Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Dec 27.
Published in final edited form as: Cell Rep. 2015 Dec 10;13(11):2386–2394. doi: 10.1016/j.celrep.2015.11.028

Actinomycin D Specifically Reduces Expanded CUG Repeat RNA in Myotonic Dystrophy Models

Ruth B Siboni 1,5, Masayuki Nakamori 2,5, Stacey D Wagner 1, Adam J Struck 1, Leslie A Coonrod 1, Shanee A Harriott 3, Daniel M Cass 3, Matthew K Tanner 1, J Andrew Berglund 1,4,✉
PMCID: PMC4691565  NIHMSID: NIHMS737970  PMID: 26686629

Summary

Myotonic Dystrophy type 1 (DM1) is an inherited disease characterized by the inability to relax contracted muscles. Affected individuals carry large CTG expansions that are toxic when transcribed. One possible treatment approach is to reduce or eliminate transcription of CTG repeats. Actinomycin D (ActD) is a potent transcription inhibitor and FDA-approved chemotherapeutic that binds GC-rich DNA with high affinity. Here, we report that ActD decreased CUG transcript levels in a dose-dependent manner in DM1 cell and mouse models at significantly lower concentrations (nanomolar) compared to its use as a general transcription inhibitor or chemotherapeutic. ActD also significantly reversed DM1-associated splicing defects in a DM1 mouse model, and did so within the currently approved human treatment range. RNA-seq analyses showed that low concentrations of ActD did not globally inhibit transcription in a DM1 mouse model. These results indicate that transcription inhibition of CTG expansions is a promising treatment approach for DM1.

Introduction

Myotonic dystrophy (DM), the most common form of adult onset muscular dystrophy, is a disease characterized by (but not limited to) myotonia, muscle wasting, insulin resistance, cardiomyopathy, and cognitive dysfunctions (Ranum et al., 2006; Cho et al., 2007). DM has two clinical manifestations: type 1 and type 2 (DM1 and DM2). DM1 is caused by an inherited expansion of CTG repeats in the 3’ UTR of the DMPK gene (Harley et al., 1992; Mahadevan et al., 1992). Unaffected individuals have between 5 and 35 CTG repeats, while those afflicted with DM1 have more than 50 and can have up to thousands of repeats (reviewed in O'rourke et al., 2009). When transcribed into RNA, the CUG repeats serve as binding sites for RNA-binding proteins, including the MBNL family of splicing factors (Miller et al., 2000). By binding to and aggregating with the CUG repeats, MBNL proteins are effectively “sequestered” away from performing their canonical functions (Ho et al., 2004; reviewed in Osborne and Thornton, 2006). Consistent with this model, in vivo fluorescent probing experiments of expanded CUG repeats demonstrated that they form nuclear aggregates, or foci, containing MBNL proteins (Fardaei et al., 2002; Ho et al., 2005).

Members of the MBNL family regulate the alternative splicing of over 100 different transcripts, and are also involved in RNA localization and processing events (reviewed in Konieczny et al., 2015; Echeverria and Cooper, 2012). Some mRNAs that are mis-spliced in DM1, including insulin receptor (INSR), cardiac troponin T (TNNT2), and muscle-specific chloride channel (CLCN1), correspond directly or are linked to symptoms experienced by DM1 patients— insulin insensitivity, cardiac defects, and myotonia, respectively (Savkur et al., 2001; Philips et al., 1998; Mankodi et al., 2002). Although there is currently no treatment, approaches are under development that reduce or eliminate CUG:MBNL aggregates using small molecules, antisense oligonucleotides, and peptides (Warf et al., 2009; Arambula et al., 2009; Nakamori et al., 2011; Lee et al., 2012; Wheeler et al., 2012). More recently, studies have indicated that small molecules that interact with CTG-rich DNA reduce CUG RNA levels, likely through transcription inhibition (Coonrod et al., 2013). The latter finding prompted us to identify transcription inhibitors that possess high affinity and specificity for CTG-rich DNA.

Actinomycin D (ActD) is a small molecule known to bind GC-rich DNA and is naturally produced by Streptomyces bacteria (Waksman and Woodruff, 1940). ActD is commonly used in mRNA stability studies as a general transcription inhibitor, with common protocols using final concentrations of 1–3 µM to achieve global transcription inhibition (Bensaude, 2011; Perry and Kelley, 1970). Importantly, it is also a potent anticancer drug that has been FDA approved since 1964 for multiple tumor types under the clinical name Cosmogen®. From a structural standpoint, ActD is a neutral molecule comprised of a planar phenoxazone ring with two cyclic pentapeptides (Figure 1A), and binds double and single-stranded DNA (but not RNA) by intercalating with GpC sequences with high specificity (Mueller and Crothers, 1968; Kamitori et al., 1992). A crystal structure by Hou and colleagues demonstrated that ActD binds CTG:CTG DNA with high affinity, implicating the importance of the destabilized T:T mismatch for binding (Hou et al., 2002; Liu and Chen, 1996). Close inspection of this crystal structure reveals that the hydrophobic cyclic pentapeptides of ActD molecules are in proximity to each other when bound to CTG DNA, possibly stabilizing the ActD:DNA complex (Figure 1B). CTG:CTG DNA duplexes are a structural feature of CTG triplet repeat expansions, often the result of DNA slippage during replication (Chi and Lam, 2005; Petruska et al., 1996). Collectively, these studies suggest that ActD may possess a higher affinity for CTG repeat expansions compared to other GC-containing targets in vivo.

Figure 1. Biochemical analysis of Actinomycin D molecules bound to CTG:CTG DNA.

