Double-stranded RNA (dsRNA) can trigger silencing of
homologous gene expression by a mechanism termed RNAi
(for RNA-mediated interference) (1). RNAi is an evolutionarily
conserved phenomenon and a multistep process that involves
generation of active small interfering RNA (siRNA) in vivo
through the action of an RNase III endonuclease, Dicer. The
resulting 21- to 23-nt siRNA mediates degradation of the
complementary homologous RNA (reviewed in refs. 2–4).
RNAi has been used as a reverse genetic tool to study gene
function in multiple model organisms, including plants, Caenorhabditis
elegans, and Drosophila where large dsRNAs efficiently
induce gene-specific silencing (1, 5–7).
One obstacle to achieving RNAi in mammals is that dsRNAs
longer than 30 nt will activate an antiviral response, leading to
the nonspecific degradation of RNA transcripts and a general
shutdown of host cell protein translation (8, 9). As a result, the
long dsRNA, with a few exceptions (10, 11), does not produce
RNAi activity, and RNAi therefore is not a general method for
silencing specific genes in mammalian cells. This obstacle has
been recently overcome by Tuschl and colleagues (12) who found
that gene-specific suppression in mammalian cells can be
achieved by vitro-synthesized siRNA that are 21 nt in length, long
enough to induce gene-specific suppression, but short enough to
evade the host interferon response.
In this article, we describe a DNA vector-based approach to
achieve RNAi in mammalian cells. With this approach, small
RNAs are predicted to be synthesized from a DNA template
under the control of an RNA polymerase III (Pol III) promoter
in transfected cells. Pol III has the advantage of directing the
synthesis of small, noncoding transcripts whose 3 ends are
defined by termination within a stretch of 4–5 thymidines (Ts)
(13). These properties make it possible to use DNA templates to
synthesize, in vivo, small RNAs with structural features close to
what has been found to be required for active siRNAs synthesized
in vitro (14).
Using this DNA vector-based RNAi approach, we show that
transfected as well as endogenous genes can be efficiently
inhibited. We have examined the effects of the in vivosynthesized
siRNAs on a transfected reporter gene, a housekeeping
gene, and genes involved in cell cycle control and DNA
methylation. In each and every case, we find that these small
RNAs efficiently and specifically inhibit the synthesis of proteins