Genome Therapy of Myotonic Dystrophy Type 1 iPS Cells for Development of Autologous Stem Cell Therapy

Abstract

Myotonic dystrophy type 1 (DM1) is caused by expanded Cytosine-Thymine-Guanine (CTG) repeats in the 3'-untranslated region (3' UTR) of the Dystrophia myotonica protein kinase (DMPK) gene, for which there is no effective therapy. The objective of this study is to develop genome therapy in human DM1 induced pluripotent stem (iPS) cells to eliminate mutant transcripts and reverse the phenotypes for developing autologous stem cell therapy. The general approach involves targeted insertion of polyA signals (PASs) upstream of DMPK CTG repeats, which will lead to premature termination of transcription and elimination of toxic mutant transcripts. Insertion of PASs was mediated by homologous recombination triggered by site-specific transcription activator-like effector nuclease (TALEN)-induced double-strand break. We found genome-treated DM1 iPS cells continue to maintain pluripotency. The insertion of PASs led to elimination of mutant transcripts and complete disappearance of nuclear RNA foci and reversal of aberrant splicing in linear-differentiated neural stem cells, cardiomyocytes, and teratoma tissues. In conclusion, genome therapy by insertion of PASs upstream of the expanded DMPK CTG repeats prevented the production of toxic mutant transcripts and reversal of phenotypes in DM1 iPS cells and their progeny. These genetically-treated iPS cells will have broad clinical application in developing autologous stem cell therapy for DM1.

Figure 1

Loss of nuclear RNA foci in genome-treated DM1 induced pluripotent stem (iPS) cell clones. (a) A typical Puromycin and Ganciclovir-resistant clone (phase contrast image). (b) Parental DM-03 iPS cells with nuclear RNA foci. (c, d) Nuclear foci were not detectable in the Puromycin and Ganciclovir-resistant clones of DM-03 iPS cells.

Figure 2

Exogenous polyA signals (PASs) were integrated in the designed transcription activator-like effector nuclease (TALEN) targeting site and were transcribed contiguous with Dystrophia myotonica protein kinase (DMPK) transcription. (a) Genomic polymerase chain reaction (PCR) analysis showing the integration of the insertion cassette into the TALEN targeting site. Primer pair I9SF/TSR amplified the whole insertion cassette, detectable in Puromycin and Ganciclovir-resistant clones and donor vector but not in parental induced pluripotent stem (iPS) cells. Primer pair IVF2/IVR amplified a portion of the insertion cassette and a portion of genomic DNA downstream of the end of the 3′ homology arm. VF/VR amplified the TK resistant gene in donor backbone. CTGSF/CTGSR amplified genomic DNA as a template loading control. (b) RT-PCR showing the effect of the integration on DMPK gene transcription. Products from primer pair E8F3/PGKR1 were detected in the two foci-negative clones but not in the parental cells. Products from E8F3/E9R1 suggested upstream mRNA was intact in all of the clones. Products from E8F2/E10R2, which spans exon 8, 9, 10, and long introns, showed normal DMPK transcription in parental cells and clone 13–3 and 33–4, indicating that the normal allele was unaffected. GAPDH was amplified as a reverse transcription control. (c) Schematic overview of primer location. (d). Southern blot digested by EcoRI demonstrated insertion of the polyA signal (PAS) cassette in the mutant allele by showing the disappearance of the corresponding expanded band. This is due to the introduction of an extra EcoRI site within the PAS cassette. The normal allele remains intact. There can be two different normal allele sizes after EcoRI digestion, one ~8.6 kb and the other ~9.6 kb. Sample Negative PBL has both alleles, and sample Positive PBL has only the 9.6 kb allele and then the expansion allele. DM1-03 subject has the 9.6 kb allele and the expansion allele. (e) Southern blot digested by NcoI showed the expanded CTG repeats in DM1 parental and genome treated iPS clones (from 1793 repeats to 2087 repeats). Positive PBL is genomic DNA from peripheral blood from a DM1 expansion positive patient with ~750 CTG repeats. Negative PBL is a control peripheral blood DNA from a normal subject with two normal allele sizes.

