

Figure 2.

Open in a new tab

(A) Pipeline of heart-targeted transcriptional profiling using TU-tagging. Note that Gal4- inducible UPRT transgene (UAS-UPRT) has been combined with UAS-MblRNAi and with UAS-Bru-3 for the purpose of TU-tagging. Hand>UPRT;lacZ is the control line used to identify pathogenic gene deregulations in Hand>UPRT;MblRNAi and Hand>UPRT;Bru-3 contexts. Flies were starved at day 4 for 6 hr before being transferred to 4TU containing food for 12 hr. (B) Volcano plot and IGV tracks from control Hand>UPRT;lacZ flies show examples of enrichment of heart-specific genes (e.g. Hand, Tin) (red, right side) and depletion of non-heart-expressed genes (blue, left side), thus validating the specificity of heart targeting.

TU-tagging-based transcriptional profiling of DM1 models with heart asynchrony reveals aberrant expression of genes regulating cellular calcium transient

To select candidates with a potential role in conduction disturbance, we focused on a pool of genes commonly deregulated in Hand>MblRNAi and in Hand>Bru-3 contexts (Figure 3A, Venn diagram), comprising 118 genes, with 64 conserved in humans (Figure 3A, heatmap and Figure 3—source data 1). Among them, gene interaction network analysis (Figure 3B) identified four candidates involved in the regulation of cellular calcium level and known to be essential for proper heart function (van Weerd and Christoffels, 2016), namely inactivation no afterpotential D (inaD), Syntrophin-like 1 (Syn1), Rad, Gem/Kir family member 2 (Rgk2) and straightjacket (stj). Their respective human orthologs are FRMPD2, SNTA1, REM1 and CACNA2D3/α2δ3. They all show high DIOPT orthology scores with their Drosophila counterparts (Figure 3—source data 1), except for FRMPD2 encoding a scaffolding PDZ domain carrying protein structurally similar to inaD but with a low DIOPT score. As inaD, Syn1 and Rgk2 were downregulated in both DM1 contexts (Figure 3A,B) we assessed the effects of their attenuation. We observed that cardiac-specific knockdowns of inaD, Syn1 and Rgk2 overall led to an increase in diastolic and systolic diameters (Figure 3—figure supplement 1). However, we did not observe asynchronous heartbeat phenotypes, indicating that decrease in inaD, Syn1 and Rgk2 expression has no role in DM1-associated conduction defects. We also simulated increased cardiac expression of stj observed in both Hand>MblRNAi and in Hand>Bru-3 contexts and found that in addition to arrhythmia and asystoles, Hand>stj flies exhibited asynchronous heartbeat, reminiscent of conduction defects observed in DM1 (Figure 1C). This observation and the fact that the expression of two other Drosophila α2δ genes (CG42817 and CG16868) remained unchanged (Figure 3—figure supplement 2) prompted us to test whether elevated expression of stj could contribute to DM1 cardiac defects.

Figure 3. Heart-specific transcriptional profiling of DM1 flies identifies deregulation of genes controlling cellular calcium level.

(A) Venn diagrams of genes deregulated in Hand>UPRT;MblRNAi and in Hand>UPRT;Bru-3 contexts (FC > 1.7) followed by heatmap of commonly deregulated genes. (B) Genemania interaction network of conserved candidates including stj, Rgk2, Syn1 and inaD known to be involved in Ca2 regulation. A scheme presenting the structure of the voltage-gated calcium channel and its regulatory component Stj/α2δ3 is included. Color code in genemania network represents up and down regulation according to the heatmap. Cyan/Blue intensity code indicates fold change of genes expression in the heatmap.

Figure 3—source data 1. A list of Drosophila genes and their Human orthologs deregulated in both Hand>Bru-3 and Hand>MblRNAi contexts.
Orthology scores were obtained from DIOPT tool (DRSC Integrative Ortholog Prediction Tool) (Hu et al., 2011) which predicts orthologs, based on sequence homology, phylogenetic trees, and functional similarity (www.flyrnai.org/cgi-bin/DRSC_orthologs.pl).

Figure 3.

Figure 3—figure supplement 1. Attenuation of inaD, Syn1 and Rgk2 all lead to an increase of diastolic (DD) and systolic (SD) heart diameters.

Figure 3—figure supplement 1.

For each candidate gene effects of attenuation were tested using two different UAS-RNAi lines crossed with Hand-Gal4 driver. More than 20 flies for each genotype were analysed by SOHA. Statistical significance of DD and SD changes was assessed using one-way ANOVA, Kruskal-Wallis Dunn’s multiple comparison post-test (* - p<0.05; ** - p<0.01; *** - p<0.001).
Figure 3—figure supplement 2. The expression levels of CG42617 and CG16868, two additional α2δ protein-coding genes are not affected in the heart of DM1 fly models.

Figure 3—figure supplement 2.

Normalized RNAseq IGV tracks in control (Hand>lacZ) and in pathogenic DM1 contexts (Hand>MblRNAi and Hand>Bru-3) are shown aligned with genomic exon/intron organization of CG42617 (A) and CG16868 (B).
Open in a new tab

To link calcium transient to heart asynchrony, we first tested the propagation of calcium waves in normal and in asynchronously beating hearts. We applied the GCaMP3 fluorescent calcium sensor that allows detection of calcium waves in vivo (Limpitikul et al., 2018). As shown in Figure 4A, calcium waves perfectly shadow cardiac contractions in wild-type hearts, each calcium peak aligning with the beginning of contraction along the heart tube (here registered in segments A3 and A4). By contrast, in asynchronously beating Hand>MblRNAi hearts, calcium peaks are not always in perfect alignment with contractions in the anterior A3 segment and are not detectable in the A4 segment that does not beat (Figure 4A). Calcium waves thus correlate with cardiac contractions in normal fly hearts, and are affected in asynchronously beating hearts, linking calcium current regulation and DM1-associated conduction defects.

Figure 4. Calcium waves in asynchronous Drosophila heart.

Figure 4.

Open in a new tab

In wild type hearts, calcium waves underlie cardiac contractions so that calcium peaks (arrows) align with the onset of contractions (arrowheads) in both A3 and A4 segments. In contrast, in asynchronously beating Hand>MblRNAi heart, in the anterior A3 segment, calcium peaks (arrows) are not always in perfect alignment with contractions (arrowheads) and could not be detected in A4 segment that does not beat. Notice that GFP signal in Hand>MblRNAi;GCaMP context is lower than in the control (Hand>GCaMP) most probably because of a lower Hand-Gal4 induction in two UAS transgene context.

Increased cardiac expression of Stj contributes to asynchronous heartbeats in DM1 flies

To determine the cardiac function of Stj, we first tested whether this protein actually accumulated in the adult fly heart. A weak cytoplasmic Stj signal associated with circular myofibers of cardiomyocytes and a strong signal in the myofibers of ventral longitudinal muscle (VLM) underlying the heart tube (Rotstein and Paululat, 2016) was visible in the adult Drosophila heart (Figure 5A and scheme Figure 5B) stained with anti-Stj antibody (Neely et al., 2010). This suggested that low stj levels in cardiac cells could have functional significance. To test this hypothesis, we overexpressed stj in the entire adult fly heart using Hand-Gal4 and in cardiomyocytes only using Tin-Gal4 (Figure 5C). When driven with Hand-Gal4, about 25% of individuals overexpressing stj displayed conduction defects, and the percentage of flies with asynchronous heartbeats increased to 30% in the Tin>stj context (Figure 5C). This suggests that in flies, a high Stj level in cardiomyocytes creates a risk of conduction disturbance.

Figure 5. Increased cardiac expression of Stj contributes to the conduction defects observed in DM1 fly models.

(A) Stj protein is detected at a low level in cardiomyocytes (CM) but at a higher level in the ventral longitudinal muscles (VLMs) that underlie adult heart tube. Stj in VLMs and in heart tube-attaching alary muscles (AMs) marks the network of T-tubules. PC denotes pericardial cells. Scale bars, 50 μm. (B) Scheme of the adult Drosophila heart with representation of Stj expression. (C) Both Hand-Gal4 (whole heart) and Tin-Gal4 (cardiomyocytes only) driven overexpression of Stj lead to asynchronous heartbeats similar to those observed in Hand>MblRNAi and Hand>Bru-3 contexts. (D) Barplot representing percentage of Hand>MblRNAi and Hand>Bru-3 flies displaying asynchronous heartbeats after reducing cardiac Stj expression via RNAi (in stj rescue conditions). Notice that lowering Stj expression efficiently reduces the risk of asynchronous heartbeats in Hand>Bru-3 context. Number of fly hearts tested (n) is indicated and statistical significance (Fisher’s exact test) denoted by ns (p>0.05) and * (p<0.05). (E) IGV tracks showing RNAseq peaks over the stj locus in healthy control and pathological contexts. Red boxes highlight stj-RC – specific exon 15 tracks enriched in Hand>MblRNAi and Hand>Bru-3 contexts and 3’UTR-specific tracks.

Figure 5—source data 1. Corrected total cell fluorescence (CTCF) of Stj signal in circular fibers of cardiomyocytes.

Figure 5.

Figure 5—figure supplement 1. Stj protein levels in different genetic contexts visualized by immunostaining.

Figure 5—figure supplement 1.

Stj protein expression (red) in circular fibers of cardiomyocytes in wild-type (wt) (A) and in different genetic contexts (B-E) tested. Circular fibers are visualized with phalloidin staining (green). Notice that Stj protein levels appear higher in Hand>Bru-3 and in Hand>MblRNAi compared to wt. High signal level in Hand>stj and loss of signal in Hand>stjRNAi demonstrate also the specificity of stj antibody. Scale bars, 50 μm. (F) Corrected total cell fluorescence (CTCF) of Stj signal in circular fibers of cardiomyocytes (Figure 5—source data 1) in different genetic contexts measured using Image J according to Burgess et al. (2010). Statistical significance (Kruskal-Wallis test) denoted by ns (p>0.05) * (p<0.05), ** (0.01) and *** (p<0.001).
Figure 5—figure supplement 2. stj-RC transcript isoform carrying long 3’UTR and alternatively spliced exon 15 is up regulated in Hand>MblRNAi and in Hand>Bru-3 hearts.

Figure 5—figure supplement 2.

(A) RT-qPCR probes for stj-RC specific exon 15 and for total stj transcripts are indicated on the scheme. (B) RT-qPCR analysis of stj-RC and total stj transcript levels in control (Hand>lacZ) fly hearts and in two DM1 contexts (Hand>MblRNAi and Hand>Bru-3). One-way ANOVA, Kruskall-wallis Dunn’s multiple comparison post-test was applied for assessing statistical significance (* - p<0.05; ** - p<0.001).
Open in a new tab

To test whether stj is a required mediator of Bru-3-overexpression and/or MblRNAi-induced conduction defects, we performed genetic rescue experiments by attenuating stj expression via RNAi in Hand>Bru-3 and Hand>MblRNAi flies (Figure 5D). These experiments revealed reductions of asynchronous heartbeats, which were statistically significant for both DM1 contexts but non-significant when referenced to DM1/UPRT lines, showing lower conduction defects (Figure 5D). Intriguingly, lowering stj transcript levels in the heart rescued asynchronous heartbeat more efficiently in Hand>Bru-3 flies than in Hand>MblRNAi flies (Figure 5D). This suggests that fine-tuning of Ca2 level by stj, but potentially also other calcium regulators, may be at work to ensure synchronous heartbeat and avoid conduction block.

