Friday, August 31, 2007

3-stranded siRNAs

The detailed knowledge about the steps that occur during the natural process of RNAi has inspired scientists to come up with new ways to harness this pathway through the use of novel trigger RNAs. DNA-directed sh-miRs and sh-RNAs, Dicer-substrate RNAs, siRNAs, and single-stranded RNAs can all be used with the aim of obtaining activated RiSC effector complexes containing the guide RNA to seek and destroy target mRNAs.

Now, Bramsen and colleagues report in an Advance Access publication in Nucleic Acid Research on August 28, 2007 on the use of modified 3-stranded siRNAs. The fundamental siRNA design comprises a guide RNA paired with a passenger RNA to form a double-stranded RNA, 21-23 base pairs in length with optimally 2 nucleotide 3’ overhangs. In the case of 3-stranded siRNAs, the passenger strand is split into two separate strands.

3-stranded siRNAs have their roots in stunningly beautiful crystal structures of the Argonaute protein, the Slicer endonuclease of RiSC that cuts the target mRNA, bound to the guide RNA and a complementary RNA (Barford and Patel groups, 2005). In that work, the complementary RNA could have been the target mRNA, but just as well the passenger RNA in an siRNA. Not long after these structures were published, a number of groups reported almost simultaneously that, indeed, the passenger strand of an siRNA is also cleaved. While passenger strand cleavage is not essential for RNAi to work, due to the existence of a bypass mechanism, the Martinez group (Vienna) demonstrated that it greatly stimulates target mRNA cleavage by RiSC. As pre-cleaved, that is 3-stranded siRNAs, could also trigger efficient RNAi, it was concluded that the disruption of the passenger strand allows RiSC to quickly free itself from the passenger RNA for it to act on its real mRNA targets.

The paper by Bramsen et al. now demonstrates and discusses some of the advantages that the use of 3-stranded siRNAs may have. Surprisingly, 3-stranded siRNAs appear to be very forgiving to nucleic acid modifications. RNA modification is considered important to endow siRNAs with the proper drug-like properties, particularly stability, but also reduced off-targeting. Importantly, the authors find that modifications in the guide RNA strand, which is typically more sensitive, tolerated modifications that would abrogate gene silencing within the context of an uninterrupted siRNA. This is especially interesting when it comes to minimising off-targeting by the guide RNA. Small modifications to the second nucleotide from the 5’ end have already been shown to greatly diminish most of the microRNA seed-dependent off-targeting, and the increased flexibility that the 3-stranded siRNA design allows promises further improvements with regard to specificity and also efficacy. Being Danish, the authors suggest the use of locked nucleic acids (LNAs) to achieve this goal. In addition, there is absolutely no off-targeting by the passenger strand as it cannot function any more as a RiSC effector RNA.

You may wonder about the commercial implications of 3-stranded siRNAs. I could imagine the Danish biotech company Santaris to become involved, an oligonucleotides therapeutics company specialising in the use of LNAs. LNAs are highly potent in binding complementary sequences and show great potential for the therapeutic inhibition of microRNAs. Santaris’ patent policy, however, has led to a situation where LNAs are still not very widely used due to their often prohibitively high cost, and for unknown reasons you can almost be assured that when there is an LNA-based publication it involves a group from Denmark.

I would further be interested in whether any of the other academic groups that have contributed to the development of 3-stranded siRNAs claim any IP rights. One of these could be the Zamore group from the University of Massachusetts with ties to CytRx and Alnylam. Martinez on the other hand came out of the Tuschl lab, a co-founder of Alnylam. Lastly, Nastech, a company apparently keen in using anything but the fundamental siRNA design, has applied for a patent involving 3-stranded siRNAs (WO2006US42978 20061103). As often in science, meaningful progress is often made simultaneously by a number of groups and it is almost impossible to determine who deserves ultimate credit.

Tuesday, August 28, 2007

Impressions from Cold Spring Harbor mRNA processing Meeting

Every 2 years scientists from around the world gather in Cold Spring Harbor, New York, to discuss the latest developments in Eukaryotic mRNA Processing.

