Cannabis Breeding Enters Its Next Era

Want to start humankind’s first cannabis farms on the Moon and Mars? Dr. Matthew Indest wouldn’t be a bad guy to have on board.

Indest has a Ph.D. in Plant, Environmental, and Soil Sciences from Louisiana State University (LSU). He was the national director of genetics and cultivation at the multistate operator Curaleaf, where his team developed hit cultivars like the Titan Express sativa. He’s currently working with PrecisionF1™ inbred hybrids at Refined Genetics. And he’s troubleshooting farms through Catalyst BC’s global consulting services.

In his own backyard, and in his lab work, Indest is doing what he loves: working with plants, helping people grow them better, and innovating in breeding. He has independent research projects under his hemp license in Louisiana, focused on fiber and grain. And he’s helping push the industry forward beyond its reliance on clones toward F1 hybrid seed lines as is done in modern tomato horticulture, among other crops.

Cannabis Business Times contributor David Downs —co-author of “Marijuana Harvest” with Ed Rosenthal—spoke with Indest for CBT’s recurring “10 Questions With …” series. In this exclusive interview, Indest reflects on the four key breeding technologies set to change cannabis the fastest; his work at Curaleaf; the living soil wave; important farm automations; the science of drying and curing; and more.

Editor’s note: This interview has been edited for style, length and clarity.

David Downs: Where is cannabis breeding today compared with traditional agricultural crops? You've seen both worlds. I get the sense that we’re still in the dark ages, but I wanted to ask you.

Dr. Matthew Indest: Formalized breeding programs [in traditional agriculture] have been going around for decades with land-grant institutions. A breeder that I had the pleasure of getting to know had retired from LSU after about 50 years. After he came back from World War II, he established some work in cotton breeding in Louisiana, and that was his life. He bred an immense amount of cotton over the decades.

That's somebody I personally relate to as far as establishing these breeding programs through the universities. A lot of the heirloom-type breeding is just keeping seed of the best plants in your locale. Some of that was a consequence of trade being limited and people not moving around very far; more the local community selecting plants and things like that.

I still feel like we're kind of in that era of seeing what cannabis germplasm exists. [But] we're coming into it with global access. And so, our ability to seek out and connect with people from all over the world and see the different varieties that exist puts us at a different starting point than some of these other varieties or other crops.

But in terms of the advanced technologies that are already adapted for and proven in other crops, I think we’re very much in the early days with the application of those known technologies in cannabis, which is exciting to see where it will take us.

The acceleration that’ll come from that, seeing what it has done for other crops means that our runway can be even shorter because we’re not inventing processes. We are adapting them to a new species that has a lot of plasticity. That’s another exciting part about cannabis.

Downs: Follow-up question. What technology is going to yield a lot the soonest for cannabis that's just table stakes in traditional ag?

Indest: For the soonest – I hadn’t ranked them. I can think of at least four different technologies that are worth mentioning. I’d say the first is doubled haploids. Once it gets adopted and applied, it will accelerate things furthest in my opinion.

If we look at other hybrid crops, you’re inbreeding to around an eighth generation or so to get the homozygosity that you need in order to make a more stable hybrid progeny, the offspring.

By going to that eighth generation, you start to approach a 100% homozygosity for specific genes – but it's kind of limited based on what you’re selecting for and kind of how genes move around.

What double haploids do is a technique where you start from the sex cells of the plant that are already half the chromosome count. And you’re doubling that DNA. Instead of chromosomes crossing over and mixing during fertilization, it’s an exact copy on both chromosome pairs of these genes across the entire DNA of the plant. So, you can approach homozygosity in a single generation.

So, basically, eight rounds … within one cycle.

Downs: Crazy.

Indest: So that's one, double haploids.

The other is another type of ploidy manipulation, which is like seedless watermelons. So, it starts by doubling the complete DNA of one parent, so the plant is a tetraploid. You’re keeping the other parent as its original diploid pair. And when you cross those two, you now have a triploid variety. It’s the averaging of the four- and the three-chromosome state. Triploid varieties are typically low fertility or sterile, so even when pollinated, they produce little to no seeds. There’s just a mismatch of chromosomes that don’t allow it to produce seed easily.

