How to overcome the challenge of model selection in DMD gene therapy?

Duchenne muscular dystrophy (DMD) is an X-linked recessive genetic disorder caused by mutations in the DMD gene, resulting in abnormal function of the dystrophin protein. This ultimately leads to progressive muscle degeneration. Duchenne muscular dystrophy (DMD) is a rare muscle disorder but it is one of the most frequent genetic conditions affecting approximately 1 in 3,500 male births worldwide. It is usually recognized between three and six years of age.

Types of Gene Therapy for DMD

The gene therapy for DMD can be classified into three main types: the first type involves AAV-mediated delivery of mini dystrophin or micro-dystrophin; the second type utilizes ASO-mediated exon skipping therapy; and the third type employs CRISPR gene editing techniques.

AAV-Mediated Delivery of Mini Dystrophin or Micro-Dystrophin

The AAV-mediated delivery of mini dystrophin or micro-dystrophin has the potential to be applicable to a wide range of patients, as this supplementation therapy may work for individuals with different types of mutations. Companies like Sarepta, Roche, and Solid are actively developing AAV-based therapies targeting mini dystrophin or micro-dystrophin. Among them, SRP-9001 has shown the most promising progress. Recently, it has gained attention in the news as the FDA advisory committee experts recommended accelerated approval for SRP-9001. Preclinical animal studies for this pipeline have utilized the classic mdx model of DMD. The mdx mouse model carries a mutation in the CAA codon in exon 23, which leads to abnormal gene expression. While mdx mice exhibit disease phenotypes similar to DMD, there are certain differences compared to real patients. First, the phenotype of mdx mice is milder, particularly in terms of fibrosis and muscle atrophy. Second, the lifespan of mdx mice is approximately 80% of normal mice, whereas the average lifespan of DMD patients is only about one-third of healthy individuals. Third, the mdx mouse mutation site in exon 23 is not a hotspot mutation, which limits its disease homogeneity. Additionally, mdx mice show some variations in phenotypes depending on different backgrounds, adding complexity to studies conducted on mdx mice.

ASO-Mediated Exon Skipping Therapy

ASO-mediated exon skipping therapy is one of the most prominent gene therapy approaches in the pipeline for DMD. This is attributed to the strong targeting ability and ease of synthesis of ASOs. Companies like Sarepta, Nippon, and DYNE have invested in this area. Exon skipping targets specific exons such as 44, 45, 50, 51, and 53, and multiple exon skipping strategies have been reported in the literature. Nippon's pipeline NS-065/NCNP-01 utilizes ASO-mediated exon skipping of exon 53 and has been tested in preclinical animal studies using mdx52 mice. These mice carry a deletion of exon 52 in the Dmd gene, resulting in premature termination of translation at exon 53. While mdx52 mice are more suitable for ASO-based gene therapy as the mutation site is closer to hotspot mutations in patients, there are risks associated with selecting ASOs specific to human DMD gene from the mouse Dmd gene due to sequence differences between humans and mice. On the other hand, mdx mice are not suitable for ASO-mediated exon skipping, particularly when targeting exons located after exon 23, as mdx mice already have a termination signal at exon 23, and skipping the exons following exon 23 does not provide behavioral efficacy readouts. In the literature, the hDMDΔ52/mdx model has been used to evaluate the efficacy of ASO-mediated exon skipping. This model involves the introduction of full-length human DMD gene with exon 52 deletion and subsequent hybridization with mdx mice to generate a DMD disease model. While the hDMDΔ52/mdx model expresses full-length human DMD protein, the insertion site of the DMD gene is unknown, it is not an X-linked gene, and further hybridization with mdx mice is required, making the model construction complex.

CRISPR Gene Editing Techniques

The progress of CRISPR gene editing therapy in DMD has been relatively slow, partly due to the lack of DMD disease models that can facilitate specific drug target screening and drug predictability. CRD-TMH-001 is the first CRISPR-based pipeline for treating DMD, and it has received Investigational New Drug (IND) approval. Several successful cases of CRISPR therapy for DMD have been reported in the literature. For example, the use of a cytosine base editor (CBE) to remove premature termination codons and enable continued translation of the Dmd gene has been demonstrated in the Δ50;h51KI model. This model involves humanization of the E51 region of the mouse Dmd gene and deletion of E50, introducing a premature termination codon at E51. It is evident that compared to mdx52, the Δ50;h51KI model used in CRISPR gene editing therapy requires a higher degree of humanization for the drug targeting sequence. This is an important factor to consider in CRISPR gene editing therapy, as off-target effects can have immeasurable negative impacts on patients. However, it should be noted that the Δ50;h51KI model has limitations in terms of humanization compared to the hDMDΔ52/mdx model.