It was my second time. Four years ago, not long after RNAi had been discovered to operate in humans, I presented some data relating to the effect of Dicer knockdown on intergenic transcript abundance in an RNAi session. This time, there was no RNAi session, not even as part of the RNA turnover session. Instead, the Meeting focussed on the more traditional topics of the field, particularly mRNA splicing. This development illustrates how fast RNAi has grown, not only in numbers, but also in diversity of research areas so that it is not possible any more to accommodate it in one or two sessions alone.

This is unfortunate as I think both fields have a lot to learn from each other. One example where such an information exchange would be fruitful is in the rapidly growing application of high-throughput cloning and sequencing techniques for the identification of RNA populations. Specifically, I have not seen the CLIP technique, which allows the identification of RNAs bound to a protein of interest, being widely applied in the RNAi arena.

Similarly, I feel that the potential of targeting misspliced RNAs with RNAi for therapy is underappreciated due to a lack of sufficient communication. Almost one third of all human disease is thought to be due to splicing defects and I could well imagine RNAi therapeutics specifically targeting misspliced RNAs that cause pathogenic gain-of-functions. Type 1 myotonic dystrophy is another condition where a toxic RNA causes a disease that is essentially due to splicing mis-regulation. Importantly, Langlois and colleagues have shown before that it is possible to down-regulate the underlying mutated DMPK mRNA by DNA-directed RNAi (J Biol Chem. 2005 Apr 29; 280:16949-54).

However, no matter where conference organizers draw the line, they will have to recognize that RNAi and eukaryotic mRNA processing are intimately intertwined. I hope this realization will help speed up research in microRNA biogenesis for example where the nuclear events in particular remain surprisingly understudied. This cannot but change given that many of the leading scientists in the RNAi field have grown up studying traditional RNA processing and almost everybody at the conference has successfully used RNAi in their research. By the way, only once was it mentioned that somebody failed with their knockdown experiments, and, after much trepidation, it turned out that this person had been using morpholino antisense oligos, not RNAi.

Monday, August 20, 2007

Quark Biotech on Course to Submit 3 RNAi-based INDs in 2007

The speed with which Quark Biotech has been turning out clinical-stage siRNA drug development programs is quite remarkable. With one ongoing phase I study for age-related macular degeneration (AMD) and one submitted IND for acute kidney injury, the Fremont, California, based biotech company today announced plans to file a third IND in 2007 for an siRNA drug candidate, this time for preventing chemotherapy-induced hearing loss. More siRNA programs are already in the pipeline.

Quark Biotech is an interesting player in the RNAi Therapeutics field as it leverages genomic screening techniques to rapidly discover novel drug targets. Indeed, the discovery of RNAi in the era of the human genome could not have come at a better time. As RNAi can target virtually any gene, efforts like those of Quark Biotech will allow the genome-wide screen for drug candidates essentially based on their biological suitability and not limited by “druggability”. Further demonstrating the cross-fertilisation between genomics and RNAi, the inhibitor screens performed by Quark Biotech, Cenix BioScience, and a number of other commercial and academic groups, are largely based on RNAi itself. Thus, RNAi simultaneously functions both as a target validation tool and potential drug candidate and allows for such fast development time-lines.

Quark Biotech sees itself as part of a new type of triangular relationship in the RNAi space where its expertise and value-creative role is target discovery. It then licenses the rights to use RNAi against these targets from companies like Alnylam Pharmaceuticals and Silence Therapeutics, and partners with typically larger, deep-pocketed companies for clinical development and hopefully commercialisation of the drug. I, for one, will be watching very closely whether Quark Biotech’s aggressive development strategy will prove successful in the end.