We then set out to correlate Stj protein expression level in Hand>Bru-3, Hand>MblRNAi and Hand>stj fly hearts with the extent of cardiac asynchrony in these contexts. We found that Stj signal in the cardiomyocytes was higher in Hand>stj flies than in Hand>MblRNAi and Hand>Bru-3 flies (Figure 5—figure supplement 1). However, the percentage of flies displaying conduction defects was higher in the Hand>MblRNAi context than in the Hand>stj context (Figure 5C,D) indicating that Stj was not the sole factor whose deregulation causes asynchronous heartbeat in the Hand>MblRNAi context. This observation is consistent with partial rescue of conduction defects in the Hand>MblRNAi;stjRNAi context (Figure 5D).

We also undertook to determine, which among three stj transcript isoforms (Figure 5E) is elevated in DM1 contexts. The analysis of RNAseq peaks over the stj locus (Figure 5E) indicated that transcripts harboring exon 15 (stj-RC) and long 3’UTR sequences (stj-RB and stj-RC) are enriched in both Hand>Bru-3 and Hand>MblRNAi fly hearts. The upregulation of stj-RC was validated by RT-qPCR on dissected wild type and DM1 hearts (Figure 5—figure supplement 2) showing that its expression level is equivalent to the expression of all stj transcripts thus providing evidence that stj-RC is the main stj isoform whose expression increases in pathological contexts.

Ventricular cardiomyocytes of DM1 patients with conduction disturbance show an increased expression of α2δ3

To determine whether our data from the fly model were relevant to DM1 patients, we first tested α2δ3 protein and transcript levels in ventricles and atria of normal mouse hearts. Like Stj, α2δ3 protein was also detected in the mouse cardiac cells, with high levels in the atrial and low levels in the ventricular cardiomyocytes (Figure 6A). This differential α2δ3 expression in atria and ventricles was also found at the transcript level (Figure 6B).

Figure 6. Increased cardiac expression of Straightjacket human ortholog α2δ3 is associated with conduction defects in DM1 patients.

(A) In mouse, α2δ3 protein is detected at a high level in atrial cardiac cells and at a low level in ventricles. LV denotes left ventricle. Scale bars, 100 mm. (B) RT-qPCR analysis of mouse α2δ3 transcripts confirms low expression level in ventricles and high expression level in atria. Statistical significance was determined by Mann-Whitney U test (*** p<0.001) (C, D) α2δ3 (C) transcript and (D) protein levels analysed by RT-qPCR (C) and western blot (D) in human ventricular muscle from normal controls and DM1 patients with conduction defects. Note the statistically relevant increase in both α2δ3 transcript and α2δ3 protein levels in ventricles of DM1 patients (Mann-Whitney U test). (E) A scheme illustrating normal and DM1 conditions with low and increased α2δ3 levels in ventricular muscle, respectively. In pathological context, the aberrant inward calcium current in ventricular cardiomyocytes could lead to conduction defects and in particular to the IVCD.

Figure 6.

Figure 6—figure supplement 1. RT-qPCR analysis showing that Cav1.2 transcripts are elevated in human cardiac samples from DM1 patients with conduction defects.

Figure 6—figure supplement 1.

Open in a new tab

We then analyzed α2δ3 transcript and protein expression in a restricted number of human ventricular muscle samples from healthy donors and from DM1 patients with conduction defects. Ventricular α2δ3 transcript and protein levels were both significantly higher in DM1 patients with conduction disturbances than in controls (Figure 6C,D). Analyzing the same human samples, we also found a higher expression of the main calcium channel α1/Cav1.2 unit (Figure 6—figure supplement 1), further evidence (Figure 6E) that Ca2 inward current deregulation underlies DM1-associated conduction defects.

Discussion

Cardiac dysfunctions decrease life expectancy in DM1, and conduction disturbances affect up to 75% of DM1 patients (McNally and Sparano, 2011). They are diagnosed by a lengthened ECG PR segment indicating atrio-ventricular (AV) block or QRS complex widening (Groh et al., 2008) that might or might not be associated with bundle-branch blocks of the His-Purkinje conduction system. Lengthened QRS without complete or incomplete bundle-branch block results from IVCD. It involves affected conduction via Purkinje fibers but also via ventricular cardiomyocytes that could contribute to the QRS widening observed in DM1 patients.

To identify gene deregulations underlying conduction defects in DM1 patients, we used the Drosophila model, which harbors a simple heart tube without a specialized conduction system, implying that synchronous propagation of contraction waves involves cardiomyocytes only. We reasoned that this model could provide insights into IVCD associated with ventricular cardiomyocyte dysfunction.

We applied a heart cell-specific transcriptional profiling approach based on TU-tagging (Miller et al., 2009) by which we identified 64 conserved genes commonly deregulated in both DM1 contexts with asynchronous heartbeat. Applying a sensitive cell-specific approach followed by RNAseq was critical for revealing discrete gene deregulations, considering that cardiac asynchrony occurred only in a subset of DM1 flies. Also, Hand>MblRNAi and Hand>Bru-3 flies developed two other DM1-associated heart phenotypes: dilated cardiomyopathy and arrhythmias. To facilitate candidate gene selection, we generated gene interaction networks for deregulated candidates and found that four of them were involved in the regulation of cellular calcium level.

Calcium regulators and heart function

We previously reported (Picchio et al., 2013) that abnormal splicing of SERCA encoding the major SR calcium pump was involved in the myotonia phenotype in DM1 flies. Analyzing two heart-specific DM1 fly models, we found that four other conserved calcium regulatory genes, inaD/FRMPD2, Syn1/SNTA1, Rgk2/REM1 and stj/α2δ3, are commonly deregulated. Genes inaD and Syn, which were downregulated, encode scaffolding PDZ domain proteins with regulatory functions on TRP Ca2 channels (Shieh and Zhu, 1996; Ueda et al., 2008); their mutations cause cardiac hypertrophy, arrhythmia, and heart block (Spassova et al., 2004). Moreover, inaD was reported to prevent depletion of ER calcium store by inhibiting Ca2 release-activated Ca2 (CRAC) channels (Su et al., 2003). Rgk2, also downregulated, belongs to the Ras superfamily encoding small GTP-binding proteins (Puhl et al., 2014). Rgk proteins including Rgk2 interact with the α− or β-subunit of CaV1.2 calcium channel and negatively regulate its function (Puhl et al., 2014; Magyar et al., 2012). However, by interacting with CaMKII and with Rho kinase, both involved in cardiac hypertrophy and heart block, Rgk proteins could also influence heartbeat Cav1.2-independently. Finally, stj, whose transcript levels were significantly increased in both DM1 contexts, encodes an auxiliary subunit of this same CaV1.2 channel, and is known to positively regulate this channel’s abundance (Hoppa et al., 2012). CaV1.2 mutations are known to lead to diverse heartbeat dysfunctions (Splawski et al., 2004), and CaV1.2 deregulation is associated with DM1 (Rau et al., 2011). Importantly, the Drosophila counterpart of Cav1.2 (Ca-1αD) also ensures cardiac contractions and calcium transients in the fly heart (Limpitikul et al., 2018), suggesting a conserved role of Cav1.2/Ca-1αD in heartbeat. Here, by following a fluorescent calcium sensor (GCaMP3), we found that calcium waves shadowed synchronous cardiac contractions but were disrupted in asynchronously beating DM1 fly hearts, implying that aberrant calcium transient in cardiomyocytes is associated with heart asynchrony. Thus even if no specialized conduction system is present in the Drosophila heart, propagation of calcium waves along the cardiac tube is finely regulated to ensure synchronous heartbeat. Among selected calcium regulators, only cardiac overexpression of stj resulted in asynchronous heartbeat, so we focused our study on this candidate gene.

Stj/α2δ3, new candidate gene for DM1-associated conduction defects

A previous large study dedicated to identifying genes involved in heat nociception (Neely et al., 2010) identified stj/α2δ3 as an evolutionarily conserved pain gene. Here, we found stj transcripts upregulated in cardiomyocytes of two DM1 fly models Hand>Bru-3 and Hand>MblRNAi exhibiting asynchronous heartbeat. We also show that reducing stj transcript levels significantly ameliorated heart asynchrony in Hand>Bru-3 hearts while improving the conduction defect phenotype of Hand>MblRNAi flies. Thus additional gene deregulations contribute to asynchronous heartbeat in Hand>MblRNAi context, attesting the complexity of heart dysfunction in DM1. At the protein level, Stj is also more abundant in cardiomyocytes of two Drosophila DM1 contexts with cardiac asynchrony, while at a very low level in the wild-type cardiomyocytes. This suggests Stj protein involvement in the DM1-associated asynchronous heartbeat in flies. The precise mechanisms causing stj transcript elevation in DM1 contexts remain to be defined, but our data suggest that a combination of splice-dependent and 3’UTR-dependent mechanisms could be in play.

Intriguingly, vertebrate stj counterpart α2δ3 also displays low transcript and protein levels in ventricular cardiomyocytes and a distinctively higher expression in atrial cardiac cells. We hypothesize that this differential expression could be related to the mechanisms regulating cardiomyocyte contraction in atria and in ventricles. The high α2δ3 expression in atria is consistent with cardiomyocyte-dependent conduction in atria that do not harbor a specialized conduction system. By contrast, its low expression in ventricles could be correlated with the presence of His-Purkinje fibers facilitating conduction in the large ventricular muscles. We were, however, unable to test α2δ3 involvement in cardiac conduction defects in a DMSXL mouse DM1 model carrying large numbers of repeats (Huguet et al., 2012) as it displays only mild cardiac involvement without conduction disturbances. We thus tested α2δ3 expression directly in human cardiac samples. Ventricular cardiomyocytes from healthy donors showed, like the control mice, low α2δ3 transcript expression, which was significantly elevated in ventricular cardiac cells from DM1 patients with conduction defects. DM1-associated increase of expression also occurs for α2δ3 protein and is accompanied by high transcript levels of the main cardiac channel unit Cav1.2/α1, suggesting that the α2δ3 elevation leads to a higher channel density and increased calcium entry to the cardiomyocytes. This could contribute to IVCD affecting the ventricular cardiomyocyte conduction rate and thereby the synchrony of cardiac contraction.

Taken together, with evidence based on heart-specific transcriptional profiling of DM1 Drosophila models and on gene deregulation in cardiac samples from DM1 patients, we propose that the conduction disturbance observed in DM1, and in particular IVCD, arises from likely impacted Ca-α1D/Cav1.2 calcium channel function in ventricular cardiomyocytes in which regulatory subunit stj/α2δ3 plays a central role.

Materials and methods

Key resources table.