Figure 1

Open in a new tab

(A) Chemical structure of Actinomycin D. (B) Crystal structure of the 2:1 ActD–(ATGCTGCAT)2 complex (reimaged from Hou et al., 2002). ActD molecules are in blue van der Waals representations, while DNA is shown in skeletal form. Intercalation of the phenoxazone ring is observed at both GpC steps, while pentapeptides remain within the minor groove. (C) Representative ITC isotherm for the interaction between ActD and (CTG)4. Raw heats of reaction versus time are imbedded in the upper left of the graph. The average thermodynamic binding parameters when fit with the sequential binding model are K1 = (9±6)*104, ΔH1 = (7±6)*103 kcal/mol, K2 = (2±1)*106, ΔH2 = (−3±1)*104 kcal/mol. (D) The lowest free energy structure of the (CTG)4 sequence (Reuter et al., 2010).

In this study, we determined if ActD could reduce or even completely reverse the toxic effects of transcribed CTG expansions in DM1. Using both cellular and mouse models, we demonstrated that ActD reduces CTG transcript levels and ameliorates DM1-associated mis-splicing at low concentrations. RNA-seq analyses of a DM1 mouse model further confirmed that the concentrations of ActD necessary for significant CTG transcript reduction did not result in global transcription inhibition and rescue mis-spliced events in DM1.

Results

Actinomycin D interacts with CTG DNA repeats, but not CUG RNA repeats

To ascertain whether ActD interacts with CUG RNA in a similar capacity as with CTG DNA, thermodynamic and stoichiometric properties associated with the formation of ActD complexes with (CTG)4 and (CUG)4 sequences were determined by Isothermal Titration Calorimetry (ITC). At least two different types of enthalpic changes occur between (CTG)4 and ActD (Figure 1C). Increasing ActD concentrations within the system resulted in an initial endothermic reaction followed by an exothermic reaction. The data were best fit with a sequential binding model, indicating cooperativity between the two binding events. The initial CTG-ActD interaction demonstrated a KD of 10 +/− 7 µM, a considerably weaker affinity than the second interaction, which produced a KD value of 0.5 +/− 0.3 µM, consistent with the lowest free-energy structure of (CTG)4 containing two ActD GpC sites (Figure 1D). Taken together with previous studies that have indicated the need for a structural rearrangement of the DNA for tight binding by ActD (Hou et al., 2002; Paramanathan et al., 2012), these ITC data support an allosteric model of ActD: DNA binding. Furthermore, as ActD did not produce an enthalpic change when titrated into the (CUG)4 RNA (data not shown), it is clear that ActD is unable to bind CUG RNA in vitro and thus unlikely that ActD preferentially binds CUG RNA in vivo.

Reduction of transcripts containing expanded CUG repeats with Actinomycin D

Two DM1 cell models were tested to determine the ability of ActD to reduce CUG repeat RNA levels. Previous studies demonstrated that overexpression of 960 interrupted CUG repeats by transfection into HeLa cells recapitulates DM1 mis-splicing (Ho et al., 2004; Warf et al., 2009). Northern blot analysis revealed that CUG RNA levels decreased by more than 50% (p=0.002) with 5 nM ActD treatment over 18 hours, decreased by ~70% with 10 nM ActD, and a dose-dependent decrease was observed from 10 to 20 nM ActD (Figure 2A). Over this range of ActD concentrations, levels of the GAPDH transcript (used as a control) remained relatively constant. To determine if ActD reduced endogenous CTG repeat transcription, a DM1 patient-derived fibroblast line containing over 2000 CTG repeats was tested (Chaouch et al., 2009). The DM1 patient-derived cells responded to ActD over a lower concentration range compared to HeLa cells, with a dose-dependent decrease in CUG RNA levels observed from 1 to 6 nM ActD treatment over an 18 hour treatment (Figure 2B). CUG RNA levels were reduced by 44% at 3 nM and 60% at 6nM (p= 0.03). Interestingly, a reproducible 20% increase in CUG RNA was observed around 7 nM treatment but was significantly lower than in untreated DM1 cells. CUG repeat RNA remained constant from 7 nM to >100 nM ActD treatment (data not shown). Once again, GAPDH control levels were unaffected by ActD levels in this concentration range. The concentrations of ActD appeared to be mildly toxic to HeLa cells at the concentrations tested, and very little to no toxicity was observed in the DM1 patient-derived fibroblast line, as confirmed by cell viability assays (data not shown).

Figure 2. Actinomycin D reduces CUG repeat levels in DM1 cell models.

Figure 2

Open in a new tab

(A-B) CUG repeat RNA levels after ActD treatment in a DM1 HeLa cell system. Northern blot analysis and quantification demonstrate a significant decrease in CUG repeats at 5 nM within the HeLa cells and a dose-dependent decrease follows thereafter. GAPDH, a housekeeping gene, is used as a control. In DM1 patient cells, a modest increase in CUG repeats is observed after 6 nM, and CUG levels persist at that amount from 7 nM to 100 nM ActD (only data points up to 10 nM ActD are shown). Asterisks indicate level of significance (*= p < 0.05, ***= p < 0.001). Error bars represent standard deviation. (C) CUG ribonuclear foci formation in response to ActD treatment in HeLa cells. (top row) Nuclei of untreated HeLa cells. DAPI staining of nucleus is on far left and nuclear MBNL1 staining is in the second column. These untreated cells were not transfected with CUG repeats. (middle row) Transfection with CUG960 gives rise to ribonuclear foci in HeLa cells. Aggregation of MBNL1 and CUG repeats observed in the second and third columns, respectively. (bottom row) Treatment with 10 nM ActD in HeLa cells results in decreased number and size of MBNL1 and CUG aggregates (second and third columns), and reduction of MBNL sequestration as evidence by reversion to more diffuse staining (second column).