Thursday, August 16, 2007

MicroRNA Sponges! A Gene Therapy Approach to Inhibiting MicroRNAs

A report from the lab of Phil Sharp, a scientific co-founder and advisor of Alnylam, in the August 12 online edition of Nature Methods demonstrates the inhibition of microRNA function in mammalian cells through the expression of so called “microRNA sponges” (Ebert et al. (2007): MicroRNA sponges: competitive inhibitors of small RNAs in mammalian cells. Nature Methods, DOI: 10.1038/NMeth1079). This comes less than a month after the publication on a similar process in plants in which microRNA activity was shown to be inhibited by target mimicry (3 August 2007 Blog: “Journal Club: Target Mimicry as a New Way to Regulate microRNA Activity”).

MicroRNA sponges are expressed RNAs that function by competing with the natural targets of a given microRNA for its binding and which consequently de-represses these targets. They are therefore functionally equivalent to synthetic antisense inhibitors of microRNAs, sometimes referred to as “antagomirs”, and by the same token should facilitate the functional investigation of microRNAs. Equally exciting is the prospect of using microRNA sponges for the therapeutic inhibition of microRNAs in a gene therapy approach. This new development is analogous to the situation in the RNAi arena where synthetic and gene therapy approaches exist to bring about gene silencing.

Importantly, Ebert and colleagues demonstrate that microRNAs sharing the same seed sequence could be inhibited by the same sponge. This illustrates that sponges work by the normal microRNA-target recognition process and the inhibition is not simply due to sense-antisense interactions. Feeding into this endogenous pathway may also explain why at least in this study competition for microRNA binding with sponges was more potent than with antisense molecules, and why sponges with bulged target sites were more effective than those with perfectly complementary target sites, at least when expressed from Pol II promoters.

Practically, one could therefore imagine a sponge silencing whole families of functionally and sequence-related microRNA families, something that the authors show cannot easily be achieved with single antisense molecules. This is likely to be therapeutically relevant, as it is known for example that members of so called “oncomir” families redundantly contribute to cancer progression. Moreover, due to the modularity of the sponge system, different microRNA binding sites could be incorporated into a single sponge expression cassette to inhibit structurally unrelated microRNAs.

Next to therapeutic applications in microRNA targeting, this study is also of relevance for the RNAi Therapeutics field as this paper confirms now for human cells that siRNA off-targeting through a microRNA mechanism not only raises safety issues, but may also reduce on-target efficacy. I think it would be well worth revisiting whether some of the siRNA modifications shown to reduce off-target gene silencing also enhance on-target gene silencing. Such studies may eventually allow for the development of even more potent siRNAs that work at lower concentrations.

Finally, I would like to point out that RNAs could also function as sponges when expressed from the quite potent RNA polymerase III U6 promoter. Similar to the normal U6 snRNA product, these sponges can only be detected in the nucleus- yet they successfully compete for microRNA binding. As it is unlikely that these RNAs can inhibit the maturation of highly structured microRNA precursors which are present in the nucleus of a cell, this data adds to the growing body of evidence that microRNAs and RNAi-related processes occur in the nucleus. What could be the function of such nuclear small RNAs? Personally, my favourite hypothesis on how microRNAs act is that they bind their targets in the nucleus following transcription but before theire export into the cytoplasm. This is in contrast to most models that hold that microRNAs mature in the cytoplasm and recognise their targets there.

It will also be interesting to find out whether Pol III-driven sponge activity, as shown for Pol II-driven sponges, is seed dependent as this would suggest that transcripts other than those generated by Pol II may functionally associate with the microRNA/RNAi silencing apparatus.

While it has taken the microRNA sponge concept almost 6 years longer than for expressed RNAi to emerge, it will now be very exciting to follow up on the potential implications for basic RNAi/microRNA biology and microRNA-based therapies. And with regards to investment opportunities, it is anybody’s guess as to who will pick up the IP on microRNA sponges since so many different technologies and corporate strategies overlap with, but do not completely cover it.

Saturday, August 11, 2007

A Tumultuous Week for RNAi Therapeutics Companies

Credit worries, earnings calls, a new significant biodefense contract, and flu RNAi news all combined to produce a highly volatile week for RNAi Therapeutics companies, both financially and scientifically.

