A Note of Caution on siRNA Studies

A new paper in the “Instant Online” section of Human Gene Therapy offers a cautionary tale about interpreting results of studies employing short interfering RNA (siRNA) as a therapeutic strategy. The authors report that these siRNAs typically trigger interferon production while the control siRNA used in many studies (at least 13, by their count) has unusually weak immune-stimulating properties. They also show that the anti-influenza activity reported for some siRNAs is entirely dependent on immune stimulation; modification of these constructs to abrogate their capacity to induce interferon caused a complete loss of the anti-influenza effect. The authors close by stressing that “to avoid the potential of misinterpreting therapeutic efficacy caused by siRNA-mediated immune stimulation for a specific RNAi effect in animal models, it is critical for researchers to appreciate the capacity for siRNA to activate such a response and to fully characterize the immunostimulatory potential of both active and control siRNA sequences.”

Free full text PDF: http://www.liebertonline.com/doi/pdfplus/10.1089/hgt.2008.131

Hum Gene Ther. 2008 Aug 19. [Epub ahead of print]

Misinterpreting the therapeutic effects of siRNA caused by immune stimulation.

Robbins M, Judge A, Ambegia E, Choi C, Yaworski E, Palmer L, McClintock K, Maclachlan I.
Protiva Biotherapeutics Inc., Burnaby, British Columbia, Canada.

Activation of innate immunity has direct effects in modulating viral replication, tumor growth, angiogenesis, inflammatory and other immunological processes. It is now established that unmodified siRNA can activate this innate immune response and therefore there is real potential for siRNA to elicit non-specific therapeutic effects in a wide range of disease models. Here we demonstrate that in a murine model of Influenza infection, the anti-viral activity of siRNA is primarily due to immune stimulation elicited by the active siRNA duplexes and not the result of therapeutic RNAi as previously reported. We show that the misinterpretation stems from the use of a particular control GFP siRNA that we identify as having unusually low immunostimulatory activity compared to the active anti-influenza siRNA. Curiously, this GFP siRNA has served as a negative control for a surprising number of groups reporting therapeutic effects of siRNA. The inert immunologic profile of the GFP sequence was unique among a broad panel of published siRNA, all of which could elicit significant Interferon induction from primary immune cells. This panel included 8 active siRNA against viral, angiogenic and oncologic targets whose reported therapeutic efficacy was based on comparison to the non-immunostimulatory GFP siRNA. These results emphasize the need for researchers to anticipate, monitor and adequately control for siRNA-mediated immune stimulation and calls into question the interpretation of numerous published reports of therapeutic RNAi in vivo. The use of chemically modified siRNA with minimal immunostimulatory capacity will help to more accurately delineate the mechanism of action underlying such studies.

Comments

[iayork]

We routinely test our siRNAs to see if they induce any of 3 or 4 interferon-inducible genes, and compare to our standard -ve control. I am almost certain that we tested that particular GFP control and decided not to use it, but I can't remember if interferon induction was the reason (my lab notes for those experiments are elsewhere). On the other hand I know we tested a different GFP -ve control that induced sky-high levels of interferon.

Off-target effects due to interferon induction is a huge concern and something that needs to be specifically tested in every siRNA experiment.