A Cat who Curates Rare Mice

Mouse models can play an essential role in allowing researchers to understand rare diseases and develop drugs to treat them. Cat Lutz, senior director of mouse repository and in vivo pharmacology genetic resource science at The Jackson Laboratory, researches mice as a model for human neurodegenerative disease. The lab’s mouse repository and Rare and Orphan Disease Center today features more than 12,000 unique strains including more than 1,700 live colonies that are distributed to the scientific community. We spoke to Lutz about the role mouse models play in rare disease research, how new gene editing technologies are changing the development of mouse models, and why new technologies are unlikely to displace their use anytime soon. 

Daniel Levine: Thanks for joining us.

Cat Lutz: Thanks for having me. It’s nice to be here today.

Daniel Levine: We’re going to talk about your work at the Jackson Laboratory, the role mouse models play in rare disease research and drug development, and how you work with both industry and patient organizations. Perhaps to begin with, we can start with the Jackson Lab itself, which has been around for more than 90 years. For people not familiar with its work, what is it and what does it do?

Cat Lutz: The Jackson Laboratory is a non-profit research institute. Our location is Bar Harbor, Maine. It’s a lovely area to work. I’ve been here for quite a number of years and it’s a pleasure to interact with all of the researchers here at the lab. We have a focus at the Jackson Laboratory on research in mouse genetics and mouse genomics. Since its inception, the Jackson Laboratory has primarily focused on working with mice and mouse models of human disease. Our mission has been steadfast over many, many years but as technology and science have developed, so have we. We have gone from a traditional mouse genetics facility and research center to now focus on genomics, as well as drug efficacy testing. We use all the technologies at our disposal to help move this research along.

Daniel Levine: You’re a neuroscientist by training. Today, you’re a senior director of the mouse repository in the in vivo pharmacology services. How did your work lead you there?

Cat Lutz: Yeah, it’s kind of interesting. I always had a passion for neuroscience and was quite drawn to it both in my undergraduate degree and in my graduate work. I started working at the Jackson Laboratory and fell into the mouse genetics part of things. It was a nice marriage, if you will, between the neuroscience and the mouse genetics. From there, I was working on a lot of neurological diseases using mice as a model. For example, epilepsy was one of the areas that I did my graduate degree in. Then over the years, I think genetic engineering got to be too tempting to not really plunge into. Looking at various mouse models, not just in neuroscience but across a number of different therapeutic areas, seemed to be a natural way to look at how the Jackson Laboratory could serve its mission to the scientific community—that is to provide these mouse models to the rest of the world. In addition to being a premier research institute for mouse genetics, we also serve as a mouse resource and repository, the idea that anybody can get a mouse that they need to work on, and people, possibly in California, or a researcher in Texas, or a researcher in Ohio, could work with the exact same mouse model. This gives a lot of credence to things like rigor and reproducibility. The mouse repository portion of that was what I was drawn into—the idea of using genetic engineering to create mouse models, but then also helping to facilitate the biomedical community’s research by providing them as a resource for their research. Then the pharmacology was something that followed naturally. When you have all these really interesting mouse models of disease, the idea that you want to treat those particular diseases with therapeutics that could be moved from the mouse models to the clinic, for me, was a natural progression as well.

Daniel Levine: Why are mouse models so critical for rare disease drug research and development?

Cat Lutz: I think that a lot of people don’t necessarily see mice as good models for human disease, but, in fact, most of the genes are conserved at a very high level in the DNA between mice and humans, both being mammals. The idea is that the biology, in general, of the pathways and the mechanisms can be better studied sometimes in the mouse then they can be in the human population. The reason for that is mice can be inbred, unlike people. You can hold the genetic background of a mouse steady, and it’s a really powerful component to research because now the only thing that you’re really looking at are mutations and environment. Whereas in people, everybody is a little bit of a mix of genetic material coming from multiple generations and DNA that is segregating. In a lot of ways, we’re able to simplify human genetics and human disease by working in mouse systems. Like I said, the reproducibility is interesting when it comes to gene conservation. Then, you’re working with the mammalian system. Other mammalian systems like non-human primates and monkeys are certainly a good model for humans as well, but they’re not as easy to work with. The lifespan, the reproductive biology, and the manipulations are not as easy to do as they are in a mouse. The mouse, over many years, has become the ideal organism for looking at genetics and studying human disease for the reasons I just mentioned. It’s fast, it’s efficient, and there’s a lot of technology around mice that has helped.