1  Reagent type (species) or resource Designation Source or reference Identifiers Additional information
2  Gene
(D. melanogaster)		 stj FLYB: FBgn0261041
3  Gene
(D. melanogaster)		 mbl FLYB: FBgn0265487
4  Gene
(D. melanogaster)		 bru3 FLYB: FBgn0264001
5  Genetic reagent
(D. melanogaster)		 UAS-stj FlyORF Zurich ORFeome Project FlyORF ID: F001252
6  Genetic reagent
(D. melanogaster)		 stj TRIP Bloomington Drosophila Stock Center RRID:BDSC25807 RNAi line
7  Genetic reagent
(D. melanogaster)		 Mbl RNAi VDRC Vienna Drosophila Resource Center VDRC: GD 28732 RNAi line
8  Genetic reagent
(D. melanogaster)		 P{XP}bru3d09837 Exelixis at Harvard Medical School Harvard: d09837 UAS carrying P element insertion line
9  Genetic reagent
(D. melanogaster)		 UAS-GFP Bloomington Drosophila Stock Center RRID:BDSC_32201
10  Genetic reagent
(D. melanogaster)		 UAS-HA-UPRT Bloomington Drosophila Stock Center RRID:BDSC_27604
11  Genetic reagent
(D. melanogaster)		 UAS-lacZ Bloomington Drosophila Stock Center RRID:BDSC_1776
12  Genetic reagent
(D. melanogaster)		 UAS-GCaMP3 Bloomington Drosophila Stock Center RRID:BDSC_32116
13  Genetic reagent
(D. melanogaster)		 Ca-alpha1D TRIP Bloomington Drosophila Stock Center RRID:BDSC_25830 RNAi line
14  Genetic reagent
(D. melanogaster)		 Rgk2 TRIP Bloomington Drosophila Stock Center RRID:BDSC_65932 RNAi line
15  Genetic reagent
(D. melanogaster)		 Rgk2 RNAi VDRC Vienna Drosophila Resource Center VDRC: GD44611 RNAi line
16  Genetic reagent
(D. melanogaster)		 Syn1 RNAi VDRC Vienna Drosophila Resource Center VDRC: GD27893 RNAi line
17  Genetic reagent
(D. melanogaster)		 Syn1 RNAi Bloomington Drosophila Stock Center RRID: BDSC_27504 RNAi line
18  Genetic reagent
(D. melanogaster)		 inaD RNAi VDRC Vienna Drosophila Resource Center VDRC: GD26211 (unavailable) RNAi line
19  Genetic reagent
(D. melanogaster)		 inaD RNAi VDRC Vienna Drosophila Resource Center VDRC: 330227 RNAi line
20  Genetic reagent
(D. melanogaster)		 Hand-GAL4 Laurent Perrin, TAGC,
Marseille, France		 GAL4 driver line
21  Genetic reagent
(D. melanogaster)		 TinCΔ4-GAL4 Rolf Bodmer, SBPMD Institute, San Diego, US GAL4 driver line
22  Antibody anti-Stj Hugo Bellen, Baylor College of Medicine, Huston, USA guinea pig polyclonal Use 1:500 (IHC)
23  Antibody anti-α2δ3 Greg Neely, University of Sidney, Australia rabbit polyclonal Use 1:200 (IHC)
24  Antibody anti-α2δ3 GeneTex GTX16618 - rabbit polyclonal Use 1:500 (WB)
25  Antibody anti-Bru3 Millegen, Toulouse, France rabbit polyclonal Use 1:1000 (IHC)
26  Antibody anti-Mbl Darren Monckton, Glasgow, UK sheep polyclonal Use 1:200 (IHC)
27  Antibody anti-GFP Abcam ab5450 – goat polyclonal Use 1:500 (IHC)
28  Antibody Anti-GAPDH Sigma Aldrich SAB2108266 – rabbit polyclonal Use 1:3000 (WB)
Open in a new tab

Drosophila stocks

All fly stocks were maintained at 25°C on standard fly food in a 12 hr:12 hr light-dark cycling humidified incubator. To generate the DM1 Drosophila models, we used the inducible lines UAS-MblRNAi (v28732, VDRC, Vienna, Austria) and UAS-Bru3 37 (bru-3d09837, BDSC, Bloomington, USA) and crossed them with the driver line, Hand-Gal4 (Han and Olson, 2005; Sellin et al., 2006) (kindly provided by L. Perrin, TAGC, Marseille), to target transgene expression to the adult fly heart (cardiomyocytes and pericardial cells). To examine the expression pattern of the Hand-Gal4 driver line, we crossed them to UAS-GFP line (#32201, BDSC). Control lines were generated by crossing the above-cited lines with w1118 flies. For TU-tagging experiments, the UPRT line (UAS-HA-UPRT 3.2, #27604, BDSC) was combined with the previously cited UAS lines and with UAS-LacZ (#1776, BDSC) as a control, generating UAS-LacZ;UAS-HA-UPRT, UAS-MblRNAi;UAS-HA-UPRT and UAS-Bru3;UAS-HA-UPRT stocks. For functional analyses of straightjacket, the lines used were UAS-Stj (FlyORF, F 001252) and UAS-stj RNAi (#25807, BDSC) and tinCΔ4-Gal4 cardiomyocyte-specific driver (Lo and Frasch, 2001; Perrin et al., 2004) (kindly provided by R. Bodmer, SBPMD Institute, San Diego). For functional analyses of the candidate genes, we used the following UAS-RNAi lines: #26211 (VDRC) and #330227 (VDRC) for inaD; #27893 (VDRC) and #27504 (BDSC) for Syn1 and #44611 (VDRC) and #65932 (BDSC) for Rgk2. The UAS-GCaMP3 line (#32116, BDSC) was crossed to Hand-Gal4 driver and applied as a sensor of calcium waves in the cardiac tube.

Optical heartbeat analyses of adult Drosophila hearts

To assess cardiac physiology in adult flies, we used the method previously described (Fink et al., 2009; Ocorr et al., 2007). Briefly, 1 week old flies were anesthetized using FlyNap (# 173025, Carolina), and then dissected in artificial hemolymph solution, removing head, legs, wings, gut, ovaries and fat body. The hearts were allowed to equilibrate with oxygenation for 15–20 min prior to filming 30 s movies with a Hamamatsu camera (>100 frames/s) under a Zeiss microscope (10X magnification). Movie analysis was performed with SOHA (semi-automatic optical heartbeat analysis) based on using Matlab R2009b (Mathworks, Natick, MA, USA) to collect contractility (diastolic and systolic diameters, fractional shortening) and rhythmicity (heart period, diastolic and systolic intervals and arrhythmicity index) parameters.

Immunofluorescence on Drosophila heart

Briefly, the fly hearts were dissected as described above and fixed for 15 min in 4% formaldehyde. The immunostaining procedure was performed as described elsewhere (Picchio et al., 2013). The following primary antibodies were used: sheep anti-Mbl antibody (1:200, kindly provided by Darren Monckton), rabbit anti-Bru-3 (1:1000; Millegen, Toulouse, France), rabbit anti-Stj (1:500; kindly provided by Hugo Bellen), rabbit anti-α2δ3 (1:200, gift of Greg Neely) and goat anti-GFP (1:500, Abcam, ab 5450). Rhodamine phalloidin (ThermoFischer Scientific) was used to reveal actin filaments. Fluorescent secondary antibodies were from Jackson ImmunoResearch. For Mbl, we used a biotinylated anti-sheep antibody (Biotin-SP-AffiniPure Donkey Anti-Sheep IgG (H L), Jackson ImmunoResearch) combined with a DTAF-conjugated streptavidin (Jackson ImmunoResearch).

RNA extraction and RT-qPCR on adult fly heart samples

Total RNA was isolated from about 10–15 hearts from 1 week old flies, using TRIzol reagent (Invitrogen) combined with the Quick-RNA MicroPrep Kit (Zymo Research) following the manufacturer’s instructions. RNA quality and quantity were respectively assessed using Agilent RNA 6000 Pico kit on an Agilent 2100 Bioanalyzer (Agilent Technologies) and Qubit RNA HS assay kit on a Qubit 3.0 Fluorometer (Thermo Fischer Scientific). RT-qPCR (three biological replicates) was performed as previously described (Picchio et al., 2013), using Rp49 as a reference gene. The following pairs of primers were used: Rp49: forward GCTTCAAGGGACAGTATCTG and reverse AAACGCGGTTCTGCATGAG; Mbl: forward CGTGGAGGTCCAGAACGG and reverse AATATCAGGTCAATATAAAGCG; Bru-3: forward CCGAAGGTTGAGTTTGCTC and reverse CTTCAGCTGTAAAGCACGG; total stj transcripts: forward AGGCTCGCGAGTTCAACC and reverse AAGGCCTGCTCCGTGTAC and exon 15 carrying transcripts: forward GACGCCCTTTACTCTGGTCA and reverse ATAAAGGGCACAAACTTGCG.

Immunofluorescence on murine heart

Mouse hearts were dissected and fixed for 2 hr in paraformaldehyde (4% in 1X PBS) at 4°C and processed for cryosectioning as described elsewhere (Beyer et al., 2011). Frozen sections were incubated overnight at 4°C with anti-α2δ3 (1:200) (Neely et al., 2010). Fluorescent images were obtained using a Zeiss AxioimagerZ1 with an Apotome module and an AxioCamMRm camera.

RNA extraction and RT-qPCR on murine heart samples

RNA extraction was performed as described elsewhere (Huguet et al., 2012). Transcripts were amplified in a 7300 real-time PCR system (Applied Biosystems) using Power SybrGreen detection (Life Technologies). Standard curves were established with mouse tissues expressing the tested genes. Samples were quantified in triplicate, and experiments were repeated twice. α2δ3 transcripts were amplified using forward primer GGGAACCAGATGAGAATGGAGTC and reverse primer TTTGGAGAAGTCGCTGCCTG. Polr2a was used as internal control (forward primer GGCTGTGCGGAAGGCTCTG and reverse primer TGTCCTGGCGGTTGACCC).

RNA extraction, RT-qPCR, and immunoblot on DM1 patient cardiomyocytes

Human ventricular cardiac muscle tissues were obtained at autopsy from five DM1 patients and four normal controls after informed consent was obtained. All experimental protocols were approved by the Institutional Review Board at Osaka University, and carried out in accordance with the approved guidelines. Total mRNA was extracted and first-strand complementary DNA synthesized using protocols described previously (Nakamori et al., 2008). RT-qPCR was performed using TaqMan Gene Expression assays (Hs01045030_m1 and 4333760F, Applied Biosystems) on an ABI PRISM 7900HT Sequence Detection System (Applied Biosystems), as described previously (Nakamori et al., 2011). Level of CACNA2D3/α2δ3 mRNA and Cav1.2/α1 mRNA were normalized to 18S rRNA. For protein analysis, cardiac muscle tissues were homogenized in a 10x volume of radioimmunoprecipitation assay buffer (25 mM Tris-HCl; pH 7.5; 150 mM NaCl; 1% NP-40; 1% sodium deoxycholate; and 0.1% sodium dodecyl sulfate) containing a protein inhibitor cocktail (Sigma-Aldrich). The homogenate was centrifuged for 10 min at 10,000 × g and the supernatant was collected. Equal amounts of protein (30 µg) were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred onto Immobilon-P membranes (Millipore), as previously described (Nakamori et al., 2008). Blots were blocked for nonspecific protein binding with 5% (w/v) nonfat milk and then incubated with a 1:500-diluted antibody against CACNA2D3/α2δ3 (Gene Tex) or 1:3000-diluted antibody against GAPDH (glyceraldehyde 3-phosphate dehydrogenase) (Sigma-Aldrich). After repeated washings, the membranes were incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG (Life Technologies). The ECL Select Western Blotting Detection Reagent (GE Healthcare) and ChemiDoc Touch Imaging System (Bio-Rad) were used to detect the proteins. Only the upper α2δ3 band (Figure 6D) corresponding to predicted protein size was quantified.

