E. coli is your gut resident, not just a food poisoning bug

Say "E. coli" out loud and most people picture a recalled bag of spinach. That reflex is understandable, and it is mostly wrong. The organism in the news headline and the organism in your colon are the same species — and the one in your colon has probably been there, quietly, for years.

What it is

Escherichia coli is a Gram-negative, rod-shaped, facultatively anaerobic bacterium of the family Enterobacteriaceae whose primary habitat is the gut of vertebrates, where it lives in symbiosis with its host ¹.

Hold on to the phrase "facultatively anaerobic." It means the organism can breathe oxygen when there is any and ferment when there is none. Almost everything E. coli does for you, and almost everything it does to you, follows from that flexibility.

Where it lives in the body

The large intestine, chiefly — and it is not an occasional guest. In Martinson and Walk's 2020 review in EcoSal Plus, E. coli belongs to the gut microbiome of more than 90 % of people ². In the authors' own series — eight healthy adults giving stool biweekly for 6 months to 2 years, 95 colonies picked per sample — 87 % of the 32,470 isolates evaluated were E. coli ². Eight people is a thin base for a species-wide claim, so hold it loosely; but so far as it goes, when we say "the gut Enterobacteriaceae," we are mostly saying E. coli.

Counts vary enormously. In a different series the same review recounts — McBurney and colleagues sampling two healthy adults monthly for a year — Enterobacteriaceae in stool ranged from 10² to 10⁹ CFU per gram ², a spread of seven orders of magnitude across just two people followed over time. Even at the high end this is a minority population; the colon is ruled by strict anaerobes. "Almost universal" and "abundant" are two different claims, and only the first one holds.

Here is the part that surprises people. You do not carry a hundred different E. coli. Typically only one or two serotypes are present at any one time ² — long-lived residents that persist for months, plus transients that pass through in days to weeks ². Your E. coli is, in a real sense, yours.

Outside the gut, E. coli shows up where it has no business being: the bladder and urethra, the bloodstream, the meninges of a newborn. Same species, wrong address. That relocation is most of the harm section below.

The name, old and new

The paediatrician Theodor Escherich found the organism in infant stool and gave it its first description. The name formal nomenclature carries forward, though, is Bacillus coli Migula 1895 — the basonym on the Approved Lists — which Castellani and Chalmers moved into the genus Escherichia, named for Escherich, in 1919. The species name has been stable since; what has moved recently is the address above it. The phylum long called Proteobacteria is now validly published as Pseudomonadota, one of 42 prokaryotic phyla whose names were formally established in 2021 ¹². Papers written before that use the old word for the same group.

There is a stranger naming problem underneath. Shigella — still carried in clinical microbiology as its own genus with four species — is genetically not a separate organism at all. Lan and Reeves showed that Shigella lineages sit inside the species Escherichia coli, having arisen independently after acquiring the same virulence plasmid ¹¹. The clinical label survives for practical reasons. The family tree does not support it.

What it does for us

Two entries, and I want to be careful with both.

It holds the aerobic niche. E. coli is the predominant aerobic organism of the vertebrate gut and lives there in symbiosis with its host ¹. The mechanism usually attached to this — that by consuming the traces of oxygen diffusing in from the gut wall it helps keep the colon anaerobic for everyone else — is a plausible and widely repeated inference from its physiology. Said out loud: I have not found a human trial demonstrating that benefit in people. Treat it as mechanism, not as proven effect.

It builds vitamin K2. E. coli carries the pathway for menaquinone synthesis, the bacterial form of vitamin K ⁴. The caveat here is not small: making menaquinone in the colon and absorbing usable amounts of it are separate questions, and the second is far less settled than supplement copy suggests.

And one strain has a real trial behind it. E. coli Nissle 1917 is a licensed probiotic. In Kruis and colleagues' 2004 trial in Gut, 327 patients with ulcerative colitis in remission were randomised to either Nissle 1917 (n = 162) or standard mesalazine (n = 165) for 12 months; in the per-protocol analysis relapse occurred in 36.4 % of the probiotic group (40 of 110) and 33.9 % of the mesalazine group (38 of 112) — equivalent efficacy ³. Two caveats, both important. This was equivalence to an established drug, not a comparison against placebo. And it is evidence about one strain in one disease, which tells you nothing about the E. coli in your own colon.

What it does against us

Now the ledger's other side, and it is heavy.

Scale first. In the Global Burden of Disease analysis of 33 bacterial pathogens, an estimated 7.7 million deaths in 2019 were associated with those bacteria — 13.6 % of all deaths worldwide ⁵. Five pathogens, E. coli among them, accounted for 54.9 % of that total ⁵. On the resistance side the picture is sharper still: of an estimated 4.95 million deaths associated with bacterial antimicrobial resistance in 2019, E. coli was the single leading pathogen ⁶.

Urinary tract infection is the everyday face of it. Foxman's review counted nearly 7 million office visits, 1 million emergency department visits and 100,000 hospitalisations a year in the United States, with nearly 1 in 3 women having had an infection requiring antibiotics by age 24 ¹⁰. Most of those are E. coli that travelled a few centimetres from the gut.

Shiga-toxin strains are the food-poisoning ones. The 2011 German outbreak of E. coli O104:H4 gave a grim natural experiment: 3,816 reported cases, 54 deaths, and haemolytic-uraemic syndrome — a kidney and blood complication — in 845 patients, 22 % of the total ⁸. Unusually for this disease, the patients who went on to develop haemolytic-uraemic syndrome were mostly adults — 88 % of them, median age 42 — and 68 % of them were women ⁸. The source was traced to sprouts; among restaurant visitors, those served sprouts had a relative risk of 14.2 for illness ⁹.

And the treatment trap. In Wong and colleagues' prospective study of 71 children with O157:H7 diarrhea, haemolytic-uraemic syndrome developed in 10 children, and antibiotic treatment carried a relative risk of 14.3 (95 % CI 2.9 to 70.7) for developing it ⁷. Said out loud: 71 children is a small study and that confidence interval is enormous. But the direction is consistent with what we know about toxin release, and it is why "just take an antibiotic" is the wrong instinct for bloody diarrhea.

How to hold both facts

E. coli is the best-studied organism in biology and one of the deadliest bacteria on earth, and it is also a quiet, near-universal resident of healthy people. Both are true, and the resolution is not "it depends on your immune system." It is that the species is a wide genetic spread. Most of its lineages, its family lines, are commensal: they live in the gut, feed on what passes through, and do the host no harm, and they are built for exactly that life. A minority carry toxin, invasion or urinary-adhesion genes — the equipment for poisoning tissue, breaking into cells or clinging to a bladder wall — and those are the ones that turn pathogen, disease-causer, somewhere else ^{1 11}. The reasonable conclusion is not to fear the organism in you. It is to respect what a few of its relatives can do when they arrive somewhere they do not belong.