It all started with Nastech’s announcement on Monday morning of a Friday webinar to discuss their flu RNAi development strategy. Coupled with hedge fund troubles, margin calls and short covering, this created the perfect scenario for an impressive run-up of more than 35% run-up in Nastech’s share price at the close on Thursday. In my last blog, I cautioned, however, that the press release accompanying Nastech’s earnings call on Wednesday contained a mere reference to positive in vitro data on flu RNAi, something that had already been demonstrated by various groups before.

Flu RNAi then received some more attention on Thursday when Alnylam announced that their flu RNAi program, in collaboration with Novartis, would be delayed due to in vivo efficacy and safety issues. An IND that had been scheduled to be filed by the end of this year will now be postponed until at least 2008. Apparently, Alnylam’s own data and data from the literature suggest difficulties in obtaining sequence-specific viral knockdown with current flu animal models. This does not come totally unexpected since there has been discussion in the field on the merits of purposefully combining the innate immune responses elicited by some, particularly liposomally formulated unmodified siRNAs with the sequence-specific inhibition of viruses such as HCV, RSV, and flu.

While I fully agree with Alnylam’s decision to ensure the highest quality and safety of their pipeline at an early stage, it would be interesting to know the exact nature of the siRNAs in question (modified or unmodified), delivery methods used (formulated or naked), and the types of animal models that were used. This is particularly relevant given that the company’s most advanced clinical program for RSV targets another virus of the respiratory tract. It should be kept in mind, however, that safe and potent pre-clinical RNAi efficacy has been demonstrated for RSV. Moreover, we should not be surprised to see such differences since although both viruses infect similar tissues, the exact biological distribution and viral kinetics may significantly differ between any two viruses.

The in vivo efficacy/safety issues were compounded by the reduced support from the US government for novel treatments in preparation of a pandemic avian flu. While federal support for RNAi addressing public health threats continues to benefit the development of the RNAi Therapeutics platform in general, this demonstrates once again that due to changing political climates, the government cannot be relied upon for direct commercial benefit. However, political interest in preparing for a possible avian flu pandemic is particularly strong in some of the emerging economies in Asia, particularly China, and only yesterday a study by a Chinese group was published in an advance online article of the journal “Antiviral Research” on the in vitro and in vivo efficacy of RNAi for H5N1 (Zhou et al.: “Effective small interfering RNAs targeting matrix and nucleocapsid protein gene inhibit influenza A virus replication in cells and mice.”)

Ironically, on the same day that the delay was announced, Alnylam reported that it had been awarded another significant US government contract of $38.6M over 33 months for the development of RNAi antivirals for the treatment of hemorrhagic fever viruses. Currently, there is no effective treatment for these viruses which are perceived as a threat to national security. RNAi meanwhile has proven to be the most promising treatment in animal models so far.

This announcement on Thursday seemed to have further stoked the fantasies in Nastech shares and the whole sector which recorded significant gains in the face of a big drop in the major indexes. However, expectations were somewhat disappointed by the Nastech webinar on Friday morning which failed to uncover major breakthroughs in flu RNAi. Also, an analyst question related to sequence-specific in vivo knockdown was not directly addressed. While I certainly appreciate the educational aspect of this webinar, it left me, and probably others, scratching my head why this company went out of its way to present these data at 5am Pacific Time as if market-sensitive data were about to be disclosed. Nastech shares gave up almost 8% that day.

Here are some more interesting bits and pieces from this busy week:

1) ISIS stated that no fatty liver (steatosis) was observed in pre-clinical animal models with their apoB100 antisense compound. This should be reassuring news for ISIS’ lead compound given the concern raised by other reports of fatty liver in mice following apoB100 knockdown.

2) Alnylam’s phase II RSV studies for naturally infected patients was postponed to the first half of 2008 after phase II experimental infection data are available. These are expected by the end of this year. This certainly makes a lot of scientific sense and illustrates Alnylam’s scientific data-driven flexible pipeline management approach.

