Kevin McKernan, chief scientific officer of Medicinal Genomics, on sequencing the cannabis genome and its implications for the cannabis industry and “personalized medicine.”
Project CBD: We’re speaking with Kevin McKernan, chief scientific officer of Medicinal Genomics, a Massachusetts-based company which a few years ago—I believe in 2011—sequenced the cannabis genome or a particular cannabis strain.
McKernan: Yes, that’s right.
Project CBD: This would seem to have some significant implications for the cannabis industry as a whole. We’d like to explore that with you. Maybe you could explain a little bit about what we mean by “sequencing the cannabis genome”?
McKernan: Sure. So, this was in 2011 and the tools we had to sequence back then were still evolving very, very quickly, but we were able to get a very draft version of this genome sequence. Now what this is, is reading every letter in the genome and really, in any cannabis samples, many know there are two genomes: there’s really the mother and father genome. The plant is known to be diploid. So it’s got 20 chromosomes and one copy of each from mother and father, there are some chloroplasts and mitochondria genome in there as well, but we want to read all of those letters so we can begin to build a map of all of the genetics that might predict cannabinoid expression, terpene expression, maybe even flavonoid expression. As it moves into hemp, maybe some of the genes that are governing either seed size and oil and fiber.
So to read the entire genome, it’s about 1 billion bases long, it’s a billion letters of genetic code. In the process, with the technology we have, we probably only got about 400-500 megabases of really nicely aligned sequence. Now that seems like it, you’ve only gotten about half of the genome. That’s probably true. There’s a lot of repetitive nature in plant genomes. They have copies of things that are identical scattered throughout them. And those end up when you sequence them, it’s kind of like putting together a big jigsaw puzzle. Those are like all the pieces that look the same. And sometimes you don’t know exactly where they go. And so they get—they’re in the sequence but they get kind of left as ambiguous when you put all this data together. But we do have is a really nice scaffolds of some of the genes everyone is very attentive to, like THC synthase is one that is of real interest, really nice sequence coverage of those, CBDA synthase and some of the genes that are governing cannabinoids.
Project CBD: When you say “synthase” you’re talking about the kind of the precursor gene for what will become CBD or the gene that encodes the enzyme that creates CBDA and THCA?
McKernan: Yeah, so it’s an enzyme involved that the DNA codes for protein, and that protein folds into little enzymes that folds the precursor molecule into either THC—and there’s another gene called CBDA synthase, and the “A” is for the acid form because it makes THCA and CBDA before it, in the plant form. That’s taking a Cannabigerol precursor, and if you look at Cannabigerol, it’s like a ring structure and it’s got a long tail on it that it kinds of folds and wraps into two more cyclical groups that either makes THC, or one more cyclical group can make CBD. And there’s two different genes responsible for folding that precursor two different ways, and they’re actually in competition for the precursor.
So we believe, and many others before us have actually done this work to demonstrate that DNA variants in those genes have some predictive capacity on how much THC it’s going to make. So, there’s been some publications showing mutations in key areas of the gene—and this is known as a FADA-binding domain, a very technical term for a really catalytic core of the enzyme, that folds those molecules. And when there’s DNA variance in there, it does a slower job at it. Sometimes it doesn’t do any job at it. So, we believe—this is kind of a [rheostat] of how much THC is produced depending on where these variants are in the gene. And if we can get a better picture of this, we can begin to predict these chemotypes of the plant at a seedling stage without having to fully grow them out.
Now, the environment is always going to play a role in this dance as well, but we’re just trying to get a really firm picture of the genetics so that we can have some understanding of the capacity of the plant. If we know the plant is going to make all CBD, you might put that in a different grow room, or you might do something different with it, breed it with something else where you’re trying to bring CBD in. But if you know it’s going to be really strong THC strain that perhaps gives you a different direction to go with it.
Project CBD: You’ve touched on some of the implications of this genetic science for cultivators, for breeders possibly, or for people who are trying to fine-tune particular strains for particular medicinal properties. Let’s get into this a little bit more. How will this really be—this kind of knowledge—other than pure science, how will it be applicable for growers or for the cannabis industry as a whole?
