A targeted antisense therapeutic approach for Hutchinson–Gilford progeria syndrome

Abstract

Hutchinson–Gilford progeria syndrome (HGPS) is a rare accelerated aging disorder characterized by premature death from myocardial infarction or stroke. It is caused by de novo single-nucleotide mutations in the LMNA gene that activate a cryptic splice donor site, resulting in the production of a toxic form of lamin A, which is termed progerin. Here we present a potential genetic therapeutic strategy that utilizes antisense peptide-conjugated phosphorodiamidate morpholino oligomers (PPMOs) to block pathogenic splicing of mutant transcripts. Of several candidates, PPMO SRP-2001 provided the most significant decrease in progerin transcripts in patient fibroblasts. Intravenous delivery of SRP-2001 to a transgenic mouse model of HGPS produced significant reduction of progerin transcripts in the aorta, a particularly critical target tissue in HGPS. Long-term continuous treatment with SRP-2001 yielded a 61.6% increase in lifespan and rescue of vascular smooth muscle cell loss in large arteries. These results provide a rationale for proceeding to human trials.

Lamins are type V intermediate filament proteins that consist of a fibrous network underlying the inner nuclear membrane and throughout the nucleoplasm¹. The major A-type lamins, lamins A and C, are generated as alternatively spliced products of the LMNA gene and differ in structure at the carboxyl terminus. After translation, prelamin A undergoes a series of posttranslational modifications before inclusion into the nuclear lamina, including farnesylation and trimming of the carboxyl terminal CAAX motif, methylation of the farnesylcysteine residue and proteolytic cleavage of the final 15 residues by the zinc metalloprotease ZMPSTE24. In forming the nuclear lamina, lamins interact with several membrane-associated proteins and cytoskeletal elements to participate in nuclear and cytoskeletal organization, mechanical stability, chromatin organization, gene regulation and genome stability^2,3.

Mutations in the LMNA gene cause a number of heritable disorders, collectively referred to as laminopathies, which range in phenotype from muscular dystrophy to premature aging³. Among these laminopathies, HGPS occurs with a prevalence of 1 in 20 million people⁴. Ninety percent of cases of HGPS (classic HGPS) are caused by a rare single-nucleotide mutation (c.1824C>T, p.G608G) that does not alter the coding sequence but instead activates a cryptic splice donor within exon 11 of LMNA^5,6. The resulting protein product, termed progerin, lacks the 50-residue region that contains the ZMPSTE24 cleavage site, resulting in the incorporation of a permanently farnesylated truncated protein within the nuclear lamina⁵. Progerin acts in a dominant negative manner by disrupting normal nuclear structure and function. The presence of progerin is associated with abnormal nuclear morphology, loss of peripheral heterochromatin, altered histone modifications, mechanical defects and increased DNA damage⁷. At the tissue level, individuals with HGPS develop growth deficiency, lipodystrophy, bone dysplasia, sclerotic dermis and atherosclerotic lesions within the vasculature, leading to mortality from heart attacks or strokes at an average age of 14.6 years^8,9.

Two therapeutic approaches to the treatment of HGPS have been investigated in clinical trials. The first seeks to prevent prenylation and, therefore, incorporation of progerin into the nuclear membrane through the use of the farnesyltransferase inhibitor lonafarnib. Despite notable gastrointestinal side effects⁹, prolonged treatment with lonafarnib resulted in decreased mortality rate of HGPS patients¹⁰. The second approach, currently the topic of a clinical trial with everolimus, utilizes mammalian target of rapamycin (mTOR) inhibitors to facilitate lysosomal degradation of intracellular progerin by activation of autophagy¹¹. The broader effects of mTOR inhibition result in immunosuppression. Thus, both of these therapeutics include side effects that could exacerbate a condition of failure to thrive¹². These limitations of current approaches to HGPS highlight the persistent need for precisely targeted gene therapeutic techniques. The strategy we present in this study utilizes antisense technology to specifically target progerin transcripts with PPMOs (Fig. 1a). These synthetic nucleotide analogs are capable of preventing translation through sequence-specific steric hindrance and can also modulate splicing by blocking spliceosome-pre-messenger RNA interactions, a strategy currently in use for the treatment of Duchenne muscular dystrophy^13,14. Thus, the potential to target and inhibit splicing and translation of LMNA transcripts containing the G608G mutation represents a new therapeutic tool for treating HGPS that exerts effects specifically on the pathological target without adversely affecting diverse cellular processes.

Fig. 1 |. Design and testing of a PPMO-based strategy for blocking the production of progerin.

a, Right, Normal LMNA transcript processing. Left: aberrant LMNA splicing. The HGPS G608G mutation position is indicated by the arrow at the junction of the retained exon 11 sequence (dark gray) and the 150-nt portion of exon 11 that is deleted (light gray) by splicing at the cryptic splice site. The antisense PPMO binds transcripts containing the mutation, thereby blocking cryptic splice site recognition by the spliceosome. b, Tiling of 11 alternatively designed PPMOs around the cryptic splice site generated by the LMNA G608G (c.1824C>T) mutation. The two controls included an oligonucleotide complementary to the normal LMNA sequence (wtLMNA, WT) and an oligo predicted to lack complementarity to any sequence in the human genome (scrambled control, CTRL). c, RT–qPCR results, normalized relative to TFRC RNA, of proband-derived fibroblasts cultured in the presence of 6 μM of alternative PPMOs for two weeks. Averages and s.d. were determined from triplicate reactions of biological triplicates. d, Immunoblots of triplicate lysates from proband-derived fibroblasts cultured in the presence of 6 μM of alternative PPMOs for two weeks. e, Quantitation of lamin A, progerin and lamin C proteins from western blot analysis, normalized relative to β-actin (ACTB). Values and error bars represent the mean ± s.d. of triplicate cultures. A, lamin A; P, progerin; C, lamin C; *P < 0.05, **P < 0.01,***P < 0.001 versus transcripts or protein from cells receiving saline.

In this study, we present the results of a preclinical study investigating the efficacy of PPMO-based RNA targeting in vitro in HGPS cell lines and then in vivo using a transgenic mouse model that expresses the human LMNA gene harboring the classic G608G mutation. That homozygous mouse model, C57BL/6-tg(LMNA*G608G) HClns/J, (referred to as LMNA^G/G) faithfully recapitulates key clinical characteristics of HGPS, including rapidly progressive vascular smooth muscle cell (VSMC) loss in large arteries, joint contractures, bone dysplasia, severe lipodystrophy, skin tightening and growth deficiency, which contribute to progressive deterioration and premature death by age 8 months¹⁵. We demonstrate that long-term intravenous treatment with the candidate PPMO, SRP-2001, extends the lifespan in these mice and is associated with preservation of VSMC in murine aortas.

