A No-Nonsense Effort to Develop a Therapy that Works Across Genetic Diseases

Nonsense mutations prematurely end the translation of a gene into protein and can result in a serious deficiency. About 10 to 15 percent of inherited genetic diseases involve nonsense mutations. Alltrna is developing transfer RNA therapies designed to correct the problem in protein synthesis caused by these aberrations. What’s particularly exciting about the approach is that a single therapy has the potential to work across nonsense mutations regardless of the gene in which they occur. We spoke to Michelle Werner, CEO of Alltrna, about nonsense mutations, how the company’s transfer RNA therapies work, and why the approach has the potential to address so many diseases at once.

Daniel Levine: Michelle, thanks for joining us.

Michelle Werner: Thanks, Danny. I’m happy to be here.

Daniel Levine: We’re going to talk about nonsense mutations, Alltrna, and its efforts to develop a therapy that can universally address nonsense mutations across thousands of diseases. Let’s start with a few concepts though. How is RNA translated into protein and what is the specific role tRNA or transfer RNA plays in that process?

Michelle Werner: Yeah, so I mean it really starts with the central dogma from our Biology 101 days where a DNA gets transcribed into an mRNA, which then is translated into a protein. But when we unpack that part of the translation process from an mRNA, the way that an mRNA is structured is that it has these three letter codes, each one of those codes, codes for an amino acid. And when you have the right sequence of amino acids, this is what makes up a protein. Now, what’s really interesting is the role that a transfer RNA plays in this protein translation process. So, if you look inside the translation machinery of the cell, you have the ribosome which travels along the mRNA strand as it’s being decoded from a code to the amino acid. And inside the ribosome is the tRNA or the transfer RNA. And its role specifically is to read each one of these triplet codes on the mRNA to then decipher or understand which amino acid is being coded for. It then finds that corresponding amino acid in the surrounding cellular environment and transfers that amino acid to the growing protein chain, hence the name transfer RNA, because it plays this important transfer process for the amino acid. So that’s what the role of the tRNA is. And if you even just think about it because of its function, the tRNA is actually the physical link between the mRNA and the protein. So, without a tRNA, we would just never have any proteins.

Daniel Levine: And in this process, when it’s working properly, what is the role of a codon?

Michelle Werner: Yeah, so when this is working properly, the codon is these three letter codes on that mRNA and it codes for an individual amino acid, and then it’s the chain of amino acids that becomes the protein. Now, the codon on the mRNA actually precisely matches what’s called the anticodon of the tRNA. So the tRNA has this clover leaf structure and at one end is the anticodon loop, and it precisely binds to each one of these codons on the mRNA to really decipher those instructions for the amino acids.

Daniel Levine: Last concept I wanted to ask you about before we talk about what Alltrna is doing, is nonsense mutations. What are they?

Michelle Werner: Yes. So unfortunately, as I’m sure Danny, that not always is the code for an amino acid perfectly correct. And sometimes errors occur and one of those types of errors is what’s called a nonsense mutation. So, a nonsense mutation happens when one of these codes for an amino acid instead is mutated and instead of coding for an amino acid, it actually codes for a stop instead. And what happens is that stop becomes a signal then that the translation process should discontinue; that it’s reached the end of the coding sequence. And why, that then that translation stops at that time actually, again, comes down to the important function of the tRNA, because in nature when there is a nonsense mutation, there are actually no tRNAs that know how to bind to these stop mutations, and therefore the translation process terminates too early. And then the result is a truncated or shortened or oftentimes actually an absent protein. And this is what causes disease when you don’t have that full length functional protein as nature intended it in the healthy state.

Daniel Levine: Nonsense mutations occur across rare genetic diseases. One of the challenges with emerging genetic medicines is that they can be targeted in ways that are very effective but don’t work on the underlying problem. In the case of a nonsense mutation, how large a problem do nonsense mutations represent? How common are they?

Michelle Werner: Yeah, so it’s actually quite fast. So when we think about all the genetic diseases, there are actually 300 million patients or people around the world that have a mutational driven disease from all kinds of different types of mutations. Now, 10 percent of that 300 million or 30 million patients, in those patients that mutation is a nonsense mutation. And we at Alltrna would call this category of patients what we would call stop codon disease. So any genetic disease diagnosis that stems from a premature stop codon or a nonsense mutation versus other types of mutations like frame shifts or missense or duplications, for example.

