By Doug Macron
A team of researchers led by Stanford University's Mark Kay earlier this month published data identifying the members of the human Argonaute family as the key players in the liver toxicity observed with high doses of shRNAs.
Notably, the findings also indicate that the toxic effects may not be solely the result of interference with the microRNA pathway, but also with other small regulatory RNAs that have yet to be identified.
The paper, which appeared in The Journal of Clinical Investigation, builds upon research Kay and his colleagues published in Nature in 2006, which showed that sustained, high-level shRNA expression in murine livers can result in severe toxicity and death (GSN 5/25/2006).
"The issues we wanted to follow up on from the Nature paper were … what the rate-limiting factors in RNAi-mediated knockdown [are], and whether or not these factors were actually involved in the toxicity events that we observed," Kay told Gene Silencing News this week.
According to that earlier paper, the investigators delivered shRNA-expression cassettes to hepatitis B transgenic mice using an optimized delivery vector based on duplex-DNA-containing adeno-associated virus type 8 and a U6 promoter, resulting in severe liver toxicity and death in a number of the animals.
The team observed a correlation between the shRNA levels and the toxicity, and conjectured that the shRNAs were overwhelming the mice's miRNA processing machinery.
Meantime, other groups, including one led by Harvard University's Debora Marks, have generated similar findings.
Describing their findings in JCI, the investigators said they had aimed to "further unravel the cellular determinants of in vivo RNAi efficacy, toxicity, and persistence, sparked by previous findings by us and others that high-level shRNA expression can cause cytotoxicity and fatalities in animals."
The research, conducted in collaboration with colleagues from the University of Heidelberg, showed that over-expression of exportin-5, which has been shown to be required for the nuclear export of small RNAs, enhanced shRNA efficacy in the liver of adult mice, but also increased hepatotoxicity, according to the paper.
The team also identified the four human Ago proteins as "downstream factors involved in saturation of endogenous cellular RNAi, all of which were able to interact with shRNAs in cells and mice," it adds. "In Ago/shRNA co-expression studies, Ago-2 was the primary rate-limiting determinant of both in vitro and in vivo RNAi efficacy, toxicity, and persistence."
In adult mice, vector-based Ago-2/exportin-5 co-expression "enhanced U6-driven shRNA silencing of exogenous and endogenous hepatic targets, reduced hepatotoxicity, and extended RNAi stability by more than three months," the paper adds.
The use of weaker promoters to minimize shRNA expression, meanwhile, reduced in vivo toxicity and resulted in greater than 95 percent persistent knockdown of hepatitis B virus and other transgenes in mouse livers for more than a year.
Kay noted that one of the questions arising from the 2006 Nature paper was whether the toxicities observed with shRNA expression at high levels also extended to siRNAs.
[ pagebreak ]
"If Ago2 is rate limiting, then the answer might be yes," he told Gene Silencing News. But in the end, "I still don't think this is in any way something that I would be that worried about in terms of [siRNA] therapeutics because there is a really wide therapeutic window."
At the same time, there are other toxicity issues facing the siRNA drugs space that should be of greater concern, such as "using siRNAs conjugated to nanoparticles, because there are a lot of issues about lipid in combination with nucleic acids causing innate immune responses through toll-like receptors," he added.
"I think that's more of an issue, and it is being well-addressed," he said.
"As a whole, [the] new data verify and expand our prior conclusion that adverse in vivo shRNA effects are highly complex and at least partly due to saturation of cellular rate-limiting components," the researchers wrote in JCI. "Still, we do not rule out additional … explanations that may also be consistent with our phenotypical observations."
Kay noted that other groups have created mice in which Dicer has been conditionally knocked out in the liver. "While there are some phenotypes observed with that, the severity of the toxicity we see with over-expression of shRNAs in our model seems to be much more than in the Dicer knockout," he said.
So, "I don't think that the toxicity we're seeing is simply the result of inhibiting microRNAs from functioning," Kay explained. "I think it's more complicated than that."
In JCI, Kay's team noted that "toxicity manifested later in the Dicer knockout mice as compared with our shRNA over-expression studies," and that this is "fully consistent with an expected much more gradual relief of miRNA-dependent gene regulation after conditional Dicer ablation, which will mainly affect miRNA processing, a consequence that may initially go unnoticed due to the long half-life of many miRNAs.
In contrast, shRNA-mediated overloading of the Ago/RISC complex will instantly block miRNA function and in the case of Ago2, moreover, impact miRNA biogenesis and maturation," they wrote. "Thus, although the phenotypes of Dicer or Ago loss are typically very similar … the underlying molecular mechanisms may differ."
"I think there are a lot of other small RNAs that are doing something, and it may not just be the lack of microRNAs [causing toxicity but also the impact on] other small RNAs or other potential points in the pathway that aren't well understood," he said.
As of right now, however, "I don't have an absolutely good idea that would explain it all."