Daniel Levine: How predictive are they for neurological disease and what are their limitations?

Cat Lutz: I think for neurological diseases, it’s always been a little bit interesting and it’s also been a little bit tough at times as well. When you think about neurological diseases that are complex, let’s use Alzheimer’s for an example. Alzheimer’s disease is a complex trait. It’s not a single gene defect, as we all know. So, studying complex diseases is even harder to do in the patient population, but if you can reduce it to practice in a mouse model by manipulating those genetic components that you know are part of the story, you can make a lot of progress. I think the other hard thing about neurological diseases is that it’s sometimes difficult, especially in pediatric neurology, to look at the differences between neurodevelopmental issues and problems. The things that happen in utero or perhaps in development of the embryo can be a little bit harder to dissect. We have a lot of genetic engineering tools that help us introduce mutations before and after embryogenesis that helps with that. Then, maybe the most challenging thing, in the past, with treating neurological disorders in general has been drugs and therapeutics that will cross the blood-brain barrier. If you’re looking to treat things like epilepsy or other neurological disorders, you have to make sure that the treatment that you’re working with has the ability to cross the blood-brain barrier in a way that can be effective. Again, I think some of the technologies that we have now to facilitate that, not just with small molecules and small compounds, but with other vectors that can transverse the blood-brain barrier, provide a lot more utility in therapeutic development than we had even just five years ago.

Daniel Levine: There are a number of rare conditions where there have been naturally occurring animal models that have really helped advance the development of new therapies. Is there a significant difference in what can be learned from an animal model of a disease that’s naturally occurring in that model compared to an animal model where the diseases are engineered?

Cat Lutz: I think we always like to think about if the disease occurs naturally in the animal model that we’re looking at. For example, if you look at macular degeneration, which is a significant problem in the aging population for humans, it’s very challenging to study in a mouse model because a mouse simply doesn’t have a macula to begin with. So, you’re looking at physiological differences. Even then, there’s so much of the structural biology and the physiology that’s similar that maybe those types of things could be considered minor hurdles that don’t necessarily stop you from doing what you need to do using the mouse model. I think that naturally occurring mutations in animal models offer sometimes a limited insight into that particular disease. For example, if you have a dog model for ALS or a hearing disorder or loss, it’s a trait that’s been acquired and possibly selected or inbred within that animal model. Sometimes it’s very limiting because it only allows you to see one particular mutation, for example, in age-related hearing loss, epilepsy, or neuromuscular disease. Whereas in the mouse models, not only can you genetically engineer those mutations in every gene that you can identify as is probably causative with that [disease], you could even go down to the different mutation level. That provides you a broader spectrum of phenotypes and clinical presentations that you can study and get a broader picture of how that disease is going to progress.

Daniel Levine: We’re at this amazing time of innovation right now, where we’re seeing the advent of many new tools for genome editing. Have these tools change the speed or process of developing mouse models?

Cat Lutz: Absolutely. I hate to date myself in looking back at some of the technologies that we had in the past. You mentioned spontaneous mouse models or spontaneous mutations. For many years, that’s what we used to study the diseases. We would find a mouse, for example, in some of our breeding facilities that maybe was ataxic or that was starting to have seizures when it was handled, and we would use those spontaneous mutations primarily for what we would study. Probably in the early nineties is when genetic engineering through embryonic stem cells started to really take hold. Now it was possible to look at things a little bit differently. Instead of just studying the genes that presented themselves spontaneously, you could now engineer them. Even then the technology wasn’t as advanced. It was a little tough, you were using mouse embryonic stem cells and manipulating them. Sometimes, in the early days, you might consider that only a few select labs could do that kind of work. The technology itself was a little time consuming and clunky. There was a lot of trial and error associated with it and a lot of limitations in what you got on the other end of the mouse model, if you were fortunate enough to introduce the mutation that you wanted to. I think with CRISPR Cas9, which is a genome editing technology that we’ve really started to use in the past few years, the technology is almost ridiculously easy in its application. It can be used on a variety of different genetic backgrounds and inbred backgrounds. So the ability to create a genetically engineered mouse is now something that can be done in just a few short weeks.

Daniel Levine: When you hear advice given to patient organizations that are interested in advancing research and finding treatments, there’s a predictable to-do list that includes things like getting a website up, starting a registry and natural history, and often high on that list is developing an animal model or a mouse model. At what point should patient organizations look to that task?