Advantages and Limitations of Existing Models

Classic mdx Mouse Model:

• Involves a mutation in the CAA codon in exon 23, which leads to abnormal gene expression
• Milder phenotype compared to DMD patients
• Lifespan is approximately 80% of normal mice
• Not suitable for ASO-mediated exon skipping, particularly when targeting exons located after exon 23, as mdx mice already have a termination signal at exon 23,
• mdx mouse mutation site in exon 23 is not a hotspot mutation, which limits its disease homogeneity
• Exhibits variations depending on backgrounds

mdx52 Mouse Model:

• Involves a deletion of exon 52 in the Dmd gene, resulting in premature termination of translation at exon 53
• more suitable for ASO-based gene therapy as the mutation site is closer to hotspot mutations in patients, but there are risks associated with ASO selection

hDMDΔ52/mdx Model:

• Involves humanization of the DMD gene with exon 52 deletion
• Expresses full-length human DMD protein but has limitations in terms of humanization, such as:
  • Unknown insertion site of DMD gene, not X-linked, and;
  • Requires further hybridization with mdx mice, making the model construction complex

Δ50;h51KI Model:

• Involves humanization of the E51 region of the mouse Dmd gene and deletion of E50, introducing a premature termination codon at E51
• Compared to mdx52, the Δ50;h51KI model used in CRISPR gene editing therapy requires a higher degree of humanization in the drug targeting sequence
• Has limitations in terms of humanization compared to the hDMDΔ52/mdx model
  
  

Humanized DMD (hDMD) Models by Cyagen

Considering the mutation characteristics of the DMD gene and the shortcomings of existing models, such as the complexity of construction, uncertainty in Transgenic (Tg)-induced gene insertion sites, insufficient humanization regions, and non-hotspot mutations, Cyagen has independently developed humanized DMD (hDMD) research models. In addition to the classic model mdx mouse, we have also independently developed humanized mice with DMD hotspot mutations, including DMD (hE8-30), DMD (hE44-45), and DMD (hE49-53). Furthermore, we have modified wild-type humanized mice to construct hotspot mutation humanized disease models, which can simultaneously obtain better control group and disease models.

Advantages of Cyagen's hDMD Models Include:

• Construction of wild-type and point-mutation humanized disease models in the hotspot mutation region
• Customization of different point mutations on existing wild-type models, improving efficiency and success rate
• Humanization region includes most drug-targeting regions to improve suitability for drug screening and pharmacological research, especially for gene therapy-related drugs such as ASO, CRISPR, and siRNA
• Human DMD gene is inserted in situ with a stable and confirmed copy number, ensuring stable inheritance
• Includes several DMD patient-relevant ‘hotspot’ mutations: DMD (hE8-30), DMD (hE44-45), and DMD (hE49-53).
  
  
  

Reference:

[1] Ryder-Cook AS, Sicinski P, Thomas K, et al. Localization of the mdx mutation within the mouse dystrophin gene. [J]. EMBO,1988.

[2] Mizobe Y , Miyatake S , Takizawa H , et al. In Vivo Evaluation of Single-Exon and Multiexon Skipping in mdx52 Mice[J].Methods Mol Biol, 2018.

[3] Hoen P A C ' ,  Meijer E J D ,  Boer J M , et al. Generation and Characterization of Transgenic Mice with the Full-length Human DMD Gene[J]. Journal of Biological Chemistry, 2008, 283.

[4]Zhang Y, Li H, Nishiyama T, McAnally JR , et al. A humanized knockin mouse model of Duchenne muscular dystrophy and its correction by CRISPR-Cas9 therapeutic gene editing[J]. Mol Ther Nucleic Acids. 2022.