Figure 3

Pluripotency assay by teratoma formation in parental Myotonic dystrophy type 1 (DM1) induced pluripotent stem (iPS) cells (DM-03) and genome-treated iPS cells. Both parental DM1 iPS cells and genome-treated iPS cells formed teratomas (upper panel) with typical three germ layer tissue structures on hematoxylin and eosin staining (middle panel). However, nuclear RNA foci were only seen in tissue derived from parental DM1 iPS cells but not in that from genome-treated iPS cells (lower panel). DM-03: parental DM 1iPS cells. 13-3: genome-treated iPS cells.

Figure 4

The integration of the polyA signals upstream of DMPK CTG repeats in DM-03 induced pluripotent stem (iPS) cells led to elimination of nuclei RNA foci in linear-differentiated neural cells. (Upper panel): nuclear RNA foci were only detected in neural stem cells (NSCs) derived from parental DM-03 iPS cells while no foci were detected in normal NSCs or NSCs derived from genome-treated iPS cells. Green: nestin. (Lower panel) nuclear RNA foci were only detected in spontaneous differentiated neurons derived from parental DM-03 iPS cells while no foci were detected in normal neurons or neurons derived from genome-treated iPS cells. Green: neurofilament H. Normal: neural cells from normal iPS cells. DM-03: neural cells from parental DM1 iPS cells. 13-3 and 33-4: neural cells from genome-treated iPS cells with polyA signals (PASs) integrated in DMPK intron 9.

Figure 5

Genome treatment reversed aberrant splicing of microtubule-associated protein tau (MAPT) and muscleblind-like splicing regulator (MBNL 1), MBNL 2 to normal patterns. (Upper panel) in early stage neural stem cells (NSCs) (passage 4 after passing from rosettes), transcripts of MAPT are mainly of the fetal isoform (exon 2, 3 exclusion) and there is no difference among NSCs from each of the induced pluripotent stem (iPS) clones. However, MBNL 1 and MBNL 2 already showed more adult isoform in normal and genome-treated NSCs even though the fetal isoform was still dominant (exon 7 exclusion ratio below 0.5), compared to parental Myotonic dystrophy type 1 (DM1) NSCs. (Lower panel) in late stage of NSCs (passage 10 after passing from rosette), most MAPT transcripts have switched to adult isoforms (exon 2 inclusion ratio over 0.5). MBNL1 and MBNL 2 have reverted to adult isoform dominant in normal and genome-treated NSCs (exon 7 exclusion ratio is >0.5) but parental NSCs continued to express mainly fetal isoform. *P < 0.05 by one-way analysis of variance.

Figure 6

Both parental and genome-treated induced pluripotent stem (iPS) cells underwent normal neuron and astrocyte differentiation. (a, b) Fourteen days after neuronal differentiation showing neural network. (c) A neuron derived from genome-treated iPS cells, showing neuronal marker (Neurofilament H) with normal neurite outgrowth. (d) Astrocytes from genome-treated iPS cells, showing astrocyte marker (GFAP).

Figure 7

Elimination of nuclear RNA foci and reversal of aberrant splicing in linear-differentiated cardiomyocytes derived from genome-treated induced pluripotent stem (iPS) cells. (Upper panel) (a) Cardiomyocyte from normal control iPS cells showing no nuclear RNA foci. (b) Cardiomyocyte from parental Myotonic dystrophy type 1 (DM1) iPS cells showing nuclear RNA foci. (c) Cardiomyocyte from genome-treated DM1 iPS cells showing disappearance of nuclear RNA foci. (d) Agarose gel images showing the reversal of aberrant splicing pattern in cardiac troponin T (cTNT), insulin receptor (INSR), muscleblind-like splicing regulator (MBNL)1 and MBNL2. (Lower Panel) Graphs showing the statistical differences of exon exclusion, in which exon 5 or 7 exclusion is adult isoform in cTNT, MBNL1 and MBNL2 while exon 10 exclusion is fetal isoform in INSR. *P < 0.05 by one-way analysis of variance.