As someone in this industry, I doubt this will become a reality. The real problem is cost. When you use yeast, you have to supply the energy for the entire system. When you grow a plant, you get free energy from the sun. The cost savings are hard to beat. Nicotine extracts aren't made from yeast either. That should be a big warning flag for the business sides of these cool projects.
> When you use yeast, you have to supply the energy for the entire system
The “entire system” is a vat with yeast in it. For starters, that’s a much more space efficient way of growing something than in plants.
Secondly, the energy you provide to the system, in brewing, is just sugar, which is incredibly cheap. In fact, given that this works well, at large scale, for beer brewing (i.e. ethanol production), why do you think it wouldn’t work well for THC production? In fact, energy consumption in commercial (especially illicit) cannabis production is a nontrivial cost factor.
The THC molecule is made from carbon, hydrogen and oxygen, and it seems to be true for all cannabinoids. So if we can extract the cannabinoids but leave the rest of the organic compounds (big "if", I know), one could theoretically recycle the other required nutrients. In which case a steady supply of sugar might indeed be enough for the most part.
You’re right. The parent comment focussed specifically on energy, so I did as well. But obviously both cannabis plants and yeast need other nutrients. That said, a minimal yeast broth is extremely simple.
My understanding of illicit cannabis production is only from films like "Lock, Stock and Two Smoking Barrels" but it seems to be heat lamps in basements and caravans etc.
And the impression the news gives me is that UK Police regularly use heat cameras from helicopters and analysis of electricity bills to find illegal growers.
Would this approach allow test-tube illegal cultivation with a much less detectable footprint?
Roughly 1/3 of cannabis is grown inside [1], primarily for reasons of quality control. I'd actually expect this number to grow as the legal market expands and tighter control is needed to keep up with regulations. Which is to say that I'm sure there's a market for this.
Making compounds with GM yeast tends to be a lot more efficient than growing whatever the original gene came from because the modified yeast tends to dedicate a lot more of its output to making whatever you've spliced in.
Land is pretty cheap. The kind of stuff used in manufacturing is what they call biomass. Basically they just chop down and grind the whole plant with a tractor. Planting will get automated too. Plant genetics are also improving.
The result is that the cost of hemp/cannabis is rapidly dropping. These new startups will have to surpass a moving target. I'm not saying it's impossible but I doubt it.
Yeah but then you're growing and refining a plant to avoid growing and refining a plant.
The article said it needs to improve 100x. I'm not sure if that's reflective of today's or future prices as they continue to rapidly drop.
It's certainly not impossible. I think it's just unlikely given the similarity between the tobacco and cannabis industries from a supply chain perspective.
>As someone in this industry, I doubt this will become a reality. The real problem is cost. When you use yeast, you have to supply the energy for the entire system. When you grow a plant, you get free energy from the sun. The cost savings are hard to beat.
It's not like beer or cheese has a profit problem...
So this accidentally made me go down a speculative rabbit hole... Cannabis is apparently a C3 plant[0]. It's also not a perennial. So while you have a point, there is still room for improving efficiency. I tried looking up what the current status with genetically modifying plants with high photosynthetic efficiency was. First thing I found was... tobacco[1]. Yeah, that's not all that useful.
More generally, I wonder if there might be some kind of "holy grail" easy-to-grow, easy to genetically modify plant out there that could be a perfect template for photosynthetic production of all kinds of organic compounds. I know people have been trying to modify algea and cyanobacteria to make synthetic biofuels directly for ages, for example. Sadly the Wiki page on photosynthetic efficiency remains fairly superficial[2], and the C3, C4 and CAM pathway articles, while a tempting rabbit hole, don't seem to help much either[3][4][5].
Also, we all know monocrop farming is not going to be sustainable: the simplicity of it makes it easier to "optimize" for farming machines, but at the cost of soil degradation, while also being more prone to diseases and weather disasters. Plus, with current improvements in automation the optimization advantage is shrinking rapidly. Could agroforestry, like the kind advocated by Geoff Lawton[6] or Willie Smits[7][8][9] be combined with this? The benefit/problem with agroforestry is that it has many different kinds of complex yields. All useful, but perhaps not so flexible. However, Smits suggests an economy based on the sugar palm as a source of bioethanol:
> And Smits said that he discovered that because of the [black sugar palm]'s special leaf structure, its year-round production and extremely efficient photosynthesis, the yield of ethanol from the sugar palm was far greater than the biofuel output from other feedstocks in use around the world. Smits says that his process can produce 19 tons (6,300 gallons/24,000 liters) of ethanol per hectare annually. That's a staggering output-to-land area ratio compared to corn, the favored ethanol crop of the United States, at 3.3 tons (1,100 gallons/4,200 liters) per hectare, by most recent U.S. Department of Agriculture yield figures. It also far outshines Brazil's sugarcane; output was assumed to be 4.5 tons (1,500 gallons/5,700 liters) per hectare in the U.S. Environmental Protection Agency's recent lifecycle analysis of renewable fuels. [A hectare is 2.5 acres.]
