Cheating at Evolution
Today's lesson from nature: Never hesitate to rip off the work of someone who spent a long time figuring something out.
Last time, we talked about the desert woodrat—a cute little critter that has a cast-iron stomach, thanks to a few specialized gut bacteria that help it digest toxic plants. These plants aren’t the ”kill you in a few bites” kind of toxic, they’re more the “don’t bother unless you’re starving” kind. The ratio of nutrition to indigestion is not good.
It’s the fact that these plants aren’t great—but good enough to still get eaten—which sets the stage for gut bacteria to pull one of their greatest tricks: horizontal gene transfer.
Typically, we think of evolution as happening vertically—i.e. you develop a fit new trait, so you have more offspring than average, and eventually that trait becomes dominant as your progeny spread their seed far and wide.
But bacteria don’t have to play by those rules: they can pick up DNA from the environment and put it straight to use. Any dead cell in their vicinity is a wellspring of potentially useful new traits.
Compared to how we traditionally think of evolution happening, this is an unbelievable gain in efficiency. Instead of inventing everything themselves via the “random walk” of mutations to existing genes, bacteria can simply adopt tried and true genes from the environment. After all, if there’s a lot of a certain gene floating around, the organism it belongs to must be doing pretty well—so it’s very likely to be useful.
All of this ends up being of great advantage to the creosote woodrats, because it effectively lets them cheat at evolution. Consider: they eat a lot of juniper leaves, which are pretty nasty. But in doing so, they inadvertently eat a lot of the bacteria and fungi that live on the juniper plant. These microorganisms wouldn’t be living there if they hadn’t figured out some way to deal with the toxic terpenes that the plant secretes. That means that, in the biochemical milieu of the woodrat’s gut, you’re likely to find the exact genes necessary to detoxify those leaves and extract nutrition from them without getting your liver destroyed.
Nine hundred ninety-nine times out of a thousand, we’re not Kirby. Simply eating an organism with a certain “power” isn’t enough to gain that power for yourself, because it’s a chancy process. But one time is all it takes: if one of your gut bacteria picks up the right gene, it gains a trait that benefits both itself and its host, because a lot of the things that are toxic to mammals are toxic to microbes, too.
With that advantage, it can out-compete all of its closest relatives in your microbiome and become the dominant strain of its species. Suddenly, eating juniper leaves has semi-permanently improved your ability to eat juniper leaves, all within one lifetime. The ghost of Jean-Baptiste Lamarck (of “giraffes got long necks by stretching them a lot” fame) does a little touchdown dance in the great beyond.
While the ability for a microbe (and by proxy its host) to gain genes like this is wildly useful, it becomes revolutionary if you can pass those useful, acquired traits on to your offspring. Fortunately, we’ve got that covered too. Certain “core” components of the mammalian microbiome are transmitted vertically, reliably passed down from generation to generation. [For a perhaps-unnecessarily heated discussion of the possible mechanisms behind this process, see “Please Stop ‘Vaginal Seeding’”.]
This evolutionary backdoor turbo-charges the acquisition of useful traits. We have reliable vertical transmission of a “core” microbiome, and horizontal gene transfer from environmental microbes to that core microbiome. Obviously, these core microbes came from somewhere, so straight-up “adoption” of environmental bacteria into the gut must also play a role—but that ventures into the question of how a person’s body compiles the list of "good” bugs, and I don’t know enough about this to speculate on it.
As Below, So Above
I set a good amount of store by the results showing that lactation is a major vector for gut-bacterial transmission, but a bacterium doesn’t need to get passed down in milk to get passed down. Plenty of people who aren’t breastfed develop healthy microbiomes, which usually resemble their families’. Our bodies put a lot of effort into creating the right “niches” for bacteria, so that all it takes is one cell finding its way into the right spot, where it can grow and flourish.
But in most of the animal kingdom, at least among our mammal cousins, they don’t just sit around waiting for one of dad’s farts to do the trick: they go right to the source. Scientists call it coprophagy, which is just a way of saying “eating shit” that’s fancy enough to publish in a journal like Nature.
Young foals and calves will seek out their mothers’ poops, inoculating themselves early-on with the bacteria that will help them do the tough work of extracting calories and other useful nutrients from grasses.
Throughout life, mice and rats will eat poo in order to get important vitamins like B12, (which is produced by bacteria in the colon, but needs to pass through the upper GI tract to be properly absorbed). This behavior tends to effectively “pool” the microbiomes of cohabitating animals, which turns out to be a useful trait in scientific studies: all you have to do is keep some rodents in the same cage for a while, and their microbiomes become relatively homogeneous. (It’s also occasionally a very annoying trait if you’re trying to model human microbiome interventions, because—as I once heard a researcher put it—“most humans are not coprophagic”.)
Even dogs! I’ve heard a friend complain that he can’t take his eyes off his pup at the dog park for a minute without it trying to gobble up another dog’s turd. “Stephen,” he asks me: “what is wrong with my dog such that he sees a literal piece of shit and thinks mmm, tasty snack?”
And I had to explain to him: It’s a feature, not a bug! He probably smells something he’s missing in there, and knows instinctively that this is the way to get bacteria which will produce it for him.
We’ve established that the microbiome enables accelerated evolution, because bacteria reproduce faster, and can acquire genes from other bacteria within a single generation without crossbreeding. Coprophagy is simply the next step up from that: horizontal gene transfer at the macroscopic scale. As below, so above.
