Researchers from Harvard Medical School’s CBR Institute for Biomedical Research have developed a new technology that can systemically deliver siRNAs across the blood-brain barrier.
The technology, called Corvus, involves joining siRNAs to short peptides derived from rabies virus glycoprotein, which binds to the nicotinic acetylcholine receptor present on neuronal cells.
In a study in which they intravenously injected siRNAs into mice, the researchers found them present at very high levels in virtually all parts of the brain, but not in any other tissues. Additional experiments confirmed that the siRNAs were functional against target genes.
The work, which was done in collaboration with investigators from the University of Iowa, Korea’s Samchully Pharm, and Seoul-based Hanyang University, appears in this week’s Nature. It has already caught the attention of a number of industry players.
Ryan Dietz, a technology transfer associate at the CBR Institute, told RNAi News this week that discussions about licensing the technology have been held with “several” companies in the RNAi therapeutic and reagent fields, although no deals have been finalized.
He expects to sign a number of license agreements over the rest of the year, noting that while the CBR Institute is open to field-specific, exclusive licenses, “most likely non-exclusive [licenses] would be the way to go.”
“Some of the industry partners I’ve spoken with very much want me to do [non-exclusives] because they don’t want complications; they just want an enabling technology,” he said.
Although RNAi holds the promise of treating a number of neurological disorders, the blood-brain barrier has hamstrung efforts to develop systemically administered RNAi therapeutics against gene targets in the brain.
According to CBR researcher Priti Kumar, who is the first author of the paper, it was this very issue she and her colleagues were up against when they began developing the Corvus technology.
“We stated out looking [to develop siRNAs against] flaviviruses like Japanese encephalitis [virus] and West Nile [virus] … because, so far, there are no drugs that exist to treat” these diseases, she told RNAi News.
Last April she and her colleagues published a paper in PLoS Medicine showing how a single siRNA could be used to suppress the encephalitis associated with both JEV and West Nile when delivered into the brains of mice via intracranial injection.
how a single siRNA could be used to suppress the encephalitis associated with both JEV and West Nile when delivered into the brains of mice via intracranial injection.
“The only problem was to get it [into the brain we needed to perform] intracranial injections because the blood-brain barrier is tightly packed with endothelial cells that do not allow [large] molecules to pass through,” she said.
Based on existing data from other labs, she and the other CBR Institute researchers began investigating whether they could use lentiviruses pseudotyped with RVG to overcome this hurdle.
“Rabies virus is a neuronal virus and it is able to get into neurons and brain cells because of its glycoprotein envelope, [which] binds to the nicotinic acetylcholine receptor … present at very high densities on neuronal cells and the endothelial cell lining of the blood-brain barrier,” Kumar said.
Though early experiments with this approach were encouraging, ultimately lentiviruses are not an option for brain tissue-specific therapies “because they integrate into the host cell, and integration into the chromosomes of brain cells is not something that is desirable at all,” Kumar noted.
Furthermore, the use of RVG itself as a delivery vehicle would likely trigger an immune response against the protein that would preclude repeated administration of the therapeutic, she added.
“So we [began] to look into what it is about the rabies virus glycoprotein that binds to nicotinic acetylcholine receptor,” she said.
As it turns out, this work had already been done. In 1982, Yale University researcher Thomas Lentz identified
a 29 amino acid-long peptide derived from the glycoprotein that mediates the binding of RVG to nicotinic acetylcholine receptor.
“We thought we could use this tiny peptide instead of the whole protein … to carry molecules across the blood-brain barrier,” Kumar said.
“But we first wanted to make sure that this peptide could get across the blood-brain barrier in vivo,” she said. “Nobody has ever done any characterization of this peptide. Lentz only tried making these peptides and checking in solution whether it could bind to nicotinic acetylcholine receptor, [but] he never actually tried it in an application model.”
After confirming that the peptide could indeed cross the blood-brain barrier, Kumar and the other researchers were then faced with the issue of linking it to siRNAs.