CUG repeat foci formation is reduced with Actinomycin D treatment

In order to directly visualize ActD’s effect on MBNL1 sequestration, CUG foci formation and nuclear MBNL1 localization were examined in the HeLa cell system in the presence or absence of ActD. MBNL1 antibody MB1a (4A8) was used to probe for MBNL1 localization as previously described, and a CAG probe was employed to bind CUG RNA (Holt et al., 2007). Untreated HeLa cells (without CUG repeats) demonstrated diffuse localization of MBNL1 throughout the nucleus (Figure 2C, top row). In the presence of CUG960, RNA foci were observed, and MBNL1 staining demonstrated aggregation (Figure 2C, middle row). Upon incubation with 10 nM ActD for 18 hours, we observed a decrease in the number of foci per cell when compared to the transfected controls (Figure 2C, bottom row). While most cells treated with this ActD concentration still possessed some nuclear foci, there was a 50% reduction observed in the number of foci per nucleus (n=66, p=0.00002). Furthermore, treated cells demonstrated a 46% decrease in the number of large foci, when large foci were defined as greater than 2 µm3 (n=66, p < 0.000025). Nuclear MBNL1 in treated cells was more diffuse compared to untreated cells, despite the persistence of several foci. Our observations of foci reduction in the HeLa model and the reduction of CUG RNA in both cell types motivated further investigations to determine if ActD treatment could correct DM1 molecular pathology in a mouse model.

Actinomycin D partially rescues mis-splicing of multiple events in a DM1 mouse model

The HSALR mouse model expresses 220 CUG repeats under the human skeletal actin promoter (Mankodi et al., 2000). Quantitative RT-PCR showed a significant reduction (p=0.01) of the HSA transgene mRNA in the HSALR mouse in response to a five day 0.025 mg kg−1 ActD treatment compared to PBS-treated controls (Figure 3A). In contrast, mRNA levels of the endogenous Dmpk gene (which does not possess long repeats in these transgenic mice) did not decrease significantly upon ActD treatment (Figure 3B), suggesting specificity for the repeat-containing HSA transgene.

Figure 3. Actinomycin D rescues mis-splicing events in HSALR mice.

Figure 3

Open in a new tab

(A) qRT-PCR analysis of HSA transgene levels in HSALR mice treated with PBS and 0.5% DMSO or indicated dosages of ActD for 5 days. Each circle represents the transcript levels within vastus muscle of a single mouse. Mice treated with any of the ActD dosages exhibited significant reduction of HSA transgene levels as compared to control treated mice. Asterisk indicate significance (* = p < 0.05). (B) qRT-PCR analysis of endogenous DMPK levels (no repeats) in HSALR mice treated with PBS and 0.5% DMSO or indicated dosages of ActD for 5 days. None of the ActD dosages caused significant reduction of DMPK levels. (C-H) Jitterplot representation of various endogenous splicing events perturbed in DM1 mice. Each symbol represents the splicing outcome for vastus muscle of a single mouse, while line represents the average of all experiments. (C) Clcn1 splicing demonstrates almost complete rescue by 0.25 mg kg−1 per day for 5 days. Gel images of two replicates per condition are demonstrated below. (D) In the Atp2a1 event, almost full rescue is observed by 0.25 mg kg−1 per day for 5 days. (E) Mbnl1 auto-regulates exon 5 inclusion, but the highest ActD dosage only reverses ~50% of the observed mis-splicing in mice. (F) ActD reverses mis-splicing of the Vps39 exon 3 event completely by 0.25 mg kg−1 per day for 5 days. (G-H) Nfix and Ldb3 events both exhibit partial but significant reversal of mis-splicing upon treatment with ActD.

Splicing rescue of two commonly examined, MBNL-regulated transcripts was observed upon ActD treatment in a DM1 transgenic mouse model: the muscle-specific chloride ion channel (Clcn1), the mis-splicing of this event is responsible for causing the myotonia phenotype in HSALR mice, and the sarco/endoplasmic reticulum Ca2+ ATPase 1 (Atp2a1 or Serca1) (Mankodi et al., 2002; Kimura, 2005). In Clcn1, ActD treatment with either 0.125 mg kg−1 or 0.25 mg kg−1 over a five day treatment regimen resulted in the robust rescue of Clcn1 splicing, as demonstrated by an increase in the exclusion of exon 7a (Figure 3C, lanes 3 and 4). The Atp2a1 event also showed significant mis-splicing rescue. In this event, treatment with ActD increased levels of exon 22 to 83 ± 10% at the highest dosage (Figure 3D, lanes 3 and 4). Although the mis-splicing of these events never reached full rescue, the observed rescue was significant at the highest dosages. Toxicity was not observed within the mouse model over this treatment range. Other important MBNL-regulated splicing rescue events were characterized in response to ActD treatment, including Mbnl1 (auto-regulates its own expression at the splicing level), Vps39 (regulator of TGF-beta signaling), Nfix (a transcription factor), and Ldb3 (important in membrane protein clustering). All of these events also demonstrated partial or complete rescue of mis-splicing (Figures 3F-3H). Collectively, these results showed that ActD partially reversed an important portion of DM1 molecular pathology.