Daniel Levine: Alltrna has developed a platform for creating tRNA therapies. What’s particularly compelling about this is that a single therapy has the potential to address nonsense mutations across rare genetic diseases. Can you explain how your therapies work?

Michelle Werner: Yeah, so what we think is really exciting about the tRNA is because that function that I just described of a tRNA is the same no matter which protein is being coded for, no matter which gene is affected, and actually it does the same job at each and every one of those codons. So when you have a nonsense mutation, we actually observe in genetics databases that these same nonsense mutations occur across many, many, many different diseases. In fact, we have seen data that says the same nonsense mutations exist across thousands of different diseases. And what we are doing at Alltrna is engineering tRNAs that actually can bind and read these premature termination codons or these nonsense mutations. So, as I described earlier, in nature there are no tRNAs that know how to bind to these premature termination codons, and that’s why that translation process terminates too early because it reads it as a stop instead of an amino acid. But what we’re doing is engineering these tRNAs that actually can bind to this premature termination codon and instead of actually reading it as a stop, it actually understands what amino acid should have been coded for should the mutation have not been there. So, as if it was intended by nature, and then it carries on the function that it would normally do. It would be recognized endogenously by the ribosome, it would find that corresponding amino acid within that cellular environment and transfer it to the growing protein chain, thereby it restores the protein translation process in a very precise way just as it should have happened if it had been a healthy cell. And the result is a full length functional protein, which is what typically would be the result if that mutation had not been there in the first place.

Daniel Levine: You’re using machine learning to look across tRNA biology. How large a world of naturally occurring tRNA are there and what’s the potential to engineer tRNAs for therapeutic benefits?

Michelle Werner: Yes. So machine learning plays a very important role component of our platform. So first and foremost, I would say in nature, tRNAs are actually the most abundant RNA that exist in cellular structure in general. But then when we think about the way that we design our tRNAs, and we design our tRNAs using two different approaches, one of which is designing to optimize the sequence of the nucleotides that make up the tRNA, and there are about 76 nucleotides in a tRNA, and there are many different variabilities within that sequence optimization that occurs. Then we also then optimize those sequences using chemical modifications. Think about this as putting decorations on the tRNA to enhance different properties, and I’ll come back to that in a second. But when we do put these decorations on our tRNAs, there are actually about 120 or so different chemical modifications that we can place at each position of the 76 nucleotide tRNA structure. So when we combine the possible numbers of sequences plus the numbers of chemical modifications at each position, the combinatorial space is actually massive. You can think about it as there are more tRNA patterns possible than there are atoms in the universe. So, that’s correct more than there are atoms in the universe. Hence without a machine learning engine to help with that design process, it’s impossible for any brilliant Nobel laureate to be able to do this on their own.

Daniel Levine: The company uses the language of synthetic biology. It talks about the cycle of design, test, and build and learn. I’m wondering though, to what extent you’re able to engineer modifications into tRNA to improve its therapeutic properties.

Michelle Werner: Yes. So that comes back to the point of why do we do these chemical modifications? And we don’t just do them because we can, we do them because we believe it’s going to serve a purpose clinically. So as we advance towards the clinic, and what we’ve learned through the high throughput screening on the machine learning guided design of our tRNAs and going through hundreds of thousands of tRNA patterns is that not all engineered tRNAs are created equal and that certain chemical modifications yield different results. For example, we can use these modifications to increase the activity of our tRNAs by activity, I mean their ability to read through the premature termination codons. We also know that we can use these chemical modifications to enhance the potency of our tRNAs. And we’ve shown data now, both in vitro and in vivo that the sort of newer generations of our tRNAs, the ones that we’re designing now versus the ones that we were designing a year ago or so, are far, far more active and more potent than the earlier generations. And we think this is really critical because now as we’re advancing towards the clinic, of course, it’s making sure that we’re able to rescue proteins to levels that we think are going to be clinically meaningful to patients and to do so at a dose and a schedule that’s also going to be manageable. So this is all part of our design process. We can also design for other features as well, such as the immunogenicity profile of these tRNAs, their ability to select with a very high accuracy that correct amino acid that should have been coded, for example. These are other things, features that we can design and screen for.

Daniel Levine: So, beyond the tRNA, what are the components of Alltrna therapy? Does it require a vector? Are there elements to affect its targeting, half-life, or other activity that are also packaged into that?