Results

Optimal progerin splice inhibition in vitro with SRP-2001.

To identify the most specific and most effective antisense target to inhibit aberrant splicing at the LMNA c.1824C>T locus, we designed 14 phosphorodiamidate morpholino oligomers (PMOs), tiling across the exon 11 cryptic splice site in 5 nucleotide intervals. Treatment of the fibroblasts from a patient with HGPS (HGADFN167) with 80 μM of PMOs for 2 weeks, followed by quantitative PCR with reverse transcription (RT–qPCR) and western blot, identified two PMOs that produced a significant reduction of progerin transcripts (33 and 72%, respectively) and a reduction of progerin protein by 52 (P < 0.003) and 95% (P < 0.0001, data not shown). We then synthesized 11 PPMOs, tiling by single-nucleotide steps centered on the best of the PMOs (Fig. 1b and Supplementary Table 1). While PMOs are capable of uptake by cell lines in vitro, PPMOs with the peptide tag more readily penetrate cells in vivo allowing for intravenous administration¹⁶. Treatment of patient HGADFN167 in vitro with 6 μM of PPMOs for two weeks, followed by analysis of LMNA/C and progerin-specific transcript levels by digital-droplet qPCR with reverse transcription (ddRT–qPCR) relative to an internal control, the transferrin receptor (TFRC), identified one PPMO, designated SRP-2001, which produced the greatest reduction of progerin transcript levels (92%, P < 0.001) and an 83% reduction in normal LMNA transcripts (P < 0.001) compared to untreated HGPS fibroblasts (Fig. 1c and Extended Data Fig. 1). Of interest, the transcript levels of LMNC increased significantly with PPMO treatment, which is consistent with some influence on alternative splicing proximal to the PPMO binding sites. Regardless of whether normalization was done relative to TFRC, as a fraction of total A-type lamin transcripts, or relative to LMNC, SRP-2001 was the most potent inhibitor of progerin transcription of the PPMOs tested (Extended Data Fig. 2). Western blot analysis of cell lysates after PPMO exposure demonstrated that, relative to β-actin control, PPMOs with comparatively stronger transcriptional inhibition generated the largest reduction of progerin protein. The most significant reduction occurred when cells were treated with SRP2001 (P < 0.01; Fig. 1d,e). Increased lamin C protein levels were observed in treatments corresponding to increased LMNC transcription. Further analysis of protein expression relative to the expression of all A-type lamins and lamin C confirmed that SRP2001 treatment resulted in the greatest reduction of progerin and lamin A protein levels (Extended Data Fig. 3).

SRP-2001 also blocks progerin splicing in nonclassical progeroid laminopathies.

In nonclassic HGPS, progerin production is caused by LMNA coding or intron 11 mutations, wherein generation of progerin transcripts and protein occurs by activation of the classic HGPS cryptic splice site (G608S, c.1822G>A) or by weakening the normal donor site at the start of intron 11 (1968 + 1G>A, +2T>C and +5G>C), favoring splicing using the cryptic exon 11 sequence at G608. We compared the two-week SRP-2001 treatment of normal and classic HGPS fibroblasts at 6 μM and 12 μM in vitro with the fibroblasts from patients with nonclassic HGPS cell lines and demonstrated by ddRT–qPCR and western blot that, despite a nucleotide mismatch, SRP-2001 reduced progerin transcripts and protein significantly (P < 0.001) in all cases (Extended Data Figs. 4 and 5). Thus, it appears that SRP-2001 can be potentially beneficial for any mutation that induces splicing at the cryptic G608G locus.

SRP-2001 restores cellular proliferation and nuclear lamina protein expression.

We also found that SRP-2001 treatment of cell lines increased their proliferative capacity in culture (Fig. 2a). In a 7-d proliferation assay, normal control cells divided approximately 12 (P < 0.05), 33 (P < 0.001) and 56% (P < 0.001) faster, respectively when cultured in 6 μM of SRP-2001, compared to receiving saline or no treatment. Although the HGPS fibroblast population doubling time took two to three times longer than parental normal control cells under normal culture conditions, we found that in the presence of 6 μM of SRP-2001 the proliferation rate of these cells increased by 29 (p < 0.05), 47 (p < 0.05) and 110% (P < 0.01), respectively.

Fig. 2 |. Effect of SRP-2001 on cellular proliferation and B-type lamin protein expression in the fibroblasts of a patient with HGPS.

a, Treatment of three independent patients with classic HGPS and sex-matched parental control fibroblast cell lines with SRP-2001 improves proliferative capacity in culture. Cells received either no treatment, saline (vehicle) or 6 μM of SRP-2001, performed in triplicate. *P < 0.05, **P < 0.01, ***P < 0.001 versus vehicle control. b, Western blot analysis of A- and B-type lamins showing partial rescue of B-type lamins, particularly LMNB2, in HGPS fibroblasts treated with SRP-2001. LMNB1 and LMNB2 levels increased 80 (P = 0.11) and 100% (P < 0.05), respectively in HGPS fibroblasts at 12 μM of SRP-2001 versus cells receiving no treatment. Values and error bars represent the mean ± s.d. of triplicate cultures.

Before initiating testing in vivo, we sought to assess the extent of functional off-target effects of SRP-2001 treatment using normal control fibroblasts. We focused primarily on B-type lamins because of the nature of the interaction of these proteins in the nuclear lamina and because alterations in the nuclear lamina can affect nuclear morphology, gene expression and cellular proliferation^17,18. Quantitation of B-type lamin expression by western blot analyses of cell lysates demonstrated that both lamin B1 and lamin B2 were reduced by 75% in HGPS versus normal control cells in the absence of any treatment (P < 0.01 for both). Although no effect of treatment was seen in normal control cells, SRP-2001 increased lamin B2 levels as much as twofold, with no significant difference in lamin B1 levels, in a dose-dependent manner in HGPS cells (Fig. 2b).

Systemic delivery of SRP-2001 in the LMNA^G/G HGPS mouse model.

We next compared the in vitro results from the fibroblasts of our human patient with the SRP-2001 treatment of mouse fibroblasts generated from our homozygous progeria transgenic mouse model (LMNA^G/G). We treated HGADFN167 and LMNA^G/G fibroblasts in culture for two weeks with increasing doses of SRP2001 and evaluated progerin splicing and protein expression using a human LMNA/C/progerin-specific ddRT–qPCR assay (Fig. 3a,d) and human-specific lamin A/C and progerin western blots (Fig. 3b,c,e,f). In both human HGPS and LMNA^G/G mouse fibroblasts, we observed significant reduction of progerin transcript levels and reduction of progerin protein expression with an estimated half maximal effective concentration (EC₅₀) of approximately 3 μM.

