[SANET-MG] RNAi kills

[jcummins <jcummins@UWO.CA>]

RNAi is a genetic technique which has begun to be employed extensively 
in genetic engineering. The USDA plum C5 honey sweet being petitioned 
for commercial release is an example of RNAi to kill viruses. The paper 
below deals with the finding that RNAi may kill. RNAi is frequently 
active against organisms in widely different kingdoms such as 
bacteria-man, etc suggesting that RNAi genetic applications require full 
safety testing of the RNAi.
Nature Reviews Drug Discovery 5, 528-529 (July 2006) | doi:10.1038/nrd2104

Safety concerns raised over RNA interference
Simon Frantz

Top of pageAbstractHigh doses kill mice, but researchers maintain 
confidence in gene-silencing method

Warnings: As more is understood about the gene-silencing capabilities of 
RNA interference, more issues are being raised about potential safety 
concerns in therapeutic applications.
Gauging the progress of the next big thing in science is often 
difficult, as failures are often greeted with similar, if not greater, 
fervour as successes. No one doubts that the current flavour of the 
month, RNA interference (RNAi), is likely to be a scientific revolution, 
but in recent months several alarms have been sounded over how 
successfully it can be developed into safe and efficacious treatments. 
The latest warning call is a paper showing that mice died from liver 
toxicity after being injected with high doses of a certain type of RNA 
(Nature published online 25 May 2006 doi:).

This latest study has generated a lot of heat, although delegates 
attending an international gene therapy meeting the week after the paper 
was published remarked that the study generated a stronger response in 
the media than the RNAi community. "People in the field understood that 
this wasn't any kind of show-stopper — if anything, it offered further 
information to move things forward," says Mark Kay, Professor of 
Pediatrics and Genetics at Stanford University, and lead researcher in 
the latest study.

The study should, however, raise awareness of toxicity issues associated 
with RNAi. "The surprise wasn't that you got toxicity — the surprise was 
what the toxicity was," says Kay.

The emergence of RNAi as a possible novel form of therapeutic has been 
nothing short of meteoric. It has only been 5 years since scientists 
found that silencing the activity of genes with short sequences of RNA 
complementary to a gene's sequence could be carried out in mammals 
without triggering the endogenous immune response to double-stranded 
RNA. In that time, RNAi treatments based on introducing small, 
synthesized pieces of RNA called small interfering RNA (siRNA) have 
already entered clinical trials (see table).

But like most areas of drug development, the more we discover about 
RNAi, the more we are beginning to understand its limitations. Studies 
have revealed that some siRNAs can silence genes other than the intended 
target and can activate the interferon defence mechanism (see Nature 
Rev. Drug Discov. 2, 763–764; 2003). Now, Kay's group has shown that 
another form of RNAi-based therapeutic — one that is more like gene 
therapy, in that it packages genes encoding RNA molecules called short 
hairpin RNAs (shRNAs) into viruses — also raises potential safety concerns.

Kay's team found that within a week or two of injecting their 
experimental RNAi therapy for hepatitis B virus into mice, most animals 
began to die of liver toxicity. According to the paper, the source of 
the toxicity doesn't seem to be the RNA sequence or the virus, but 
rather the dose. The original promoter sequence used for driving 
production of shRNA created large amounts of RNA that overwhelmed the 
RNAi system. Switching to a promoter that made fewer RNA molecules 
reduced toxicity dramatically.

John Rossi, Chairman and Professor of the Division of Molecular Biology 
at the City of Hope's Beckman Research Institute in Duarte, California, 
tells a similar story about his studies on an shRNA-based therapeutic 
for HIV. Rossi found that high levels of shRNAs are toxic in tissue 
culture cells, but that a different promoter system can circumvent this 
problem. "Like previous safety issues with RNAi, all of this points to 
the need to optimize your dose and the fact that careful, controlled 
studies are needed to assess toxicity," says Rossi. "You've got to be 
careful when you are using any component of RNAi. It's an endogenous 
system that can be saturated at a variety of different levels." Rossi's 
team manipulate haematopoietic stem cells from a patient ex vivo, and 
this, says Rossi, would detect any toxicity issues with the treatment 
before the cells are re-introduced into the patient.