”Although RVG peptide can bind to neuronal cells, it does not bind nucleic acids and therefore cannot be used to transport siRNA,” the researchers wrote in their Nature paper. “However, short, positively charged, cell-penetrating peptides bind negatively charged nucleic acids by charge interaction.”
Previous reports in the literature have shown that oligo-arginine can facilitate the cellular uptake of nucleic acids, while the cholesterol-conjugated oligo-d-arginine has been shown
to be effective in delivering siRNAs into transplanted tumors in mice.
In light of these data, the Harvard team synthesized a peptide that was a fusion of the 29 amino acid-long RVG peptide and oligo-d-arginine “with the idea that RVG would target the blood-brain barrier and the oligo-arginine … should act as a carrier for the siRNA,” Kumar explained.
The researchers found that this fusion peptide, which they designated RVG-9R, could specifically deliver siRNAs inside neuronal cell lines in vitro and bring about “very effective” silencing.
“Some of the industry partners I’ve spoken with very much want me to do [non-exclusive licenses] because they don’t want complications; they just want an enabling technology.”
To evaluate the delivery approach in vivo, Kumar and her colleagues systemically administered into mice fluorescein isothiocyanate-labeled siRNAs complexed to RVG-9R via tail-vein injections.
“What we found was that we could pick up [the label] only in brain tissue but not in any other tissue,” Kumar said.
Brain-sectioning analyses done in collaboration with Beverly Davidson at the University of Iowa revealed that the peptide was able to penetrate “almost any part of the brain that is supplied by vasculature,” she added.
“The next thing,” Kumar said, “was to find out if [the siRNAs] were functional — are they being released by the RVG peptide and do they actually mediate gene silencing?”
To test this, the researchers administered the RVG-9R bound to GFP-targeting siRNAs into a transgenic mouse that expresses high levels of the fluorescent protein in its organs.
“We found a selective reduction of GFP levels only in the brain and not in any other tissue showing that the siRNAs were not only being delivered to the target [cell] type but that they were also functional,” she said.
To see if Corvus could be used to silence an endogenous gene, Kumar said that she and her colleagues designed siRNAs against the SOD1 gene, certain mutations in which are associated with the familial form of amyotrophic lateral sclerosis.
“Both messenger RNA and protein levels of SOD1 were significantly decreased in the brain, but not in other organs in … treated mice,” they wrote in Nature.
Extending these findings to their initial work with flaviviruses, the researchers then complexed JEV-targeting siRNAs to RVG-95 and administered them intravenously into immunodeficient mice, which are uniformly susceptible to peripheral infection with flaviviruses.
“We find that we can get up to 80 percent of [the JEV-infected] mice protected with siRNA targeting JEV when we use” the Corvus technology, Kumar said.
“Taken together, our results suggest that RVG-9R peptide may enable transvascular delivery of siRNA to the central nervous system,” the investigators wrote in Nature.
Although the gene knockdown obtained using the Corvus technology was significant and comparable to silencing achieved by other labs examining prolonged infusion of siRNA in the central nervous system, the researchers conceded that it is “relatively modest” overall.
Kumar said that this is likely due to the fact that the siRNAs used in the in vivo experiments were not modified or encapsulated, and therefore subject to nuclease degradation.
As such, this issue is expected to be easily resolved through the use of chemical modifications to the siRNA and the use of protective carriers such as liposomes — something Kumar said is being developed at the CBR Institute with a number of collaborators, including Alnylam Pharmaceuticals.
“Further studies to localize the presence of siRNA and gene silencing in different cell types within the brain are also needed to understand the mechanism by which RVG-9R enables delivery to the brain,” the researchers added in their paper.
Despite the need for additional studies, the data presented in Nature highlight “the potential of RVG-9R to mediate transvascular delivery of siRNAs to the central nervous system,” they wrote.
“RVG-mediated delivery might also allow the use of RNA interference for the systematic analysis of gene function in brain cells under experimental settings … [and] might also be used for the brain-directed transport of other therapeutic molecules such as gene therapy vectors and small-molecule drugs.”