Michelle Werner: Yeah, so our engineered tRNAs are an oligonucleotide, and as with oligonucleotides, so you’re familiar with such as mRNAs and siRNAs, which you’ve probably heard about, they do require a delivery vehicle in order to be able to get to the site of action. And so we have made a strategic choice, at least in the first instance, to use a lipid nanoparticle as a delivery vehicle. And the reason why we’ve chosen LNPs is because LNPs now have obviously been in millions of people around the world, especially part of these Covid vaccines. We know their profile very well, we know the safety associated with them and they’re readily available. And what we want to do, because tRNA is a brand new modality, it’s never been used in patients before. We wanted to, I’d say, de-risk delivery by leveraging a very well-known and understood delivery vehicle, then combining with a more novel delivery vehicle. So that’s our strategic choice for the first program. Now that being said, that does mean that our initial program will focus on liver diseases because LNPs go to the liver very selectively. But we do believe that our engineered tRNAs can and will play a very important role for diseases outside of the liver. And muscle diseases or CNS diseases are very high on the list of those that we’d like to tackle next. And so we’re already starting to explore different delivery mechanisms that might allow us to access those. I should also say, coming back to the synthetic biology, that the chemical modifications that we’re doing on our engineered tRNAs also help us make progress towards conjugating our tRNAs or being able to have ligands or linkers attached to these tRNAs, which may also enable delivery through other mechanisms in the future.

Daniel Levine: And what’s known about your ability to deliver it to various cell types and tissues in the body? Is delivery the biggest challenge?

Michelle Werner: Yes. I mean, I’d say that’s not anything specific to Alltrna or the work that we’re doing with tRNAs, but definitely in the world of RNA based therapies, and oligonucleotides broadly, delivery is absolutely crucial. And so the good news is that the industry as a whole is also really trying to tackle that and to help unlock and enable broader delivery mechanisms. And I do think that we are starting to see some of those emerge. And so, I’m really looking forward to having that technology advance at the same time that our tRNA technology advances. And I do think the combination of those innovations will be able to bring a meaningful difference to patients across a number of different diseases and tissue types.

Daniel Levine: You mentioned pursuing CNS diseases. Will these cross the blood-brain barrier or do you have to consider something like intrathecal delivery?

Michelle Werner: On their own they won’t, but certainly when it comes to the delivery mechanism, we would certainly be taking that into consideration to be able to tackle that challenge. One of those very challenging things related to the CNS, but also there are others that are making progress on that front as well, which we look forward to learning from.

Daniel Levine: One of the interesting aspects of developing a treatment across indications is the potential for conducting basket trials that look at a single therapy in multiple indications at once. What’s the potential here to do that? And have you had any discussions with regulators about that potential?

Michelle Werner: Yeah, so what I would say is this is one of the things that I believe is so unique about tRNA is that we have a real opportunity to tap into. And basket trials are not novel. They have been around the block before. In fact, in the oncology space, which is where much of my background is, there are many medicines, close to a dozen, that have been approved using basket trials. And this is where we have patients who technically have a different disease diagnosis, but they all have the exact same mutation. So they’re common in that way and they’re given the same therapeutic in that trial. And we do think that’s something that we can leverage also across rare diseases. So, our plan, and again, coming back to our formulation with a lipid nanoparticle, would be to unlock or to bring forward a basket trial of rare genetic liver diseases, of which there are about 400 of these types of diseases that we can choose from. Now across all of those 400 rare genetic liver diseases, we see the exact same nonsense mutations across the board. And so, we are looking to bring together a group of patients with the same mutation, same amino acid stop combination that our first engineered tRNA would be able to address and to be able to give them the same engineered tRNA in that clinical trial and to look at outcomes across the board. What’s really great about this strategy is when you think about other types of technologies, even novel technologies like gene therapy or gene editing or mRNA, all of those must take a gene by gene or disease by disease approach. Now, this is a massive problem when you have close to 10,000 genetic diseases. If we’re tackling them one by one, it’s going to be forever. We’re going to be long gone, Danny, before we really make a dent in all 10,000 of those. So, we really have to think about ‘a many diseases at a time’ strategy for us to be able to do that. And because of the way that the tRNA works, we have this unique opportunity to do exactly that. So yes, this is our strategy for our engineered tRNAs. We are already starting to make progress in designing what our initial clinical development plan will look like. And what I’m very excited about is that even when we hear the narrative of health authorities, be it the FDA, but also EMA and the MHRA, so our European and our UK counterparts as well, they all have already recognized that ‘a many diseases at a time’ strategy must be the approach to really tackle rare diseases as a whole. Because so many of these rare diseases, unfortunately, are just too rare in order to warrant their own innovation specific for their own disease and to have an own clinical trial. That’s devastating if you’re a patient that has one of those diagnoses. So what we are hoping to do is really to enable clinical innovation and exploration and opportunity for some of these patients and families who may not otherwise be able to have that, and basket trials are the way to go for that.