Fig. 3 |. Determination of SRP-2001 efficacy in culture.

a–f, Fibroblasts derived from a proband with classic HGPS (a–c) and mice carrying two copies of the human mutant LMNA transgene (LMNA^G/G) (d–f) were treated for two weeks with SRP-2001 and analyzed for LMNA expression by qPCR and western immunoblot. For transcript quantitation, the averages and s.d. were determined from triplicate reactions of biological triplicates. The values for protein analyses were determined from three independent cultures. The EC₅₀ value was estimated at 3 μM based on intracellular progerin transcript (a) and protein (b,c) levels in HGPS cells and progerin transcript (d) and protein (e,f) levels in transgenic mouse cells. *P < 0.05; **P < 0.01; ***P < 0.001 versus the same transcripts or proteins in cells receiving saline only.

In preparation for in vivo studies with SRP-2001, we employed the EGFP-654 transgenic reporter mouse to investigate the ability of PPMOs to penetrate tissues relevant to HGPS pathology¹⁹. The EGFP transgenic mouse contains a minigene construct comprised of the EGFP cDNA interrupted by the insertion of the β-globin gene intron 2 under a strong promoter allowing ubiquitous expression. The intron is engineered with a mutation that promotes splicing of the minigene, inserting an intronic sequence into the mRNA and preventing the expression of the EGFP protein. Injection of the EGFP-654 antisense oligomer (Extended Data Table 1) into the mouse resulted in restoration of correct splicing of the EFGP pre-mRNA and EGFP protein production (Fig. 4a). We injected the PPMO-EGFP-654 antisense oligomer, conjugated with the same peptide as SRP-2001, by intravenous route at 40 mg kg⁻¹ per mouse 3 times per week and collected quadriceps, heart and aorta tissues on day 7 for RT–qPCR analysis and immunofluorescence histology. Intravenous control assays were included using the same EGFP-654 antisense target sequence with a previously studied B-peptide conjugate with known characteristics in this assay²⁰. The most serious pathology in children with HGPS occurs in the medial layer of large arteries^21–23, which is recapitulated in the LMNA^G/G mice (W. Cabral et al., manuscript submitted for publication). We performed immunofluorescence histochemical analysis on the aortae of PPMO-EGFP-654-treated mice to assess the penetration of the PPMO-EGFP moiety into the vascular media. Immunofluorescence staining by anti-green fluorescent protein (GFP) demonstrated immunoreactivity in the vascular media (Fig. 4b), heart, quadriceps and liver (data not shown). RT–qPCR analysis of the quadriceps and heart demonstrated near complete correction of the aberrant β-globin splicing. In the aorta, EGFP-654 administration resulted in partial correction of aberrant β-globin splicing compared to the untreated control (Supplementary Fig. 1). Thus, we demonstrated that PPMOs penetrate the essential vascular disease target tissues in HGPS.

Fig. 4 |. Efficient in vivo delivery of PPMO to murine vascular cells and tissues.

a, The EGFP-654 splicing assay demonstrates efficient in vivo delivery of PPMOs to murine vascular cells and tissues. The schematic of the EGFP-654 reporter system illustrates that the mutation at nucleotide 654 of the intron results in partial intronic retention in spliced mRNA and prevention of proper translation of EGFP. The targeted PPMO blocks the aberrant splice site and restores proper splicing allowing for EGFP expression. b, Immunohistochemical analysis of sectioned ascending aortas after intravenous injection of EGFP-654 PPMO into reporter mice demonstrates restored expression of EGFP in VSMCs. Saline served as the negative control. Two mice (one male, one female) were injected in a single experiment and eight sections per aorta were analyzed. c–e, Single-dose pharmacodynamic analysis of SRP-2001 was performed on female and male mice expressing two copies of the human LMNA transgene harboring the classic HGPS mutation, G608G (LMNA^G/G). The efficacy of intravenous drug delivery was determined by comparing progerin transcripts and PPMO levels in the aorta (c), heart (d) and kidney (e) after single intravenous doses of 60 mg kg⁻¹. Values and error bars represent the mean ± s.d. for n = 6 per treatment group.

We investigated the pharmacodynamics of administering SRP2001 in LMNA^G/G mice after intravenous injection at 60 mg kg⁻¹ to inform the periodicity of dosing in a preclinical trial. Six mice (three females/three males) injected at time zero were killed at time points from 5 min to 48 h, followed by immediate necropsy, analysis by ddRT–qPCR for progerin transcript levels and SRP-2001 concentration by mass spectrometry (Fig. 4c–e). Although variability was high, the average levels of progerin transcript expression were moderately decreasing in the aorta over the time course of treatment. In the heart, a modest decrease in the first 30 min recovered to initial levels over 48 h. There was no clear trend of progerin transcript levels in the kidney. Intravenous injections of SRP-2001 achieved maximal concentration at 5 min and dissipated over 4 h in the aorta and heart, while SRP-2001 accumulated over the following 8h in the kidney before dissipation.

SRP-2001 reduces VSMC loss and extends longevity in LMNA^G/G mice.

We next conducted a longevity trial using intravenous administration of 60mg kg⁻¹ twice per week. Twelve mice (6 females/6 males) were injected beginning at 3 weeks of age, matched with saline-injected control mice for the duration of the study. Mice were observed for health throughout the course of the study until established humane end points were evident. End points included severe weight loss of ≥15% in one week, lethargy, failure to thrive, or moribund state, as established in an approved animal care and use protocol. After necropsy, tissues were preserved for ddRT–qPCR, western blot analysis and histological analysis. Untreated LMNA^G/G mice did not gain weight as rapidly or extensively as their wild-type (WT) littermates. As they approached 7–8 months, they began to lose weight, presented with progressive kyphosis, joint contracture and reduced activity and died prematurely at an average age of 30 weeks (Fig. 5a; W. Cabral et al., manuscript submitted for publication). SRP-2001-treated mice at the same age presented with less severe kyphosis, were more flexible, attentive and active and more similar to WT C57BL/6 mice. While SRP-2001-treated mice did not improve in weight gain to WT levels, they kept the weight they gained throughout the duration of the longevity study (Fig. 5b). Survival analysis of the longevity study indicated that the saline control group survived an average of 213.5 d while the mice treated with SRP-2001 lived an average of 346.5 d achieving a 61.6% increase in longevity (Fig. 5c; P < 0.0001). There was a small difference in survival between male and female mice treated with SRP-2001 (325.5 versus 357 d, P < 0.04) but not significantly when compared to the difference in the saline control mice (199.5 versus 238 d, P < 0.10).

