Gene Therapy for Hemophilia and Duchenne Muscular Dystrophy in China

Introduction

There are 58,000 monogenic diseases caused by mutations in a single gene, and these diseases lack sufficient treatment. ¹ Gene therapy holds enormous potential for treating these diseases through gene addition or in situ gene correction. To date, 10.5% of gene therapy trials worldwide focus on monogenic diseases (www.wiley.com), and Chinese scientists have made important contribution to this field. In fact, the first gene therapy trial in China for hemophilia B (HB) was conducted early in 1995. ² This review aims to outline recent progress in gene therapy for hemophilia and Duchenne muscular dystrophy (DMD) in China.

Hemophilia

Hemophilia A (HA) and HB are X-linked bleeding disorders caused by mutations in genes encoding for either clotting factor VIII (FVIII) or factor IX (FIX), which affect 1/5,000 males (HA) and 1/25,000 (HB) worldwide. Patients often suffer from recurrent bleeds and chronic joint damage. Current treatment for hemophilia consists of lifelong protein replacement with recombinant or plasma-derived clotting factors. This treatment is extremely costly and invasive, as infusion is required two to three times per week to maintain minimal therapeutic clotting factor levels. Gene therapy has the potential to provide lasting treatment for these conditions.

Adeno-associated virus (AAVs) vectors are promising for hemophilia gene therapy. Studies have shown that AAV vectors expressing a codon-optimized FIX intravenously injected into patients with severe HB corrected the phenotype. ^3,4 However, high does are required to induce detectable FIX expression. The presence of neutralizing antibodies (NAbs) is another challenge for AAV applications. ^{5 –7} Therefore, either AAV transduction must be enhanced or NAb activity must be avoided.

Chinese researchers have found that incubation of human serum albumin with AAV8/FIX-OPT can produce fivefold higher FIX levels in HB mice. ⁸ In addition, this group described modifying the AAV capsid as a novel approach to enhance AAV transduction. They co-transfected AAV2, AAV8, and AAV9 helper plasmids to produce polyploidy AAV vectors, and observed a twofold higher transduction in the mouse liver using the triploid vector AAV2/8/9 compared to AAV8 following systemic administration. More importantly, further analysis indicated that the NAb escape ability of triploid AAV2/8/9 vector was improved by approximately 20-fold, 32-fold, and 8-fold, respectively, when compared to AAV2, 8, and 9, indicating potential for use in future clinical trials. ⁹

As an alternative to codon-optimization, a single R338L variant in the FIX gene was found to have eight times higher levels of clotting activity compared to wild-type FIX. ¹⁰ An optimized AAV8 vector expressing FIX R338L from liver-specific promoter-1 (LP1), scAAV8-LP1-hFIX, was injected into HB mice through systemic injection or hydrodynamic injection, and normal levels of clotting activity were achieved 30 days after injection. ¹¹

Genome editing using programmable nucleases has also shown promise in gene therapy for hemophilia. ^{12 –15} Zinc finger nucleases (ZFNs) to correct HB in mouse models have been inserted partial cDNA (exons 2–8) or full-length cDNA into hFIX mutant locus in hFIX mutant transgenic mice or into mouse albumin gene loci, respectively. ^{12 –14} Additionally, a novel FIX Y371D mutation was developed in an HB mice model in China. Hydrodynamic tail-vein (HTV) injection of naked Cas9-sgRNA plasmid and donor DNA into FIX Y371D mice has demonstrated homology-directed repair (HDR). Importantly, correction of 0.56% of FIX alleles in hepatocytes was sufficient to restore hemostasis. ¹⁶ In a similar study, gene correction (1%) in FIX alleles of hepatocytes was observed in 62.5% of treated HB mice following HTV injection of naked Cas9-sgRNA plasmid and donor DNA, which was sufficient to remit the coagulation deficiency. ¹⁷

Two major types of chromosome inversions involve introns 1 and 22 of the FVIII gene and account for 1–4% and 50% of severe HA cases, respectively. Park et al. initially used transcription activator-like effector nucleases (TALENs) to generate an HA iPSC model with intron 1 inversion and then reverted the inversion. ¹⁸ They corrected inversions that involve introns 1and 22 of FVIII in HA patient-derived induced pluripotent stem cells (iPSCs) using the clustered regularly interspaced short palindromic repeat (CRISPR)-associated nuclease caspase 9 (Cas9) system and achieved a low targeting efficiency (up to 6.7%). ¹⁹ Additionally, the TALEN sites remained intact after gene correction, leading to the possibility of using second-round inversion, or small insertions and deletions (INDELs).

