Silencing eRNA – towards novel therapeutics for heart disease lncRNA silencing using Antisense LNA™ GapmeRs
Dr. Jaya Krishnan
Dr. Jaya Krishnan
is a group leader at the MRC Clinical Sciences Centre and Imperial College London, UK.
What is the main focus of the research conducted in your lab?
We are interested in hypoxia-sensitive enhancer templated RNAs (eRNAs) and their role in heart disease. eRNAs are a sub-group of long non-coding RNAs, templated from enhancer regions, thought to be involved in cis-mediated gene regulation.
How did you come to be interested in long non-coding RNAs?
This all started from some studies we did years ago where we observed that increased levels of Hypoxia Inducible Factor α (HIFα), a master regulator of oxygen sensitization and adaptation, are critical in the transition to heart disease. HIFα is ubiquitously expressed, but in spite of this, there exists a cell-type specific signature of HIFα function. Herein lies the mystery. One possibility is that HIFα cooperates with other transcription factors to establish this cell-type specific regulation. The other possibility is that cell-type specific gene expression is mediated through control of enhancer function. Using RNA sequencing we have identified eRNAs that may be HIFα-dependent.
My group is exploring if HIFα-dependent eRNAs play a role in mediating the cell-type specific signature of HIFα function and if these eRNAs contribute to the establishment of cardiac pathology or the progression to heart failure. The end goal here would be to identify eRNAs expressed only in cardiac cells of the diseased heart and to develop antisense RNA-based strategies to neutralize the effects of these pathology-induced eRNAs. Due to the restricted expression of the targeted eRNAs, we would hope that this approach would have the added benefit of achieving cell-type specific therapeutics. Our data so far suggests that this is indeed the case, at least in mice.
What is the specific aim of the current project?
The aim was to use LNA™ GapmeRs to knock down a specific eRNA in vivo
and see if this has an effect on the development of heart disease.
How did you design and select the Antisense LNA™ GapmeRs to use in your in vivo studies?
We sent the sequence of the eRNA to Exiqon to design the LNA™ GapmeRs and we tested 10 LNA™ GapmeRs in vitro
by transfecting them into 293 cells (ectopically expressing the eRNA) and primary heart cells. Two LNA™ GapmeRs worked very nicely and a third worked quite well. Then we selected the LNA™ GapmeRs to be synthesized in larger quantity for in vivo
experiments in mice.
How successful were your experiments using Antisense LNA™ GapmeRs in vivo?
We have promising data. We applied the LNA™ GapmeRs to wild type mice subjected to two different in vivo
- Transverse Aortic Constriction, TAC: a pathological protocol to mimic the human condition of pressure overload in the heart where the aorta leading to the heart is directly constricted
- One-kidney one-clip: a pathological protocol to mimic human hypertension which involves the clipping one of the large vessels of the kidney
With the pressure overload protocol, we injected the LNA™ GapmeRs at 10 mg/kg just before the surgery (day 0), and again on days 1 and 2, and on day 3 at a lower dose (5 mg/kg). Two weeks later we looked at cardiac function and the growth of the heart. We found that the LNA™ GapmeR conferred a partial rescue in the growth of the heart, however, in terms of heart function we saw a very nice complete rescue. Control mice subjected to the pressure overload protocol show about 15% decrease in contractile function, whereas the LNA™ GapmeR-treated mice perform as well as normal mice, so normal cardiac function is maintained.
The hypertension protocol gave us the most exciting and elegant data in vivo
. The whole experiment lasts much longer - about 13 weeks - because the heart needs 8 - 9 weeks to enter the disease state. We asked the question, can knockdown of this eRNA reverse a pre-existing disease phenotype – along the lines of therapeutic application. After about 8 and a half weeks, when we observed a clear shift to the pathological state of heart failure, we started delivering the LNA™ GapmeRs at 10 mg/kg every 7 days. What is nice about this experiment is that we saw a gradual improvement in cardiac function - both in terms of the dimensions of the heart and also in the function of the heart. “The hypertension protocol gave us the most exciting and elegant data [in vivo].”