Fig. 5 |. Long-term treatment of HGPS mice with SRP-2001 significantly extends their lifespan.

LMNA^G/G mice received twice weekly treatments via intravenous tail injection at 60 mg kg⁻¹ beginning at 2 weeks of age. a, Left: 7.5-month-old LMNA^G/G mouse treated with SRP-2001. Right: 7.5-month-old LMNA^G/G mouse receiving saline (vehicle). b, Growth curves of LMNA^G/G mice treated with SRP-2001 or saline (vehicle) compared to C57BL/6 WT mice. Values and error bars represent the mean ± s.d. for n = 6 mice per treatment group. c, Kaplan–Meier plots for saline-treated controls (black line, n = 12) and SRP-2001-treated mice (blue line, n = 12) illustrate the 61.6% increase in lifespan of the treated group, from a median of 30.5 weeks to a median of 49.5 weeks (P < 0.0001).

Histological analysis of the ascending aorta in the saline-treated animals revealed loss of VSMCs that is typical of patients with HGPS and the HGPS mouse model^21,23,24. Treatment with SRP-2001 demonstrated the relative preservation of VSMCs with reduced proteoglycan accumulation (Fig. 6a). Further analysis revealed a greater than sevenfold increase of VSMC density in SRP-2001-treated mice versus matched saline controls (P < 0.001) and 53% decrease in the adventitial area (P < 0.01), indicating less vessel damage (Fig. 6b). Evaluation of VSMC density suggested a potential gradient where more VSMCs were present in the media closer to the vessel lumen than the adventitial side of the media. In SRP-2001-treated mice 71% of the VSMCs of the aortic medial layer were found in the luminal half compared to the adventitial half. In comparison, five-month-old untreated LMNA^G/G aortic media presented 49.9% of the VSMCs in the luminal half relative to the adventitial half (P < 0.08). Progerin transcript measurement in the ascending aorta resulted in a 76% reduction (P < 0.05) paired with a twofold increase in LMNA transcript levels (P < 0.01) (Fig. 6c and Extended Data Fig. 6a). In the heart, although progerin transcript levels were reduced by 62% with SRP-2001 treatment (P < 0.01), LMNA transcript levels were decreased by 56% (P < 0.05) (Fig. 6c and Extended Data Fig. 7a). Concurrent analysis of protein expression in the aorta revealed a 25% reduction of progerin overall (P < 0.05) when compared to total A-type lamins (Extended Data Fig. 6b,c). In the heart, lamin A, C and progerin proteins were reduced by 58, 80 and 84%, respectively relative to smooth muscle actin (P < 0.05; Extended Data Fig. 7b,c).

Fig. 6 |. Partial rescue of vascular tissue degradation in mice treated with SRP-2001.

End point analysis of vascular tissues of LMNA^G/G mice treated twice weekly until the end point at an average of 30.5 weeks with saline (vehicle) compared to the end point at an average of 49.5 weeks with 60 mg kg⁻¹ of SRP-2001. a, Representative images from Movat’s pentachrome-stained ascending aorta sections illustrate VSMC loss, elastin degradation, thickened medial layer and proteoglycan accumulation observed in LMNA^G/G mice receiving saline only. The aortas of mice treated with SRP-2001 contained less proteoglycan (blue) and retained VSMCs (red). H&E stain of the ascending aorta showing VSMC nuclei (blue) presence in SRP-2001-treated mice nearer the lumen of the tunica media. For each sample and stain, a total of four sections were imaged for analysis. b, Left: end point VSMC density within the tunica media of the ascending aortas from mice receiving PPMO or vehicle (saline) only. Right: adventitial area of ascending aortas at the end point. c, SRP-2001 reduces expression of progerin in vivo as determined by qPCR analysis of LMNA transgene expression in vascular tissue from mice receiving saline (vehicle) or SRP-2001. Values and error bars represent the mean ± s.d. for n = 12 mice per treatment group (6 males, 6 females); *P < 0.05, **P < 0.01.

SRP-2001 treatment did not result in the reduction of progerin transcript levels or protein expression in skin tissue. The nonsignificant trend toward reduction of progerin transcript levels was accompanied by a 53% (P < 0.05) decrease of LMNA transcripts and a mild (17%, P < 0.05) increase of LMNC transcript levels when quantitated as a fraction of A-type lamins (Supplementary Fig. 2). In the liver, long-term treatment with SRP-2001 resulted in significant reduction of progerin (51% decrease, P < 0.01) and LMNA (93% decrease, P < 0.05; Extended Data Fig. 8a) with a concomitant 41% increase (P < 0.001) of LMNC transcripts relative to total LMNA transcripts, which correlated with an 85% (P < 0.05) and 87% (P < 0.05) reduction of lamin A and progerin protein expression relative to the β-actin reference. Most significant in the liver was the 37% reduction of lamin A (P < 0.05), the 57% reduction of progerin (P < 0.001) and the 49% increase in lamin C (P < 0.001) protein levels relative to total A-type lamins (Extended Data Fig. 8b,c). Analysis of A-type lamin transcript levels in mesenchymal-derived tissues with SRP-2001 treatment revealed significant reduction of progerin transcript levels in quadriceps (decreased 80%, P < 0.001) and kidney (decreased 61%, P < 0.001) and a trend in progerin transcript levels in the bone (decreased 52%, P < 0.05) when compared as a fraction of all A-type lamins (Extended Data Fig. 9). We note that, in all tissues, treatment with SRP-2001 resulted in a greater reduction of progerin transcript levels than the level of progerin. This may reflect a long half-life of progerin, as well as greater efficacy of splicing inhibition in terminally differentiated tissues with little or no effect on progerin protein that is already incorporated into the existing nuclear envelope.

Systemic autopsies of treated animals did not reveal any gross or microscopic abnormalities of major organs, with the exception of the kidney. SRP-2001-treated mice developed irregular cortical surfaces, occasional degenerated tubules and tubular epithelial cell basophilic granules (Extended Data Fig. 10). Assessment of renal function as measured by creatinine in terminal bleeds was normal, suggesting normal glomerular filtration rate, while blood urea nitrogen was moderately elevated in the treated mice, most likely a result of terminal dehydration.