McKernan: So, where we’ve seen this take foothold in other marketplaces is with a process known as “marker assisted selection.” I’m a much bigger fan of that type of breeding than let’s say let’s go monkey around with the genes, when we don’t really understand the whole genome yet. So the genome stuff, I think, is a little bit arrogant right now, and maybe even ever, depending on how much we ever get to know about this genome. But what people do tend to do with the sequence information in breeding is they use the DNA markers to track the traits. And it’s just kind of a measuring tool. Instead of measuring the chemotypes as they grow out—which is a great way, I don’t think that’s ever going to go away. But if you want to measure what chemotypes you might get at the first sign of the leaf, there’s information to be had there that can tell you, okay, here’s the terpene profile we think it’s going to have.
Now we don’t know the markers that do that today, but this is what they’ve done in other very valuable crops, is that they get an understanding of maybe 100,000 to a million of these different single letter changes in the genomes, and they track those in all the plants. And they use those as sort of a proxy for, okay all of these changes over here tend to track with terpinolene and these ones tend to track with beta-caryophyllene , and these ones tend to track with maybe pinene. And so, we know that those markers are predictive that this plant’s going to be, maybe a myrcene dominant indica, and this one’s going to be a pinene or something that’s more Jack Herer—like with a terpinolene.
So those markers we think at a hole-punch we could distill. You could get a hole-punch of the leaf, all you need is about 30 nanograms of DNA to do this. This is 30 billionths of a gram of DNA that you could get this information from, and it will give you maybe 100,000 to 1 million different data points on the strain. Ultimately, there’ll be databases to help you correlate that all right, this pattern is from a plant that’s on this part of the phylogenic tree and we know that it tends to make these types of terpenes. And we’re going to track those SNIPS, they’re called single-nucleotide polymorphism. So there’s a lingo here what people call them SNIPS. Once we have that, we can then make some more informed decisions about, okay, we have a wonderful plant over here making some cannabigerol and some cannabichromene and we really want to bring in limonene to this, and what do we cross it with to make that happen in the fastest way possible. So we think it becomes just a tool set to help guide breeding. And this process called “marker assisted selection” is simply a fancy term for we’re going to measure this with something other than photography to perhaps guide where we go with the breeding.
Project CBD: So Medicinal Genomics has actually been gene sequencing several different strains, many different samples?
McKernan: We have, yes.
Project CBD: And what have you found in terms of diversity or lack thereof, and how has prohibition affected that?
McKernan: So it’s a fascinating question. There’s certainly signs of there being a genetic bottleneck toward high-THC plants. But at the same time of that having happening, there’s been a tremendous amount of inter-breeding that’s going on, that if you were to compare the variation we might see in the drug type plants, next to those that we see in the plants that have had less prohibition on them, like hemp plants (hemp plants are registered and there’s process in some countries to actually legally grow those), you actually see far more genetic diversity in the drug types, the ones that are underground. And, although they might have more THC in them, there has been perhaps a lot more exchange to find different terpenes or to breed different traits into them. I’ve always found that very fascinating: that the ones that are a little bit more above board and leveraging perhaps the registries and the patent systems actually have less genetic diversity then what we’re finding in the markets that perhaps that are out here.
Project CBD: Let’s talk a little bit about the implications for the medical use of marijuana. Personalized medicine is a buzzword today within the cannabis therapeutics world, and in terms of what that means, I think, on the ground for patients—patients are sort of groping in the dark to find what strain might work best for them, what helps them the most, maybe what ratio of CBD to THC is going to be their entry point into utilizing cannabis medicinally, and that always comes down to well, it’s about that person, how that person is going to relate to a particular strain or the plant in general. But I think what you’re talking about with the science of genetics sort of takes this idea of personalized medicine to a whole different level. So, how—and I know also that Courtagen, your sister company, has been working on the human side of the genome—so how do the two come together and what does this mean for personalizing medicinal cannabis?
