Nonviral delivery of nCas9 for “safe harbor” integration to treat MPS IVA

Mucopolysaccharidosis type IVA (MPS IVA) is a rare lysosomal storage disorder resulting from mutations in the GALNS gene and subsequent accumulation of glycosaminoglycan, i.e., keratan sulfate (KS) and chondroitin 6-sulfate. MPS IVA leads to severe skeletal dysplasia and cardiovascular and respiratory complications. The current standard of care is enzyme replacement therapy, which involves life-long, frequent infusions but has poor outcomes in bones.¹ Other treatment options being evaluated include hematopoietic stem cell transplantation, substrate reduction therapy, and gene therapy.² However, all of these approaches are limited in use due to either low to moderate therapeutic benefits (especially in bones) or high risk (e.g., immune responses, toxicity). There remains a significant unmet need to develop a novel therapy that provides long-lasting therapeutic benefits with minimal risk. CRISPR-mediated gene editing is a revolutionary breakthrough in the field of biomedical technology. On December 8, 2023, the first CRISPR-mediated gene editing medicine (Casgevy) was approved for treating sickle cell disease,³ representing a milestone of CRISPR-mediated gene editing for medical applications (Table 1). Previously, CRISPR-mediated gene editing has been tested preclinically for treating other lysosomal storage disorders⁴ but has not been for MPS IVA. In this study, Leal et al. employed Cas9 nickase (nCas9) coupled with iron oxide nanoparticle (IONP) nonviral delivery to insert a GALNS transgene gene at the ROSA26 locus in a murine model of MPS IVA.⁵ The treatment led to increased GALNS levels, a significant reduction of storage materials, and partial recovery of bone pathology. In addition, the treatment was well tolerated without significant IONP-related toxicity and immune activation. This study represents a novel approach for treating MPS IVA and lays a solid foundation for clinical development.

Table 1.

Currently approved cell and gene therapy drugs

Drug name	Indication	Sponsor	Modality	Approvals
Elevidys	DMD	Sarepta	AAV	USA (2023)
Glybera	lipoprotein lipase deficiency	UniQure	AAV	EU (2012)
Hemgenix	hemophilia B	CSL	AAV	USA (2022), EU (2022)
Luxturna	RPE65 mutation-associated retinal dystrophy	Spark Therapeutics	AAV	USA (2017), EU (2018)
Roctavian	hemophilia A	BioMarin	AAV	EU (2022), USA (2023)
Upstaza	AADC deficiency	PTC Therapeutics	AAV	EU (2022)
Zolgenmsa	spinal muscular atrophy	Novartis	AAV	USA (2019), EU (2020)
Adstiladrin	bladder cancer	Ferring Pharmaceuticals	adenovirus	USA (2022)
Imlygic	myeloma	BioVex	herpes simplex virus	USA (2015), EU (2015)
Vyjuvek	epidermolysis bullosa	Krystal Biotech	herpes simplex virus	USA (2023)
Abecma	myeloma	Celgene	CAR-T	USA (2021), EU (2021)
Breyanzi	B cell lymphoma	Juno Therapeutics	CAR-T	USA (2021)
Carvykti	multiple myeloma	Janssen	CAR-T	USA (2022)
Kymriah	acute lymphoblastic leukemia	Novartis	CAR-T	USA (2017)
Tecartus	mantle cell lymphoma	Kite Pharma	CAR-T	USA (2020), EU (2020)
Yescarta	large B cell lymphoma	Kite Pharma	CAR-T	USA (2017)
Casgevy	sickle cell disease	Vertex	CRISPR cell therapy	USA (2023)
Lyfgenia	sickle cell disease	Bluebird Bio	lentiviral cell therapy	USA (2023)
Skysona	cerebral adrenoleukodystrophy	Bluebird Bio	lentiviral cell therapy	USA (2022)
Zynteglo	beta thalassemia	Bluebird Bio	lentiviral cell therapy	EU (2019), USA (2022)
Lantidra	type 1 diabetes	CellTrans	cell therapy	USA (2023)
laViv	nasolabial fold wrinkles	Fibrocell Technologies	cell therapy	USA (2011)
MACI	knee cartilage defects	Vericel Corporation	cell therapy	USA (2016)
Omisirge	blood cancer	Gamida Cell	cell therapy	USA (2023)
Provenge	prostate cancer	Dendreon Corporation	cell therapy	USA (2010)

DMD, Duchenne muscular dystrophy; AADC, aromatic L-amino acid decarboxylase; CAR-T, chimeric antigen receptor T.

