SHAPE Really Matters When it Comes to Short Hairpin RNAs
John J. Rossi, PhD
Posted: April 18, 2012

SHAPE-directed Discovery of Potent shRNA Inhibitors of HIV-1
Justin T Low, Stefanie A Knoepfel, Joseph M Watts, Olivier ter Brake, Ben Berkhout and Kevin M Weeks

RNA interference (RNAi) has become a powerful tool for blocking gene expression via sequence specific degradation of the targeted RNA. Soon after, it was demonstrated in 2001 that RNAi could be used in human cells; RNAi was applied to a variety of biological systems to understand gene function or to inhibit viral and other deleterious gene expression.

One of the earliest applications of RNAi was to use promoter expressed short hairpin RNAs (shRNAs) which are processed into RNAi triggers called small interfering RNAs (siRNAs). The transcribed shRNAs are exported to the cytoplasm via the microRNA export pathway and once in the cytoplasm they are processed by the cellular RNAi machinery into siRNAs. The siRNAs guide a complex of proteins termed the RNA induced silencing complex, or RISC, to the target mRNA via Watson- Crick base pairing. RISC then cleaves the target and blocks gene expression. Designing potent shRNAs is not straight forward, and rules that apply to the design of chemically synthesized siRNAs often do not apply to shRNAs.

In the April issue of Molecular Therapy, Dr. Justin T. Low and colleagues have addressed this problem by evaluating the contribution of target site accessibility to shRNA function. They used a large training set of 84 shRNAs with known potencies that are complementary to sites along the HIV-1 genome and correlated the efficiencies of these shRNAs in inhibiting HIV replication with data obtained from SHAPE (selective 2’ hydroxyl acylation analyzed by primer extension) probing of the HIV genomic RNA. This structural probing method is used to determine regions of secondary and tertiary structure. Not surprisingly they observed that the most active shRNAs targeted regions with the least secondary and tertiary structure. Looking more deeply into the highly accessible target sequences, they observed that the target structure complementary to the first seven bases, the seed and extending to thirteen bases from the 5’ end of the shRNA derived siRNA guide strand also needed to be unstructured and accessible to hybridization with the shRNA derived siRNA guide strand. Finally they noted that there was a strong correlation between the thermodynamic stability of the guide strand-target strand and shRNA function.

The surprising conclusion from these studies was that features such as thermodynamic end properties of the shRNA derived siRNAs, and other sequence motifs in the guide strand previously reported to be important for synthetic siRNA activity were not associated with the potency of the shRNAs. To further test their findings, they went on to design 26 additional shRNAs based upon the SHAPE analyses and found all of these to have potent inhibitory function.

The significance of this study is that it not only provides a facile method for designing anti-HIV-1 shRNAs, but provides a strategy for the design of shRNAs in for the more ambitious use of shRNAs to target the transcriptome in large scale studies of gene function.