TU-tagging experiments

The TU-tagging protocol was adapted from previously published studies (Gay et al., 2014; Miller et al., 2009).

4TU treatment, fly collection, and total RNA extraction

TU-tagging experiments were performed on the following 1-week-old adult flies: Hand>LacZ;HA-UPRT, Hand>MblRNAi;HA-UPRT and Hand>Bru3;HA-UPRT. Briefly, flies were starved for 6 hr at 25°C and transferred to fresh food media containing 4TU (Sigma) at 1 mM. After 12 hr incubation at 29°C, about 30–40 flies of the described genotypes were dissected (removing head, wings, legs, ovaries and gut) in DPBS 1X (Gibco, Life Technologies), immediately transferred to Eppendorf tubes, and snap-frozen in liquid nitrogen. Total RNA isolation was performed as described above in TRIzol following the manufacturer’s instructions (ThermoFischer Scientific).

Purification of TU-tagged RNA

Bio-thio coupling was performed on about 30 µg of total RNA with 1 mg/mL biotin-HPDP (Pierce) for 3 hr followed by two chloroform purification steps to eliminate the unbound biotin, as previously described (Miller et al., 2009). To verify the efficiency of the biotinylation step, an RNA dot blot was performed using streptavidin-HRP antibody. The streptavidin purification step then served to collect the thiolated-RNA fraction (cardiac-specific 4TU) and the unbound fraction (non-cardiac RNAs) using a µMACS streptavidin kit (MACS, Miltenyi Biotec) following the manufacturer’s instructions. The purified fraction was precipitated as previously described (Miller et al., 2009). RNA quality and quantity were assessed using Bioanalyzer and Qubit systems according to the manufacturer’s instructions. RT-qPCR was performed to check for enrichment of cardiac-specific transcripts (Hand and H15) by comparing 4TU fraction relative to input fraction (i.e. the biotinylated fraction containing cardiac and non-cardiac RNAs), with Rp49 gene used as reference.

RNA sequencing

Library preparation

Library preparation was performed on about 50 ng of cardiac-specific RNA from the previously described conditions, using the Ovation RNA-Seq Systems 1–16 for Model Organisms adapted to Drosophila (NuGEN, #0350–32) as per the manufacturer’s instructions with a few modifications. First-strand synthesis was performed with the integrated DNAse treatment (HL-dsDNAse, ArcticZymes) and cDNA was fragmented using a Bioruptor sonication system (Diagenode) via 30 cycles of 30 s ON/OFF on low power. 15 PCR library amplification cycles were performed on 50 ng of starting RNA, and a size-selective bead purification step was done using RNAClean XP beads (Agencourt AMPure, Beckman Coulter). Quantitative and qualitative assessment of the library was performed using an Agilent dsDNA HS kit on an Agilent 2100 Bioanalyzer (Agilent Technologies).

Sequencing parameters

The indexed Illumina libraries prepared from the cardiac-specific Hand>LacZ;HA-UPRT, Hand>MblRNAi;HA-UPRT and Hand>Bru3;HA-UPRT RNA fractions were mixed at equal concentrations and sequenced (100 bp paired-end reads) on the same lane of a HiSeq 2000 system (EMBL Gene Core Illumina Sequencing facility, Heidelberg, Germany). All four genotype sets of 1-week-old flies were sequenced in duplicate to give a total of eight samples sequenced.

Data analysis

The RNA-sequencing data were checked for good quality using the FastQC package (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Each sample was aligned to the reference Dmel genome release 5.49 using Bowtie2 (Langmead and Salzberg, 2012). The aligned reads were sorted and indexed using SAMtools (Li et al., 2009). Pearson correlation was determined for each duplicate of each condition. Differential gene expression was obtained using a pipeline based on the Deseq2 package. Rlog transformation was applied on raw count data obtained for the duplicates of each condition before computing differential expression. Genes were tested individually for differences in expression between conditions. We set a fold-change threshold at 1.7 and p-value threshold at 0.05 for meaningful differential up-expression, and fold-change threshold at 0.59 and p-value threshold at 0.05 for meaningful differential down-expression.

In vivo calcium transient analyses

We performed short (10 s) time lapse recording of beating hearts, dissected as for SOHA experiments, and mounted inverted on 35 mm glass-bottom dishes (IBIDI ref 81218) for imaging on LEICA SP8 confocal microscope. Twenty control hearts (Hand>GCaMP) and 35 DM1 hearts dissected from Hand>MblRNAi;GCaMP 1-week-old flies were analyzed. Two channel films recorded from a single optical level (GCaMP-GFP channel representing calcium waves and transmitted light channel representing contractions) were analyzed using Image J. For each film, we first transformed Frames on Image stacks and used ‘Plot Z axis profile’ function to generate graphs.

Statistical analyses and access to RNAseq data

All statistical analyses were performed using GraphPad Prism v5.02 software (GraphPad Inc, USA). Fisher’s exact test was applied to analyze statistical significance of conduction defect phenotypes, and Mann-Whitney U test to analyze RT-qPCR data from mouse and human heart samples and Western blot α2δ3 expression data in human samples. We also applied one-way ANOVA, Kruskal-Wallis Dunn’s multiple comparison post-test to statistically analyze RT–qPCR data for stj isoforms and changes in diastolic and systolic heart diameters. All RNAseq data reported here were deposited with the GEO-NCBI tracking system under accession number #18843071.

Acknowledgements

This work was supported by AFM-Téléthon (MyoNeurAlp Strategic Program), Agence Nationale de la Recherche (Tefor Infrastructure Grant), Fondation pour la Recherche Médicale (Equipe FRM Award) and the Intramural Research Grant for Neurological and Psychiatric Disorders (NCNP grant to MN).

Funding Statement

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Contributor Information

Monika Zmojdzian, Email: monika.zmojdzian@uca.fr.

Krzysztof Jagla, Email: christophe.jagla@udamail.fr.

Margaret Buckingham, CNRS UMR 3738, Institut Pasteur, France.

Utpal Banerjee, University of California, Los Angeles, United States.

Funding Information

This paper was supported by the following grants:

  • AFM-Téléthon MyoNeurAlp strategic program to Krzysztof Jagla.

  • Agence Nationale de la Recherche Tefor Infrastructure to Krzysztof Jagla.

  • Fondation pour la Recherche Médicale Equipe FRM to Krzysztof Jagla.

  • National Center of Neurology and Psychiatry Intramural Research Grant for Neurological and Psychiatric Disorders to Masayuki Nakamori.

Additional information

Competing interests

No competing interests declared.

Author contributions

Formal analysis, Investigation, Methodology, Writing—original draft.

Funding acquisition, Investigation, Methodology, Writing—original draft.

Formal analysis, Investigation, Methodology, Writing—original draft.

Formal analysis, Investigation, Methodology.

Formal analysis, Investigation, Methodology.

Methodology.

Formal analysis, Methodology, Writing—review and editing.

Supervision, Funding acquisition.

Supervision, Investigation, Methodology, Writing—review and editing.

Supervision, Methodology, Writing—review and editing.

Formal analysis, Investigation, Methodology.

Conceptualization, Formal analysis, Investigation, Visualization, Methodology, Writing—review and editing.

Conceptualization, Formal analysis, Supervision, Funding acquisition, Investigation, Methodology, Writing—review and editing.

Ethics

Human subjects: Human ventricular cardiac muscle tissues were obtained at autopsy from five DM1 patients and four normal controls after informed consent was obtained. All experimental protocols were approved by the Institutional Review Board at Osaka University (Protocol Number 14183-6) and carried out in accordance with the approved guidelines.

Additional files

Transparent reporting form

Data availability

All data generated or analysed during this study are included in the manuscript and supporting files. Sequencing data have been deposited with the GEO-NCBI tracking system under accession number GSE109370.

The following dataset was generated:

Plantie E, Picchio L, Renaud Y. 2018. Deregulation associated with cardiac conduction defects in Myotonic Dystrophy Type 1 using TU-Tagging. NCBI Gene Expression Omnibus. GSE109370