Cat Lutz: That’s a good question. It’s something that, as more patient organizations start to assemble, there is definitely the understanding that the end goal is a therapeutic. The quicker they have a mouse model that can recapitulate some of the key clinical presentations that would be seen in their patient population, the better off they’re going to be. Certainly the things that you mentioned, self-assembling, getting a natural history, and really understanding, for rare diseases, what the basis of the commonalities and underlying clinical presentations are, is really critically important. A lot of times these rare diseases are pretty heterogeneous in the way they present. Sometimes the patients, from one individual to another, look very different. It is important to understand what the full spectrum of the clinical presentations are. Patient organizations that assemble for the purpose of curing their children or at the very minimum finding a treatment, very quickly now recognize that animal models can help in the research—not just to understand the disease itself, but to actually be a model for directly testing therapeutics. If you had a mouse model, for example, that had epilepsy and you were looking for a therapeutic to treat that particular disease, you would want to genetically engineer that precise genetic mutation into the gene that is causative in your patient population. If the mouse model that you genetically engineered also had epilepsy like you saw in your patient population, now you have a very promising model to help test different therapeutics and know if this particular therapeutic is going to be helpful in the amelioration of that particular epileptic phenotype. They’re quite powerful tools. I think that most rare disease patient organizations understand that they’re working with a small population of individuals and getting FDA approval for particular therapeutics is going to require a concerted effort through research to have outcome measures that are going to be convincing enough for the FDA to approve a clinical trial. A lot of the times that convincing data comes from the mouse models themselves.

Daniel Levine: How expensive is this to develop a mouse model for a patient organization?

Cat Lutz: It’s a lot less expensive than it used to be. I’ll say that. The genetic engineering approaches that I mentioned with CRISPR Cas9 have been reduced to practice and a pipeline approach. I think making the mouse model itself is not very expensive. I think the research that goes into characterizing the mouse model is where some of the expense can come in. It requires a team of people—not just one particular laboratory, but a team of individuals who are committed to working together to advance the understanding of the mouse model as quickly as possible. Having one individual working on the characterization of a mouse model can be slow and time translates into dollars. I don’t think that’s what a lot of these patients communities are looking for. Thus the idea of creating consortiums and individuals who can take the making of the mouse model in a pipeline approach, get it characterized as much as possible, and then have the specialists come in, who work on that particular disease area or specialize either clinically or in a research component. That really is the key to success that is efficient in terms of time and money.

Daniel Levine: Once a model is created, what’s got to be done to sustain it and share it with researchers?

Cat Lutz: I think that this is where the Jackson laboratory really shines in its research mission. Not only do we have the capabilities of genetically engineering these mouse models, it really is our mission to provide these to researchers around the world. We have over 12,000 strains of mice that we’ve acquired with various genetic mutations or spontaneous mutations, genetically engineered, inbred strains, you name it, we are probably the largest repository for mouse models. We also have recognized that the advancement of therapeutics for any disease is really hampered by the accessibility of resources. In addition to all the great research that goes on at the Jackson Laboratory, we’re very well known for the distribution of our mouse models as a resource component. We have a catalog, if you will, that individual researchers who are working on a particular disease can simply call and order and have that mouse model delivered to their vivarium so that they can conduct the research. It doesn’t involve a collaboration. It doesn’t involve a lot of paperwork or legal documentation. It’s simply there to try to be as efficient as possible in terms of getting these resources into the hands of the people who can use them.

Daniel Levine: In that regard, how unique is what the Jackson lab does?

Cat Lutz: I think it’s pretty unique. You definitely see that the model of a mouse repository is definitely mimicked around the world. You see mouse repositories, for example, in the European Union, you see them in Asia, but I think the Jackson Laboratory was definitely the first. A lot of individual repositories around the world have worked to mimic this idea of availability, efficiency, and resource sharing, which, if imitation is the best form of flattery, I think that we could consider ourselves very well thought of. I think it’s important to have these resources in different countries. Although I should say that we have really tight communications with these individuals around the world to make sure that we’re not duplicating resources either. The Jackson Laboratory distributes mouse models not just to the scientific community in the United States, but all across the European Union and Asia. We even have a facility now that we’re opening in China so that we can better serve that community.