How did you validate the knockdown using Antisense LNA™ GapmeRs in vivo?
We checked that our target eRNA is downregulated in the LNA™ GapmeR treated animals using qPCR on RNA isolated from the heart. Since eRNAs regulate flanking genes or genes in their vicinity, we also validated that the flanking genes were also altered in the same way that we observed in vitro.
Were there any specific challenges in this project?
We had a high rate of mortality when we tried to deliver the LNA™ GapmeRs intravenously (IV), so we moved to intraperitoneal (IP) delivery instead, and that worked much better for us. The mice are undergoing a very severe procedure anyway, so we didn’t want to add an additional stress with multiple IVs. When you do IV there is always a risk that some of the material you inject doesn’t actually go into the veins but comes out, so we couldn’t be confident that every single injection was done perfectly every time. The nice thing about IP is that it makes it much more reproducible in our hands at least. I have heard from my collaborator, who has done it for other groups, that IP delivery of LNA™ oligos works very nicely. We also did this with LNA™ microRNA inhibitors for a microRNA project, and there it also worked nicely.
How do you feel about your results so far?
When we first started we were not optimistic – we didn’t think it would work. So we were pleasantly surprised that it actually worked!
What, if any, was your previous experience with gene silencing?
Our main in vitro
model system is rodent cardiomyocytes and these cells don’t transfect well at all with siRNAs, so we have used a lentiviral shRNA system for gene knockdown in this system.
Why did you decide to use Exiqon’s LNA™ GapmeRs instead of any other techniques e.g. siRNA or shRNA to knock down your target of interest?
The problem is that translating viral systems to the clinical setting is not so simple. LNA™ GapmeRs were the only thing we could think of with possible clinical translational value – an antisense oligo that is stable and can actually enter and function in the nucleus of the cell.
The reason why we used LNA™ GapmeRs is because it’s fairly straightforward and we can use some of the established protocols that have been used for LNA™ oligos. That has actually worked quite well. “We used LNA™ GapmeRs because it’s fairly straightforward and we can use some of the established [in vivo] protocols.”
In parallel we have developed a procedure using adeno associated virus to get cardiac specific shRNA expression in mice, but these results have not come in yet.
What would be your advice to others about getting started with gene silencing experiments?
- Start with a simple experiment.
- Get good funding! In the ideal world you would do many experiments to figure out the ideal conditions, but this is quite expensive. We surveyed the literature in relation to heart disease and what LNA™ oligos had been used before to resolve heart disease in terms of dose. We used that as a guide, and were fortunate that we saw an effect with minimal optimizations.
- Validate your LNA™ GapmeRs in vitro very well before going in vivo. Our non-coding species is present in very low abundance, so it can be challenging to validate the results. This stage for us was the longest, to be confident we had a few LNA™ GapmeRs that we really trusted.
- If you can, start with a high dose that has been reported to be used in the literature then potentially go down if you think it’s too high. We haven’t observed any liver toxicity (we have not done any bioassays, or creatinine levels, but it seems that liver function is maintained just by looking superficially at the liver). In our case, we did not spend a lot of time optimizing the dose, but I suspect we could have seen better results with the pressure overload protocol if we had just used the high dose of the LNA™ GapmeR – 10 mg/kg.
- Our situation is quite special - the eRNA we are interested in is expressed only in the heart cells. I would imagine that if you are knocking down a coding or non-coding gene that is expressed in a few different tissue types, you might want to address if the knockdown occurred also in other tissue types that you may not be interested in. You may wish to rule out that knockdown in a different tissue type could be responsible for the phenotype you see, or confirm that it is the main organ you’re interested in that mediates whatever changes you see.
- Use other methods (e.g. adenoviral shRNA or mouse genetic knockouts) to complement the LNA™ GapmeR method and indirectly rule out any questions about off-target effects.
What are the next steps in the current project?
We are also generating a conditional mouse knockout model. It would be very nice to complement the LNA™ GapmeR data with a genetic model.
When and where will be hear /read more about your studies?
We are preparing a manuscript for publication.