3) On the patent front, Alnylam announced the issuance of Kreutzer-Limmer I in Canada, and Tuschl I in Australia. Given the complexity in the ownership of this particular patent, especially with regard to CytRx and the University of Massachusetts, I wonder why the Press Release failed to mention UMass as a licensor for Tuschl I to Alnylam while acknowledging Max Planck Innovation Gmbh, the MIT, and the Whitehead Institute for Biomedical Research.

4) Alnylam management states continued interest by Big Pharma and biotech for a piece of the RNAi platform. More significant deals are possible, if not certain. Also, Novartis is more likely than not to exercise its adoption license for Alnylam’s IP. Such deals should strengthen the company’s balance sheet beyond their guidance of over $435M in cash at the end of this year.

What a week!

Wednesday, August 8, 2007

Back to Basics: Does RNAi Exist in Humans?

Whenever I try to explain the promise of RNAi for human biology and medicine to friends and family, I feel torn by a conflict of staying scientifically accurate or running the risk of losing my audience by burdening them with too many details. You may ask “why” now- isn’t RNAi this elegant straightforward mechanism in human cells whereby double-stranded siRNAs silence complementary mRNAs? How would you react when I said that a number, if not the majority of notable molecular biologists hold that RNAi does not exist in humans?

RNAi is most easily explained by starting with long double-stranded RNA (dsRNA) that is then chopped up by Dicer into short interfering RNAs which then get incorporated into the RiSC complex. The siRNAs loaded in RiSC consequently seek out and destroy their targets by a slicing mechanism. However, long double-stranded RNAs are not suitable for specific gene suppression in humans due to an innate immune system called the interferon response. The interferon response recognises dsRNA greater than 30bp in length and consequently shuts down almost all protein translation in a cell. This is a mechanism by which the cell protects itself from viruses which often produce dsRNA intermediates as part of their life-cycle.

I therefore vividly remember myself in an elective class (“RNA World”) as an undergrad student at Edinburgh University, when my lecturer, Dr. David Tollervey, predicted that it would only be a matter of time that RNAi would be discovered in humans- 1 year before Tuschl and colleagues published their seminal paper on siRNAs. In retrospect, it is clear that his confidence must have come from the fact that the human genome was littered with RNAi-related genes, such as some coding for Dicer and Argonaute proteins. He also may have been familiar with recent publications on the discovery of RNAi-related small RNAs in plants (Hamilton and Baulcombe) and the finding that double-stranded RNAs were processed into similarly sized small RNAs in an Drosophila in vitro system (Zamore, Tuschl, Sharp, Bartel).

Of course, we all know now that it was Tuschl who had the genius to conclude that siRNAs would be the natural effectors of RNAi in humans without inducing the interferon response. Except that they were not. Of course, synthetic siRNAs would work, but years of searching for bona fide naturally occurring siRNAs in humans derived from long dsRNA failed to convincingly prove their existence. It turns out that the RNAi enzymes are all present in vertebrates for another, albeit related reason, namely microRNA-mediated gene regulation. MicroRNAs are a major class of small RNAs that are not derived from long dsRNA, but hairpin precursors and that have turned out to be ubiquitous mediators of post-transcriptional gene regulation and that require RNAi-related proteins for their biogenesis and function. It is in this pathway that experimentally introduced siRNAs perform their gene silencing function.

It therefore appears that in humans the interferon system may have rendered antiviral roles for RNAi unnecessary. This is unlike in many invertebrates which lack an interferon system and where the ancient antiviral function is still alive and kicking. A fascinating story of evolving overlapping biological pathways with a happy ending for keeping the therapeutic promise of RNAi alive, but when you ask a scientist about RNAi in humans- beware!