The authors' group developed a novel strategy for in situ genetic correction of FVIII intron 22 inversion in HA patient-derived iPSCs using TALENs. The sequence of exons 23–26 plus a polyA signal was precisely inserted at the junction of exon 22 and intron 22 with a high targeting efficiency of >50% via HDR. The gene correction restored FVIII transcription and protein expression and rescued functional FVIII secretion in the gene-corrected iPSC derived endothelial cells (ECs) and mesenchymal stem cells (MSCs). ²⁰ This was the first report of an efficient in situ genetic correction of large gene mutations using targeted addition of small sequences. Nonviral and integration-free approaches for reprogramming urine cells are now available in China_, indicating the potential of this approach for clinical use.

Work from the authors' laboratory has also exploited genetically modified cells with site-specific integration of exogenous therapeutic genes for the development of hemophilia therapies. Human ribosomal DNA (rDNA) locus consists of nearly 400 copies of the 45S pre-RNA (rRNA) gene clustered on the short arms of all five acrocentric chromosomes (13, 14, 15, 21, and 22). ²¹ It is well-known that human rDNA copy number variations are common among healthy individuals, and loss or gain of some copies of the rRNA gene is very common, without any phenotypic effects, such that they could be stably inherited. Balanced translocations involving short arms of the acrocentric chromosomes containing rDNA cluster are usually clinically normal. Therefore, disrupting one or several copies of the rRNA gene might be tolerable for cells because of the high copy number. In addition, the rDNA locus may represent a recombination hotspot, as it exhibits high intrinsic recombinational properties during both meiosis and mitosis. For these reasons, it is proposed that the rDNA locus might be a potential “safe harbor” for transgenes.

Previously, FVIII or FIX was targeted into the rDNA locus of several cell lines, including HT1080, HL7702, and human embryonic stem cells, using nonviral vectors. ^22,23 The results indicated that integrated FVIII or FIX could be expressed efficiently at the rDNA locus. An efficient site-specific integration of hFVIIIBD was also achieved at the multi-copy rDNA locus in HA patient iPSCs with no detection of off-targets using TALENickases. ²⁴ In addition, EPCs and megakaryocyte platelets differentiated from integrated iPSCs could efficiently express integrated FVIII and show promise for clinical use.

A number of cell types have been investigated in the search for hemophilia gene therapies, including liver sinusoidal endothelial cells (LSECs), skeletal muscle cells, and hematopoietic stem cells. ^{25 –27} Chinese scientists isolated MSCs from transgenic mice expressing hFVIIIBD and transplanted them into HA mice. hFVIIIBD was detectable in multi-organs in the recipient HA mice, resulting in an effective phenotypic correction in HA mice. ²⁸ In addition, ZFNs were used to target the hFIX gene in the AAVS1 locus in human MSCs and the genetically modified MSCs effectively expressed hFIX in vitro and in vivo. ²⁹

Dmd

DMD is a recessive X-linked disease that affects approximately 1/5,000 newborn males with the dystrophin gene mutation, which leads to loss of functional dystrophin expression. Dystrophin is a critical component of muscle cells, as it integrates cytoskeleton with the extracellular matrix and maintains the integrity of cell membrane during contraction. Absence of dystrophin leads to progressive skeletal muscle wasting and fatal heart failure. ^{30 –32} Currently, there is no effective treatment to prevent the occurrence and progression of this fatal disease.