Keep in mind that once a tree has grown it requires less energy and material to maintain. Non-perennial plants have to grow every year, which comes with large energy and material overheads (both from the plants itself and from the labour involved). That makes the claim more plausable. (Cool little tangent: looking for more information on this Arenga Pinnata the first thing I came across was its listing as an invasive species on BioNET-EAFRINET, a knowledge database run by Kenyan and Ugandan research institutes[10]. This is the kind of work that never reaches the Western media, and I would guess it's partially because it does not fit our existing narrative biases that are still very much stuck in colonial times)
Now, bioethanol is a fuel source for cars, but sugar is a fuel source for living things. So why not use that as a form of "energy currency" for biohacking? It would have the benefit of increased flexibility. Imagine having a food forest centered around sugar palms, like Smits suggests, with a few genetic modifications for even higher energy yields, and with a lab nearby that can grow the more complex organic compounds with simpler but easier to genetically modify life-forms, like yeast. The sugar palm provides a steady source of biofuel, while the yeast (or whatever) being cultivated can be changed quickly to meet shifting market demands.
Anyway, this is all just speculation, and perhaps the losses of energy involved are so great it would not work out. OTOH, going the agroforestry route yields so many other integrated benefits like local climate control, CO2 sinks and a fresh water supply in the groundwater that I expect it will be a net win when viewed holistically.
Given that tobacco is also a C3 plant your link to the post about improving tobacco photosynthesis by reducing the cost of photorespiration is exactly the mechanism I expect to be exploited by any scientist/company seeking to improve C3 photosynthesis.
This work is based around is tobacco because it is an excellent model crop for transformation and in situ agricultural studies. Not because this lab is explicitly seeking to improve tobacco.
This work is probably the most impressive improvements made to the science of practically improving photosynthesis.
It was published by one of the best labs in the best department for the study of photosynthesis in the world.
I apologize for being too negative about the tobacco, you are absolutely right. The paper concludes that it is hopeful for improving grain crop yields:
> We are optimistic that similar gains may be achieved and translated into increased yield in C3 grain crops because photorespiration is common to all C3 plants and higher photosynthetic rates under elevated CO2, which suppresses photorespiration and increases harvestable yield in C3 crops.
However, does not refute the other papers I have linked about C3/C4/CAM photosynthesis that suggest that C4 may not be universally beneficial, but an optimization in certain contexts only. I do hope enhancements to plants in agroforestry settings, or restorative agricultural practices in general, are also possible. Those are badly needed if we want to regenerate topsoil and perhaps even make food production a net carbon-sink.
By the way, as always it turns out that things are not so simple as just adding a more efficient photosynthesis pathway to a life-form and everything will be better. There are complex trade-offs between the different photosynthesis pathways, and some results are counter-intuitive. Despite having evolved independently multiple times, practically no tree species with the C4 pathway seem to have evolved: only "grasses, herbs, bushes and shrubberies"[0]. According to the summary on Wikipedia[1], C4 has an advantage in high temperatures and low rainfall environments, and needs more sunlight. Furthermore, as a linked article on the researchgate discussion[3] mentions:
> While growth and photosynthetic potentials are typically less in the woody species relative to the C4 plants, the woody seedlings often tolerate low light within the C4 canopy, and will steadily grow until they overtop the C4 plants, unless they are reduced or killed by an episodic disturbance (Bond and Midgley, 2000; Wedin, 2004; Bond, 2008)
So there is a reason the trees win in the long-run if left undisturbed, and that could mean that adding C4 to other plants isn't a panacea as it is less robust under poor sunlight conditions. And in another article[3] comparing a C3 tree with one of the few known C4 trees:
> The results show that the carbon-gaining capacity of E. forbesii is comparable to that of a C3 species in a moderately cool, shaded forest environment. There appears to be no particular advantage or disadvantage associated with the C4 photosynthetic pathway of E. forbesii in this environment.