Irvin Chen, Professor of Microbiology, Immunology & Molecular Genetics 
at the University of California, Los Angeles, has also found toxicity 
with an shRNA directed against CCR5 for HIV therapy. But reducing 
toxicity with a less efficient promoter also reduces potency, he says. 
Finding better treatments therefore involves screening numerous RNA 
sequences to identify those that strike the best balance between potency 
and toxicity. "There is no way of predicting whether an shRNA is going 
to be toxic or not, or potent or not for that matter," says Chen.

Dose-dependent: Mice given RNA interference therapy died of liver 
failure resulting from the saturation of the RNAi process.
Chen has also found that toxicity occurs in primary cells, but not in 
established cell lines. "It emphasizes the fact that it is important to 
look in a system that is as close to the real life situation as 
possible," says Chen. Rossi agrees. "Most people are doing cell-culture 
work, and cells in culture are pretty hardy compared with primary 
cells," he says.

Why and how RNAi causes toxicity isn't fully understood. One proposal is 
that the lethal shRNAs compete with endogenous RNA sequences called 
microRNAs (miRNAs) for the same cellular machinery that allows the 
RNAi-mediated silencing process to take place. The prime suspect, 
according to the authors of the latest study, is the protein exportin 5, 
which shuttles RNAs into the cytoplasm, and is thought to be a 
rate-limiting step for shRNA and miRNA processing. High levels of shRNAs 
saturate exportin 5, preventing miRNAs that are vital for normal 
cellular development from binding. Kay stresses that he isn't certain 
that exportin 5 is the cause of the toxicity, and his team is now trying 
to overexpress and silence exportin 5 in the liver to see how it affects 
RNA processing.

The big question is whether the siRNA-based treatments that are in the 
clinic could have the same risk. In theory, this shouldn't be the case 
because siRNAs enter the RNA-processing machinery downstream of exportin 
5. Kay says that it is still worth looking to see whether siRNAs are 
toxic in this manner, though he concedes that it is difficult to get 
equivalent amounts of siRNAs into cells in a mouse liver to answer that 
question.

According to John Maraganore, CEO of Alnylam, researchers at the company 
have never seen this effect with siRNAs. Maraganore says that a study in 
more than 50 mice showed that siRNAs targeting the apolipoprotein B gene 
in the liver and jejunum produced no effect on endogenous miRNA levels, 
despite silencing the target mRNA by around 80% (Nature 432, 173–178; 
2004). When the same siRNA was tested in primates, transient and 
reversible liver injury were observed (Nature 441, 111–114; 2006), but 
Maraganore says this seems to be related to the silencing of 
apolipoprotein B, rather than the type of toxicity observed with the 
shRNA study. "In our experiments we have never seen an effect on miRNAs 
as it relates to the use of siRNAs," says Maraganore.

Kay is quick to point out that he still has confidence in RNAi, and 
hopes that his findings do not stall the development of RNAi-based drugs 
because researchers divert their energies towards investigating the 
mechanisms underlying RNAi toxicity. "Like any drug, if we had to 
understand total mechanism, we wouldn't have any drugs on the market," 
says Kay. "So mechanistic studies must go on in parallel, and any new 
idea gives you a rationale for redesigning your strategies and approaches."

********************************************************
To unsubscribe from SANET-MG:
1- Visit http://lists.sare.org/archives/sanet-mg.html to unsubscribe or;
2- Send a message to <listserv@sare.org> from the address subscribed to the list. Type "unsubscribe sanet-mg" in the body of the message.

Visit the SANET-MG archives at: http://lists.sare.org/archives/sanet-mg.html.
Questions? Visit http://www.sare.org/about/sanetFAQ.htm.
For more information on grants and other resources available through the SARE program, please visit http://www.sare.org.