Natural Compound Boldine Lessens Myotonic Dystrophy Type 1 Phenotypes in DM1 Drosophila Models, Patient-Derived Cell Lines, and HSALR Mice

Abstract

Myotonic dystrophy type 1 (DM1) is a complex rare disorder characterized by progressive muscle dysfunction, involving weakness, myotonia, and wasting, but also exhibiting additional clinical signs in multiple organs and systems. Central dysregulation, caused by an expansion of a CTG trinucleotide repeat in the DMPK gene's 3' UTR, has led to exploring various therapeutic approaches in recent years, a few of which are currently under clinical trial. However, no effective disease-modifying treatments are available yet. In this study, we demonstrate that treatments with boldine, a natural alkaloid identified in a large-scale Drosophila-based pharmacological screening, was able to modify disease phenotypes in several DM1 models. The most significant effects include consistent reduction in nuclear RNA foci, a dynamic molecular hallmark of the disease, and noteworthy anti-myotonic activity. These results position boldine as an attractive new candidate for therapy development in DM1.

Keywords: Drosophila; boldine; drug development; myotonic dystrophy; natural small molecule; patient-derived cells; rare disease.

Figure 1

Phenanthrene-based molecules. (a) Common structure and presence of phenanthrene in alkaloid drugs (morphine, codeine). (b) Chemical structure of boldine. Similar regions are highlighted with a red dotted line. Created with BioRender.com (accessed on 24 May 2023).

Figure 2

Evaluation of phenotypes after treatment with boldine in DM1 Drosophila models. (a) In situ hybridization of cross-sections of the musculature of flies treated with DMSO (0.25%) or boldine (12.5 µM). Cell nuclei were labelled in blue, while the CAG probe labelled the foci in red. White arrows indicate examples of merging red and blue signals. Boldine decreases the number of nuclei with foci. *** p-value < 0.001 calculated with an unpaired Student’s t-test using the program GraphPad Version 8.4.2. The error bars correspond to the standard error. (b) Longevity curves representing the percentage of the surviving flies as a function of time in days. Representation from the survival curves of the MHC-Gal4>UAS-i(CTG)480 flies (pink) treated with boldine (50 µM), or with (blue) DMSO (0.25%) and yw flies (green) treated with DMSO (control). *** p value < 0.001, ** p-value < 0.01 calculated with the Kaplan–Meier test using the GraphPad program. 40× optical microscope magnification. Created with BioRender.com (accessed on 24 May 2023).

Figure 3

Evaluation of DM1 phenotypes after treatment with boldine in cells derived from patients. (a) Fluorescent in situ hybridization showing the ribonuclear foci in fibroblasts differentiated to myoblasts derived from patients treated for 24 h with 100 μM boldine or with DMSO (1%). Nuclei are shown marked with DAPI in blue, while the foci are shown marked by the CAG probe in red. No foci were found in the myoblasts of healthy individuals (no DM1). Examples of foci detection (white arrows) in DM1 cells without treatment. Quantification of the number of foci contained in the nucleus of patient-derived myoblasts treated with DMSO or boldine, *** p-value < 0.001. Quantification of the number of cells without foci after the treatment of patient-derived fibroblasts with DMSO or boldine * p-value < 0.01. Quantifications were carried out by counting at least 100 independent nuclei (25 nuclei in 2 fields) and four biological replicates. p-value calculated with an unpaired Student’s t-test using GrapPhad. The error bars correspond to the standard error. (b) Detection of MBNL1 through immunohistochemistry using an anti-MBNL1 antibody in myoblasts derived from patients treated for 24 h with boldine at 100 μM or with DMSO (1%) and in myoblasts derived from healthy individuals (no DM1). Nuclei are shown marked with DAPI in blue, while MBNL1 is shown labelled by a specific antibody in green. (b) Nuclear distribution of MBNL1 quantified as pixels present in the two channels (blue and green) per nucleus area, using confocal microscopy images and the Image J program (NIH). Percentage of free MBNL1 normalized with respect to the mean distribution of MBNL1 in non-DM1 cells. The study was carried out by counting at least 150 nuclei. No statistically significant differences were detected with Student’s t-test using the GrapPhad. The error bars correspond to the standard error. 63× optical microscope magnification. Created with BioRender.com (accessed on 24 May 2023).

Figure 4

Intramuscular myotonia quantification after boldine and mexiletine in a HSA^LR mouse model. (a) To assess the efficacy of boldine, we injected different concentrations of boldine (60 mg/kg, 50 mg/kg, 25 mg/kg, 12.5 mg/kg, and 6.25 mg/kg) into the left quadriceps of the hind paw of mice between four and six weeks of age (n = 3 to 7). The right quadriceps served as an internal control and received DMSO (20%). Mice were treated with the compound or DMSO for three consecutive days and were sacrificed two days after the last dose. Myotonia was evaluated before treatment and the day of sacrifice. (b) Additionally, we compared boldine vs. mexiletine anti-myotonic activity. Top: myotonia levels in HSA^LR mice 2 days after intramuscular treatment with mexiletine (n = 4) or boldine (n = 4) at 60 mg/km in left quadriceps (LQ) or with the vehicle (DMSO 20%) in right quadriceps (RQ). The same experimental procedure in (a). Bottom: values of myotonia of the left quadriceps of HSA^LR mice treated with boldine or mexiletine measured for 15 days after treatment. The degree of myotonia is represented on a scale, where 0 means the absence of myotonia and 3 means the manifestation of myotonia in 90% or more of the times analyzed. The error bars represent the standard error. Created with GraphPad and BioRender.com (accessed on 24 May 2023).

Figure 5

Evaluation of boldine and mexiletine activity after chemically induced myotonia. TRR values in mouse control at 30, 60, 90, 120, 180, and 240 min, after the injection of 9-AC (60 mg/kg) together with the DMSO vehicle (20%) (Pink curves), and in mice treated with 9-AC (intraperitoneal, 60 mg/kg) and boldine (intragastric, 10 mg/kg) (blue curve) or mexiletine (intragastric, 10 mg/kg) (green curve). For both, a TRR reduction was observed. Every curve shows the mean plus standard deviation of the TRR of 7–9 mice. Created with GraphPad and BioRender.com (accessed on 24 May 2023).

Figure 6

RNA–molecule binding assays. (a) Gel retardation of RNA formed by 4-CUG repeats labeled with FAM is not observed, either retained in the gel wells or after the quantification of free fluorescent RNA after an addition of boldine at different concentrations. The error bars correspond to the standard error. (b) Adding pentamidine (positive control) to the highest concentrations increased FAM CUG23 polarization caused by the union between both molecules. Adding boldine failed to result in an increase in FAM-CUG23 polarization compared to the negative control (DMSO) in any of the concentrations used. *** p-value < 0.001 calculated with an unpaired Student’s t-test using the program GraphPad. The error bars correspond to standard error. Created with BioRender.com (accessed on 24 May 2023).