ActD does not globally modify expression profiles of treated DM1 mice

Because ActD has a well-characterized ability to bind DNA ubiquitously at higher concentrations and induce global transcription inhibition, we wanted to assess ActD’s effects on transcript levels at the low dosages used in these studies. Therefore, RNA-seq analyses were performed on 0.125 mg kg−1, and 0.25 mg kg−1 ActD treated HSALR mouse vastus muscle RNA. Only 4.3% of genes were differentially expressed (p < 0.1) with 0.125 mg kg−1 ActD treatment, compared to the untreated controls; a similarly low proportion of genes were affected with the higher treatment (0.25 mg kg−1, 5.1%.) (Figure 4A). Approximately 3% of genes decreased in expression with either dosage, possibly a direct result of ActD-mediated transcriptional inhibition. Meanwhile, about 2% of genes increased in expression with either dosage. Collectively, these data demonstrated that ActD does not broadly inhibit transcription within these treatment concentrations, and does not affect a relatively large percentage of the transcriptome. Furthermore, we observed consistency in the way genes changed across the two dosages (Table S1, Figure 4 B-C). In total, 631 genes were upregulated by either dosage, and 401 of these genes were shared across the two concentrations (Figure 4B). When comparing the fold change of these values, we observed a correlation (Pearson r= 0.967) between the two ActD treatments, indicating that the degree of change did not vary significantly at the higher dosage (Figure 4C). Therefore, it is clear that the 0.25 mg kg−1 dosage, which is required for maximal splicing rescue, does not affect significantly more transcripts compared to the 0.125 mg kg−1 dosage, although we anticipate that a larger dosage difference may affect a greater proportion of genes. The functional categories of genes that are changed in both ActD treatments are found in Table S2 and indicated that the gene ontology term "extracellular matrix" had a significant p-value (p=9.22x10−21).

Figure 4. Less than 5% of genes are differentially expressed in mice upon Actinomycin D treatment.

Figure 4

Open in a new tab

(A) MA-plots of 0.125 mg kg−1 compared to no-treatment (PBS) control mice (left panel) and 0.250 mg kg−1 compared to no-treatment (PBS) control mice (right panel). Gene log2 fold change is plotted against the mean of normalized counts. Red circles denote genes with adjusted p values less than 0.1. Points that do not fall within the window are denoted with triangles. (B) Venn diagram depicts the number of differentially expressed genes shared by the two treatments compared to the no-treatment control. (C) Genes are plotted with the log2 fold change for mice treated with low (0.125 mg kg−1) verses high (0.250 mg kg−1) dose of ActD.

RNA-seq data was also used to validate our previous findings that ActD rescued mis-splicing in the HSALR mouse, as well as identify additional splicing changes that resulted from ActD treatment. Differential splicing analyses were used to compare “percent spliced in” (PSI or Ψ) values of alternatively spliced cassette exons between WT and ActD-treated mice (change in splicing denoted as ΔΨ). In total, 265 skipped exon events showed evidence of mis-regulation between WT and HSALR mice (Δ Ψ > 0.05, Bayes factor > 5, Table S3). This finding was consistent with previously published results (Wang et al., 2012). Of those events, 70 were rescued to some degree with 0.25 mg kg−1 ActD, 53 events were rescued with 0.125 mg kg−1 ActD and 40 showed rescue in both dosage groups. Importantly, of the 70 events rescued at the 0.25 mg kg−1 dosage, the majority demonstrated greater than 50% mis-splicing rescue upon ActD treatment (Figure 5A). Three events previously examined using qRT-PCR (Mbnl1, Ldb3, and Atp2a1) demonstrated a clear trend toward splicing rescue as did three previously unexplored events (Capzb, Igtb1, and Usp47) (Figure 5B). Interestingly, the gene ontology term "alternative splicing" was among those with a low p-value for both ActD concentrations (p=4.00x10−10) for gene expression changes (Table S2).

Figure 5. Numerous mis-spliced events are rescued upon Actinomycin D treatment in DM1 mice.

Figure 5

Open in a new tab

(A) Distribution of the average percent rescue observed for the 70 splicing events that were mis-regulated in HSALR mice and showed evidence of rescue in littermate mice treated with 0.25 mg kg−1 ActD. (B) Jitterplot representation of various splicing events perturbed in HSALR mice that showed evidence of rescue. Each symbol represents the splicing outcome estimate generated by MISO in the vastus muscle of a single mouse and the line represents the 95% confidence interval on that estimate.

Discussion

Our results demonstrate that ActD, a known transcription inhibitor with affinity and specificity for GC-rich DNA (Kamitori et al., 1992), reduces CUG RNA transcript levels in both DM1 cell and mouse models (Figures 2A-2B), and rescues MBNL1 dependent mis-splicing events in a mouse model (Figure 3). At low nanomolar concentrations, ActD reduced CUG levels in HeLa and DM1 cells by 50%, and a similar effect was observed with a 0.25 mg kg−1 dosage over 5 days in a DM1 mouse model. Importantly, RNA-seq results from ActD treated DM1 mice demonstrated that ActD did not affect transcription globally; at 0.25 mg kg−1 less than 5% of genes were altered; however, multiple mis-spliced events were partially or fully rescued with this ActD dosage (Figure 5). Taken together, these results suggest that ActD possesses specificity for the CTG/CUG repeats at this concentration, and furthermore has the ability to reverse DM1 molecular pathology.

Currently, ActD is most commonly used as a key component in the multimodal treatment of several cancer types, including Wilms tumors in children and gestational trophoblastic neoplasias in adult women (Hill et al., 2014; Osborne et al., 2011). Standard dosages of treatment for children and adults vary between 0.015-0.045 mg kg−1. In this study, the lowest and highest dosages administered to mice were 0.025 mg kg−1 and 0.25 mg kg−1. To extrapolate the human equivalent dosage (HED) of this treatment range, we used a body surface area normalization method, recommended as the appropriate conversion method when entering phase I and II clinical trials (Reagan-Shaw et al., 2008). This method calculates the HED of a treatment by multiplying the animal dosage by an appropriate species-specific conversion factor, or Km, which has been previously established for a number of species used as animal models (Freireich et al., 1966). Our results demonstrated that ActD is able to reduce CUG RNA in mice with a HED value of 0.002 mg kg −1 and corrected mis-splicing at a HED value of 0.02 mg kg −1. Both of these concentrations are below the approved therapeutic range used for Wilms tumors in standard adult patients (body weight average = 60 kg). Overall, our findings demonstrate that inhibiting transcription of CTG expansions is a promising approach for potential treatment of DM1. It will be interesting to determine if ActD can inhibit transcription of other GC-rich toxic RNAs such as CCUG repeats that cause DM2, CGG repeats that cause FXTAS and GGGGCC repeats implicated in causing ALS (Mohan et al., 2014).