Discussion

Since the identification of the classic HGPS c.1824C>T, p.G608G mutation in LMNA, a variety of therapeutic approaches have been proposed. This includes farnesyltransferase inhibitors to reduce the production of the toxic progerin protein^25,26 and autophagy activation with rapamycin analogs to promote its clearance^12,27. While the former approach has proven clinically beneficial^9,10, and the latter is currently the subject of a clinical trial, these approaches are focused on reducing the effect of the dominant negative progerin protein at the posttranslational level and relieving specific downstream effects of the dominant negative progerin protein, respectively. A genetic therapeutic approach of blocking progerin production at the level of transcription comes closer to correcting the fundamental defect. In this study, we demonstrated the efficacy of SRP-2001, a new PPMO that inhibits aberrant splicing at the c.1824C>T, G608G of the LMNA gene in HGPS fibroblast cell lines and results in extended longevity in a human-BAC transgenic mouse model of HGPS.

Comprehensive screening of antisense oligomers led to the identification of SRP-2001, which inhibited the level of progerin transcripts in human HGPS and transgenic LMNA^G608G fibroblasts by >95% resulting in a significant reduction (>50%) in the levels of progerin protein in long-term cell culture and responding in a dose-dependent manner. The lack of similar reduction in progerin protein relative to transcript levels has been previously suggested due to elevated resistance to cellular degradation²⁸. This inhibition of progerin production results in the significant restoration of cellular proliferation of human HGPS fibroblasts.

In addition to the effects of progerin expression on cellular proliferation, reduction of LMNB and aberrant distribution of other components of the nuclear lamina are well-described aspects of the cellular disease phenotype in HGPS; restoration of these alterations has been achieved by morpholino-mediated progerin knockdown^29,30. We also noted reduced levels of B-type lamins in immunoblots of HGPS fibroblast lysates, which were increased on reduction of progerin by SRP-2001 treatment. Interestingly, cellular senescence has been associated with transcriptional repression of B-type lamins in human lung and dermal fibroblasts, as well as keratinocytes^31,32, which may be associated with lamin B localization at intranuclear sites of late synthesis phase DNA replication³³. Thus, the rescued levels of B-type lamins in HGPS fibroblasts after treatment may contribute to the increased proliferative capacity when progerin production is inhibited.

We also documented significant reduction in lamin A and increase in lamin C transcript expression after exposure of normal and HGPS fibroblast cells to SRP-2001. The improvement in proliferation of normal control fibroblasts in the presence of SRP-2001 was unexpected but likely a result of the changes in lamin A and lamin C expression. The reduction of lamin A transcript expression occurs presumably because SRP-2001 still binds to the WT sequence analysis of exon 11, even with a mismatch. The mechanism by which reduced lamin A leads to increased proliferation of WT cells is not entirely clear, but it has been demonstrated that Lmna^−/− mouse embryonic fibroblasts exhibit increased proliferation that is reversed by the reintroduction of Lmna/c³⁴. Lamin A protein expression is higher in tumor versus benign tissue in prostate cancer³⁵. It has also been reported that the ratio of LMNC to LMNA transcript expression is increased in breast cancer tissue³⁶. The overexpression of lamin C is not presumed to be pathological because mice that have been engineered to make only lamin C are entirely healthy³⁷. Proper formation of the nuclear lamina is dependent on the total expression of lamins. If expressed at sufficient levels, lamin A, B1 or B2 can each assemble into an organized nuclear laminar structure³⁸. Therefore, these changes in lamin A and lamin C are not expected to represent a major risk of SRP-2001 treatment.

We have demonstrated that the effective in vivo delivery by intravenous injection of SRP-2001 to predominantly affected tissues promotes the improvement of the vascular pathology of LMNA^G/G mice by promoting the relative preservation of VSMCs. Note that the preservation shown in Fig. 6 is the result of comparing 30.5-week-old saline-treated mice versus 49.5-week-old SRP2001-treated mice. It is intriguing to note that VSMC loss is less severe near the lumen in SRP-2001-treated mice, suggesting a gradient effect of intravenous exposure to SRP-2001 splice inhibition of progerin production. Although long-term treatment of LMNA^G/G with SRP-2001 does not restore growth characteristics to normal, it results in sustained weight and an increase in longevity of these mice by 61.6% from a median of 30.5–49.5 weeks, an improvement rate greater than any previously seen in the HGPS literature with other treatment strategies. Deposits of basophilic material (possibly PPMOs) in the kidney are noted after 120 mg kg⁻¹ per week intravenous administration for the entire life of the animal but do not appear to have harmed renal function.

One previous study has shown the beneficial effects of antisense oligomer therapies in a Lmna^G609G/G609G knock-in mouse model with similar characteristic cardiovascular defects, vascular smooth muscle loss and premature death. This model utilized two octa-guanadine dendrimer-conjugated morpholino oligomers targeting exons 10 and 11 and achieved significant extension of the lifespan³⁰. However, that study did not result in notable improvement of cardiovascular pathology, a prominent phenotype associated with HGPS. This might reflect a difference in the tissue bioavailability of the morpholino used in that study compared to SRP-2001. In the course of our investigation, we learned of results of a similar study in this same mouse model, using lipid-conjugated antisense oligonucleotides (M. Puttaraju et al., manuscript submitted for publication). That approach, which tested several different antisense oligonucleotide targets, also demonstrated significant reduction of progerin expression in multiple tissues, including the aorta, and resulted in modest extension of lifespan, further validating the approach of progerin splice inhibition as a potential therapeutic to treat patients with HGPS.

By extension from cell culture results, it is possible that the efficacy of this treatment in blocking progerin and lamin A production at the transcriptional level might allow for therapeutic application to any of the mutations that result in splicing at the cryptic G608G locus, including c.1822G>A, G608S and intron 11 mutations that weaken the exon 11 splice donor site.

We conclude that treatment of HGPS with SRP-2001 can be an effective alternative or additive to drug therapeutics that work at the protein level. While this genetic approach to HGPS therapy does not completely ameliorate the effects of progerin in our mouse model, it surpasses the level of preclinical disease remediation by current drug therapies. Given that this same approach is already being applied for other human disorders such as Duchenne muscular dystrophy³⁹ and spinal muscular atrophy⁴⁰, the path toward an approvable human clinical trial appears promising.

Online content

Any methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at https://doi.org/10.1038/s41591-021-01274-0.

Methods

Mouse strains and animal care.

