Gut Reactions to Studying the Microbiome

A microbiome analysis reveals the present and, surprisingly, the past.

As a human genomicist, I long knew bubkes about microbes, except that they’re dauntingly diverse and do serious magic, inside us and out. Having guided family, friends, and even Ozzy Osbourne through the analysis of their own murky genomes, I now know we stick with complexity when it’s personal. Like others, I wondered what many kinds of inner data a microbiome analysis might reveal. Having one done seemed a great next step.

But which microbiome? Like our planet, with its reefs, forests, and tundra, our bodies harbor tissue-distinctive lifescapes. We’ve explored spots open to swabs: skin, ear, nose, and genital tracts, plus the mouth and gut ends of our food canal.

The last intrigued me the most. Our guts’ alchemy, after all, may instinctively engross and inform us. Happily, the crowd-funded scientists at American Gut offered a deal: $129 bought one swab for my privileged, obsessively quantified self.

If microbial selfies seem faddishly new, know that we’ve made them since 1683, when microscopist Anton van Leeuwenhoek found tiny swimmers in his own tooth gunk. Though Anton marveled at that first human glimpse of lone bacteria, the following centuries jaded us. Pox after plague were traced to infections — big, easily tracked blooms of microorganisms — and germs were typecast as foes more than fellows, let alone friends.

But if microscopes unmasked some long-imagined devils (Anton apparently sketched Campylobacter rectus, a goblin of gum rot), cheap DNA sequencing has sanctified other germs, finding them at home in healthy tissues. Some longtime tenants may even help keep us well by repelling deadly rivals, processing nutrients and drugs, and perhaps even steering us to prolong our relationship with our germs through personal behaviors that benefit both parties.

But with microbiomes, as with human genomes, beware pat stories, because caveats abound.

First, we often census microbes only roughly, reading just one DNA tract (encoding ribosomal RNA) as a blunt ID that lumps together diverse members of a typical bacterial family. Doing so leads us to miss some clans (archaea, molds, and tiny plants and animals) and leaves viruses, which lack such IDs, wholly undocumented.

Second, like the stick figures in a human census, microbiome data rarely say who’s naughty, nice, or both. Bacteria with the same ID but genomes differing elsewhere can act quite differently. One’s knack for infecting or poisoning may even trace to its own microbiome: tiny phages, the viruses who hop among bacteria, and whose own genomes vary importantly, too.

Last, as communities, microbiomes vary from tissue to tissue, person to person, and over time. Our first fuzzy views of these worlds won’t fully show how germs act and interact to shape our health. Imagine Martians studying complicated earthlings by counting blurry pixels on a street — brunette, redhead, hat, Dalmatian — and you’ll grasp how in microbiomes, as in human genomics, we risk mistaking correlation or even coincidence for causation.

To really know our germs, we’ll have to read their diverse whole genomes, and their RNA too, in many tissues, from many people, over and over. That’s a lot more data to gather and process; but this is the only way to reveal how microbes shape and track our bodies’ states, beyond simple hero and villain caricatures. For better or worse, those data movies start with frames from you and me.

What did this first snapshot show? Compared to a typical American gut, mine harbored lots of firmicutes and actinobacteria, few bacteroidetes, and hardly any proteobacteria. Overall, many stiff-celled firmicutes and actinobacteria live as stodgy chemical wizards, a bit like our macrobiome’s plants and fungi. Some squishy bacteroidetes and proteobacteria, meanwhile, prowl around boldly, more like nimble animals (some proteobacteria, including our mitochondria’s ancestors, have even colonized within lasting lines of bigger cells).

But these big, gut-typical branches of the bacterial family thicket bristle with stunningly diverse twigs. Within each branch, some twigs harm us, others don’t, and only a few fit in any one person. So why should I carry one mix and you another? Our gut’s genetically distinctive human cells may turn out to matter in ways that could be revealed by surveying more people. But their direct influence likely pales next to the powerful influence of our diet.

My own palate runs coarse and earthy. Greens, nuts, kelp, and fatty fish galore. Little dough or sugar. After boiling an egg, I may eat the shell (calcium, folks!). But most tellingly, I scarf yogurt, sauerkraut, natto, and other fermented stuff. I precook starches and proteins, and many such foods teem with firmicutes — which show up on cue in me, and other American Gut folk who eat and exercise similarly.

My actinobacteria thicken the plot. Some thrive in soil (when soaked, they give rain its pungent scent) and could hitchhike on the radishes I love. But my most plentiful ones, corynebacteria, don’t fit that profile. Rare even in vegan guts, they greatly puzzled American Gut co-founder and biologist Rob Knight — and revived an old mystery for me.

Seventeen years ago, in post-Soviet Georgia, an unusually nasty corynebacterium, C. diphtheriae, reupholstered my tonsils, halting a long-planned trip. The diagnosis was startling, as I had been vaccinated and boostered to boot. How had diphtheria caught me?

Cured by an antibiotic, I had long quit guessing. But like word from an old sweetheart, the new gut data rekindled my wonder. Had that throat infection snuck downstream, quietly hiding out for years? Perhaps, but my gut housed various corynebacteria, not just C. diphtheriae. Have my tissues always welcomed that whole clan especially warmly?

Color me curious but skeptical. Diphtheria vaccine, after all, trains our immune systems not on C. diphtheria cells but on a toxin they spew. Even if my body flunked such training, I needn’t coddle corynebacteria themselves. Indeed, my whole genome — even a close look at immune-crucial HLA genes on chromosome 6 — shows no obvious soft spot for any particular bacteria.

That may reflect our slowly fading genomic ignorance, so I’ll certainly keep my eye out for new scientific research. But my main next step is to confirm whether I chronically harbor corynebacteria at all — or if my gut’s bumper crop was just a blip. To find out, the American Gut lab will more thoroughly study a stool sample I sent this spring. We’ll check other tissues typically friendlier to corynebacteria, such as my (ever-stuffy) nose.

In coming years, microbiomes will likely anchor our personal big data. Because they change readily, in health-relevant ways, we may read and tweak them more often, with less concerns about privacy, than our own genomes (cue drooling clinical and direct-to-consumer companies).

Each of us will find personal puzzles in such data, like my quirky corynebacteria. Such surprises will guide us as we explore deeper — what we read and pester our doctors about — and may, excitingly, connect us with others, to pool enough personal data to yield new scientific insights.