24 Sep 2002

In theory, the specificity provided by RNA interferencethe silencing of genes by small interfering RNAs (siRNA)makes it a potentially powerful therapeutic (see ARF related news story). In practice, however, there are many hurdles that must be overcome before this potential can be realized (see ARF related news story), not least being the problem of getting the siRNAs into target cells and organs. One solution to this problem is to produce the RNAs in situ by exploiting the cell’s own RNA polymerase, a method that has facilitated siRNA production from cells transiently transfected with plasmids. Now, in the current Nature Biotechnology online, researchers from Beverly Davidson’s lab at the University of Iowa, show that viral vectors can be used to facilitate this approach in vivo.

Joint first authors Haibin Xia and Qinwen Mao used adenoviruses harboring an siRNA producing construct that consisted of a small section of enhanced green fluorescent protein (eGFP) DNA coupled to a cytomegalovirus promoter. They tested the efficacy of their system in transgenic cells and mice expressing eGFP. In vitro, eGFP expression in HEK cells was almost completely abolished by the virus, while in vivo, virus injected into mice brain reduced eGFP expression near the site of injection. The authors were also able to use their vector to silence endogenous genes such as b-glucuronidase, by exchanging the eGFP part of the construct for one targeting the native gene. Mice injected with these viruses had a statistically significant lowering of b-glucuronidase activity in the liver (12 percent decrease).

The authors demonstrated the therapeutic potential of this technique by inhibiting the expression, in neural cell lines, of eGFP chimeras that contained a polyglutamine (polyQ) tract. PolyQ proteins are notorious for forming aggregates and are responsible for numerous neurodegenerative diseases, including Huntington’s disease. Silencing the chimeras substantially reduced the number and size of these aggregates.

The key to the success of this system lies in the construct. A hairpin siRNA is coded by adjacent 21 base pair segments of DNA, one being the reverse complement of the other, which are tailed by a minimal polyA sequence. These are placed in close proximity to the promoter resulting in a complete transcript that is only 63 base pairs long. Other arrangements proved much less effective.—Tom Fagan