McKernan: So I’m really excited about this topic because we—on the flip side of our business we’re sequencing 700-800 patients a month to try and find variants that help dosing patients with different drugs; sometimes it does implicate that you might benefit if you go straight to CBD. The classic case is sodium channel 1 mutations or Dravet syndrome, and they are the class of patients that are responding very well—not just anecdotally on the Internet, but even in the FDA-based trials you can’t deny this, that they’re having a real positive affect from [GW Pharmaceuticals’s] Epidiolex, that they’re testing this on.
Project CBD: You’re speaking about CBD or CBD-rich product, it’s helping these particular children with epilepsy and in terms of Dravets, but there are many different epileptic diseases
McKernan: We tend to sequence 500 genes for epilepsy patients because it’s a long-tail, and this is really the lesson of personalized medicine. It’s kind of a lesson of the FDA, is that they’re not designed for personalized medicine. They’re designed for one-size-fits-all drugs, that aspirin and things just hit everybody with the same thing. Those are gone. There are no more of those. There are very few of them. And it’s pretty clear now that the reason drugs are failing in the FDA and the reason the costs are going up is that they’re trying to apply this model of blockbuster drugs to shove it through the FDA. What we’re really seeing in cancer and all these fields is that everyone’s genome is, we differ by probably 4 million variants in our own genome. And so what you might describe as chronic pain could be a completely different molecular mechanism if I have chronic pain. And we begin to understand that when we dive in and sequence people’s genomes. Oh yeah, you have chronic pain because you have TRP receptors that are off. And I have chronic pain because I have a TRAP1 variant. And those might require different mechanisms for dosing. It might require different cannabinoids.
And so, I’m very excited about bringing those two fields together because we can see the personalized medicine—there’s only one market in the country that I think has grown faster than personalized medicine, and it’s cannabis. And cannabis is this market that has, what looks to be, about to become, the largest open source set of pharmaceutical drugs that grow to high concentrations in plants that are weeds. I mean this is a total revolution in medicine that’s about to become over-the-counter for physicians; that if we can line up the right patient variants with the right cannabinoid and terpene profiles, the cost of health care is just going to go right down. And we can do—what I would like to call it as, it’s almost like a technological leapfrogging event, right. You see these cases where two different technologies kind of—the two different tsumanis that are coming together and they create a kind of a perfect storm of a wave.
Project CBD: The two different technologies are now the—?
McKernan: I think it’s the personalized medicine insight. Knowing the patient’s genome and then also knowing the entire cannbinoid profile of these plants—and the reason I’m fairly focused and myopic on the cannabinoids is because it’s really not that myopic at all! There’s just a whole portfolio of these things, and they all grow in the best factory you can imagine. There’s certainly a lot of interest in ripping these genes out and putting them into yeast and seeing if you can keep up with the plant.
Project CBD: That’s not your focus.
McKernan: No, I think that’s harder. I think it’s a harder job to do. And the plant’s very good and it grows—we have all this infrastructure to grow this plant. What you do find, in some places the pharmaceutical industry, they want to put their pathways into plants like this because they become medicinal factories for them. And then, you know, we have a science of growing this plant that’s quite mature and produces tons of this stuff. So, I’m actually quite optimistic about that side of the equation being done just perfectly today. But to bring the two together, you do see these destructions happen in other markets that I’m always very attentive to, that you see things like new batteries that are coming out these days can now drive things on the ground that are gyroscopic scooters now, and cars like Tesla, and then there’s solar panel technologies—and when those things come together, you see third world’s skipping landlines right now. They’re not installing landlines, they’re going right to Nokia cellphones. They’re not going to build central banks because they have cryptocurrencies or M-Pesa like currencies that are happening in Kenya.
The same thing I think is going to happen in medicine with the FDA. The FDA is one form of doing medicine, but it can’t scale with the innovation that’s out there. And the innovation I see really happening in medicine is, at this level, is the level of trying to get cannabinoids to the right patients. And that model is much more democratized, it’s much more individualistic, it’s much more focused on and respectful of each patient being unique. And we’re losing that in the health care system. The health care system is getting very one-size-fits-all.