We also observed a significant decrease of cellular foci in our HeLa-based DM1 model (Figure 2C). However, we interpreted these data cautiously because of the noted distinction in how ActD reduces CUG RNA in HeLa versus DM1 cellular systems. It is not clear why the reduction of CUG repeats by ActD reaches a plateau and modestly increases at higher ActD concentrations in the DM1 patient cells. Another small molecule could be added to target the remaining CUG repeats to determine if a cocktail approach could have a significant impact on CUG repeat levels. These results combined with previous work showing that diamidines reduced levels of CUG repeat RNAs in DM1 cell and mouse models (Coonrod et al., 2013; Siboni et al., 2015), suggest that targeting transcription of expanded CUG repeats and other repeat expansions should be pursued to develop lead compounds for potential therapeutics for DM and other toxic RNA diseases.

Experimental Protocols

Isothermal Calorimetry Methods

DNA and RNA sequences were purchased from Integrated DNA Technologies. Actinomycin D was dissolved in ITC buffer Stock DNA, RNA, and ActD solutions were prepared by dissolving in ITC buffer (1x Phosphate Buffered Saline, 5 mM MgCl2). The oligomer samples were then heated to 90 degrees Celsius for 5 minutes followed by cooling on ice. ITC measurements were carried out on a MicroCal titration calorimeter (GE Healthcare) at 20 degrees Celsius. The solution in the cell (the nucleic acid) was stirred at 293 rpm by the syringe during the entire course of the experiment and the reference power of the instrument was held at 25 µCal/s. ActD was titrated in 8 µL injections with a wait time of 300 sec between injections. A smaller initial injection of 4 µL was made after the initial wait time of 60 s to guarantee that only fully-concentrated ActD was injected in subsequent injections. All ITC data were analyzed using Origin 7.0 software supplied by MicroCal. A non-linear least squares fit algorithm was used with a two site, sequential binding model. One site and two independent sites models were also tested but did not fit the data as well as the sequential model.

Northern Blot Analysis

Tissue culture methods were performed as previously described (Coonrod et al., 2013). A total of 1 µg of cellular RNA was loaded onto a 1% formaldehyde denaturing agarose gel with 1x MOPS buffer (20mM MOPS, 8 nM NaOAc, 1 mM EDTA) in Northern Sample Buffer (67% formamide, 21% formaldehyde solution, 1x MOPS buffer). The gel was run in 1x MOPS buffer at 100 V for 4.5 hours. Gel was transferred to a 0.45 µm nylon Magna membrane (GE) and the cross-linked using a UV Stratalinker (Stratagene) on the optimal cross-link setting. Membranes were probed at 55°C overnight using radiolabeled oligos: 5’- (CAG)2TCGAG(CAG)4 for CUG repeats in CUG960 and 5’- TCCACCACCCTGTTGCTGTAGCCAAATTCG for GAPDH. For DM1 patient-derived cells, the following radiolabeled oligo was used: 5’- (CAG)7. Blots were quantified using ImageQuant (Molecular Dynamics). The relative levels of RNA were calculated by first normalizing lanes within the same gel using the GAPDH signal, and then gels were normalized by setting the levels of repeats in the untreated cells to 1. Error was determined by calculating the standard deviation of triplicate data.

Fluorescent Microscopy

Imaging of foci was performed as previously described (Warf et al., 2009). Briefly, HeLa cells were plated in 6-well plates onto coverslips and transfected with 500 ng of CUG960 plasmid for each experiment. After transfection, ActD was added in the indicated concentrations and the cells were fixed with 4% PFA 16 hours later. Cells were permeabilized, then probed overnight at 37 °C with a Cy3 CAG10 probe (IDT, IA). The next night, cells were probed overnight with the MB1a anti-MBNL1 antibody (1:5000 dilution, MDA Monoclonal Antibody Resource) in 1× PBS. Cells were washed with 1× PBS and incubated with donkey anti-mouse Alexa 488 (1:500 dilution). After being mounted onto glass slides using hard-set mounting media that contains DAPI (Vectashield), cells were imaged on an Olympus Fluoview FV1000 with a Bx61 scope. Nuclear foci for each cell were quantified using ImageJ. The number of cells scored was 35 for the CUG960 transfection, and 33 cells were scored when cells were treated with ActD. F test analyses were performed to determine if the data sets possessed equal or unequal variances. Appropriate T-test analyses were then performed using Excel.

ActD Treatment of Mice

HSALR transgenic mice in line 20b (FVB inbred background) were previously described (Mankodi et al., 2000). Gender-matched homogeneous HSALR mice of 6–7 weeks age were treated with ActD at the indicated dose by daily intraperitoneal injection for 5 days. Control group received PBS with 0.5% DMSO. Mice were sacrificed 1 day after the final injection and vastus lateralis muscles (quadriceps) were obtained for splicing analysis. RNA extraction, cDNA preparation, PCR amplification, and analyses on agarose gels using a fluorimager were all performed as previously described (Warf et al., 2009).

qRT-PCR

Quantitative real time PCR was performed using TaqMan Gene Expression assays on an ABI PRISM 7900HT Sequence Detection System (Applied Biosystems) and analyzed using the Quantitative- Comparative (CT) method. The levels of human skeletal actin (HSA) mRNA and mouse endogenous DMPK mRNA were normalized to Gtf2b mRNA and displayed graphically as relative mRNA levels.