Perhaps most disturbing is this opening sentence in the abstract of a paper from 2006:
> Plants with the C4 photosynthetic pathway dominate today's tropical savannahs and grasslands, and account for some 30% of global terrestrial carbon fixation. Their success stems from a physiological CO2-concentrating pump, which leads to high photosynthetic efficiency in warm climates and low atmospheric CO2 concentrations.
Since "low atmospheric CO2 concentrations" are not exactly our biggest concern at the moment, does C4 actually help then?
Mixed lighting greenhouses are common because you have a better final product, more control over the environment, and less risk of contamination from nearby farms that are using banned pesticides. There is still quite a bit of sungrown but it requires the right climate. Typically sungrown is only suitable for manufactured products.
I'm not sure. I'd produce something you cannot grow with a plant that grows so easily they call it weed. I think it's a cool technology, they just chose an interesting target b/c of the hype about the industry.
Depends, greenhouses are pretty common too, but yes, in general, you wouldn't want someone seeing what you're growing. I find it unlikely though that these funded startups are targeting the underground market.
One of the reasons why the prohibition of alcohol failed but cannabis is illegal has to do with the DIY-supply. Everyone could brew some alcohol at home. Growing a hemp field is not so easy to do undetected and therefore much easier to enforce. Cheap artificial lighting did have some impact but this yeast (if it works well) could be a real game changer.
The prohibition of cannabis also failed. More people than ever are using it, and where it’s illegal it’s run by gangsters, the predictable outcome, exactly the same as prohibition of alcohol.
Actually I think it’s easy to grow weed in your backyard with the sun, just make sure to kill the male plants and you’re good. Whereas say growing tobacco requires much more onerous and is generally done on large farms.
I just had a vision of a bunch of cops growing exclusively male plants, and concentrating the pollen in a centrifuge device, to be used later in neighborhood foggers.
No, scientists brew THC and CBD using modified beer yeast.
Much as how coffee or tea have a lot of interesting things in them that are not caffeine, cannabis has a host of interesting things in it that are neither of those two primary cannabinoids.
yup I think that's something I'd actually call a true lifehack!
but that being said, it would be really interesting to see how a development like this could influence the nation wide development of cannabis legislation, especially down here in the Bible belt. granted I don't think it'll necessarily have a huge impact but i think having a cheap and efficient source of cannabinoids equalizing the market value and providing a larger pool to experiment upon to help support the idea of cannabis legalization
but hey even if not it's cool to see what we can hack nature to do!
Until the next hacker crosses yeast and poppies to create home-brew opiods. Very soon you might not need to be a chemistry teacher to found your own meth empire. Some sugar water and a few drops of the right yeast might suffice.
This seems like a nice end-around for the argument that weed isn't medically pure. I remember one of the arguments against medical marijuana was that CBD and THC aren't the only active chemicals that weed is producing. In this case you can create only the chemicals you want when conducting trials and manufacturing pharmaceuticals.
You take a gene from a plant, change the codons to whatever yeast uses, and then you put it into the yeast DNA.
The hard part is that presumably multiple genes are required to produce a compound, and getting all of them to produce stuff at the right rate is probably difficult as well, since we don't 100% understand how to regulate gene expression.
> You take a gene from a plant, change the codons to whatever yeast uses
This isn't necessary; codon usage is very basal. The biggest challenge you might face is prokaryotic vs. eukaryotic expression, but that's due to processing of the transcript (e.g. splicing) and the nascent peptide (e.g. all kinds of stuff). Messing with codons is simply optimization.
Expression is indeed the hard part, but not because of regulation. The general strategy is to use inducible promoters: grow the yeast, add chemical, get protein/metabolite.
Because this isn't constitutive, you can get away with a lot; even high levels of expression that will kill the cell will still produce enough of the desired product that it's an effective approach. And even relative gene-product dosage isn't too bad because you can associate the different gene products with particular known promoters and enhancers, etc.
What you have to account for is the general state of the cell; is it producing enough precursors, are intermediate products toxic, do they need to be confined for modification/processing, etc. etc. Doing complicated biochemistry gets messy.
Making simple peptides, e.g. insulin, is easy, and that's why it was the first GM pharmaceutical. Making a whole biochemical processing facility in a cell is .. a bit more of a challenge.
Step 1 is you hope that someone already identified the gene(s) that do what you want. You look it up in a database and copy-paste the DNA sequence into a word document.