Daniel Levine: You’ve been involved on the creation of many mouse models. I thought as a practical example, you can discuss the work you did with the SMA Foundation and the role that the mouse model played in the development of the first therapy for a neurodegenerative condition. For listeners not familiar with spinal muscular atrophy, what is it, and how does it manifest itself and progress?

Cat Lutz: Spinal muscular atrophy is a devastating pediatric disease with a relatively high incidence in the population, almost as high as cystic fibrosis. You don’t necessarily hear a lot about spinal muscular atrophy because most of the babies that are born with SMA usually die before they’re about two years of age. There are varying degrees of severity in SMA, but essentially it’s a neurological disease very much like Lou Gehrig’s disease where motor neurons in the body gradually die. So, babies are normally born healthy, or somewhat healthy, then lose their ability to have basic functions. Most of those babies never sit, and some babies will sit but never stand. In severe cases babies, first described as floppy baby syndrome, because their muscle tone and their overall weakness, at the end of their lifespan the motor neurons stop innervating the muscles usually resulting in the inability for those children to breathe. Most of the children with SMA will die of respiratory arrest. Again, it looks very much like ALS and Lou Gehrig’s disease in babies.

Daniel Levine: What was the work you did and how did it impact the development of Spinraza, the antisense therapy that would become the first approved therapy for the condition?

Cat Lutz: SMA was a huge success story. A lot of individuals, including the National Institutes of Health, recognize that because of the genetics of SMA there were a number of potential therapeutics that one could develop that might have a chance at working in the clinic. It was an approach with therapeutics that had a lot of shots on goal. Those shots on goal were antisense oligonucleotides that corrected the genetic mutation at the DNA level, they were gene therapy where the missing protein was replaced using an AAV vector-based system, and then there were lots of other small molecules that were involved in altering the splicing of that particular gene. When you think about one disease with multiple opportunities for therapeutics, it was extremely attractive. I think the thing that did accelerate Spinraza, in addition to a number of other FDA-approved drugs now for SMA, were the mouse models. Not only the mouse models that were developed by Arthur Burgess at Ohio State, but those that were improved upon in different areas, including the ones that we had here at the Jackson Laboratory. The biggest contribution that we made was being pivotal in the dissemination of those mouse models. We distributed those models to researchers by the thousands. It wasn’t necessarily a resource that was difficult to come by because they were at the Jackson Laboratory and anybody could get them. In addition to creating some of the mouse models and distributing the mouse models that were already in existence, we started doing a lot of drug efficacy testing on those mouse models because we have incredible resources here in terms of being able to work on these mouse models in ways that may be prohibitive to an academic institution. Our mouse breeding facilities as well as our capabilities are pretty fantastic. I often refer to the Jackson Laboratory as a sort of this mini hospital for mice where any diagnostic treatment or readout that you need to have can occur. The making of the mouse models, the distribution of the mouse models, the characterization of the mouse models, and then ultimately the testing of a lot of the drugs really came together to help facilitate some of what we now have as Spinraza and some of the other FDA-approved drugs now for SMA.

Daniel Levine: As new technology emerges and through a combination of organ-on-a-chip technology, organelles, lab-grown organs, or artificial intelligence, do you ever imagine mouse models being replaced?

Cat Lutz: No, I don’t think so. I would say that throughout the years, throughout the decades, there’s always been advancements in either cell-based biology or lower model organisms. A lot of the work that we do for pharmacology and drug testing actually occurs first in cell lines. Then sometimes we’ll work in Drosophila fruit flies, and sometimes we’ll work in zebra fish or C. elegans. These are lower model organisms that aren’t mammalian in nature, but still can provide you with a lot of information. Organoids are now another variation of that, which go beyond just cell lines, but have the composite around them to really be, as they say, organs-on-a-chip or organoids in their structure. They represent an incredible advancement and a tool that certainly the scientific community is going to use. At the end of the day, you want an intact in vivo system. You still have to connect the liver with the rest of the body, to the heart, to the brain, and to the vasculature system. So, when we think about the disease in its totality, you have to think about the whole organism. When you think about therapeutics and their totality, you have to think about biodistribution and various cell types that it’s going to hit. I think that all of these technologies are hugely helpful and beneficial, but I don’t think that one replaces the other.

Daniel Levine: Cat Lutz, senior director of mouse repository and in-vivo pharmacology genetic resource science at the Jackson Laboratory. Cat, thanks so much for your time today.

Cat Lutz: Thank you.