Market Watch: Nastech reported “positive non-clinical study results for [their] siRNA program for treatment of influenza [in tissue culture in vitro studies].” Following an announcement this Monday of a webinar on flu RNAi to be hosted by Nastech this Friday, frantic trading led to big gains in Nastech stock in anticipation of major RNAi news. Nastech and Alnylam vie for a potentially lucrative government contract for flu preparedness and progress in this area will be closely watched as flu RNAi may turn out to be the first commercial RNAi Therapeutics product. It is likely that NSTK will give up some of its gain in the wake of today’s news, while ISIS may gather momentum as their RNA modification IP estate is starting to yield a rich harvest. Alnylam is to follow with their report tomorrow.

Saturday, August 4, 2007

Journal Club: Target Mimicry as a New Way to Regulate microRNA Activity

Just how intricately RNAi-related pathways can be wired, is demonstrated by a study on the fate a nutrient-regulated microRNA in plants [Franco-Zorrilla et al. (2007). Target mimicry provides a new mechanism for regulation of microRNA activity. Nature Genetics, Advance Online Publication, doi: 10.1038/ng2079].

Unlike in animals where microRNAs recognise targets sites of incomplete base-pair complementarity to promote translational repression, microRNAs in plants largely target fully complementary sites in an mRNA. Like with perfectly paired siRNAs in animals, this leads to the destruction of the targeted transcript by endonucleolytic cleavage at the target site. This process is thought to be rapid and the small RNA is then free to recycle and target a new message. By contrast, the turnover rate for animal microRNAs is much less well understood. However, as was often the case in the relatively short history of RNAi-related research, this study in plants may offer us a clue about the kinetics of microRNA activity in animals.

Franco-Zorrilla and colleagues observe that the induction of a microRNA that is regulated based on phosphate availability and that has known target transcripts based on perfect microRNA-target site complementarities around the expected cleavage sites, is accompanied by another non-coding RNA (RNA that is not translated into proteins) that also has high base-pair complementarity to the microRNA, but with telling mismatches around the otherwise predicted cleavage site. Sure enough, this non-coding RNA is recognised by the microRNA, but not degraded. Through a series of elegant genetics, the authors demonstrate that this sufficiently diverts the microRNA, which is of low abundance to start with, so that it cannot act on their “normal” mRNA substrates any more. Hence, the non-coding RNA functions as a sink and regulates microRNA activity by tricking it to bind to itself. This strongly suggests that the turnover of the microRNA complex on incompletely based-paired targets is very slow and suppression requires a one-to-one microRNA-target site relationship.

If you have cared to read all this, I will now tell you what I think the implications are for the development of RNAi Therapeutics. When thinking about off-target effects, we have been largely concerned about the detrimental effects of suppressing unintended mRNAs, largely through microRNA-like translational suppression. However, if the plant system reflects RNA silencing kinetics in humans, then another consequence of off-targeting may be decreased on-target activity due to decreasing the pool of available siRNAs. Consequently, if it were possible to prevent such off-targeting, for example through chemical modifications and bioinformatics, then one could think about lower siRNA doses in the clinic. It is not clear whether the 2’O-methylation strategy pioneered by Dharmacon scientists to limit off-target silencing also prevents microRNA-like binding of an siRNA to incompletely based-paired RNAs, but the strategy certainly points in the right direction.
By Dirk Haussecker. All rights reserved.

Disclaimer: This blog is not intended for distribution to or use by any person or entity who is a citizen or resident of, or located in any locality, state, country or other jurisdiction where such distribution, publication, availability or use would be contrary to law or regulation or which would subject the author or any of his collaborators and contributors to any registration or licensing requirement within such jurisdiction. This blog expresses only my opinions, they may be flawed and are for entertainment purposes only. Opinions expressed are a direct result of information which may or may not be accurate, and I do not assume any responsibility for material errors or to provide updates should circumstances change. Opinions expressed in this blog may have been disseminated before to others. This blog should not be taken as investment, legal or tax advice. The investments referred to herein may not be suitable for you. Investments particularly in the field of RNAi Therapeutics and biotechnology carry a high risk of total loss. You, the reader must make your own investment decisions in consultation with your professional advisors in light of your specific circumstances. I reserve the right to buy, sell, or short any security including those that may or may not be discussed on my blog.