Once a single woodrat gets the bacterial gene that lets it eat juniper leaves without any ill effects, all the other woodrats in its den are going to get that gene in a few weeks, too. And when they scatter out to meet other woodrats from other dens, and shack up together, their partners and children can gain that trait as well.
Those that don’t are slated to be outcompeted pretty quickly, by their cousins who suddenly have unfettered access to a unique and abundant source of calories.
In that sense, you could say it’s “eat shit or die”.
And when we look around at the rest of the mammals, and they’re all doing it—when dogs, informed by a sense of smell that is incomprehensibly exquisite compared to ours, regularly decide to chow down on a new friend’s turd…you have to wonder: are we the ones being dumb here? What do they know that we don’t?
What I’m Not Saying
I’m not saying we should all start greeting each other by sniffing assholes, eating poop willy-nilly, for the same reason that sensible parents don’t let their kid download things via BitTorrent on the family computer. It’s risky: you might get a free copy of Photoshop, or you might get a virus that fries the device. Our disgust response is there to protect us from illness.
Thus, the dream of next-generation probiotics: An “app store” for the biological software system that is your microbiome, where if you have high cholesterol, you can go to the drugstore and grab a bacterium that will sequester it as it’s cycled through the GI tract—letting you shit it out rather than reabsorbing it. High cortisol? Rummage around, you can probably find something with an 11-β-hydroxysteroid dehydrogenase. Entering menopause? A Slackia isoflavoconvertens will convert compounds from your diet into a natural estrogen-mimetic that reduces the severity of hot flashes. It's genes a la carte, and it’s so much easier than drug development or gene therapy, because it’s how the body is naturally supposed to work: these are human genes, just modular and mobile.
Because evolution happens fastest at the microbial scale, the most recently-evolved traits—i.e. the interpersonal phenotypic differences within our species—are disproportionately likely to be microbial in origin. This means that, for practically any function that would be good for a human body, there’s likely already a microbe that does it, out there in someone’s gut. If you want to treat anxiety, you could put someone on drugs, or try and hack their GABA receptors to be more responsive with gene therapy…or you could just give them a Gammaproteobacterium that produces natural benzodiazepine-like compounds from the food they eat, or a bacteriophage that targets whatever pathogen is turning their food into p-cresol.
There are situations where this isn’t applicable, of course—things like sickle cell anemia or cystic fibrosis come to mind. But these actual genetic diseases, where you’ve got a definite mutation that’s causing an illness, are such a small percentage of chronic illnesses. Autism, schizophrenia, depression, anxiety, ankylosing spondylitis, Parkinson’s, multiple sclerosis—in the vast majority of these cases, we can’t pin down a gene, or even a set of genes responsible. It’s frustrating for the patients who’ve been begging their doctors for answers to no avail, but it’s also a reason to hope.
What I AM Saying
Earlier, I compared eating shit to downloading random files off a P2P server, but this is unfair to shit. See, you can usually tell a happy, healthy person just by looking at them and talking to them for a while. And, because the microbiome has such a drastic influence on a person's overall well-being, a healthy person is almost sure to carry a healthy microbiome.1 Add in our ability to test for things like parasites and major pathogens, and poop is generally a lot less dangerous than your disgust response would have you believe.2
Point being: the dogs are onto something. Coprophagy is the easiest and most impactful way to alter an animal’s biology, and being able to observe the donor’s traits gives you a good idea which direction that transfer is going to push things. The same is true of fecal microbiota transplant, which isn’t that different from coprophagy at the end of the day, when capsules are used.
Whatever you want to call it, it’s a kind of black magic—inasmuch as we don’t fully understand all the traits the microbiome is responsible for. For instance: there are reports of patients with alopecia arreata receiving FMT to treat C. diff, and suddenly starting to grow hair again. Why? Fuck if I know! A poop is like a grab-bag of 20 semi-random traits from the donor. Your risk of colon cancer might go up or down, depending on what theirs is relative to yours. You might get a sweet tooth, or more patience, or bad PMS.
Despite the chaos it entails, black magic still has its practitioners for one simple reason: it delivers results, and fast. If you have Parkinson’s or multiple sclerosis and your donor doesn’t, that grab bag is worth opening, because it’s likely your best shot at stopping the disease. You might get your donor’s high blood pressure in the process, but I’d call that a bargain.
The “app store” is a long way off. It’s going to be a long time before all the functions in a healthy microbiome are understood. Even longer until precision solutions based on that understanding are available. In the meantime, there’s nothing stopping people from trying microbiome modulation themselves. All you need is a very healthy (and very understanding) friend.
🖖🏼💩
There are exceptions—there are subtle pathogens, and viruses like hepatitis can be transferred in stool. This has a long latency before it starts to do visible damage.
I should point out that poop gets drastically more dangerous the longer it’s been out of a person’s body. The species that make up a healthy gut are mostly anaerobic, meaning they die upon exposure to oxygen. These guys have no incentive to make you sick; their only ticket to a continued existence is to keep you healthy. Things like Salmonella, E. coli, and Vibrio cholera are much better at dealing with O2, so they have the option of finding their way into another host by giving you diarrhea.
Talked to a VC who was excited about a proposed Shit Bank. There is money behind some of these ideas. I wonder how much it matters that you eat similarly to the donor. Surely quite a bit!