Daniel Levine: As you think about a basket trial, what are the elements you’re considering for determining what indications you might include?

Michelle Werner: Yeah, so I mean there’s several factors that go into the patient selection for any of these clinical trials. What we really need to be able to do is identify patient populations that we think can benefit from this technology of reading through these premature termination codons; that the disease is well understood enough; that has to have biomarkers, for example, that can measure an impact of a therapeutic like ours. And so those are the ones that we’re looking to include in the baskets. Now, I don’t think we will include all 400 diseases in a basket, but I do think there’s about 20 or so different diseases that we will choose from in order to be able to have this initial basket trial designed. And what we’re hoping to be able to achieve in the end for patients is to have a disease agnostic but mutation specific type indication that would allow broad use across these rare genetic liver diseases.

Daniel Levine: And what’s the potential for tRNA therapies beyond nonsense mutations? Are you looking beyond those?

Michelle Werner: Yeah, I mean, so it’s still really, really early days. And for now, at Alltrna we’re very much focused on developing initial clinical proof of concept in nonsense mutations. But as I mentioned before, there are other types of mutations such as missense mutations or frameshift mutations, and these are ones that we also agree an engineered tRNA very well could be used to address those populations as well. And so that would certainly be an area of interest for us for a future exploration. But we need to walk before we can run Danny, and that’s why we’re focusing on nonsense mutations first.

Daniel Levine: Alltrna is a Flagship Pioneering company. Flagship has launched a number of platform-based companies. Is there an opportunity to work collaboratively with other technologies in the Flagship portfolio?

Michelle Werner: Absolutely, and I will say this is actually one of the things that drew me to Flagship in the first place, versus maybe taking a similar role at a number of different companies. I do think there are unique benefits from being within the Flagship ecosystem. So first and foremost, I should say, Flagship has lots of experience with platform companies, especially RNA based technologies. Of course, everybody has heard of Moderna by now, which is an mRNA company. And I do like to think that Alltrna stands on the shoulders of giants when it comes to the RNA field, including companies like Moderna who have been working in this space for over a decade. So, a lot of the learnings, a lot of the challenges that have been overcome by Moderna are also things that we believe we can already get a head start on in the work that we’re doing here at Alltrna. So, that’s one particular advantage. The other advantage is right now there are almost a hundred Flagship companies that have been brought forward. Being part of the ecosystem like we are with Alltrna gives us a line of sight to some of these emerging technologies, be it sort of RNA based technologies or delivery technologies, all of which could be a true asset or very complementary to the work that we’re doing at Alltrna. And by having an early line of sight to that, it does allow for those opportunities to start exploring how some of these platforms may complement each other, either from a discovery perspective, but maybe ultimately from a clinical perspective. And those are all things that certainly are encouraged within the ecosystem, and I’m sure that Alltrna will benefit from as well.

Daniel Levine: In August 2023, the company announced a series B financing of $109 million. How far will existing funding take you?

Michelle Werner: Yeah, so we were really delighted to be able to bring on a number of top tier investors across the board and to have that financing put in place last year, we are using the proceeds from our Series B in order to advance the most promising of our tRNA candidates towards the clinic. And we do believe that we will be able to achieve that with these proceeds. And what I will say is certainly with the data that we’ve presented recently at ASGCT, we’re starting to share, as I mentioned, some of the second and third generation that are being designed now. And certainly the proceeds from that series B have done wonders for us to be able to advance our ability to design better and better. And these are the ones that we’re advancing towards the clinic.

Daniel Levine: Michelle Werner, CEO of Alltrna. Michelle, thanks so much for your time today.

Michelle Werner: It’s been a delight. Thanks so much Danny.

This transcript has been edited for clarity and readability.

The RARECast podcast is made possible through support from the Global Genes’ Corporate Alliance. The members of the Corporate Alliance support Global Genes’ mission and programs, work to meet the vital needs of people with rare diseases, and address inequities they face. To learn more about the Corporate Alliance or how your organization can become a member, click here.