Giving a Virus a New Postcode Redirecting AAV Gene Therapy to Muscle

This application covers adeno-associated virus capsid polypeptides carrying a short binding peptide with the sequence motif RGD followed by four variable amino acids, for treating and preventing muscular disease and for muscle regeneration. The applicant is Heidelberg University, Germany’s oldest university.

The Problem

AAV is the workhorse of gene therapy. It carries a therapeutic gene into cells and keeps it expressed for a long time, and it has an established safety record across approved products. Where it struggles is aim.

Each natural AAV serotype has a tropism, a set of tissues it happens to infect well, determined by which receptors its capsid binds. AAV9 has the broadest range. But for most cell types, the transduction efficiency of an unmodified capsid is far too low to be useful, whether in the laboratory or the clinic.

The obvious response is to engineer the capsid, and there is a well developed technique: insert a short peptide, typically seven amino acids, into an exposed loop of the capsid protein known as variable region VIII. Change the loop, change what the capsid binds, change where it goes. Prior work using this approach produced variants with improved uptake by astrocytes and by breast cancer cells.

But retargeting has a stubborn side effect. Even when an inserted peptide successfully confers a new specificity, substantial infectivity for the liver usually remains. The liver is AAV’s default destination, and dose that ends up there is dose that does not reach the target while still contributing to the immune and toxicity burden. Muscle is a particularly demanding target because it is a very large tissue, so any inefficiency multiplies into an enormous required dose.

What This Invention Does

The invention is an AAV capsid polypeptide bonded to a binding peptide comprising the amino acid sequence RGDX1X2X3X4, where X1 through X4 are independently selected amino acids, for use in treating or preventing muscular disease and in muscle regeneration.

RGD is not an arbitrary choice. Arginine-glycine-aspartate is the canonical recognition motif for integrins, the cell surface receptors that mediate attachment to the extracellular matrix, and integrins are abundantly present on muscle cells. Building the motif into the capsid loop gives the virus a handle on a receptor family that muscle displays heavily. The four variable positions following it allow the sequence to be tuned, which is where the breadth of the claim and the specific disclosed variants sit.

The application extends beyond the capsid polypeptide itself to the polynucleotides encoding it, host cells, assembled AAV capsids, pharmaceutical compositions, and the uses and methods that follow. That spread is characteristic of platform biotechnology filings, where the commercially important element may be the vector, the manufacturing route or the therapy depending on how the field develops.

Key Features

• RGD binding peptide. The capsid carries the integrin binding RGD motif followed by four independently selected amino acids.
• Variable region VIII insertion. The peptide is presented in the exposed capsid loop that established engineering methods use for retargeting.
• Muscle directed use. The stated indication is treatment and prevention of muscular disease and support of muscle regeneration.
• Integrin mediated uptake. Targeting exploits integrin receptors that muscle tissue displays in abundance.
• Full platform coverage. Claims extend to polynucleotides, host cells, assembled capsids and pharmaceutical compositions.
• Reduced reliance on natural tropism. The engineered specificity is intended to improve on what unmodified serotypes achieve in muscle.

Who Is Behind It

The applicant is Universität Heidelberg. The named inventors are Josef El Andari, Jonas Weinmann and Dirk Grimm, whose laboratory at Heidelberg University Hospital is a well known centre for AAV capsid engineering and vector development. The specification cites a doctoral thesis from the same institution among the prior work on peptide insertion into AAV9.

The application is a divisional of Australian application 2019258830 filed in April 2019, relates to PCT/EP2019/060790, and claims priority from European application 18169822.6 filed on 27 April 2018.

Why It Matters

Muscular diseases are among the most sought after and most difficult gene therapy targets. Duchenne muscular dystrophy is the emblematic case: a single well understood gene, a devastating and predictable course, and a treatment problem that comes down almost entirely to delivery. Skeletal muscle is roughly forty percent of body mass, so systemic dosing at the efficiencies natural capsids achieve requires quantities of vector that strain both manufacturing and safety.

Dose is not an abstract concern here. Serious adverse events, including deaths, have occurred in high dose systemic AAV trials, and liver toxicity features prominently among them. A capsid that reaches muscle more efficiently and the liver less is therefore not a convenience. It changes what dose is needed and what risk that dose carries.

The academic origin is also notable. Much foundational capsid engineering has come out of university laboratories rather than companies, and this filing, a divisional pursued years after the 2018 priority date, suggests continuing commercial interest in the family.

Related Concepts

• Adeno-associated virus – the viral vector platform being engineered.
• Gene therapy – the treatment approach this vector serves.
• Integrin – the receptor family the RGD motif binds.
• RGD motif – the three amino acid sequence at the heart of the targeting strategy.
• Duchenne muscular dystrophy – the muscular disease that drives most of this research effort.
• Viral tropism – the concept of which tissues a virus naturally infects, and what capsid engineering changes.

---

AU 2026202070 was published in the Australian Official Journal of Patents on 9 April 2026 and is open for public inspection. Patent applications represent inventions that are sought to be protected and do not necessarily reflect commercially available products.