A transgenic mouse model of HGPS was developed by retrofitting a human bacterial artificial chromosome harboring the LMNA gene containing the classic G608G mutation (c.1824C>T). The 164-kilobase transgene was incorporated into the germline of C57BL/6J mice as described previously²⁴. Single-copy mice, designated as LMNA^G/+, were bred within the C57BL/6J line for 20 generations before the generation of double-copy mice, designated as LMNA^G/G, which were used in the experiments. Mice were housed in barrier facilities with a 12-h light–dark cycle at temperatures of 65–75 °F (approximately 18–23 °C) with 40–60% humidity at the National Institutes of Health (NIH) animal facilities. Murine growth curves were determined by weekly weights. Animal care and experiments were performed in accordance with protocol G-03–5 approved by the National Human Genome Research Institute Animal Care and Use Committee.

PPMOs.

All PPMOs were designed, synthesized, purified and provided by Sarepta Therapeutics^16,41. See Supplementary Table 1 and Extended Data Fig. 1 for the oligomer base sequences and PPMO structure, respectively. Verification of oligomer uptake in vivo was ascertained in EGFP-654 reporter mice, which express a transgene encoding EGFP interrupted by the insertion of the mutated intron 2 of the human β-globin gene (Fig. 4a)¹⁹. The mutation at nucleotide 654 activates aberrant splice sites that are preferentially utilized during pre-mRNA processing, resulting in partial intronic retention and inhibition of EGFP translation. Targeted antisense-mediated steric hindrance of the aberrant splice site restores normal splice site use and EGFP expression and can be utilized for qualitative and quantitative assessment of alternative PPMO delivery models⁴². Accordingly, mice were injected intravenously (n = 3–5 per group) daily with 40 mg kg⁻¹ of PPMO EGFP-654, 30 mg kg⁻¹ of B-peptide-conjugated PPMO EGFP-654 or 200 μl of saline for 3 d before being killed at 7 d post-initial injection. Tissues were collected at necropsy, followed by RNA isolation and RT–qPCR analysis as described previously¹⁹. For immunohistofluorescence analysis of vascular tissues, mice were perfused with 2% paraformaldehyde before being killed and samples were flash-frozen to minimize autofluorescence. GFP was detected using anti-GFP antibody at 1:50 dilution (GFP polyclonal antibody Alexa Fluor 594, catalog no. A-21312; Thermo Fisher Scientific). Fluorescence emission images were obtained with a confocal microscope system (LMS 510; ZEISS) and collected with ×20 oil lens.

Cell cultures.

Fibroblasts from patients with HGPS (HGADFN167, HGADFN496, HGADFN367) and fibroblasts from normal controls (HGFDFN168, HGMDFN718, HGMDFN368) were acquired from the Progeria Research Foundation Cell and Tissue Bank. Murine fibroblast cultures were derived from dermal tissue dissected from the abdomens of newborn mice. Cells were allowed to grow out from dermal samples for two weeks before release by trypsin digestion. Cell lines were cultured in DMEM containing 10% FCS, 2 mM of glutamine and 1% penicillin-streptomycin at 37 °C in 5% CO₂.

Proliferation assays.

Cells were plated at 10,000 cells per well in 24-well culture plates (day 0) in triplicate. At 24 h post-plating, daily replenishment of media commenced with fresh medium, medium containing saline (vehicle) or medium containing 6 μM of SRP-2001. Each cell line was collected by trypsinization at 24-h increments with cell concentrations determined using a NucleoCounter with NucleoView 3000 software (version 2.1.25.12, Chemometec). Total cell numbers were calculated and plotted using Microsoft Excel (Mac version 16.43).

Gene expression analyses.

Murine tissues were collected into TRIzol reagent (Thermo Fisher Scientific), homogenized and immediately flash-frozen until ready for total RNA isolation. RNA was subsequently digested for 20 min at 37 °C with recombinant DNase I (Thermo Fisher Scientific), then analyzed for integrity and concentration on a nucleic acid Bioanalyzer (Agilent Technologies).

For qPCR, synthesis of cDNA utilized 1 μg of RNA, which was reverse-transcribed using the iScript cDNA Synthesis kit (Bio-Rad Laboratories) according to the manufacturer’s protocol. Droplets for each cDNA sample were generated in triplicate using 50 ng of cDNA, 900 nM of primers, 250 nM of probes in 1× ddPCR Supermix for Probes (Bio-Rad Laboratories) on a QX200 Droplet Generator (Bio-Rad Laboratories), followed by PCR amplification. The sequences of primers and probes used for the quantitation of LMNA transcripts are listed in Supplementary Table 2. PCR cycling conditions consisted of an initial enzyme activation step for 10 min at 95 °C, followed by 40 cycles of 94 °C for 30 s and 59 °C for 30 s with a 2 °C per second ramp rate and a 10 min enzyme deactivation step at 98 °C for 10 min. A- and B-type lamins were analyzed with the following additional primer-probe assays as indicated (all PrimePCR Probe Assays (Bio-Rad Laboratories)): mouse Lmnb1 (assay ID no. qMmuCIP0033595, FAM); mouse Lmnb2 (assay ID no. qMmuCIP0028595, FAM); mouse Hprt (assay ID no. qMmuCEP0054164, HEX); mouse Tfrc (assay ID no. qMmuCIP0041876, HEX) or human LMNB1 (assay ID no. qHsaCIP0029571, FAM); human LMNB2 (assay ID no. qHsaCIP0027822); and human TFRC (assay ID no. qHsaCIP0033292, HEX). Reactions were analyzed on a QX200 Droplet Reader (Bio-Rad Laboratories) to obtain expression levels relative to murine Hprt, murine Tfrc or human TFRC and transcript-specific copy numbers, then further analyzed using Microsoft Excel.

Western blot analyses.

Cell cultures and tissues were homogenized in high-salt radioimmunoprecipitation assay buffer (20 mM of Tris-HCl, pH 7.4, 0.5 M of NaCl; 1 mM of EDTA; 0.1% SDS, 1% Triton X-100, 1× protease inhibitor cocktail (catalog no. P8340; Sigma-Aldrich)), 100 mM of AEBSF (catalog no. SBR00015; Sigma-Aldrich), 20 μM of caspase VI inhibitor (catalog no. 219007; Sigma-Aldrich) and quantitated by bicinchoninic acid assay. For analysis of tissue extracts, mice were necropsied at the end point and tissues were stored frozen until further analysis. After homogenization, target proteins were pulled down using mouse antibodies and protein G sepharose (catalog no. 37478; Cell Signaling Technology). Immunoprecipitates were electrophoresed, transferred to nitrocellulose and probed using the appropriate rabbit primary antibodies. Samples were electrophoresed on 8% Bis-Tris gels (Thermo Fisher Scientific), transferred to nitrocellulose membranes, which were blocked in 5% milk or 5% BSA overnight at 4 °C.