References

  1. Beyer S, Kelly RG, Miquerol L. Inducible Cx40-Cre expression in the cardiac conduction system and arterial endothelial cells. Genesis. 2011;49:83–91. doi: 10.1002/dvg.20687. [DOI] [PubMed] [Google Scholar]
  2. Birse RT, Choi J, Reardon K, Rodriguez J, Graham S, Diop S, Ocorr K, Bodmer R, Oldham S. High-fat-diet-induced obesity and heart dysfunction are regulated by the TOR pathway in Drosophila. Cell Metabolism. 2010;12:533–544. doi: 10.1016/j.cmet.2010.09.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bodi I. The L-type calcium channel in the heart: the beat Goes on. Journal of Clinical Investigation. 2005;115:3306–3317. doi: 10.1172/JCI27167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Burgess A, Vigneron S, Brioudes E, Labbé J-C, Lorca T, Castro A. Loss of human greatwall results in G2 arrest and multiple mitotic defects due to deregulation of the cyclin B-Cdc2/PP2A balance. PNAS. 2010;107:12564–12569. doi: 10.1073/pnas.0914191107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Davis BM, McCurrach ME, Taneja KL, Singer RH, Housman DE. Expansion of a CUG trinucleotide repeat in the 3' untranslated region of myotonic dystrophy protein kinase transcripts results in nuclear retention of transcripts. PNAS. 1997;94:7388–7393. doi: 10.1073/pnas.94.14.7388. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. de Die-Smulders CE, Höweler CJ, Thijs C, Mirandolle JF, Anten HB, Smeets HJ, Chandler KE, Geraedts JP. Age and causes of death in adult-onset myotonic dystrophy. Brain. 1998;121:1557–1563. doi: 10.1093/brain/121.8.1557. [DOI] [PubMed] [Google Scholar]
  7. Dixon DM, Choi J, El-Ghazali A, Park SY, Roos KP, Jordan MC, Fishbein MC, Comai L, Reddy S. Loss of muscleblind-like 1 results in cardiac pathology and persistence of embryonic splice isoforms. Scientific Reports. 2015;5:9042. doi: 10.1038/srep09042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dolphin AC. The α2δ subunits of voltage-gated calcium channels. Biochimica Et Biophysica Acta (BBA) - Biomembranes. 2013;1828:1541–1549. doi: 10.1016/j.bbamem.2012.11.019. [DOI] [PubMed] [Google Scholar]
  9. Fink M, Callol-Massot C, Chu A, Ruiz-Lozano P, Izpisua Belmonte JC, Giles W, Bodmer R, Ocorr K. A new method for detection and quantification of heartbeat parameters in Drosophila, zebrafish, and embryonic mouse hearts. BioTechniques. 2009;46:101–113. doi: 10.2144/000113078. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Freyermuth F, Rau F, Kokunai Y, Linke T, Sellier C, Nakamori M, Kino Y, Arandel L, Jollet A, Thibault C, Philipps M, Vicaire S, Jost B, Udd B, Day JW, Duboc D, Wahbi K, Matsumura T, Fujimura H, Mochizuki H, Deryckere F, Kimura T, Nukina N, Ishiura S, Lacroix V, Campan-Fournier A, Navratil V, Chautard E, Auboeuf D, Horie M, Imoto K, Lee KY, Swanson MS, de Munain AL, Inada S, Itoh H, Nakazawa K, Ashihara T, Wang E, Zimmer T, Furling D, Takahashi MP, Charlet-Berguerand N. Splicing misregulation of SCN5A contributes to cardiac-conduction delay and heart arrhythmia in myotonic dystrophy. Nature Communications. 2016;7:11067. doi: 10.1038/ncomms11067. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Galindo K, Smith DP. A large family of divergent Drosophila odorant-binding proteins expressed in gustatory and olfactory sensilla. Genetics. 2001;159:1059–1072. doi: 10.1093/genetics/159.3.1059. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Gay L, Karfilis KV, Miller MR, Doe CQ, Stankunas K. Applying thiouracil tagging to mouse transcriptome analysis. Nature Protocols. 2014;9:410–420. doi: 10.1038/nprot.2014.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Groh WJ, Groh MR, Saha C, Kincaid JC, Simmons Z, Ciafaloni E, Pourmand R, Otten RF, Bhakta D, Nair GV, Marashdeh MM, Zipes DP, Pascuzzi RM. Electrocardiographic abnormalities and sudden death in myotonic dystrophy type 1. New England Journal of Medicine. 2008;358:2688–2697. doi: 10.1056/NEJMoa062800. [DOI] [PubMed] [Google Scholar]
  14. Han Z, Olson EN. Hand is a direct target of tinman and GATA factors during Drosophila cardiogenesis and hematopoiesis. Development. 2005;132:3525–3536. doi: 10.1242/dev.01899. [DOI] [PubMed] [Google Scholar]
  15. Hoppa MB, Lana B, Margas W, Dolphin AC, Ryan TA. α2δ expression sets presynaptic calcium channel abundance and release probability. Nature. 2012;486:122–125. doi: 10.1038/nature11033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hu Y, Flockhart I, Vinayagam A, Bergwitz C, Berger B, Perrimon N, Mohr SE. An integrative approach to ortholog prediction for disease-focused and other functional studies. BMC Bioinformatics. 2011;12:357. doi: 10.1186/1471-2105-12-357. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Huguet A, Medja F, Nicole A, Vignaud A, Guiraud-Dogan C, Ferry A, Decostre V, Hogrel JY, Metzger F, Hoeflich A, Baraibar M, Gomes-Pereira M, Puymirat J, Bassez G, Furling D, Munnich A, Gourdon G. Molecular, physiological, and motor performance defects in DMSXL mice carrying >1,000 CTG repeats from the human DM1 locus. PLOS Genetics. 2012;8:e1003043. doi: 10.1371/journal.pgen.1003043. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Kalsotra A, Singh RK, Gurha P, Ward AJ, Creighton CJ, Cooper TA. The Mef2 transcription network is disrupted in myotonic dystrophy heart tissue, dramatically altering miRNA and mRNA expression. Cell Reports. 2014;6:336–345. doi: 10.1016/j.celrep.2013.12.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kino Y, Washizu C, Oma Y, Onishi H, Nezu Y, Sasagawa N, Nukina N, Ishiura S. MBNL and CELF proteins regulate alternative splicing of the skeletal muscle chloride channel CLCN1. Nucleic Acids Research. 2009;37:6477–6490. doi: 10.1093/nar/gkp681. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Koshelev M, Sarma S, Price RE, Wehrens XH, Cooper TA. Heart-specific overexpression of CUGBP1 reproduces functional and molecular abnormalities of myotonic dystrophy type 1. Human Molecular Genetics. 2010;19:1066–1075. doi: 10.1093/hmg/ddp570. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kuyumcu-Martinez NM, Wang GS, Cooper TA. Increased steady-state levels of CUGBP1 in myotonic dystrophy 1 are due to PKC-mediated hyperphosphorylation. Molecular Cell. 2007;28:68–78. doi: 10.1016/j.molcel.2007.07.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Langmead B, Salzberg SL. Fast gapped-read alignment with bowtie 2. Nature Methods. 2012;9:357–359. doi: 10.1038/nmeth.1923. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abecasis G, Durbin R, 1000 Genome Project Data Processing Subgroup The sequence alignment/Map format and SAMtools. Bioinformatics. 2009;25:2078–2079. doi: 10.1093/bioinformatics/btp352. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Limpitikul WB, Viswanathan MC, O'Rourke B, Yue DT, Cammarato A. Conservation of cardiac L-type Ca2 channels and their regulation in Drosophila: a novel genetically-pliable channelopathic model. Journal of Molecular and Cellular Cardiology. 2018;119:64–74. doi: 10.1016/j.yjmcc.2018.04.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Lo PC, Frasch M. A role for the COUP-TF-related gene seven-up in the diversification of cardioblast identities in the dorsal vessel of Drosophila. Mechanisms of Development. 2001;104:49–60. doi: 10.1016/s0925-4773(01)00361-6. [DOI] [PubMed] [Google Scholar]
  26. Magyar J, Kiper CE, Sievert G, Cai W, Shi GX, Crump SM, Li L, Niederer S, Smith N, Andres DA, Satin J. Rem-GTPase regulates cardiac myocyte L-type calcium current. Channels. 2012;6:166–173. doi: 10.4161/chan.20192. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Mathieu J, Allard P, Potvin L, Prévost C, Bégin P. A 10-year study of mortality in a cohort of patients with myotonic dystrophy. Neurology. 1999;52:1658. doi: 10.1212/WNL.52.8.1658. [DOI] [PubMed] [Google Scholar]
  28. McNally EM, Sparano D. Mechanisms and management of the heart in myotonic dystrophy. Heart. 2011;97:1094–1100. doi: 10.1136/hrt.2010.214197. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Mesirca P, Torrente AG, Mangoni ME. Functional role of voltage gated Ca(2 ) channels in heart automaticity. Frontiers in Physiology. 2015;6:19. doi: 10.3389/fphys.2015.00019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Miller MR, Robinson KJ, Cleary MD, Doe CQ. TU-tagging: cell type-specific RNA isolation from intact complex tissues. Nature Methods. 2009;6:439–441. doi: 10.1038/nmeth.1329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Nakamori M, Kimura T, Kubota T, Matsumura T, Sumi H, Fujimura H, Takahashi MP, Sakoda S. Aberrantly spliced alpha-dystrobrevin alters alpha-syntrophin binding in myotonic dystrophy type 1. Neurology. 2008;70:677–685. doi: 10.1212/01.wnl.0000302174.08951.cf. [DOI] [PubMed] [Google Scholar]
  32. Nakamori M, Gourdon G, Thornton CA. Stabilization of expanded (CTG)•(CAG) repeats by antisense oligonucleotides. Molecular Therapy. 2011;19:2222–2227. doi: 10.1038/mt.2011.191. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Neely GG, Hess A, Costigan M, Keene AC, Goulas S, Langeslag M, Griffin RS, Belfer I, Dai F, Smith SB, Diatchenko L, Gupta V, Xia CP, Amann S, Kreitz S, Heindl-Erdmann C, Wolz S, Ly CV, Arora S, Sarangi R, Dan D, Novatchkova M, Rosenzweig M, Gibson DG, Truong D, Schramek D, Zoranovic T, Cronin SJ, Angjeli B, Brune K, Dietzl G, Maixner W, Meixner A, Thomas W, Pospisilik JA, Alenius M, Kress M, Subramaniam S, Garrity PA, Bellen HJ, Woolf CJ, Penninger JM. A genome-wide Drosophila screen for heat nociception identifies α2δ3 as an evolutionarily conserved pain gene. Cell. 2010;143:628–638. doi: 10.1016/j.cell.2010.09.047. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Ocorr K, Reeves NL, Wessells RJ, Fink M, Chen HS, Akasaka T, Yasuda S, Metzger JM, Giles W, Posakony JW, Bodmer R. KCNQ potassium channel mutations cause cardiac arrhythmias in Drosophila that mimic the effects of aging. PNAS. 2007;104:3943–3948. doi: 10.1073/pnas.0609278104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Perrin L, Monier B, Ponzielli R, Astier M, Semeriva M. Drosophila cardiac tube organogenesis requires multiple phases of hox activity. Developmental Biology. 2004;272:419–431. doi: 10.1016/j.ydbio.2004.04.036. [DOI] [PubMed] [Google Scholar]
  36. Picchio L, Plantie E, Renaud Y, Poovthumkadavil P, Jagla K. Novel Drosophila model of myotonic dystrophy type 1: phenotypic characterization and genome-wide view of altered gene expression. Human Molecular Genetics. 2013;22:2795–2810. doi: 10.1093/hmg/ddt127. [DOI] [PubMed] [Google Scholar]
  37. Picchio L, Legagneux V, Deschamps S, Renaud Y, Chauveau S, Paillard L, Jagla K. Bruno-3 regulates sarcomere component expression and contributes to muscle phenotypes of myotonic dystrophy type 1. Disease Models & Mechanisms. 2018;11:dmm031849. doi: 10.1242/dmm.031849. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Puhl HL, Lu VB, Won YJ, Sasson Y, Hirsch JA, Ono F, Ikeda SR. Ancient origins of RGK protein function: modulation of Voltage-Gated calcium channels preceded the protostome and deuterostome split. PLOS ONE. 2014;9:e100694. doi: 10.1371/journal.pone.0100694. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Rau F, Freyermuth F, Fugier C, Villemin JP, Fischer MC, Jost B, Dembele D, Gourdon G, Nicole A, Duboc D, Wahbi K, Day JW, Fujimura H, Takahashi MP, Auboeuf D, Dreumont N, Furling D, Charlet-Berguerand N. Misregulation of miR-1 processing is associated with heart defects in myotonic dystrophy. Nature Structural & Molecular Biology. 2011;18:840–845. doi: 10.1038/nsmb.2067. [DOI] [PubMed] [Google Scholar]
  40. Rotstein B, Paululat A. On the morphology of the Drosophila heart. Journal of Cardiovascular Development and Disease. 2016;3:15. doi: 10.3390/jcdd3020015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Savkur RS, Philips AV, Cooper TA. Aberrant regulation of insulin receptor alternative splicing is associated with insulin resistance in myotonic dystrophy. Nature Genetics. 2001;29:40–47. doi: 10.1038/ng704. [DOI] [PubMed] [Google Scholar]
  42. Sellin J, Albrecht S, Kölsch V, Paululat A. Dynamics of heart differentiation, visualized utilizing heart enhancer elements of the Drosophila Melanogaster bHLH transcription factor hand. Gene Expression Patterns. 2006;6:360–375. doi: 10.1016/j.modgep.2005.09.012. [DOI] [PubMed] [Google Scholar]
  43. Shieh BH, Zhu MY. Regulation of the TRP Ca2 channel by INAD in Drosophila photoreceptors. Neuron. 1996;16:991–998. doi: 10.1016/S0896-6273(00)80122-1. [DOI] [PubMed] [Google Scholar]
  44. Sicot G, Gourdon G, Gomes-Pereira M. Myotonic dystrophy, when simple repeats reveal complex pathogenic entities: new findings and future challenges. Human Molecular Genetics. 2011;20:R116–R123. doi: 10.1093/hmg/ddr343. [DOI] [PubMed] [Google Scholar]
  45. Spassova MA, Soboloff J, He L-P, Hewavitharana T, Xu W, Venkatachalam K, van Rossum DB, Patterson RL, Gill DL. Calcium entry mediated by SOCs and TRP channels: variations and enigma. Biochimica Et Biophysica Acta (BBA) - Molecular Cell Research. 2004;1742:9–20. doi: 10.1016/j.bbamcr.2004.09.001. [DOI] [PubMed] [Google Scholar]
  46. Splawski I, Timothy KW, Sharpe LM, Decher N, Kumar P, Bloise R, Napolitano C, Schwartz PJ, Joseph RM, Condouris K, Tager-Flusberg H, Priori SG, Sanguinetti MC, Keating MT. CaV1.2 Calcium Channel Dysfunction Causes a Multisystem Disorder Including Arrhythmia and Autism. Cell. 2004;119:19–31. doi: 10.1016/j.cell.2004.09.011. [DOI] [PubMed] [Google Scholar]
  47. Su Z, Barker DS, Csutora P, Chang T, Shoemaker RL, Marchase RB, Blalock JE. Regulation of Ca2 release-activated Ca2 channels by INAD and Ca2 influx factor. American Journal of Physiology. Cell Physiology. 2003;284:C497–C505. doi: 10.1152/ajpcell.00183.2002. [DOI] [PubMed] [Google Scholar]
  48. Taghli-Lamallem O, Plantié E, Jagla K. Drosophila in the heart of understanding cardiac diseases: modeling channelopathies and cardiomyopathies in the fruitfly. Journal of Cardiovascular Development and Disease. 2016;3:7. doi: 10.3390/jcdd3010007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Taneja KL, McCurrach M, Schalling M, Housman D, Singer RH. Foci of trinucleotide repeat transcripts in nuclei of myotonic dystrophy cells and tissues. The Journal of Cell Biology. 1995;128:995–1002. doi: 10.1083/jcb.128.6.995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Theadom A, Rodrigues M, Roxburgh R, Balalla S, Higgins C, Bhattacharjee R, Jones K, Krishnamurthi R, Feigin V. Prevalence of muscular dystrophies: a systematic literature review. Neuroepidemiology. 2014;43:259–268. doi: 10.1159/000369343. [DOI] [PubMed] [Google Scholar]
  51. Ueda K, Valdivia C, Medeiros-Domingo A, Tester DJ, Vatta M, Farrugia G, Ackerman MJ, Makielski JC. Syntrophin mutation associated with long QT syndrome through activation of the nNOS-SCN5A macromolecular complex. PNAS. 2008;105:9355–9360. doi: 10.1073/pnas.0801294105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. van Weerd JH, Christoffels VM. The formation and function of the cardiac conduction system. Development. 2016;143:197–210. doi: 10.1242/dev.124883. [DOI] [PubMed] [Google Scholar]
  53. Voght SP, Fluegel ML, Andrews LA, Pallanck LJ. Drosophila NPC1b promotes an early step in sterol absorption from the midgut epithelium. Cell Metabolism. 2007;5:195–205. doi: 10.1016/j.cmet.2007.01.011. [DOI] [PubMed] [Google Scholar]
  54. Wang GS, Kuyumcu-Martinez MN, Sarma S, Mathur N, Wehrens XH, Cooper TA. PKC inhibition ameliorates the cardiac phenotype in a mouse model of myotonic dystrophy type 1. Journal of Clinical Investigation. 2009;119:3797–3806. doi: 10.1172/JCI37976. [DOI] [PMC free article] [PubMed] [Google Scholar]

