Forget Gene Therapy. Microbes are the Future of Medicine.
Why rewire when you can reprogram?
Your genome is a lot like the hardware of the device that you’re reading this on. As the name implies, it’s hard-coded. Static. Sure, epigenetic processes turn components of it off or on—the same way your computer will send power to a certain antenna when you turn Bluetooth on, or crank up the fans when the temperature hits a certain point—but unless you drop the device and break a wire, the equipment inside stays pretty much the way it was when it came off the factory line.
But let’s say you want to do some complicated math, graph something, and you don’t have the program to do it. One solution to that problem is to give your device the hardware: Crack open its case, gut a TI-83 graphing calculator, and wire it in to communicate with your motherboard.
Sounds a little extreme, when you can download a graphing calculator app and call it a day.
See, if your genome is the hardware, your gut microbiome is the software. It’s mutable, flexible. It was made to be changed, shared, rewritten and optimized. The whole point of a computer as we know it today is that it can simulate almost any logic circuit, run arbitrary programs. And the more we learn about our biology, the more it looks like the mammalian body plan parallels this function—that its most powerful feature is the ability to host a complex and variable ecosystem.
Most people have over a hundred different species inside them, from all across the kingdom Bacteria. Each one is a modular packet of genes, producing enzymes and proteins that perform useful functions which your body would otherwise have to simply cope without. For example, look at the desert woodrat.
No, really, look at it; they’re very cute.
Their diet is something that baffled scientists for years. Its main components, juniper and creosote, are toxic to most mammals in any substantial quantity. A little hint of juniper’s terpenes is a nice touch in a gin & tonic, but try chewing on a leaf from an actual juniper bush and you’ll see what I mean.
But the woodrat munches away, perfectly happy—it’s got some metabolic edge that gives it free access to a food source which most of its competitors wouldn’t touch.
Maybe you see where this is going. Guess what happens if you give one antibiotics.
Suddenly, juniper is as toxic to the woodrat as it would be to you or me. Genomic studies on the woodrat had turned up nothing, the same way cracking open your computer case and looking at the hardware won’t usually tell you much about what it can and can’t do.
And when you think about it, of course it’s not in the genome. Why fuck around with your own chromosomes when you have a whole biological sandbox in the gut?
De-risking Discovery
If one of your intestinal cells develops a mutant enzyme that lets it process and detoxify some dietary poison, that doesn’t do you much good, does it? It’s one cell out of billions. For a genomic mutation to work to its host’s advantage, it has to happen in the gamete, the sperm or egg cell, in order to end up in every cell of the body. And if that mutation breaks some important enzyme, now you’ve got a congenital disorder. Because any evolved organism is a finely tuned system, a random change is much more likely to break something than to improve it.
But imagine that same mutational process occurring in the microbiome. It’s a population in constant flux; cell numbers doubling every twenty minutes, dying or leaving your body by the billions with each bowel movement. Here, if one bacterial cell develops a mutant enzyme that makes it better than its peers at getting the calories out of complex carbohydrates, or destroying a plant toxin that inhibits its growth, the constant turnover gives that cell the opportunity to grow faster and replace its closest relatives—which are functionally identical, except for the lack of that new enzyme.
The power of the microbiome as an evolutionary tool lies in its ability to propagate favorable mutations without assuming the risk of unfavorable ones. If a mutation in a single bacterial cell breaks an important enzyme, the bacterium dies—but who misses it? You’ve got forty million clones that are pretty much identical, except for that fatal mutation.
What’s favorable for an individual microbial species isn’t always favorable for the host, but it tends to be. Nature is not kind to creatures that ravage their ecosystem in pursuit of relentless growth—at least not for very long.
Long story short: Evolution happens over generations. That’s an inevitable fact of nature, because natural selection is a progress engine that runs on death. The smart solution, for creatures like us, is to adopt organisms that have a generation time of minutes, rather than decades, and let them do the dirty work of serving as the proving ground for new mutations.
As a bonus, the epigenetic regulation is built-in by default. Are you eating a lot of tree bark because your crops failed and you’re starving? Bacteroides that are good at digesting xylan and cellulose will proliferate, and with every meal you’ll get better at extracting calories from bark. One interesting implication of this is that sudden changes to your diet might be a little problematic; if your tree-bark-eating microbiome is suddenly flooded with animal protein, the dominant bacteria might not know what to do with it. Protein-degraders will bloom, but they’ll have to fight for turf—a “rumble in the jungle”, if you will.
There are a few more layers worth exploring here, and once you take them into account, things really start to go hockey-stick compared to typical evolutionary timelines: horizontal gene transfer, vertical transmission, and coprophagy, but we’ll get to those next time—and then dive into why microbiome modulation is a vastly more promising approach than gene therapy for most diseases.
Update: Part II of this post series: Cheating at Evolution