So that allows for some pollen control. It allows for stacking some traits because now you have three gene sets to work with. That’s another one.

A third one would be marker-assisted selection. This is where we map out the traits of interest for what cannabis is out there, whatever your breeding pool is, and you say, “This area of this chromosome is related to purple color or cannabinoid content or a component of yield.” There’s usually a lot of genes associated with yield. But if you can map out those specific targets, instead of physically going into the field and quantifying it … at the point of seed germination, you can take a leaf punch, look for seedlings with the associated trait markers, and basically screen out [what you don’t want]. You can pop 10,000 seeds, test all of them for this set of markers, and say, 9,000 of them, throw them away. “I’m only interested in this group of 1,000, for example, with markers of powdery mildew resistance and terpinolene.” So, what would have taken 10 times as much space, I can now grow on a smaller footprint and know that I’m already filtering out the noise that was never of interest because it didn't have those specific targets already. That’s a big accelerator.

Downs: I just heard about this for the first time with the autoflowering gene this year.

Indest: That’s a good example. Breeding for autoflowers is fun, and a pain. Having to filter through plants that just aren’t carrying that trait takes up space. And because cloning or tissue culture of autos is difficult, there’s some unique challenges to autoflower breeding and selecting plants for being a pollen donor or a receiver without really knowing what they’re going to do yet. There’s some strategy that’s required to make sense of that.

The fourth is one that I was talking to a tomato grower about. Tomatoes are often grown from F1 hybrid seeds, and you can buy plenty of these off the shelf. Every seed will be true to type and very uniform.

Even with highly uniform F1s, a technique that is still used in tomato production is grafting – and cannabis is known to graft well. Grafting is taking a good fruit wood, the scion, to a disease-resistant, or fertilizer- or water-use-efficient rootstock. There’s specific rootstock for tomatoes that are disease-resistant, but they don’t produce good fruit. So, growers at mass scale – I mean massive greenhouses, many, many acres – are germinating two seeds and then throwing away the tops of one, throwing away the roots of the other, combining them together to get better rooting and better fruiting into that single plant.

Downs: So cool.See plants being cultivated at scale.Photo courtesy of Matthew Indest

Indest: That was something that came to mind. I don’t think it’s brought up very much.

Downs: Just to recap: Faster homozygosity, seedlessness, finding traits faster like autos with the markers and then grafting. Those could all be big game changers. I see them more about helping with scale. Speaking of scale. One of my top questions was, can you share a success story from your time at Curaleaf? For example, a cultivar that broke out or a best-selling product that you helped dial in or, anything that’s public from there … something that you’re proud of from that era.

Indest: Sure. … Titan Express is one successful variety my team did the breeding work on. We were looking for a standout sativa that hit high operational KPIs, had a great nose, and finished in a shorter time frame. That variety checked a lot of boxes, we came up with a great name, and I believe it's still on their Grassroots flower menu among a ton of other great work. That’s just one that I like to remember.

Downs: Do you pay attention to who wins awards in Cannabis Cups and stuff at all?

The strain-hype world. Like, “Oh, Super Boof this year.” Or “Toad Venom.”

Indest: [Laughs] I am interested. I think there’s utility in that information.

I don't keep as much of a pulse on it as I could, but I think it’s important. There are other people I know that are more in tune with that, and it informs their processes.

But I also have the mindset that as a breeder, you are so far upstream of when a product makes it to market that you need to be innovating and crafting and breeding for agronomic properties and looking forward, not to the side, if that makes sense.

The varieties that if you’re making a commercial release of a cutting or seeds today, scaling that up to get it into a market for the growers is going to be six months or a year or longer depending on their operational setup.

Downs: So, it’s less about Instagram and who won in a Cup. You got a spreadsheet full of plants that are just numbers right now. And you're like, “This gal, oh my God. What’s going on here?”

