Ahead of our 2022 Virtual KAND Family & Scientific Engagement Conference on August 13, 2022, KIF1A.ORG interviewed core KIF1A Research Network members to discuss their #relentless efforts to understand KIF1A and help KAND patients in this special “Meet the Research Network” series on the KIF1A.ORG blog.
KIF1A.ORG’s Research Engagement Director Dylan Verden, PhD, had the pleasure of talking with Arne Gennerich, PhD, professor of Biochemistry at Albert Einstein Medical College. Dr. Gennerich discusses how understanding KIF1A’s structure can help inform better-targeted therapeutics for KAND.
Dr. Gennerich: So my background is actually electric engineering and physics, but I switched early on to biophysics and came to America in 2003 to do a post-doc at UCSF where I started to work with microtubule associated motor proteins. But it was really cytoplasmic dynein and their mutations across neurdevelopment diseases. But when I moved to Einstein while I continued to work on cytoplasmic dynein I became interested in KIF1A as well because it’s a really highly interesting motor in terms of its biophysical properties; we talk about the high processivity that KIF1A has that we call superprocessive.
Dylan: What techniques are you using and what are you learning about KIF1A through them?
Dr. Gennerich: In my lab we use mainly single molecule technology even technology that we built on our own, Total Internal Reflection Microscopy to visualize fluorescent KIF1A molecules that move along microtubules, and we use optical tweezers, which allows us to study forces that KIF1A motors generate. The general questions we have are really of a basic nature. We want to determine using structure-function studies why is KIF1A superprocessive, why it can move more than 20 micrometers along microtubules. Also KIF1A has a high velocity, it is 2.5 times faster than kinesin-1, the founding member of the kinesin family, and we want to determine why it’s so fast. Also, KIF1A has really intriguing force generation ability, which we also want to dissect; it easily gives up under load but it has a high on-rate so it rapidly reengages so the cargo transport can continue. We became interested in determining disease mutations due to our collaboration with Wendy Chung at Columbia University. We have an R01 from the NIH together to look at disease mutations both in vitro and in vivo and in a clinical setting. And so we have started to screen disease mutants to determine what the effects of particular mutations are.
Dylan: Talking about the elements that make KIF1A unique, its superprocessivity, its quick on-rate, why are these things so important to understand in the context of KIF1A mutations and trying to find therapeutics that either tackle KIF1A mutations directly or impact kinesins more generally?
Dr. Gennerich: In general I think the high on-rates are really important to ensure cargo transport happens over long distance. If the on-rate would be really low, this would lead to delay in transport. And we talk about particular degenerative effects on the children, and degenerative effects are effects that accumulate, so even if you have reduced velocity or reduced on-rate, while cargo transport overall still functions but is slower, it’s slower and takes more time, and this has a cumulative effect on the children. And so our aim is to screen for drugs that help these mutants to recover their velocity and on-rate and force generation. And we use two different approaches for this. One approach is highly advanced. We work together with Atomwise in California, which is a company that does virtual screens. And they have successfully identified 80 molecules, 80 compounds, and they are being shipped already, they’re already on the way to our lab. And we will test these small molecules on the P305L mutant, which is the first mutant that we have characterized in detail. And we hope that some of these molecules can activate or recover functions of the P305L mutant. And the idea is, and I can show you I have here the list of 83 different molecules that we will receive, and what we will do is once we have a molecule that looks good or promising, we can order the molecule, we can talk to a chemical company to optimize the molecule based on structures we also solved.
And this brings me to the XSeed Award that we received a couple of months ago. The XSeed Award supports our effort to create this startup company that designs drugs for microtubule associated motor molecules, but the initial project is only focused on KIF1A. Our collaborator Hernando Sosa at Einstein here, who is the leading expert on the cryo-EM structures of microtubule associated proteins, he has solved successfully the wild-type KIF1A molecule a little bit better than 3 angstroms, which is an atomic resolution structure, and he has succeeded to solve the structure of the P305L mutant. The resolution is more 4 angstroms but is still good. Our hope is to use the structures, put them in the computer, and perform virtual screens.