Decision letter

Editor: Margaret Buckingham1
Reviewed by: Rolf Bodmer2, David Brook3

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

This paper provides new insights into the underlying causes of cardiac defects in myotonic dystrophy. The authors use the fruit fly Drosophila as a model system which permits genetic manipulation of the misbalance between two RNA binding proteins, which triggers the disease. They examined the perturbations in gene expression in the heart that this produced and identified genes controlling calcium levels and notably a subunit of a calcium channel, α2δ3 known in Drosophila as straightjacket, which is present at an abnormally high level. They go on to show that overexpression of this gene in normal flies leads to asynchronous beating of the heart, whereas its downregulation in fly models for myotonic dystrophy improves the performance of the heart. Observations on mice and humans indicate that α2δ3 is normally low in the cardiac ventricles but is elevated in patients with myotonic dystrophy. These findings therefore suggest that reducing the level of this calcium channel sub-unit may be of therapeutic benefit for the prevention of cardiac conduction defects in patients suffering from myotonic dystrophy.

Decision letter after peer review:

[Editors’ note: a previous version of this study was rejected after peer review, but the authors submitted for reconsideration. The first decision letter after peer review is shown below.]

Thank you for submitting your work entitled "Straightjacket/α2δ3 deregulation is associated with cardiac conduction defects in Myotonic Dystrophy type 1" for consideration by eLife. Your article has been reviewed by a Senior Editor, a Reviewing Editor, and three reviewers. The following individuals involved in review of your submission have agreed to reveal their identity: Rolf Bodmer (Reviewer #1); David Brook (Reviewer #3).

Our decision has been reached after consultation between the reviewers. Based on these discussions and the individual reviews below, we regret to inform you that your work will not be considered for publication in eLife as it stands. However, if you can respond to the comments then we encourage a resubmission of the revised manuscript which will be sent to the same reviewers. They and the editor thought that responding to their comments would take more than the statutory revision period. They appreciated the fly work, with some points to be addressed, but were more critical of the data showing a link with DM. In addition to improving the results on patient material, the link would be reinforced by showing that in the fly model that you had developed, with the expanded repeat RNA, similar effects on stj are seen as with CELF1 and MBNL1 changes in expression.

Reviewer #1:

This manuscript by Plantié et al., studies a potential role of straightjacket/α2δ3 deregulation in cardiac conduction defects in Myotonic Dystrophy Type 1 (DM1, taking advantage of the fruit fly as a genetic model. DM1 is a muscular dystrophy disease caused by a CTG triple repeat in dystrophia myotonica protein kinase (dmpk) which causes imbalance between Muscleblind-like 1 (MBNL1 (Mbl D. mel)) and CUGBP and ELAV-like family member 1 (CELF1 (Bru-3 D. mel)). Importantly, as observed in flies the authors confirm transcriptional upregulation of α2δ3 in cardiac ventricles of DM1 patients supporting their findings. In their approach they used two different DM1 fly models exhibiting conduction defects and performed heart-specific TU-tagging based transcriptional profiling experiment, where they identified 4 genes involved in calcium handling (among them straightjacket, stj). Converging evidence from two different DM1 fly models with conduction defects that reveal a common subset of differentially expressed genes, demonstrates scientific rigor and strengthens the conclusions, including the link between the conduction defect and regulation of calcium transients. Overall, the findings are of high interest and relevance for understanding the mechanisms of DM1. However, several concerns need to be addressed.

Essential revisions:

1) Given that both genes together found to exhibit altered expression in DM1 patients (Mbl/MBNL1 down and Bru-3/CELF1 up) and that the fly models exhibit phenotype only in 25% (Bru-3 OE) or 58% (Mbl KD) of the flies, it would be of interest, whether there is an additive or synergistic effect on heart function by combining Mbl KD and Bru-3 OE.

2) More information is needed on the phenotypes of the other 3 genes, as they are potentially also contributing to the DM1-like phenotypes. For example, use different drivers for inaD, since lethal with Hand-Gal4.

3) Statistical methods need to be explained better and appropriately applied to all quantitative measurements. For example, the categorical data (how many flies show defects out of how many) need to be statistically evaluated, such as by Chi-square test (e.g. Figure 1C, Figure 5C,D), or corrected for non-normal distribution in other situations. In Figure 5D, is Hand>MblRNAi;stjRNAi statistically different from Hand>MblRNAi alone?

Reviewer #2:

The authors produced MBNL/CELF imbalance in Drosophila heart by decreasing the fly homologs Mbl/Bru-3. This produced asynchronous heart beats. They used heart-targeted TU-tagging specifically in heart cells followed by RNA sequencing to identify genes with altered expression that correlated with aberrant heart rhythm. They focus on straightjacket (stj)/CACNA2D3 (α2δ3), a subunit of the Cav1.2 voltage-gated calcium channel. Overexpression of stj in flies results in asynchronous heartbeats in 25-30% of flies and loss of stj function weakly rescues Mbl and rescues more convincingly rescues Bru-3.

One issue is the difficulty in establishing the specificity of the observed changes upon loss of Mbl or gain of Bru-3 as these relate to changes observed in DM1. For example, in both Mbl loss and Bru-3 gain, three other genes were shown to be down regulated and subsequent down regulation of each gene caused abnormal heart beats. MBNL and CELF families are highly conserved with antagonistic activities even in planaria so altered express will cause many abnormalities most of which are probably not related to DM. In this regard the study lacks experiments in a DM1 model where changes in straightjacket levels are observed in response to expression of CUG repeat RNA in the heart; CUG repeat RNA is the direct cause of the cardiac arrhythmias in DM1.

Another issue is that the level of Stj does not necessarily correlate with the extent of abnormal heart beats in Mbl knock down hearts (Figure 5C). While this could indicate additional contributing factors as suggested by the authors, it could also reflect a lack of cause effect relationship between MBNL/CELF, stj and cardiac arrhythmia.

The results attempting to show regulation of Stj mRNA by binding of Bru-3 is not convincing. There are no controls presented (mutation of putative binding sites) and what is the basis for identifying the Bru-3 and Mbl binding sites? A demonstration that Bru-3 regulates Stj mRNA levels by binding to the indicated element in the 3' UTR will need a good deal more work to establish the proposed hypothesis.

The statement that the results presented in Figure 6 "indicating a link between DM1 associated conduction defects and cardiac α2δ3 transcript regulation" is overstated. Given the large number of gene expression changes in DM1 heart, these data is not sufficient to make that case.

The data presented as immunofluorescence is difficult to interpret.

Reviewer #3:

Myotonic dystrophy type 1 is caused by a CTG repeat expansion in the 3'UTR of the DMPK gene. The repeat is transcribed but not translated into protein. The repeat expansion transcripts remain trapped in the nucleus where they sequester muscleblind-like proteins and stimulate the stabilization of CELF1. MBNL1 and CELF1 are counterbalanced factors involved in RNA splicing and an imbalance in these factors leads to a perturbation of RNA splicing in multiple transcripts and this is a key factor in DM1 pathophysiology. Electrocardiographic abnormalities are common in DM1, though this is rarely complete heart block.