RNA-seq

Whole-tissue RNA was TRIzol extracted from quadricep muscle of two age-matched mice treated with either no ActD (PBS), 0.125 mg kg−1, or 0.25 mg kg−1 actinomycin D. For each sample, 10 µg of RNA were prepped, depleted of ribosomal RNA (rRNA) using a Ribo-Zero™ rRNA removal kit (Epicentre), and used to generate sequencing libraries using the ScriptSeq™ v2 kit (Epicentre). After PCR amplification, each of the six libraries was combined in equimolar amounts to a total of 20 nM and submitted for single-end, 75-base pair sequencing on the Illumina Next-Seq 500 massively parallel sequencer at the University of Oregon Genomics Core Facility. Illumina RNA-sequencing reads were aligned against the mm10 mouse reference genome with GSNAP, using a splice sites map file that was generated from Ensemble mm10 gene models (GRCm38.80) and allowing for splicing detection (Wu et al., 2010) (http://researchpub.gene.com/gmap/). Between 48.5 and 98 million unique reads were acquired for each sample, of which ~84% mapped to the mouse genome. Gene read counts were generated with HT-Seq (Anders et al., 2015) using the Ensemble mm10 (GRCm38.80) gene model annotations and used as the input to DESeq2. Differential expression analysis was performed with DESeq2 (Love et al., 2014) using a FDR = 0.1 as a cutoff for statistical significance. Isoform abundances were estimated for each HSALR sample and compared to the pooled average of 4 wild-type samples (Wang et al., 2012) using MISO version 0.5.3 (Katz et al., 2010) and the mm10 alternative splicing event annotations provided by the developers of MISO (http://miso.readthedocs.org/en/fastmiso/). Functional annotations were obtained using DAVID bioinformatics resources (Huang et al., 2009a, Huang et al., 2009b).

Supplementary Material

1
2
3
4

Acknowledgements

This work was supported by NIAMS/NIH grants AR0599833 and F32AR063565. We appreciate helpful advice and suggestions from members of the Berglund lab and Eric Wang. The RNA seq data are available through NCBI SRA under accession number SRP064429.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Author Contributions

R.B.S., M.N. and J. A. B conceived the project, analyzed results, and wrote the manuscript. R.B.S characterized ActD treatment in cellular systems, established treatment ranges, and prepared the RNA-seq library. M.N. characterized ActD treatments in mouse systems and collected samples for RNA-seq analysis. S.D.W., M.K.T., and A.J.S. performed RNA-seq bioinformatics analysis. L.A.S performed Northern blots on DM1 cell lines. S.A.H. and D.M.C. performed isothermal calorimetry analyses.