Dylan: That’s really an incredible thing because, speaking of KIF1A-Associated Neurological Disorder as an umbrella that spans over 100 mutations, I think one question that often comes up is, “these two mutations are so close to one another in the gene, or they’re so far apart, why do the symptoms correlate or not correlate? Why do we have such different diseases across? So the hope is to be able to determine that structure, not just in wild-type KIF1A, which is a huge hurdle in a lot of approaches, “do we know the structure do we know the structure?” But to understand how the shape of KIF1A changes with a given mutation.
Dr. Gennerich: Yeah, that’s a really good point. I can tell you that based on the structure, our structure shows things that nobody has seen before, like the K-loop for instance. We even have evidence that the K-loop might assume different conformations depending on the nucleotide state. But correlating structure, looking at the sites of mutations and the amino acids and correlating them with the output, the functional output and the defects we see is very important. It not only allows us to identify drugs that binds specifically to the mutant, but it also allows us to get more information on the molecular function of KIF1A.
But for us it’s really important that we do a structure-guided drug development, and this is a different approach. Many scientists use high throughput screens with billions of molecules, they have cell culture and slide chamber or multichannel systems, where they throw on drugs and see whether they can rescue KIF1A organelle transport for instance. But our approach is we take the structure of the mutant, and drop the small molecules that we identify in our in vitro assays that have a positive effect; we dock them, or we let them bind to the motor, and then we solve the core structure of the mutant with the small molecule bound. And the idea is that we can identify the conformational changes that are induced due to the drug binding, and in addition we can talk to chemists, show the core structure, and say, “what can we do increase the affinity?” Because typically for these small molecule compounds that we receive, we have micromolar affinities, which are not high. The lower the affinity is, the less tightly it binds, the more likely it is that there are side effects. So we really try to increase the binding strength dramatically by solving the core structure with the promising molecule, talk to a chemist, who will redesign and modify the molecule based on the chemical knowledge to increase the binding strength. And that’s our hope, that down the road we will have a high-affinity binder that has positive effects and recovers its velocity, processivity, and maybe even force generation. And this will allow us to have a drug in our hands that will prevent further degeneration in those children that have P305L mutations.
But importantly we want to develop a high throughput pipeline so we want to go from mutant to mutant to mutant to mutant, because obviously there are children who have other mutations that have devastating effects. And initially these mutants are in the motor domain because we solved the structure of kinesin, full kinesin sitting on the microtubules, and so initially we will really screen the mutants that are in the motor domain. But in my lab we also have the effort to express full length KIF1A in insect cells and in collaboration with Hernando Sosa we will also try to solve the structure of full-length KIF1A, in particular the autoinhibitory state, when it’s in the monomeric state where the tail drops onto the motor domain. This in the wild-type background as well as the mutant background. These efforts will take more time but the cryo-EM work on the motor domain of KIF1A bound to the microtubules is not only promising, it works. And it’s so promising that Hernando and I have tried to submit an R01 grant that covers the basic mechanism of KIF1A, which is my part, and for Hernando even screening or determining the cryo-EM high resolution structures of several mutants. We tried to get NIH funding to cover this line of research.
Dylan: Going from this very basic science lens of structure and changes in response to therapeutics, and thinking of the end goal which is to find treatments for the KAND community, do you have any thoughts that you’d like to share with our patient community?
Dr. Gennerich: I can only say that I’m aware of how devastating these KIF1A mutations are for the children as well as for the parents who raised these children. And I must say I’m really gracious that my lab can be part of it, and we really hope that we can identify components earlier, even in the next 12 months hopefully. And I just say hang in there, and support each other, and I hope also that the medical support in particular in relation to Wendy Chung, will really help all families who are affected by KIF1A mutations.