In the present paper Plantié et al. identify Straightjacket (stj) deregulation as a factor in DM1-associated cardiac conduction defects. The authors overexpressed Bruno-3 (Bru-3) the fly ortholog of CELF1 and knocked down Mbl Muscleblind ortholog of MBNL1. Figure 1 shows the level of increase and or knockdown. In the case of Bru-3 this appears to be very highly expressed, whereas in DM1 CELF1 shows a comparatively modest increase and is stabilized via phosphorylation. Recent work indicates that number of repeat expansion transcripts is relatively small, possibly in the 6-20 range, which suggests that the amount of MBML1 sequestered may be relatively low.

Using Drosophila genetics the authors were able to perform heart-specific transcriptional profiling, which led to the identification of a subset of genes, including stj. The authors provide supporting evidence on calcium waves and rescue experiments for the relevance of this gene to heart conduction. However, is there any evidence that stj is mis-spliced as this is thought to be the primary consequence of the MBNL1/CELF1 imbalance? I am not aware of compelling evidence to support upregulation of genes as a consequence of MBNL1/CELF1 imbalance. A paper from Ebralidze et al. (2004) proposed RNA leaching of transcription factors in DM1 causing reduced gene expression, but this was not linked to MBNL1/CELF1 imbalance. In subsection “Increased cardiac expression of stj involves 3’UTR regulation and contributes to asynchronous heartbeats in DM1 flies”, the authors suggest a possible mechanism involving stj isoforms with extended 3'UTRs but it is unclear how this is contributing to increased stj expression.

The final part of the paper provides the key link to DM1 as the authors show increased α2δ3 (stj ortholog) transcript expression in ventricular biopsies. It would be reasonable to expect Western blot data to demonstrate increased protein levels and possibly some evidence of effects in tissues from other regions of the heart.

Overall this is an interesting observation but a little more work is needed to strengthen two aspects: first the molecular mechanism underlying an increase in stj levels and second evidence that the α2δ3 protein is increased in DM1 heart.

[Editors’ note: what now follows is the decision letter after the authors submitted for further consideration.]

Thank you for submitting your article "Straightjacket/α2‎δ3 deregulation is associated with cardiac conduction defects in myotonic dystrophy type 1" for consideration by eLife. Your article has been reviewed by Utpal Banerjee as the Senior Editor, a Reviewing Editor, and three reviewers. The following individuals involved in review of your submission have agreed to reveal their identity: Rolf Bodmer (Reviewer #2); David Brook (Reviewer #3).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

The reviewers are largely satisfied with the considerable additional work and modifications introduced in this re-submitted manuscript. However, before publication they require you to address the following points. There are still concerns of over-interpretation. In particular, while it is reasonable to propose that Mbl and Bru-3 bind and regulate the stj 3' UTR, the evidence is just not there. Also, the evidence that exon 15 is regulated by Mbl and Bru-3 is not convincing. The western blots showing the increase in α2δ3 protein are helpful and add support for the contention that the ortholog is affected in the disease. Overall the results of the effects of altered Mbl and Bru-3 are well done and are worth reporting with several modifications.

The authors need to indicate in the manuscript that the expression of CUG RNA repeats in the fly model does not impact stj expression. Also putting this information in the context that the DMSXL mouse model does not show conduction defects does not provide a justification for the lack of an effect on stj in their fly model; rather it means that the mouse model is not representative of DM1 heart phenotype.

The RNAseq tracts in Figure 5E are not sufficient to demonstrate a change in splicing of the exon indicated by the authors. The sequencing is not sufficiently deep to even detect all of the constitutive exons in their entirety. This analysis requires RT-PCR across the alternative exon. The RT-PCR in Figure 5E—figure supplement 2 is not the standard (or correct) approach to determine PSI – this requires using primers on flanking constitutive exons. The authors claim that Bru-3 gain or Mbl protein loss 'promotes' inclusion of exon 15 specific for stj-RC. However, since "stj total" is augmented by the same relative level, it is overall expression that is augmented, and in this approach does not speak to an alteration in splicing pattern. Exon 15 does not seem to be preferentially selected for, there is just overall more expression. The conclusions should reflect that. Alternatively, since the alternative splicing of stj is not a main point for the paper this part can be removed.

Figure 5E—figure supplement 2 is not sufficient to implicate that there are Mbl and Bru-3 binding sites in the indicated regions. There needs to be an analysis of statistical enrichment; one can find short sequences matching RNA binding protein binding motifs given they are short and variable. There is no evidence to support the contention that Mbl and/or Bru-3 regulate stj levels. This aspect too could be removed.

The request for more thorough statistics has been done in most instances, except for Figure 1—figure supplement 1A,B, Figure 5—figure supplement 1F and Figure 6B. In Figure 1C, brackets should indicate exactly which bar is compared which bar, not as a vertical line across three bars, which is confusing.

eLife. 2019 Dec 12;8:e51114. doi: 10.7554/eLife.51114.sa2

Author response

[Editors’ note: the author responses to the first round of peer review follow.]

Reviewer #1:

[…]

Essential revisions:

1) Given that both genes together found to exhibit altered expression in DM1 patients (Mbl/MBNL1 down and Bru-3/CELF1 up) and that the fly models exhibit phenotype only in 25% (Bru-3 OE) or 58% (Mbl KD) of the flies, it would be of interest, whether there is an additive or synergistic effect on heart function by combining Mbl KD and Bru-3 OE.

As suggested we generated UAS-Bru-3;UAS-MblRNAi double transgenic line and scored asynchronous heartbeat phenotypes in Hand>Bru-3;MblRNAi flies. To take into account double UAS context we also scored conduction defect phenotypes in Hand>Bru-33;UPRT and Hand>MblRNAi;UPRT flies. We found an increase in percentage of flies showing asynchronous heartbeats (partial block) from 33.3% in Hand>MblRNAi;UPRT up to 38.5% in Hand>Bru-3;Mbl1RNAi flies. Thus, this analysis reveals a moderate additive effect of Bru-3 overexpression and Mbl knockdown on conduction defects phenotype (new Figure 1—figure supplement 2). We add an appropriate comment stating this in the Results section.

2) More information is needed on the phenotypes of the other 3 genes, as they are potentially also contributing to the DM1-like phenotypes. For example, use different drivers for inaD, since lethal with Hand-Gal4.

We are grateful for this comment as it allowed us to more accurately characterize inaD, Syn1 and Rgk2 cardiac phenotypes. We tested additional RNAi lines for each of them (listed in Materials and methods section). Notice that for inaD we tested two additional RNAi lines and found they do not display lethality when crossed with Hand-Gal4. We thus took into account only the newly tested inaD lines and consider that the previously tested TRIP 52313 line generated an off target effect. Also notice that increased arrhythmia index and asystole initially reported for Rgk2 and fibrillations for Syn1 have not been confirmed by analyzing additional RNAi lines. In new Figure 3—figure supplement 2 we present observed phenotypes detected in at least two RNAi lines per gene. We report that cardiac-specific knockdowns of inaD, Syn1 and Rgk2 all lead to an increase in diastolic and systolic diameters a phenotype also observed in our fly DM1 models.

However, we do not observe asynchronous heartbeat phenotypes in inaD, Syn1 and Rgk2 RNAi contexts indicating that decrease in their expression has no role in DM1-associated conduction defects.

3) Statistical methods need to be explained better and appropriately applied to all quantitative measurements. For example, the categorical data (how many flies show defects out of how many) need to be statistically evaluated, such as by Chi-square test (e.g. Figure 1C, Figure 5C,D), or corrected for non-normal distribution in other situations. In Figure 5D, is Hand>MblRNAi;stjRNAi statistically different from Hand>MblRNAi alone?

As suggested by the reviewer we provide now statistical evaluation of asynchronous heartbeat data in all presented genetic contexts. We applied Fisher’s exact test. In Figure 5DHand>MblRNAi;stjRNAi rescue was compared with Hand>MblRNAi;UPRT double UAS condition and it is not significantly different. However, it is (p=0,02), when compared with Hand>MblRNAi alone (as suggested by the reviewer). This is also true for Hand-Bru-3;stjRNAi rescue condition. We thus modified Figure 5D by adding comparisons of Stj rescues with Hand>MblRNAi alone and Hand>Bru-3 alone.

We also applied one-way ANOVA, Kruskal-Wallis Dunn’s multiple comparison post-test to evaluate statistical significance of changes in diastolic and systolic heart diameters presented in new Figure 3—figure supplement 2.

Reviewer #2:

[…]

One issue is the difficulty in establishing the specificity of the observed changes upon loss of Mbl or gain of Bru-3 as these relate to changes observed in DM1. For example, in both Mbl loss and Bru-3 gain, three other genes were shown to be down regulated and subsequent down regulation of each gene caused abnormal heart beats.

Regarding relation to DM1, knockdown of Mbl mimics sequestration of Mbl while gain of Bru-3 mimics an increase of Bru-3 level, both observed in DM1. In our previous study (Picchio et al., 2018) by using transcriptional profiling we already demonstrated that 85% of genes deregulated in Drosophila DM1 model expressing 960 CTG repeats are common with Mbl RNAi and about 50% with Bru-3 gain of function context. We thus consider that analysing Mbl loss and Bru-3 gain of function and in particular genes commonly deregulated in these two contexts could provide valuable cues to DM1. This strategy was applied in the present study.

To better characterize cardiac phenotypes of three other candidate genes (inaD, Syn1 and Rgk2) and avoid potential off target effects we analyzed heart physiology in additional RNAi lines for those genes. Newly generated data are included to revised Figure 3—figure supplement 2.

MBNL and CELF families are highly conserved with antagonistic activities even in planaria so altered express will cause many abnormalities most of which are probably not related to DM. In this regard the study lacks experiments in a DM1 model where changes in straightjacket levels are observed in response to expression of CUG repeat RNA in the heart; CUG repeat RNA is the direct cause of the cardiac arrhythmias in DM1.

This issue needs some clarifications indeed as not all cardiac phenotypes observed in DM1 patients are easily reproduced in animal models expressing toxic CTG repeats. This is the case of DM1 associated conduction defects which were previously described i double KO mice for MBNL1 and MBNL2 (Lee et al., 2013) and in cardiac CELF1 gain of function context (Koshelev et al., 2010) but not in DMSLX mice harboring long CTG repeats (Algalarrondo et al., 2015).

Like in mice, in Drosophila, cardiac conduction defects are only present in loss of Mbl and gain of Bru-3 contexts but not in CTG repeats DM1 model. This indicates that loss of MBNL1/Mbl and gain of CELF1/Bru-3 models are more adapted than CTG repeat DM1 models to reproduce this severe cardiac phenotype observed in DM1 patients. Taking this into account we directly asked whether the increase in stj expression (contributing to conduction defects in fly models) could be detected in cardiac samples of DM1 patients with conduction defects.

It turned out that this is the case both at Stj/CACNA2D3 transcript and as we show now by Western blot at protein levels as well. Consistent with lack of cardiac conduction phenotypes, we do not detect changes in stj expression in Drosophila CTG repeat DM1 model.

Another issue is that the level of Stj does not necessarily correlate with the extent of abnormal heart beats in Mbl knock down hearts (Figure 5C). While this could indicate additional contributing factors as suggested by the authors, it could also reflect a lack of cause effect relationship between MBNL/CELF, stj and cardiac arrhythmia.