References

  1. Anders S, Pyl PT, Huber W. HTSeq A Python framework to work with high-throughput sequencing data. Bioinformatics. 2015;31:166–169. doi: 10.1093/bioinformatics/btu638. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Arambula JF, Ramisetty SR, Baranger AM, Zimmerman SC. A simple ligand that selectively targets CUG trinucleotide repeats and inhibits MBNL protein binding. Proc. Natl. Acad. Sci. U.S.A. 2009;106:16068–16073. doi: 10.1073/pnas.0901824106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bensaude O. Inhibiting eukaryotic transcription: Which compound to choose? How to evaluate its activity? Transcription. 2011;2:103–108. doi: 10.4161/trns.2.3.16172. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Chaouch S, Mouly V, Goyenvalle A, Vulin A, Mamchaoui K, Negroni E, Di Santo J, Butler-Browne G, Torrente Y, Garcia L, et al. Immortalized skin fibroblasts expressing conditional MyoD as a renewable and reliable source of converted human muscle cells to assess therapeutic strategies for muscular dystrophies: validation of an exon-skipping approach to restore dystrophin in Duchenne muscular dystrophy cells. Hum Gene Ther. 2009;20:784–790. doi: 10.1089/hum.2008.163. [DOI] [PubMed] [Google Scholar]
  5. Chi LM, Lam SL. Structural roles of CTG repeats in slippage expansion during DNA replication. Nucleic Acids Res. 2005;33(5):1604–1617. doi: 10.1093/nar/gki307. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cho DH, Tapscott SJ. Myotonic dystrophy: Emerging mechanisms for DM1 and DM2. Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease. 2007;1772:195–204. doi: 10.1016/j.bbadis.2006.05.013. [DOI] [PubMed] [Google Scholar]
  7. Coonrod LA, Nakamori M, Wang W, Carrell S, Hilton CL, Bodner MJ, Siboni RB, Docter AG, Haley MM, Thornton CA, Berglund JA. Reducing levels of toxic RNA with small molecules. ACS Chem Biol. 2013;8(11):2528–2537. doi: 10.1021/cb400431f. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Echeverria GV, Cooper TA. RNA-binding proteins in microsatellite expansion disorders: Mediators of RNA toxicity. Brain Res. 2012;1462:100–111. doi: 10.1016/j.brainres.2012.02.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Fardaei M, Rogers MT, Thorpe HM, Larkin K, Hamshere MG, Harper PS, Brook JD. Three proteins, MBNL, MBLL and MBXL, co-localize in vivo with nuclear foci of expanded-repeat transcripts in DM1 and DM2 cells. Hum. Mol. Genet. 2002;11:805–814. doi: 10.1093/hmg/11.7.805. [DOI] [PubMed] [Google Scholar]
  10. Freireich EJ, Gehan EA, Rall DP, Schmidt LH, Skipper HE. Quantitative comparison of toxicity of anticancer agents in mouse, rat, hamster, dog, monkey, and man. Cancer Chemother Rep. 1966;14(4):219–244. [PubMed] [Google Scholar]
  11. Harley HG, Brook JD, Rundle SA, Crow S, Reardon W, Buckler AJ, Harper PS, Housman DE, Shaw DJ. Expansion of an unstable DNA region and phenotypic variation in myotonic dystrophy. Nature. 1992;355(6360):545–546. doi: 10.1038/355545a0. [DOI] [PubMed] [Google Scholar]
  12. Hill CR, Cole M, Errington J, Malik G, Boddy AV, Veal GJ. Characterisation of the Clinical Pharmacokinetics of Actinomycin D and the Influence of ABCB1 Pharmacogenetic Variation on Actinomycin D Disposition in Children with Cancer. Clin Pharmacokinet. 2014;53(8):741–751. doi: 10.1007/s40262-014-0153-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Ho TH, Charlet BN, Poulos MG, Singh G, Swanson MS, Cooper TA. Muscleblind proteins regulate alternative splicing. EMBO J. 2004;23:3103–3112. doi: 10.1038/sj.emboj.7600300. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Ho TH, Savkur RS, Poulos MG, Mancini MA, Swanson MS, Cooper TA. Colocalization of muscleblind with RNA foci is separable from mis-regulation of alternative splicing in myotonic dystrophy. Journal of Cell Science. 2005;118(13):2923–2933. doi: 10.1242/jcs.02404. [DOI] [PubMed] [Google Scholar]
  15. Holt I, Mittal S, Furling D, Butler-Browne GS, Brook JD, Morris GE. Defective mRNA in myotonic dystrophy accumulates at the periphery of nuclear splicing speckles. Genes to Cells. 2007;12:1035–1048. doi: 10.1111/j.1365-2443.2007.01112.x. [DOI] [PubMed] [Google Scholar]
  16. Hou MH, Robinson H, Gao YG, Wang AHJ. Crystal structure of actinomycin D bound to the CTG triplet repeat sequences linked to neurological diseases. Nucleic Acids Res. 2002;30:4910–4917. doi: 10.1093/nar/gkf619. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Huang DW, Sherman BT, Lempicki RA. Systematic and integrative analysis of large gene lists using DAVID Bioinformatics Resources. Nature Protoc. 2009;4(1):44–57. doi: 10.1038/nprot.2008.211. [DOI] [PubMed] [Google Scholar]
  18. Huang DW, Sherman BT, Lempicki RA. Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res. 2009;37(1):1–13. doi: 10.1093/nar/gkn923. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kamitori S, Takusagawa F. Crystal structure of the 2:1 complex between d(GAAGCTTC) and the anticancer drug actinomycin D. J. Mol. Biol. 1992;225:445–456. doi: 10.1016/0022-2836(92)90931-9. [DOI] [PubMed] [Google Scholar]
  20. Katz Y, Wang ET, Airoldi EM, Burge CB. Analysis and design of RNA sequencing experiments for identifying isoform regulation. Nat Methods. 2010;7:1009–1015. doi: 10.1038/nmeth.1528. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kimura T, Nakamori M, Lueck JD, Pouliquin P, Aoike F, Fujimura H, Dirksen RT, Takahashi MP, Dulhunty AF, Sakoda S. Altered mRNA splicing of the skeletal muscle ryanodine receptor and sarcoplasmic/endoplasmic reticulum Ca2+ATPase in myotonic dystrophy type 1. Hum Mol Genet. 2005;14:2189–2200. doi: 10.1093/hmg/ddi223. [DOI] [PubMed] [Google Scholar]
  22. Konieczny P, Stepniak-Konieczna E, Sobczak K. MBNL proteins and their target RNAs, interaction and splicing regulation. Nucleic Acids Res. 2015;1;42(17):10873–10887. doi: 10.1093/nar/gku767. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lee JE, Bennett CF, Cooper TA. RNase H-mediated degradation of toxic RNA in myotonic dystrophy type 1. Proc. Natl. Acad. Sci. U.S.A. 2012;109:4221–4226. doi: 10.1073/pnas.1117019109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Liu C, Chen FM. Actinomycin D binds strongly and dissociates slowly at the dGpdC site with flanking T/T mismatches. Biochemistry. 1996;35:16346–16353. doi: 10.1021/bi961060d. [DOI] [PubMed] [Google Scholar]
