Despite all the time and attention RNAi has been receiving from those trying to use it therapeutically, the gene-silencing technology continues to face a fundamental problem: delivery.
But recently published data, based on the work of researchers from Harvard Medical School and the University of Pennsylvania, indicate that the key to overcoming systemic delivery issues may be just around the corner.
In a paper published this week in the online version of Nature Biotechnology, Judy Lieberman, a senior investigator at Harvard's CBR Institute for Biomedical Research, and colleagues describe how an antibody-mediated siRNA delivery approach can be used to trigger RNAi systemically with high specificity and no off-target effects.
According to the paper, the system comprises a fusion protein designed with the protein-coding sequence of protamine, which nucleates DNA in sperm, linked to the C terminus of the heavy chain Fab fragment of an HIV-1 envelope antibody.
"Basically, if you mix the fusion protein with siRNA, the siRNA binds by charge and [is] delivered very efficiently into just cells that have the receptor recognized by the antibody," Lieberman told RNAi News this week.
She explained that the delivery approach was first developed by Wayne Marasco at the Dana Farber Cancer Institute as a means to deliver plasmid DNA.
"It was a kind of gene-therapy way of getting plasmid DNA into target cells, and one of my postdocs, Erwei Song, thought that we could use it for delivering small hairpin plasmid DNA to induce RNA interference," Lieberman noted. "I said, 'Why don't we just try it with small [interfering] RNAs? I bet it will work.'"
To test her theory, Lieberman and her colleagues used the system to deliver naked siRNAs in HIV-infected cells or cells transfected to express HIV envelope glycoprotein gp (HIV env), the Nature Biotechnology paper states. They found that siRNAs bound to F105-P "induced silencing only in cells expressing HIV-1 envelope. Additionally, siRNAs targeted against the HIV-1 capsid gene gag, inhibited HIV replication in hard-to-transfect, HIV-infected primary" T lymphocytes, the paper notes.
"Using B16 melanoma cells transfected with an expression vector for HIV env, intravenous or intratumoral injection of F105-P-complexed siRNAs delivered siRNAs only into env-expressing (env+) tumors, but not into normal tissues or env-negative tumors," the paper states.
Furthermore, tumor outgrowth was inhibited when the siRNAs targeted the oncogenes c-myc, MDM2, and VEGF, the paper states, and the researchers were able to use an anti-ErbB2- protamine fusion protein to deliver siRNAs specifically to ErbB2-expressing (ErbB2+) breast cancer cells.
While Lieberman and her colleagues used "an artificial system to target melanoma cells by transfecting them to express HIV env," the paper states, the "delivery strategy could be modified to target any of a variety of cells via different types of cell-surface receptors. Specific tumor markers, often indicators of a poorly differentiated state or of lineage commitment, have been identified for many human tumor cells."
The paper adds that "normal cells, whose functions need to be regulated, could also be targeted by this method. Examples might be T lymphocytes in autoimmune disease, dendritic cells or macrophages during inflammatory diseases, or hepatocytes for hypercholesterolemia. The latter have been recently targeted in vivo by systemic administration of a chemically modified siRNA covalently linked to a cell-receptor ligand (cholesterol binding to the apoB receptor)."
This hypercholesterolemia research was performed by scientists from Alnylam Pharmaceuticals and appeared in Nature late last year (see RNAi News, 11/12/2004).
The antibody-mediated delivery system "is really efficient," Lieberman said. "With 100 picomoles, we're getting silencing equivalent with what you get with transfection — but we can use it to introduce siRNAs into cells that you can't transfect, like primary lymphocytes."
She added that "the delivery was very specific — [siRNAs] didn't go into any cells that didn't express the receptor. We found no induction of interferon or interferon-response genes, and no obvious inflammation or other kinds of toxicity."
Lieberman pointed out, however, that the pharmacokinetics of fusion antibody-delivered siRNAs has not yet been determined.
"Filtration of naked siRNAs by the kidney is the rate-limiting factor responsible for the short in vivo half-life of unmodified siRNAs," the paper states, but "the estimated size of the complex is 100 kilodaltons, well above the cut-off for kidney filtration."
Another issue, Lieberman said, is serum RNases, "and we haven't looked to see if being bound to the protein in any way protects the siRNAs from" degradation. The paper adds that "chemical modification of the siRNA in the complex should reduce vulnerability to serum degradation, but whether modifications would enhance in vivo efficacy is uncertain, since chemical modifications appear to come at the price of efficiency for intracellular silencing."
Additionally, the paper states that "the trafficking pathway of F105-P-delivered siRNAs into the cytoplasm remains to be understood … [and] a cell biology study to follow the delivery of fluorescent siRNAs is needed."
Despite the work that remains to be done, Lieberman said that reactions to her work has been extremely positive.
"I've been approached by a lot of people who are very interested in this technology, both small biotechs and some large pharmaceutical companies," she said. "I'm also considering forming a company.
"A lot of people are giving me advice at the moment," Lieberman added. "Basically, there's been a lot of interest, [and] right now I'm considering the options."
— Doug Macron ([email protected])