Step 2 you adjust the DNA sequence for the target organism. This might mean using different codons, changing the start of the gene, etc. Usually you also append a few extra genes like GFP or antibiotic resistance genes that make working with your DNA sequence easier.
Step 3 upload your DNA sequence to an online store for DNA synthesis. A few days later you'll get an envelope with a small plastic vial with your synthesized DNA.
Step 4 Now you need to get your DNA into the target organism. I'm not sure how you do that for yeast, but there's probably a well known protocol you can follow. You might also first put your DNA into bacteria to make more of it.
Step 5 now grow your cells, and put them under the microscope, check if they glow green (that's why you put the GFP in there) and perform various other analysis to see if what you did works. It probably didn't, so you go back and change something, and try again.
I know it's the headline, that's why I complained about it. Nature should know better.
> Check the etymology of `hack` and you'll see it's a suitable word for gene editing.
I'm not making any statement based on the etymology of the word, though I don't see what in the etymology of the word 'hack' would make it an accurate description.
If a user complains about a title being baity and they have a point, we tend to change the title on the grounds that many other readers probably feel the same way.
This was a borderline case but I think they had a small point that 'hacked' was a bit baity. Also, the GP comment was heavily upvoted, indicating that a lot of users agreed.
But they don’t: the usage of “hack” in the title was squarely within the original definition of “to hack” (for instance, it’s covered by definition 6 in the Jargon File if we accept [as is generally done in hacker culture] that such hacks aren’t restricted to computer usage). In non-technical usage the term “hack” tends to mean something else but given that HN literally has “hacker” in its title, shouldn’t we accept correct usage of the word on this site, even if it doesn’t fall outside its non-technical mainstream definition?
Speaking as a biologist, the usage of “hack” in the title is completely idiomatic within the field (compare: “genome hacking”). The claim that this usage is bait is simply factually incorrect.
> Also, the GP comment was heavily upvoted
On the contrary, it was downvoted (greyed out) at the time the title was changed. It’s now back to black.
The baitiness comes from the over-use of the word in headlines. I don't disagree with you but unfortunately accuracy is not the only concern.
Re the GP comment, I was referring to https://news.ycombinator.com/item?id=19299892, which was always upvoted and quickly reached a high positive score. When users are expressing an allergic reaction like that, we've learned that it's best to just yield. A modified title can be just as accurate and usually reduces inflammation.
That said, I think your argument is at least as persuasive. We did the standard moderation thing in this case because, as I said, it's proved to be globally optimal. But it doesn't seem to have been locally optimal here.
It wasn't meant as commentary on you using it as the title. It's more that it's a stupid pop-sci term that has no place in the communication of a giant such as Nature.
Unless it changed in the three years since I left the field: no it's not. We tended to write stuff like 'genetically modified'. If one of my students would've written 'hacked' in a paper, I would've told them to change it.
THC and CBD is kind of meh. The real fun (for perverse values of fun) is when someone starts making heroin starter that can reliably turn sugar into enough opiate to risk an overdose.
And then bootleg biologics and other patent drugs, I would totally buy a yeast that let me make my own Humira and let me control the supply; even if possessing it meant a hefty prison term.
THC and CBD have a much broader user base, though. Lots of folks will use one recreationally and will try the other medicinally before considering medical-grade heroin or even recreational heroin. Even in that case, occasional users are more willing to take a lesser amount by mouth and as a bonus, the yeast gives us another way to make legal opiods.
IIRC, we also use yeast to make insulin now.
Which does mean that things like Humira - or at least precursers to it - might some day be made like this. The bigger issue with Private Person making it rather than the labs is that Private Person doesn't always do safety procedures. Sure, You the Individual might, but random Private Person probably won't. Better to buy treatment from somewhere else and look for solutions to high drug prices in other ways until we can make this safe for the general public. (Also, if you are using Humira: Sorry about that bad health luck).
> The real fun (for perverse values of fun) is when someone starts making heroin starter that can reliably turn sugar into enough opiate to risk an overdose.
I personally think "the real fun" will be when it is possible to create electronic devices that "interface" with the brain to create similar-opioids-type "highs".
If you do some digging (it's been a while since I last found it) - supposedly some research who at one time worked at ASU actually did come up with such a method; I can't recall what it stimulated with - I want to say ultrasound, but it might have been transcranial magnetism.
I honestly don't know if you can still find his work out there; I wish I had more information...