  25. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-Seq data with DESeq2. Genome Biol. 2014;15(12):550. doi: 10.1186/s13059-014-0550-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Mahadevan M, Tsilfidis C, Sabourin L, Shutler G, Amemiya C, Jansen G, et al. Myotonic dystrophy mutation: an unstable CTG repeat in the 3’ untranslated region of the gene. Science. 1992;255:1253–1255. doi: 10.1126/science.1546325. [DOI] [PubMed] [Google Scholar]
  27. Mankodi A, Takahashi MP, Jiang H, Beck CL, Bowers WJ, Moxley RT, Cannon SC, Thornton CA. Expanded CUG repeats trigger aberrant splicing of ClC-1 chloride channel pre-mRNA and hyperexcitability of skeletal muscle in myotonic dystrophy. Molecular Cell. 2002;10:35–44. doi: 10.1016/s1097-2765(02)00563-4. [DOI] [PubMed] [Google Scholar]
  28. Mankodi A, Logigian E, Callahan L, McClain C, White R, Henderson D, Krym M, Thornton CA. Myotonic dystrophy in transgenic mice expressing an expanded CUG repeat. Science. 2000;289(5485):1769–1773. doi: 10.1126/science.289.5485.1769. [DOI] [PubMed] [Google Scholar]
  29. Miller JW, Urbanati CR, Teng-Umnuay P, Stenberg MG, Byrne BJ, Thornton CA, Swanson MS. Recruitment of human muscleblind proteins to (CUG)(n) expansions associated with myotonic dystrophy. EMBO J. 2000;19:4439–4448. doi: 10.1093/emboj/19.17.4439. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Mohan A, Goodwin M, Swanson MS. RNA-protein interactions in unstable microsatellite diseases. Brain Res. 2014;1584:3–14. doi: 10.1016/j.brainres.2014.03.039. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Müller W, Crothers DM. Studies of the binding of actinomycin and related compounds to DNA. J Mol Biol. 1968;35(2):251–290. doi: 10.1016/s0022-2836(68)80024-5. [DOI] [PubMed] [Google Scholar]
  32. Nakamori M, Gourdon G, Thornton CA. Stabilization of expanded (CTG)•(CAG) repeats by antisense oligonucleotides. Mol. Ther. 2011;19:2222–2227. doi: 10.1038/mt.2011.191. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. O'Rourke JR, Swanson MS. Mechanisms of RNA-mediated Disease. J Biol Chem. 2009;284:7419–7423. doi: 10.1074/jbc.R800025200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Osborne RJ, Thornton CA. RNA-dominant diseases. Human molecular genetics. 2006;15(Spec No 2):R162–R169. doi: 10.1093/hmg/ddl181. [DOI] [PubMed] [Google Scholar]
  35. Osborne RJ, Filiaci V, Schink JC, Mannel RS, Alvarez Secord A, Kelley JL, Provencher D, Scott Miller D, Covens AL, Lage JM. Phase III trial of weekly methotrexate or pulsed dactinomycin for low-risk gestational trophoblastic neoplasia: a gynecologic oncology group study. J Clin Oncol. 2011;29:825–831. doi: 10.1200/JCO.2010.30.4386. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Paramanathan T, Vladescu I, McCauley MJ, Rouzina I, Williams MC. Force spectroscopy reveals the DNA structural dynamics that govern the slow binding of Actinomycin D. Nucleic Acids Res. 2012;40:4925–4932. doi: 10.1093/nar/gks069. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Perry RP, Kelley DE. Inhibition of RNA synthesis by actinomycin D: characteristic dose-response of different RNA species. J Cell Physiol. 1970;76:127–139. doi: 10.1002/jcp.1040760202. [DOI] [PubMed] [Google Scholar]
  38. Petruska J, Arnheim N, Goodman MF. Stability of intrastrand hairpin structures formed by the CAG/CTG class of DNA triplet repeats associated with neurological diseases. Nucleic Acids Res. 1996;24:1992–1998. doi: 10.1093/nar/24.11.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Philips AV, Timchenko LT, Cooper TA. Disruption of splicing regulated by a CUG-binding protein in myotonic dystrophy. Science. 1998;280:737–741. doi: 10.1126/science.280.5364.737. [DOI] [PubMed] [Google Scholar]
  40. Ranum LP, Cooper TA. RNA-mediated neuromuscular disorders. Annu Rev Neurosci. 2006;29:259–277. doi: 10.1146/annurev.neuro.29.051605.113014. [DOI] [PubMed] [Google Scholar]
  41. Reagan-Shaw S, Nihal M, Ahmad N. Dose translation from animal to human studies revisited. Faseb J. 2008;22:659–661. doi: 10.1096/fj.07-9574LSF. [DOI] [PubMed] [Google Scholar]
  42. Reuter JS, Matthews DH. RNAstructure: software for RNA secondary structure prediction and analysis. BMC Bioinformatics. 2010;11:129. doi: 10.1186/1471-2105-11-129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Savkur RS, Philips AV, Cooper TA. Aberrant regulation of insulin receptor alternative splicing is associated with insulin resistance in myotonic dystrophy. Nat Genet. 2001;29:40–47. doi: 10.1038/ng704. [DOI] [PubMed] [Google Scholar]
  44. Waksman SA, Woodruff HB. Bacteriostatic and bacteriocidal substances produced by soil actinomycetes. Proc Soc Exper Biol. 1940;45:609–614. [Google Scholar]
  45. Wang ET, Cody NA, Jog S, Biancolella M, Wang TT, Treacy DJ, Luo S, Schroth GP, Housman DE, Reddy S, Lécuyer E, Burge CB. Transcriptome-wide regulation of Pre-mRNA splicing and mRNA localization by muscleblind proteins. Cell. 2012;15:710–724. doi: 10.1016/j.cell.2012.06.041. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Warf MB, Nakamori M, Matthys CM, Thornton CA, Berglund JA. Pentamidine reverses the splicing defects associated with myotonic dystrophy. Proc. Natl. Acad. Sci. U.S.A. 2009;106:18551–18556. doi: 10.1073/pnas.0903234106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Wheeler TM, Leger AJ, Pandey SK, MacLeod AR, Nakamori M, Cheng SH, Wentworth BM, Bennett CF, Thornton CA. Targeting nuclear RNA for in vivo correction of myotonic dystrophy. Nature. 2012;488:111–115. doi: 10.1038/nature11362. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Wu TD, Nacu S. Fast and SNP-tolerant detection of complex variants and splicing in short reads. Bioinformatics. 2010;26:873–881. doi: 10.1093/bioinformatics/btq057. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

1
NIHMS737970-supplement-1.pdf (180.7KB, pdf)
2
NIHMS737970-supplement-2.xlsx (210.3KB, xlsx)
3
NIHMS737970-supplement-3.xlsx (157.4KB, xlsx)
4
NIHMS737970-supplement-4.xlsx (389.2KB, xlsx)

RESOURCES