Potential Genome-editing Therapy Shows Promise in Fabry Mouse Model

Steve Bryson, PhD avatar

by Steve Bryson, PhD |

Share this article:

Share article via email
Fabry mouse model, genome-editing therapy

Janson George/Shutterstock

An experimental genome-editing therapy safely and effectively increased levels of alpha-galactosidase A (Gal A) — which is lacking in people with Fabry disease — in a mouse model of the disease, a study has shown.

By only changing the genome of about 10% of cells in the liver — where Gal A is produced — the approach was able to resolve the toxic globotriaosylceramide (Gb3) accumulation in all key tissues that are damaged in Fabry.

These findings suggest that genome editing has the potential to be an effective one-time treatment for Fabry disease.

Genome editing is an alternative treatment approach being investigated by Sangamo Therapeutics in parallel with its experimental gene therapy ST-920, which is currently being tested in a Phase 1/2 clinical trial (NCT04046224) in about 48 men with Fabry. Enrollment information can be found here.

The study, “ZFN-mediated in vivo gene editing in hepatocytes leads to supraphysiologic α-Gal A activity and effective substrate reduction in Fabry mice,” was published in the journal Molecular Therapy.

Recommended Reading

ST-920 Gene Therapy Found Safe, Effective in Mouse Model; Supports First-In-Human Trial

Fabry disease is caused by mutations in the GLA gene, which leads to an absence or deficiency in the encoded Gal A enzyme. Without this enzyme, its precursor molecule (substrate) Gb3 builds up to toxic levels and causes damage to organs, particularly the heart and kidneys.

Standard Fabry treatment replaces the missing Gal A enzyme, which helps break down Gb3 but requires regular lifelong infusions. Oral chaperone therapy can restore the activity of certain forms of faulty Gal A, but only about half of Fabry patients are amenable to this treatment. Thus, long-lasting, effective treatment to fully restore Gal A, such as gene therapy, is needed.

In standard gene therapy, such as ST-920, the GLA gene is delivered to the nucleus of cells — but is not inserted in the genome — to produce a fully active Gal A enzyme. In contrast, genome editing is designed to insert a functioning copy of the GLA gene directly into the human genome.

Genome editing can be achieved by delivering the GLA gene along with a specially created enzyme called a zinc-finger nuclease (ZFN) — a DNA-binding protein with a special conformation called a zinc finger, which targets specific locations in the genome, attached to a protein that cuts DNA to allow the functional GLA gene to be inserted.

Researchers at the Icahn School of Medicine at Mount Sinai in New York, with scientists from Sangamo, tested this genome-editing approach in a Gal A-deficient mouse model of Fabry disease.

DNA carrying instructions for the zinc-finger nuclease as well as the GLA gene were each delivered to liver cells using modified, harmless adeno-associated viruses (AAV) — three in total.

The GLA gene was designed to be inserted into another gene (Alb) that provides instructions for making the protein albumin. This protein is made by the liver in large amounts and secreted into the bloodstream to maintain fluid balance and carry various substances, including hormones and vitamins. Inserting GLA into Alb in a small portion of liver cells would allow for high production of the Gal A enzyme.

In the first set of experiments, the team tested the impact of various signal peptides on Gal A production. Signal peptides are short protein fragments attached to Gal A that guide the enzyme to its proper location in the cell, where the fragment is then removed.

The treatment was injected into the veins of Fabry mice, which were examined over the next two months. With the naturally occurring Gal A signal peptide, Gal A levels in the blood were up to 81 times higher than in normal non-Fabry control mice for the two-month study period.

Notably, using a signal peptide from a different human enzyme known as IDS, Gal A levels were 250 times higher than normal in the bloodstream for two months. Mice treated with this version — named IDSsp-hGLA — led to greater than normal Gal A activity and mostly no detectable Gb3 in the blood, liver, heart, kidney, and spleen.

Fabry mice were then given a high and a low dose of IDSsp-hGLA. After about two weeks, the activity of Gal A in the bloodstream had increased to about 60 times above normal with the low dose and 220 times higher with the high dose. Gal A activity also increased in the liver, heart, kidney, and spleen at both doses.

Further analysis found the higher-dose mice had normal levels of Gb3 in their bloodstream, liver, and heart and slightly higher amounts in the kidneys after two months.

The treatment was well tolerated at both doses, as mice were clinically healthy with only slightly elevated liver enzymes, a sign of inflammation or damage to liver cells. Moreover, high-dose Fabry mice had mean albumin levels comparable with untreated or healthy control mice, “indicating that this therapeutic approach did not disrupt normal production of albumin in the liver,” the researchers wrote.

Finally, the team found that about 6% of the liver’s cells (hepatocytes) were producing Gal A after the low dose, and about 10% at the high dose.

“These results indicate that near-complete substrate reduction in the Fabry mice was achieved with less than 10% of hepatocytes being stably modified to express the human [Gal A] enzyme,” and suggested that “ZFN-mediated in vivo genome editing has the potential to be an effective one-time therapy for Fabry disease,” the investigators concluded.