Project CBD: In a sense what you’re talking about, is kind of eliminating the “crap shoot.” If you’ve got a drug, and you have a person with a certain genetic constitution, and you’re throwing CBD at an epileptic kid, why does it stick with some kids, why does it work so amazingly and why doesn’t it work with others? So you would be able to determine this in advance?
McKernan: We’re starting to see some signals on this. And this isn’t any coordinated effort, necessarily, with FDA or with GW [Pharmaceuticals]. Because we sequence so many patients at Courtagen, we happen to have a lot of patients that are in those trials, and they tell us. And when we take their data and compare the responders to the non-responders, we can see that there’s a pattern. They’re starting to show a pattern of the responders have variants, there are four particular variants that we’ve been showcasing that they have. And interestingly enough, they’re in—one’s in a drug metabolism gene known as a CYP2C9 gene, it’s a gene involved in the metabolism of CBD. The other three genes that are showing some predictive power here are sodium channel genes that we know anandamide interacts with. So there’s a thesis here that kind of makes sense. And the reason it’s important is, in some of those patients, CBD isn’t their molecule. They get 2-400 percent, smaller, 15 percent of the kids get worse. When seizure patients, like Dravet patients get worse, it’s extensively worse. It’s like they get intubated. They end up in the ER. Sometimes they get Status Epilepticus. And often times you hear anecdotally, some of those parents switch into THC or THCA, and they get better.
So there’s different mechanisms of causing seizures, and I think as we understand those mechanisms from a genetic level, we get more challenging about which cannabinoids to hand them. We don’t know the answer to how to mix and match that perfectly today. That’s kind of the vision. That’s where we want to go. But I think what we’re going to see over time is that puzzle is going to slowly fill in piece by piece, publication by publication, that these variants, very predictive of success and this class of patients. This class of patients really needs maybe a THCA, which is a completely different molecular pathway.
Project CBD: It’s very exciting. Presumably that could be applicable to other conditions, and not just the epilepsy.
McKernan: I think so. I think chronic pain is one. We’ve published a paper on the chronic pain front. And some of these patients were very responsive to antioxidants. And we all know that some of the best lipid soluble antioxidants in the world are cannabinoids. So these patients didn’t happen to jump on that. But the fact that they were responsive to N-acetyl cysteine, which is more of a—not as much as a lipid soluble antioxidant, so it could be some improvement there. It’s very exciting to us because this variant is in 1-2 percent of the population. It’s in a gene known as TRAP1, that has some implications in cancer, but it’s never really been pinned to chronic pain or chronic fatigue. And so, we refer about 1,000 patients and we summed up all the patients that had chronic pain and chronic fatigue and Bang! A big signal pops out saying TRAP1. And when that gene’s broken the hypothesis is it’s making lots of reactive oxygen species [ROS]. So your body is churning through energy and creating all the side products of energy consumption, but it’s not actually functional. And that’s creating, we think, the pain and the fatigue, it’s that this enzyme is broken, it’s eating up all the ATP [adenosine triphosphate].
So we’re really hopeful that that will translate into maybe making a dent into this opiate problem, right. These people are just getting handed opiates and it’s probably not really addressing the core issue of creating lots of ROS, it’s just dulling the pain that’s coming from. And we’ve seen where that’s gone. It’s just gone into total and complete opiate epidemic right now.
So, we’re hopeful that some of the sequencing will come in and objectify what is the subjective problem for a physician. Patient shows up and says I have pain. Sure you do. Then they have to really monitor how much dosage they give them and whether it becomes chronic, whether they get GI issues with the opiates. But if you had a molecular marker that said no, these are the patients that actually have a broken gene that we know responds with antioxidants, you can suddenly change that conversation to “we should be considering cannabinoids for this.” I’m hopeful that’s going to happen in autism, in mitochondrial disease. We’ll probably see it in Parkinson’s and Alzheimer’s. A whole host of diseases that you can see on the Project CBD website that are playing a role. And they probably all have different molecular mechanisms as to how CBD is benefiting those things.
Project CBD: Well I think Medicinal Genomics is changing the conversation. And that’s great. It’s amazing the work that you’re doing. I appreciate Kevin McKernan for being with us today.
McKernan: Thank you. We’re excited about the field.