Nanoparticles resembling viruses can carry gene-editing tools into muscle and help restore a critical protein in a mouse model of Duchenne muscular dystrophy, according to a new preprint.
The approach currently works for muscle. But the team behind the work is exploring similar delivery options that could help treat a range of conditions, including neurological diseases.
“We have to understand how well it will get into the brain and also how well it can get into specific types of neurons,” says study investigator Feng Zhang, professor of neuroscience, brain and cognitive sciences and biological engineering at the Massachusetts Institute of Technology.
Duchenne muscular dystrophy stems from variants in the DMD gene that lead to the loss of dystrophin protein, causing muscle weakness that worsens with time. People with Duchenne are more likely to have autism and other neurobehavioral conditions than the general population, with an autism prevalence of about 7 percent, studies estimate.
In the new research, Zhang and his lab used a virus-like particle called dARC to deliver the gene editor Cas9 into mouse muscle. The strategy helped the muscle cells skip over genetic errors and restore dystrophin expression.
“This opens a totally, very interesting new avenue for gene delivery, particularly for gene editing,” says Guangping Gao, professor of biomedical research and director of the Horae Gene Therapy Center at UMass Chan Medical School, who was not involved in the work. “I really love the manuscript and its potential,” he adds.
Current gene-delivery methods, including lipid nanoparticles and adeno-associated viruses (AAVs), each have their own limitations, such as being difficult to direct at organs beyond the liver or requiring a complex manufacturing process, often using human embryonic kidney (HEK) cells. AAVs also only carry cargo up to a certain size, which can make gene-editing tools challenging to transport, and they can trigger immune reactions.
To break through the bottlenecks, Zhang and other researchers are exploring the use of particles that mimic a virus’s structure without the genetic material. Zhang is focusing on proteins derived from ancient viruses or other elements that integrated into animal (including human) genomes. Some of those integrated genes retained the ability to create proteins that can assemble into protective, virus-like shells or capsids that can help shuttle information across cells.
“They have been co-opted as part of our natural biological system,” Zhang says. “They are presumably well adapted by our immune system, so they’re not overtly immunogenic, and that makes them quite interesting.”
O
ne protein, encoded by the gene Drosophila ARC1 (dARC1), transfers messenger RNA (mRNA) across the junction between motor neurons and muscle fiber. The virus-like particles can also assemble in vitro, which could make manufacturing relatively simple, Zhang says.
In the new study, the team used E. coli to produce the dARC protein. The proteins then self-assembled into shells in test tubes in the presence of mRNA, wrapping around the genetic material because of its negative charge, as well as a negatively charged guide RNA (gRNA) and Cas9 complex. The team also attached a piece of positively charged peptide to the proteins to make the packaging more efficient.
“The scalability of it, the ability to produce something in E. coli as opposed to the HEK cells, is a big deal,” says Jerzy Szablowski, associate professor of bioengineering at Rice University, who was not involved in the study.
The dARC particles showed a strong affinity to muscle cells, which the researchers observed by packaging mRNA with fluorescent labeling inside the capsids. They found that the particles interacted with SORCS2, a receptor protein expressed in multiple tissue types in the body, as well as in the brain.
“That was really nice work, showing how these are binding to cells,” says Beverly Davidson, director of the Center for Cellular and Molecular Therapeutics at the Children’s Hospital of Philadelphia, who was not involved in the study. “It also allows you to iterate to try and target different receptors, if your goals are different cell types.”
Delivering a ribonucleoprotein (RNP) complex of gRNA and Cas9 via the dARC capsids restored 18 percent of the dystrophin protein in a mouse model of Duchenne muscular dystrophy, whereas delivering a mix of Cas9, mRNA and sgRNA restored 3.5 percent. The 18 percent rescue is comparable to results from exon-skipping antisense oligonucleotides, the team wrote in their preprint, which was posted on bioRxiv in May.
T
he tool, however, has a major limitation: the dARC capsids are unstable and break down in blood, the team found, preventing them from being easily delivered throughout the body.
“If you do systemic delivery, this is going to be a problem,” Gao says.
Zhang’s lab is now working on different strategies to stabilize the particles in serum, he says, and is exploring similar virus-like particles. To treat neurological diseases, the team will likely need to engineer a way for the particles to engage with receptors that can shuttle them across the blood-brain barrier, he adds.
An alternative route to the brain would be to inject the particles into cerebrospinal fluid, if the team can find a way to stabilize them there, Davidson says.
“If you could alter where they go and that you could show their stability when they’re introduced into the blood or other biofluids, I think that would be a big advance,” she says. “That’s where I’m hoping they take the technology.”