Indest: Yes, let’s find a fun name that’s going to stand out and match what it actually produces in the garden and win the next Cup. You know? I think if you’re just looking and kind of replicating what the current winners are, then you’re behind all that. Right? It’s trying to recreate that, as opposed to standing out with something unique. It’s just a different strategy for me.

Downs: Has there been a cultivar that’s most impressed you recently, and why? Anything from the named market or something upstream we can talk about?

Indest: I like to talk more about the process and the seed side again because a lot of these are sort of early stages. So, I think where seeing the uniformity from seeds is what's impressing me – where the industry is going, how operations will need to adapt to keep up with the technology and remain at pace with other stakeholders, so you don't fall behind. I think there’s going to be continued competition in the market from having varieties that stand out, but also from remaining efficient.

We can’t afford to be growing pounds that cost $800 a pound to produce in a market where they’re selling for less than that or even at price. I think anticipating how that's going to impact the market is more exciting for me than any specific one cultivar.

Downs: Do you ever worry about what we’re going to lose when we don’t have any [commercially unviable] 12-weekers left [in existence]?

Indest: No. I don’t. And I think this is a good thing you bring up in distinguishing different kinds of pieces of the process.

It's easy to look at the commercial operators and be like, “Oh, they’re only growing six, 12, 30 varieties, and we’re losing diversity.”

Seedling trays. "Seeding trays [is] simple automation. You’re not individually poking holes in every one of your plugs and putting the seed in a very specific way. There’s vacuum seeders and some other tools that really make the process efficient so you can bring down your cost for starting plants. So, I think that’s an easy adoption ...." - Matthew Indest on automation in cannabis.Photo courtesy of Matthew Indest

I think that’s a misconception, versus understanding that a commercial cultivator has a specific focus and goal for their operation, which is to produce product at scale and a menu that’s diverse. So, they’re feeding the market based on what they want, but doing it efficiently and focused, but that menu changes a lot.

The breeders upstream of commercial operators are looking at the germplasm from around the world and continuing to develop varieties that are going to be the best next fit for their downstream partners and customers.

But even further upstream of the breeders, you have people dedicated to germplasm collection and curation. That's something that is done in traditional crops. I can reach out to – well, even in cannabis, the hemp side, at least – the USDA's germplasm collection and request, as a hemp breeder, a set of varieties that have background from all over the world, and I can use that as a start of a breeding program.

Even in high-THC cannabis, there are people that have curated this germplasm. There are tissue culture labs and seed collections. People that are excited about that work will continue to do that work, and it has a place, and it’s supported by breeders that are seeking out that type of stuff and growers who are wanting to have small batches of something more exotic.

So, I think it’s all through partnership to have that whole chain happen. Sure, there’s some varieties that will be lost or things like that, but it's important for us to curate that material and understand what it is that’s actually there. As a breeder, you can’t keep everything. It's the art of knowing what you can throw away.

Downs: I know you’re a fan of monitoring EC (measuring electrical conductivity in feed, substrate, and runoff to see how much the plants are actually feeding and thus help steer the crop). What do you make of the “living soil” wave and people growing in soil indoors and are just kinda, like, driving while looking out the back window?

Indest: That’s interesting to me. It’s a practice that I use in all my raised beds at home for conventional gardening, not 100% organic, but more of the slower-release fertilizers.

And I am not pumping CO₂ outdoors. But it’s just a different grow environment and model. I think living soil is a little bit more relaxed way of growing because you're kind of letting the plant develop on its own course. However, EC and nutrient balance is still important. You’re working with a different release curve as nutrients are available over time rather than being immediately steerable.

There’s certainly risks from a pest and disease pressure perspective that makes scaling production beds and reusing substrate a challenge.

And then there’s the act of amending the beds and heavy metal accumulation. I think those are the areas where I exercise caution.

Downs: It’s like the opposite of crop steering. It’s like driving a car in the backseat and saying, “We’re going where we’re going.”