Significant effort has been made to develop treatments for DMD. At the genomic level, full-length dystrophin cDNA, the mini/micro-dystrophin gene, has been delivered by viral or nonviral vectors to restore the production of the dystrophin protein, and ZFNs and TALENs have been adapted to modify mutant dystrophin gene and correct dystrophin gene expression. ^{33 –35} Recently, the CRISPR/Cas9 system has been used to edit the somatic mutant dystrophin gene in vivo, representing the forefront of the field.

The mdx mouse carries a nonsense mutation in exon 23. This premature stop codon leads to disruption of dystrophin expression and causes DMD. Based on an exon skipping strategy and nonhomologous end joining mediated by CRISPR/Cas9, the mutant exon 23 of the dystrophin gene has been excised through local or systematic transduction of Cas9/gRNA constructs by electroporation, adenovirus, or AAV 8/9 in neonatal or adult mdx mice. Genetic editing using CRISPR/Cas9 has been used to restore truncated but functional dystrophin protein expression in skeletal, cardiac muscle, and endogenous myogenic precursor cells to enhance muscle function. ^{36 –39}

DMD iPSCs derived from DMD patients could serve as cellular models or cell sources for further exploration of autologous cell therapies. By generating iPSCs using fibroblasts from DMD patients and modifying the dystrophin gene using the CRISPR/Cas9 system, Chinese researchers have developed DMD models. ⁴⁰ These animal models share greater similarities with humans than other types of models. ^41,42

Transplantation of cells with or without genetic modification represents another promising approach to treat DMD as well as hemophilia. Adipose-derived stem cells (ADSCs) have shown potential in promoting the regeneration and survival of muscle cells following transplantation into mdx mice to alleviate muscle damage in muscular dystrophy. ⁴³ In one study, ADSCs were induced to differentiate into myogenic progenitors in vitro. Following transplantation into mdx mice, these progenitor cells successfully engrafted in skeletal muscle for up to 12 weeks and generated new muscle fibers with dystrophin expression. ⁴⁴ Embryonic-like stem cells (ELSCs), which are isolated from human bone marrow, can differentiate into multinucleated myotube-like cells more efficiently than MSCs. Transplantation of ELSCs into mouse models improved motor function, reduced serum creatine kinase (CK) activity, and upregulated expression of dystrophin 2 months after transplantation. ⁴⁵

A clinical trial to treat Becker muscular dystrophy using allogeneic human umbilical cord–derived MSCs has been conducted in three patients. At 12 weeks of follow-up after transplantation, all three patients exhibited improved motor function and had normalize serum CK and lactate dehydrogenase, without any apparent adverse reactions. ⁴⁶ As for genetically modified cells, bone marrow–derived MSCs from mdx were transduced with viral vectors expressing human microdystrophin gene and were then transplanted into mdx mice. Persistent dystrophin restoration in the skeletal muscle of mdx mice was evident for up to 12 weeks. ⁴⁷

Antisense oligonucleotides (AOs) have shown great potential in restoring dystrophin expression though inducing exon skipping, as indicated by cellular and animal model data. ^48,49 Evaluation of AOs for treating DMD are currently in clinical trials. However, limited systemic delivery efficiency remains a major challenge. ^{50 –53} A group of Chinese researchers have explored the effects of many chemicals, including fructose, hexose, peptide nucleic acid, and some special peptides, on increasing the efficiency of AOs delivery to muscle coupling. ^{54 –57}

Conclusion

Recent outstanding progress, especially in genome editing, has accelerated the pace of developing gene therapies for hemophilia and DMD. However, efforts to enhance the efficacy of genome editing, to reduce off-target effects associated with nuclease, and to facilitate the delivery of a genome editing tool are still underway. Although encouraging results have emerged from experiments in AAV-mediated gene therapy, a number of critical issues related to the virus, including immune response, dose requirement, low packaging capacity, and non-permanent expression, remain unresolved. The very strict approval process of gene therapy clinical trials has slowed progress in China compared to other countries. Nonetheless, Chinese researchers are making significant contributions to genome editing, and the future of gene therapy for hemophilia and DMD is promising.