The following primary antibodies were used: mouse anti-human lamin A + lamin C (clone JoL2, 1:50 dilution, catalog no. ab40567; Abcam); rabbit anti-lamin A + lamin B1 + lamin C (clone EPR4068, 1:500 dilution, ab108922: Abcam); rabbit anti-lamin B1 (clone EPR8985(B), 1:500 dilution, catalog no. ab133741; Abcam); rabbit anti-lamin B2 (clone E1S1Q, 1:500 dilution, catalog no. 13823; Cell Signaling Technology); mouse anti-α-tubulin (clone DM1A, 1:1,000 dilution, catalog no. 3873; Cell Signaling Technology); rabbit anti-β-actin (clone D6A8, 1:1,000 dilution, catalog no. 8457; Cell Signaling Technology); mouse anti-β-actin (clone 15G5A11/E2, 1:100 dilution, catalog no. MA1140; Thermo Fisher Scientific); rabbit anti-α-smooth muscle actin (1:1,000 dilution; catalog no. ab5694; Abcam); mouse anti-α-smooth muscle actin (clone 1A4, 1:100 dilution, catalog no. 48938; Cell Signaling Technology); mouse anti-GAPDH (clone 6C5, 1:100 dilution, catalog no. MAB374; Sigma-Aldrich); and rabbit anti-GAPDH (clone 14C10, 1:1,000 dilution, catalog no. 2118; Cell Signaling Technology). Blots were incubated with primary antibodies for 4 h at room temperature, followed by 2 h incubation with IRDye 800CW donkey anti-mouse IgG or IRDye 680RD donkey anti-rabbit IgG fluorescent secondary antibodies (LI-COR), then imaged on an Odyssey CLx Imaging system with the Image Studio software (version 5.2, LI-COR).

Mass spectrometry.

SRP-2001-treated mouse tissue samples were collected and homogenized in 50 mM of Tris-HCl (pH 7.5) buffer. Tissue homogenates were then subjected to proteinase K/trypsin digestion to convert all potential metabolites to one end product of PMO-Gly before the solid-phase extraction (SPE) cleanup. The digested tissue homogenates with the internal standard were loaded and washed with ammonium acetate buffer and then eluted with H₂O/ACN/FA (70/30/5) twice by using Waters HLB SPE 96-well plates. The eluates were concentrated under nitrogen gas and then injected to ultra-performance liquid chromatography-high-resolution mass spectrometry followed by parallel reaction monitoring mass spectrometry quantitation (Sciex API5000 or Thermo Quantiva; NovaBioAssays). A Thermo Q Exactive Plus high-resolution mass spectrometer was used for the assays and the lower limit of quantitation of all tissues was 100 ng g⁻¹ in mouse aorta, kidney, heart and skin tissue.

Histology.

Tissues were fixed in 2% paraformaldehyde for 24 h before dehydration with graded alcohols and embedded in paraffin. Cross-sections (4-um thick) were cut and mounted on charged slides and visualized by H&E or Movat’s pentachrome staining (CV Path). Images were captured on an Axio Scan imaging system (ZEISS) at 20× magnification and processed with the ZEN 2.0 (blue edition) software. Additional trimming and analysis for VSMC counts and adventitial area were performed in Photoshop (version 21.2.3, Adobe).

Sample measurement and statistical analyses.

Unless otherwise stated, in vitro data were generated from technical triplicates performed on biological triplicate samples. In vivo data were generated from technical triplicate measurements of samples from individual mice. For the transcript and protein quantitation analyses, statistical analyses were performed using two-tailed Student’s t-tests. For the in vivo studies, sample size was calculated for 80% power to achieve significance in the extended lifespan of 42 d in treated mice versus untreated mice and 90% power to achieve significance in the extended lifespan of 48.5 d. Kaplan–Meier survival analysis significance was determined by Mantel–Cox log-rank test using 1 d.f. with χ² = 26.17. The results of the statistical analyses are reported in Supplementary Table 3.

Extended Data

Extended Data Fig. 1 |. PPMO structure.

SRP-2001 is a peptide-conjugated phosphorodiamidate morpholino oligomer (PPMO) comprising a cell-penetrating peptide (CPP) covalently linked to a phosphorodiamidate morpholino oligomer (PMO). The PMO contains 25 morpholino subunits each bearing a nucleobase forming the sequence 5’-GAGGAGATGGGTCCACCCACCTGGG-3’. The conjugated peptide comprises the amino acid sequence R₆G, where R is arginine and G is glycine. The glycine residue covalently links the CPP to the PMO by an amide bound to the 3’ end of the PMO. Mass spectrometric characterization was performed to confirm the structure and sequence of SRP-2001.

Extended Data Fig. 2 |. Quantitative PCR screening of candidate PPMOs for targeted reduction of progerin in proband fibroblasts.

Quantitation of LMNA (A), Progerin (P), and LMNC (C) expression by ddPCR analysis of Classic HGPS (LMNA c.1824C > T) fibroblasts treated with 6uM candidate PPMOs. a, Transcript levels relative to TFRC. b, Each isoforms’ fraction relative to all LMNA transcripts. c, Each isoforms’ transcript level relative to LMNC transcripts. The greatest reduction of progerin transcripts in culture was achieved with the centrally located morpholino (SRP-2001). A, LMNA; P, Progerin; C, LMNC; WT, oligo complementary to normal LMNA sequence; CTRL, scrambled control; Values and error bars represent mean ± SD determined from triplicate reactions of biological triplicates. * p < 0.05, ** p < 0.01, *** p < 0.001 versus transcripts or protein from cells receiving saline (NT).

Extended Data Fig. 3 |. Western immunoblot screening of candidate PPMOs for targeted reduction of progerin in proband fibroblasts.

Quantitation of LMNA (A), Progerin (P), and LMNC (C) protein from immunoblots of lysates from Classic HGPS (LMNA c.1824C > T) fibroblasts treated with 6uM candidate PPMOs. Data was derived from the analysis shown in Fig. 1d and is expressed as isoform levels relative to ACTB (a), the isoforms’ fraction relative to all LMNA gene products (b), and the isoforms’ level relative to LMNC protein (c). The greatest reduction of progerin protein in culture was achieved with the centrally located morpholino (SRP-2001). A, LMNA; P, progerin; C, LMNC; WT, oligo complementary to normal LMNA sequence; CTRL, scrambled control; Values and error bars represent mean ± SD determined from biological triplicates shown in Fig. 1d. * p < 0.05, ** p < 0.01, *** p < 0.001 versus transcripts or protein from cells receiving saline NT).