Indest: Without having practiced it in this crop at scale, I’m sure that there’s still ways to steer. You’re just a bit more limited unless you’re taking, like, I’ll say, a synganic approach. I’ve heard that term used enough to where I think it’s OK. You can baseload things and then feed with your mineral salts or apply other approaches. So that’s probably where I would go: take the best of both worlds and adapt it to the infrastructure.

Downs: Two more questions. One is, what automation should folks be on the lookout for? I saw my first terp combine harvest where [the grower] had a combine going down a row of plants in a field and grabbing them all and taking them to get steam distilled right on the property. That blew my mind. But, I imagine you are thinking about automation. For the most part, this is a pretty medieval crop. Hands touch everything. But what automation should folks be on the lookout for if at all?

Indest: Good question. So, you say medieval. It’s a horticultural crop. That’s how I look at it. And those tend to be much more labor involved – tomatoes and other produce, things that are manually harvested and still a lot of human touch points.

I think cannabis in its current state fits that model, but what you described, that mechanical harvesting and mechanical planting, when once we take it to machine row crop style, now it's like an agronomic crop. It’s just a distinction I had to learn as I went from an undergrad in horticultural science to a doctorate in agronomy. It’s like, “Oh, okay. There's this line there, and some crops somewhat straddle that transition.” That’s where we are: We're seeing cannabis as it scales up. We're looking for efficiencies.

You won’t get the best-quality smokable flower out of a combine-harvested field crop. But when it comes to edibles and other formulated products, the scale and efficiency that you can harvest, it absolutely makes sense to utilize those different pieces of equipment.Terpene Belt Farms harvest. The company provides a scaled supply of hemp essential oil.Photo courtesy of David Downs

I’d say for current growers and where the industry sits today, I'll kinda bring it back to that seed aspect. Seeding trays: It's simple automation. You’re not individually poking holes in every one of your plugs and putting the seed in a very specific way. There’s vacuum seeders and some other tools that really make the process efficient so you can bring down your cost for starting plants.

So, I think that’s an easy adoption, again, for people moving into growing seeds at scale more efficiently – running that seeding process and germination protocols.

I wanted to share that as [a standard operating procedure] (SOP) related to seeds that I think growers may be stumbling at the starting block is in understanding how seeds are different from clones and running the conditions differently in order to get the best seed in.

If you try to grow seeds like you do clones, you’re going to get, typically, a spindly plant. The stem’s not going to be great. It will take a little while to establish, and that's typically because growers are not giving them the amount of light that seeds can handle.

So, germinating under a higher light intensity, say, 500 PPFD versus what you would clone at – 100, 150 PPFD – makes a completely different plant, and it's not something everybody is prepared for as they transition to seeds.

So just those two things. Again, having an SOP that’s tailored toward seeds and then seeing, “Okay, can I scale this up with a tool like a seeding tray,” and not trying to individually pluck thousands of seeds into a start will help that scale faster.

Downs: Great call-out. Lastly, I wrote a book with Ed Rosenthal called “Marijuana Harvest.” It’s, like, nine years old now. And I was struck by how the art of harvesting, drying, and curing weed was still art more than science, and that there were so many Ph.D. papers to write about what was happening in drying and curing.

Are you keeping an eye on this space at all? The coolest thing I saw recently was people measuring vapor pressure of individual buds in their drying room, or the water mobility in the buds as a new measurement for dryness, and then even going further than a moisture target, but then adding back moisture after they got to a spot.

What’s your read on improving the science of drying and the curing where we really screw up a lot?

Indest: That’s, again, yes, a fun challenge here: drying/curing, understanding what is actually happening in that process. I give a shout out to Dr. Allison Justice and her work on the science of smokeability. [She and the Cannabis Research Center and Coalition are], I think, one of the few groups that has actually presented research looking at gas exchange on what’s going on in the plant material.

We can look at these theoretical curves as far as moisture content or water activity, like how much water is available for enzymatic processes or biological growth as opposed to water that's just inaccessible in the structure of the material.

So, that active water is a much more important measurement because it tells you what biological activity could happen, whether that’s microbial growth or just improvement to the crop or aging, and some of the enzymatic or chemical changes that can happen even though the plant's dead at that point.