As stated above, conduction defects have been previously observed in MBNL1/2 DKO (Lee et al., 2013) and CELF1 gain of function (Koshelev et al., 2010) mouse models and we show here that this DM1-associated cardiac phenotype is also present in analogous contexts in Drosophila. As suggested by partial rescue of Mbl knockdown Stj is not the sole player, but it appears to be an important contributor.

The results attempting to show regulation of Stj mRNA by binding of Bru-3 is not convincing. There are no controls presented (mutation of putative binding sites) and what is the basis for identifying the Bru-3 and Mbl binding sites? A demonstration that Bru-3 regulates Stj mRNA levels by binding to the indicated element in the 3' UTR will need a good deal more work to establish the proposed hypothesis.

We agree that presented data do not sufficiently support 3’UTR-based regulation of Stj expression and that amount of work required to establish our hypothesis is not compatible with the time frame of all other revisions. However, to make a step forward we performed in silico search for MBNL1 and CELF1 binding sites (Lambert et al., 2014) within the 3’UTR stj region. We found that the long 3’UTR sequence contains several predicted CELF1/Bru-3 and MBNL1/Mbl binding sites (Figure 5—figure supplement 2), suggesting that an interplay between Bru-3 and Mbl proteins bound to 3’UTR could regulate stj isoforms abundance. In parallel, as demonstrated by Exon15 targeted qRT-PCR analysis (Figure 5—figure supplement 2) the Mbl loss and Bru-3 gain of function, both promote inclusion of Exon 15 leading to the upregulation of stj-RC. Thus, we hypothesize that in DM1 context a combination of splice-dependent and 3’UTR-dependent mechanisms control elevation of stj transcripts in cardiac cells. We include an appropriate comment to the text.

The statement that the results presented in Figure 6 "indicating a link between DM1 associated conduction defects and cardiac α2δ3 transcript regulation" is overstated. Given the large number of gene expression changes in DM1 heart, these data is not sufficient to make that case.

We agree and modify this Results section as follow:

“Ventricular α2δ3 transcript and protein levels were both significantly higher in DM1 patients with conduction disturbances compared to controls (Figure 6C, D). We also found, by analyzing the same human samples a higher expression of the main Calcium channel α1/Cav1.2 unit (Figure 6—figure supplement 1) further supporting the view (Figure 6E) that Ca2 inward current deregulation underlies DM1-associated conduction defects.”

Reviewer #3:

[…]

In the present paper Plantié et al. identify Straightjacket (stj) deregulation as a factor in DM1-associated cardiac conduction defects. The authors overexpressed Bruno-3 (Bru-3) the fly ortholog of CELF1 and knocked down Mbl Muscleblind ortholog of MBNL1. Figure 1 shows the level of increase and or knockdown. In the case of Bru-3 this appears to be very highly expressed, whereas in DM1 CELF1 shows a comparatively modest increase and is stabilized via phosphorylation. Recent work indicates that number of repeat expansion transcripts is relatively small, possibly in the 6-20 range, which suggests that the amount of MBML1 sequestered may be relatively low.

We agree that knocking down Mbl or overexpressing Bru-3 will not perfectly reproduce DM1 condition and in consequence all aspects of gene deregulations observed in DM1. Nevertheless, loss of Mbnl1 and gain of CELF1 functions have already been used to model DM1 by other labs. In our previous study (Picchio et al., 2018) by using transcriptional profiling we demonstrated that 85% of genes deregulated in Drosophila DM1 model with muscle targeted expression of 960 CTG repeats are common with Mbl RNAi and about 50% with Bru-3 gain of function context. We thus consider that analysing Mbl loss and Bru-3 gain of function and in particular genes commonly deregulated in these two contexts could provide valuable cues to DM1.

Using Drosophila genetics the authors were able to perform heart-specific transcriptional profiling, which led to the identification of a subset of genes, including stj. The authors provide supporting evidence on calcium waves and rescue experiments for the relevance of this gene to heart conduction. However, is there any evidence that stj is mis-spliced as this is thought to be the primary consequence of the MBNL1/CELF1 imbalance? I am not aware of compelling evidence to support upregulation of genes as a consequence of MBNL1/CELF1 imbalance. A paper from Ebralidze et al. (2004) proposed RNA leaching of transcription factors in DM1 causing reduced gene expression, but this was not linked to MBNL1/CELF1 imbalance. In subsection “Increased cardiac expression of stj involves 3’UTR regulation and contributes to asynchronous heartbeats in DM1 flies”, the authors suggest a possible mechanism involving stj isoforms with extended 3'UTRs but it is unclear how this is contributing to increased stj expression.

We provide now a more compelling evidence for the up regulation of one particular stj isoform – stj-RC in both MblRNAi and Bru-3 gain of function contexts (Figure 5—figure supplement 2). Stj-RC carries long 3’UTR but also contains alternatively spliced Exon 15. As demonstrated by Wang et al., (2015), both MBNL1 and CELF1 do not only act as alternative splice regulators but also display 3’UTR binding activities with potential role in regulation transcripts stability. In the case of stj-RC transcript its 3’UTR sequence contains several predicted CELF1/Bru-3 and MBNL1/Mbl binding sites, suggesting that an interplay between Bru-3 and Mbl proteins bound to 3’UTR could regulate stj-RC abundance (Figure 5—figure supplement 2). In parallel, Bru-3 gain and Mbl loss, both promote inclusion of Exon 15 (Figure 5—figure supplement 2). Thus, we hypothesize that a combination of splice-dependent and 3’UTRdependent mechanisms is involved in up regulation of stj. We add appropriate comments to the text.

The final part of the paper provides the key link to DM1 as the authors show increased α2δ3 (stj ortholog) transcript expression in ventricular biopsies. It would be reasonable to expect Western blot data to demonstrate increased protein levels and possibly some evidence of effects in tissues from other regions of the heart.

We thank the reviewer for this suggestion. With limited amount of available tissue from cardiac biopsies we were able to perform Western blot on 3 out of 5 DM1 ventricular cardiac samples analysed for transcript expression. We found that similarly to transcripts, α2δ3 protein is significantly elevated in DM1 condition. These new data are included to Figure 6.

Overall this is an interesting observation but a little more work is needed to strengthen two aspects: first the molecular mechanism underlying an increase in stj levels and second evidence that the α2δ3 protein is increased in DM1 heart.

We thank the reviewer for this comment and hope our additional data included to Figure 5 and Figure 6 strengthen two above aspects.

[Editors' note: the author responses to the re-review follow.]

The authors need to indicate in the manuscript that the expression of CUG RNA repeats in the fly model does not impact stj expression.

We introduced the following sentence to the Introduction: “We did not observe asynchronous heartbeats in flies expressing in the heart 960CTG repeats. This DM1 model (Picchio et al., 2013) developed other cardiac phenotypes such as arrhythmia.”

Also putting this information in the context that the DMSXL mouse model does not show conduction defects does not provide a justification for the lack of an effect on stj in their fly model; rather it means that the mouse model is not representative of DM1 heart phenotype.

We agree. We are not making parallels between mouse DMSXL and our Drosophila DM1 – 960CTG model.

The RNAseq tracts in Figure 5E are not sufficient to demonstrate a change in splicing of the exon indicated by the authors. The sequencing is not sufficiently deep to even detect all of the constitutive exons in their entirety. This analysis requires RT-PCR across the alternative exon. The RT-PCR in Figure 5E—figure supplement 2 is not the standard (or correct) approach to determine PSI – this requires using primers on flanking constitutive exons. The authors claim that Bru-3 gain or Mbl protein loss 'promotes' inclusion of exon 15 specific for stj-RC. However, since "stj total" is augmented by the same relative level, it is overall expression that is augmented, and in this approach does not speak to an alteration in splicing pattern. Exon 15 does not seem to be preferentially selected for, there is just overall more expression. The conclusions should reflect that. Alternatively, since the alternative splicing of stj is not a main point for the paper this part can be removed.

We agree that alternative splicing is not a main point here and modify a portion of the text subsection “Increased cardiac expression of stj contributes to asynchronous heartbeats in DM1 flies” related to Figure 5E and Figure 5—figure supplement 2, which were also modified accordingly. Revised text reads as follow:

“We also undertook to determine, which among three stj transcript isoforms (Figure 5E) is elevated in DM1 contexts. The analysis of RNAseq peaks over the stj locus (Figure 5E) indicated that transcripts harboring exon 15 (stj‐RC) and long 3’UTR sequences (stj-RB and stj‐RC) are enriched in both Hand>Bru‐3 and Hand>MblRNAi fly hearts. The up regulation of stj‐RC was validated by RTqPCR on dissected wild type and DM1 hearts (Figure 5—figure supplement 2) showing that its expression level is equivalent to the expression of all stj transcripts, thus providing evidence that stj‐RC is the main stj isoform whose expression increases in pathological contexts."

Figure 5E—figure supplement 2 is not sufficient to implicate that there are Mbl and Bru-3 binding sites in the indicated regions. There needs to be an analysis of statistical enrichment; one can find short sequences matching RNA binding protein binding motifs given they are short and variable. There is no evidence to support the contention that Mbl and/or Bru-3 regulate stj levels. This aspect too could be removed.

Having no evidence that some of predicted Mbl and/or Bru‐3 binding motifs present within 3’UTR stj region are functionally relevant we follow the reviewer’s suggestion and remove the map of potential Mbl and Bru‐3 binding motifs from Figure 5E and Figure 5—figure supplement 2A.

The request for more thorough statistics has been done in most instances, except for Figure 1—figure supplement 1A,B, Figure 5—figure supplement 1F and Figure 6B. In Figure 1C, brackets should indicate exactly which bar is compared which bar, not as a vertical line across three bars, which is confusing.

All indicated points are now amended.

Associated Data

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

    Data Citations

    1. Plantie E, Picchio L, Renaud Y. 2018. Deregulation associated with cardiac conduction defects in Myotonic Dystrophy Type 1 using TU-Tagging. NCBI Gene Expression Omnibus. GSE109370

    Supplementary Materials

    Figure 3—source data 1. A list of Drosophila genes and their Human orthologs deregulated in both Hand>Bru-3 and Hand>MblRNAi contexts.

    Orthology scores were obtained from DIOPT tool (DRSC Integrative Ortholog Prediction Tool) (Hu et al., 2011) which predicts orthologs, based on sequence homology, phylogenetic trees, and functional similarity (www.flyrnai.org/cgi-bin/DRSC_orthologs.pl).

    elife-51114-fig3-data1.xlsx (31.1KB, xlsx)
    Figure 5—source data 1. Corrected total cell fluorescence (CTCF) of Stj signal in circular fibers of cardiomyocytes.
    elife-51114-fig5-data1.xlsx (38.2KB, xlsx)
    Transparent reporting form
    elife-51114-transrepform.pdf (321.2KB, pdf)

    Data Availability Statement

    All data generated or analysed during this study are included in the manuscript and supporting files. Sequencing data have been deposited with the GEO-NCBI tracking system under accession number GSE109370.

    The following dataset was generated:

    Plantie E, Picchio L, Renaud Y. 2018. Deregulation associated with cardiac conduction defects in Myotonic Dystrophy Type 1 using TU-Tagging. NCBI Gene Expression Omnibus. GSE109370

    Articles from eLife are provided here courtesy of eLife Sciences Publications, Ltd

    RESOURCES