Extended Data Fig. 4 |. Quantitative PCR screening of SRP-2001 for targeted reduction of progerin in Classic and Non-classic HGPS cell lines.

a, Transcript copy number in normal control (NL Control), Classic HGPS (c.1824C>T), and Non-classic HGPS (c.1822G>A, c.1968+1G>A, c.1968+2T>C, c.1968+5G>C) fibroblasts following 2 weeks of treatment with SRP-2001. b, Individual LMNA transcripts expressed as a fraction of all total LMNA-derived transcripts in each cell line. c, Full-length LMNA and progerin transcript levels relative to LMNC transcripts. A, LMNA; P, progerin; C, LMNC; Values and error bars represent mean ± SD determined from triplicate reactions of biological triplicates. * p < 0.05; ** p < 0.01; *** p < 0.001 vs same transcript in cells receiving saline only (NT).

Extended Data Fig. 5 |. Reduction of progerin protein by SRP-2001 in Classic and Non-classic HGPS cell lines.

a, Western immunoblots of normal control (NL Control), Classic HGPS (c.1824C>T), and Non-classic HGPS (c.1822G>A, c.1968+1G>A, c.1968+2T>C, c.1968+5G>C) fibroblasts following 2 weeks of treatment with SRP-2001. b, Quantitation of LMNA gene products relative to ACTB. c, Quantitation of LMNA, progerin and LMNC isoforms expressed as a fraction of all total LMNA-derived gene products in each cell line. d, Full-length LMNA and progerin protein levels relative to LMNC. A, LMNA; P, progerin; C, LMNC; Graphs represent mean ± SD determined from biological triplicates shown in panel A. * p < 0.05; ** p < 0.01; *** p < 0.001 vs same protein isoform in cells receiving saline only (NT).

Extended Data Fig. 6 |. SRP-2001 reduces expression of transgene-derived progerin in murine aortas.

a, Quantitative PCR analysis of transgene transcript levels in aortas of LMNA^G/G mice receiving long-term treatment with saline (vehicle) or 60 mg/kg SRP-2001. Transcripts are quantitated relative to Hprt expression, as a fraction of all transgene-derived LMNA transcripts or relative to transgene-derived LMNC transcripts. Data are presented as mean values ± SD determined from 12 independent samples per treatment group (n = 6 males, 6 females). b, Western immunoblots of A-type lamins and beta actin (ACTB) immunoprecipitated from aorta tissue. c, Quantitation of A-type lamins from immunoblots expressed relative to beta actin (ACTB), as a fraction of all A-type lamins and relative to LMNC. Data are presented as mean values ± SD determined from 12 independent samples per treatment group (n = 6 males, 6 females), shown in panel b. A, LMNA; P, progerin; C, LMNC; * p < 0.05; ** p < 0.01; *** p < 0.001 vs same transcript or protein in mice receiving saline only (vehicle).

Extended Data Fig. 7 |. SRP-2001 reduces expression of transgene-derived progerin in murine hearts.

a, Quantitative PCR analysis of transgene transcript levels in hearts of LMNA^G/G mice receiving long-term treatment with saline (vehicle) or 60 mg/kg SRP-2001. Transcripts are quantitated relative to Hprt expression, as a fraction of all transgene-derived LMNA transcripts or relative to transgene-derived LMNC transcripts. Data are presented as mean values ± SD determined from 12 independent samples per treatment group (n = 6 males, 6 females). b, Western immunoblots of A-type lamins and smooth muscle actin (SMA) immunoprecipitated from heart tissue. c, Quantitation of A-type lamins from immunoblots expressed relative to smooth muscle actin (SMA), as a fraction of all A-type lamins and relative to LMNC. Data are presented as mean values ± SD determined from 12 independent samples per treatment group (n = 6 males, 6 females), shown in panel b. A, LMNA; P, progerin; C, LMNC; * p < 0.05; ** p < 0.01; *** p < 0.001 vs same transcript or protein in mice receiving saline only (vehicle).

Extended Data Fig. 8 |. SRP-2001 reduces expression of transgene-derived progerin in murine liver.

a, Quantitative PCR analysis of transgene transcripts in liver tissue of LMNA^G/G mice receiving long-term treatment with saline (vehicle) or 60 mg/kg SRP-2001. Expression of LMNA (A), progerin (P) and LMNC (C) is quantitated relative to Hprt expression, as a fraction of all transgene-derived LMNA transcripts or relative to transgene-derived LMNC transcripts. Data are presented as mean values ± SD determined from 12 independent samples per treatment group (n = 6 males, 6 females). b, Western immunoblots of A-type lamins and beta actin (ACTB) immunoprecipitated from liver homogenates. c, Quantitation of A-type lamins from immunoblots expressed relative to beta actin (ACTB), as a fraction of all A-type lamins and relative to LMNC. Data are presented as mean values ± SD determined from 12 independent samples per treatment group (n = 6 males, 6 females), shown in panel b. * p < 0.05; ** p < 0.01; *** p < 0.001 vs same transcript or protein in mice receiving saline only (vehicle).

Extended Data Fig. 9 |. SRP-2001 reduces progerin transcripts in multiple mesenchyme-derived murine tissues.

Quantitative PCR analysis of transgene transcripts in a, quadriceps, b, femoral bone, and c, kidneys of LMNA^G/G mice receiving long-term treatment with saline (vehicle) or 60 mg/kg SRP-2001. Expression of LMNA (A), progerin (P) and LMNC (C) is quantitated relative to Hprt expression, as a fraction of all transgene-derived LMNA transcripts or relative to transgene-derived LMNC transcripts. Data are presented as mean values ± SD determined from 12 independent samples per treatment group (n = 6 males, 6 females). * p < 0.05; ** p < 0.01; *** p < 0.001 vs same transcript or protein in mice receiving saline only (vehicle).

Extended Data Fig. 10 |. Histologic analysis of kidneys shows minimal toxicity due to long-term treatment with SRP-2001.

Representative images of haemotoxylin and eosin (HE) staining of sectioned kidneys from treated (n = 12) and untreated (n = 12) mice. Mice receiving long-term treatment with SRP-2001 developed irregular cortical surfaces (ic), degenerate tubules (dt) and basophilic granules (bg) in tubular epithelial cells. Two sections per tissue sample were analyzed.

Data availability

All requests for raw and analyzed data and materials will be promptly reviewed by the National Human Genome Research Institute and Sarepta Therapeutics to verify whether the request is subject to any intellectual property or confidentiality obligations. Any data and materials that can be shared will be released via a Data/Material Sharing Agreement. All requests should be made to the primary or corresponding author. All detailed PPMO sequence and assay primer data are presented in the extended and supplementary data.