In this study, this gene-editing system, which targets the ROSA26 locus with two single guide RNAs (sgRNAs) and inserts a GALNS transgene using nCas9, was first validated in cultured fibroblasts. Off-target effects were assessed using in silico prediction and targeted sequencing of the predicted top 10 sites. Successful integration was confirmed by Sanger sequencing and measurement of GALNS activities. Notably, DpnI assays showed the protection effects of IONPs from enzymatic degradation of nCas9 plasmids. Then, nCas9 coupled with IONPs were intravenously administered into MPS IVA mice. Based on ex vivo imaging results, IONPs were mainly localized to the liver. Measurements of hepatic and renal biomarkers in blood samples showed no significant changes, and there were no histopathological observations. These results indicate a solid safety profile of IONP delivery of nCas9 in MPS IVA mice. Further, GALNS activity increased in the liver (75.4% wild-type levels), plasma (27.7%), and other peripheral tissues (34.8%–67.4%). More interestingly, GALNS activity in the tibia and trachea increased to 6.8% and 12.6% wild-type levels, respectively. As expected, the increase in GALNS activity led to a significant reduction of storage materials (mono-sulfated KS) in plasma, liver, and humerus. Further, in treated mice, chondrocyte size in the articular cartilage was normalized, while a vacuolization decrease was also observed in chondrocytes in the growth plate, articular cartilage, and meniscus. These results demonstrate a partial pathological recovery provided by the treatment. Additionally, no significant increase in anti-GALNS immunoglobulin G (IgG) was observed in treated mice, supporting the immunotolerance to human GALNS protein in this model. There was no increase in anti-nCas9 antibodies, despite the persistence of nCas9 plasmids.

This study, for the first time, tested in vivo CRISPR-mediated gene editing to treat MPS IVA mice, paving the way for developing a novel gene-editing therapy for patients with MPS IVA. The significance of this study also resides in the application of nonviral delivery of Cas9. Previously, adeno-associated virus (AAV) vectors were often used for Cas9 delivery, creating the risk of continuous cleavage activity of Cas9, as AAV provides a relatively long-term transgene expression and may even integrate into the host genome.⁶ Another drawback of AAV delivery of Cas9 is the preexisting antibodies against AAV capsids and immune-related toxicity.⁷

In this study, transgene expression mainly from the liver achieved increased GALNS activity in many tissues including bones, presumably through the “cross-correction” phenomenon that exists in lysosomal storage disorders. Enzymes expressed from transduced cells were secreted out, traveled through blood circulation, and were taken up by other cells. This phenomenon circumvents the challenge of bone targeting through viral or nonviral delivery. It would be more interesting to see the results of expressing a fusion protein of GALNS and a bone-targeting peptide using the same strategy, which may further improve the efficacy in bones. Another benefit of this gene-editing strategy is that by targeting a safe-harbor locus, it avoids the issue of fixing heterogeneous mutations in patients. In that case, an sgRNA may have to be designed for each individual mutation, which involves additional assessment of on-target efficiency and off-target effects, as well as manufacturing and drug release tests.

Yet, some challenges still remain. For instance, the homologous sequence to the target locus in mice probably exists in humans and other species; however, the exact sequence may be different. Therefore, in additional preclinical studies in another species and in clinical studies, identification and assessment of species-specific sgRNAs may be required, and the correlation between dose, editing efficiency, and efficacy needs to be established. Also, immune responses against Cas9 and off-target effects should be addressed.