So, we’re just scratching the surface with that.

I think getting to an agreement within a group on ”what quality is” is sort of step one to inform how your processes change. I think there’s a lot of guesswork and feeling about a process that gives you the “best weed,” and it’s not quite as formal. So, people get comfortable with a process. They think that this is the way you have to do it, because that's the way it was done.

But I’d like to see a much more analytical approach of understanding what processes or what chemistry is happening in the drying process and then understanding how, as we remove moisture from the plant, that is affected.

There’s plenty of research that could be developed and published on that part alone: the final weeks of life and maintaining the best material there.

Downs: [I just saw the first] published papers on how, like, the color of the trichome is in fact related to how much THC is in there, confirming that a visual cue is relevant.

And we find as growers, we’re still snapping stems and snapping buds and crunching with our hands and then jarring or not jarring, and burping or not burping.

And then I'm seeing people curing not jars or totes, but entire rooms. Like, you don't need to cure a jar. You can cure a room if you keep the whole thing at the right temperature.

Indest: Right. And, again, it’s looking at other established industries—you can’t just grab everything from them. I worked in sweet potatoes, and there is a curing step for sweet potatoes. When you look at other crops, like curing tobacco, and preparing all these products so they’re at a shelf-stable place that’s not going to go rancid or degrade in quality, it’s absolutely a science. It starts with the art of, “We've done this. This is what I interpret to be the best,” but then the science part is now quantifying each of these different elements and developing processes that look at it in a controlled way in order to make that informed decision and then repeat it.

The hard part is – just look at domestic US production, a grow in Arizona, [versus] Illinois, New York, Florida, and Nevada – all of these places have a much different macro-environment and different shelf conditions depending on how they’re in transit, storage, or on retail shelves.

So, maintaining that quality throughout the time of harvest till it gets into the customer’s hands is extremely variable because of just where that product sits.

So that’s a whole other element.

But it’s, again, a fun puzzle to understand what’s actually going there? What contributes to quality that we can measure? And then from there, what can we manipulate to influence that measurement?

Downs: I'm bummed to hear there hasn’t been rapid progress in nine years.

Indest: Oh, there’s progress. It’s just how do you demonstrate that and compare?

One fun thing that I’ve thought about a lot: you mentioned moisture content versus a water-activity aspect.

Look at moisture-content measurement devices. Typically, it’s just heating up the sample on a sensitive scale. And as that weight leaves, you’re watching how the weight starts to leave less fast, if you’re running one of these pieces of equipment.

So I’m grinding flower. I put it on the scale, and I heat it up above the boiling point of water.

And the assumption is that everything evaporating is water. And so, it’ll tell me, “You lost 15% of this material’s weight.” Therefore, the moisture content is 15%.

Well, what we do know about cannabis is it has many volatile chemicals. And you look at a COA, depending on the flower, I’ve seen anywhere from 1% to 5% terpene concentration.

Downs: Ideally, yes.

Indest: The more, the better, right? THC level is also driving the market. So, if we compare two plants, one of them has a higher THC and terpene content, say 30% THC, 5% terpenes. And, take a gram of that, and then have a gram of something else that's lower cannabinoid content, lower terpene content, put them on the scale.

Even if the actual moisture content is exactly the same, the reported loss of drying from the heated scale will be different because the terpier, more potent THC cannabinoid one loses weight through decarboxylation and terpene loss in that step.

That’s like a standard way the industry is measuring moisture content.

But it overestimates moisture content, especially in samples that are like that.

Downs: Yes. We need to prove that.

Indest: Exactly. It’s something that I think more people need to be attentive of, or they're not going to actually hit the target moisture content that they think they are hitting. And, again, another reason to focus on water activity.

Downs: For the future Ph.D.s out there, I want to see them investigate the chlorophyll degradation curve versus smokeability; and then show the effect of different temperature and humidity levels on terpenes and THC in a reference sample over time. We don’t really have any of this.

Indest: There’s a lot to do. Let’s get funding.