← All articles
A researcher seen from behind walking a dim clinical corridor at night under cool fluorescent light, with no labels or signs.

One gut enzyme sharpened the insulin response of old male mice

ResearchMetabolism

Cell Metabolism reported on 16 November that raising one bacterial enzyme in aged male mice improved an insulin test about 20 weeks later. A similar bile-acid pattern shows up in centenarians, and the paper does not measure lifespan.

Male mice at 82 weeks received one oral dose of a gut bacterium built to make an extra enzyme. About 20 weeks later their insulin tolerance was better than that of mice given the same bacterium without the enzyme. The enzyme is bile salt hydrolase, which gut microbes use to cut bile acids free from taurine. A chemical pattern in the same direction appears in people who have lived to 100. The paper does not say that any mouse or any person lived longer because of it.

This article is information, not a substitute for a physician. It is one experiment in male mice plus a second look at samples already collected from people. It is not a bile-acid supplement, not a probiotic to buy, and not a diet to start. A product marketed as a copy of this result is something to raise with a doctor.

What ordinary aging looked like in these males

Amir Zarrinpar is the corresponding author. The group at the University of California, San Diego studied male wild-type C57BL/6J mice with free access to a standard chow, Inotiv Teklad 7912. In the age comparison, metabolic-cage readings used 8 mice per group, taken when the young mice were 8 weeks old and the aged mice were 84 weeks old. The aged mice weighed more and had more gonadal fat. They ate more during the light phase and less during the dark phase, with a similar amount of food across the day, and they moved less. Insulin tolerance was worse and glucose tolerance was altered. Those tolerance tests were not on the same day as the cage readings: glucose tolerance was tested at 12 and 88 weeks, and insulin tolerance at 14 and 90 weeks. Insulin after a meal was higher. Fasting insulin was not.

In the liver, more than 3,000 genes changed with time of day in the young mice, and about 200 did so in the aged mice. Core clock genes lost that daily swing. Expression of Cyp7a1, the gene for the enzyme that limits how fast the liver makes primary bile acids, was lower in the aged liver. In stool, aged mice had higher ratios of several unconjugated bile acids to the taurine-conjugated forms: deoxycholic acid and the alpha, beta, and omega muricholic acids. Stool lithocholic acid was higher. Ursodeoxycholic acid was not. Transcripts related to bile salt hydrolase were higher in aged mice during the dark phase. The authors read that from sequencing, not from a direct test of enzyme activity in the gut.

Which bacterial species were present is a poor witness for age in this colony. The young and aged mice came from different mothers and different cages, so a shift in the species list cannot be blamed on aging. The host measurements the authors checked did not track with cage.

The centenarian pattern is a reanalysis

No new volunteers were recruited. The human figures come from a second analysis of bile-acid measurements and stool metagenomes in a Japanese study that Nature published in 2021, with Y. Sato as first author (PMID 34325466). Centenarians had a higher ratio of deoxycholic acid to taurodeoxycholic acid than young adults or elderly adults, and a higher ratio of chenodeoxycholic acid to its taurine conjugate than elderly adults. A gene annotated as this same hydrolase, cbh (KEGG K01442), was lower in elderly adults than in young adults and higher in centenarians than in elderly adults.

The authors call the human result corroboration. People who reach 100 are a selected group, so the pattern can be survivor bias. They also say aged mice and centenarians are not the same thing, and that other cohorts are needed before the pattern can be treated as general.

One extra enzyme, one dose, male mice only

The intervention used 82-week-old males. Twelve received EcAZ-2, a native E. coli strain from the mouse gut, engineered to express a bile salt hydrolase from Lactobacillus salivarius JCM1046. Eleven received the same strain without that enzyme. Both strains were still present after a single gavage. In the dark phase, the enzyme group had higher ratios of deconjugated to conjugated bile acids. The total amount of bile acids was similar.

About 20 weeks after that dose, insulin sensitivity was significantly better in the enzyme group, in the paper's own wording, and glucose tolerance was modestly better. By 32 weeks the enzyme group showed changes consistent with lower body weight, a lower percentage of fat, and a higher percentage of lean mass. The respiratory exchange ratio was lower, which fits greater use of fat as fuel. Serum from the enzyme-treated mice increased maximal mitochondrial respiration and spare respiratory capacity in C2C12 mouse muscle cells. In the ileum, signaling related to NF-kB was lower. The authors mark that immune change as hypothesis-generating, not as the cause of the metabolic result. In the liver, the day-night difference in Clock, Cry1, and Arntl (Bmal1) was clearer with the enzyme than without it.

They describe the effects as modest, in naturally aging wild-type mice on standard chow, not in a model built to create a large disease effect.

Taurine, the amino acid these conjugated bile acids carry, has its own contested record in aging research. A separate mouse study is what extra taurine did in old mice. This paper's point is different: it is about splitting taurine off bile acids, not about swallowing more taurine.

What a reader should not take from it

Lifespan was not an endpoint. A better insulin test in male mice is not a longer life.

There was no arm of aged mice given only the fluid used to deliver the bacteria. The comparison is the enzyme versus the same bacterium without the enzyme. The authors cite their earlier work, in which this bacterial chassis did not change microbiome composition, body weight, or food intake compared with a vehicle. That comparison was not repeated in the present experiment.

The study was done only in males. Several endpoints used small groups and, the authors write, need independent replication. One Lactobacillus enzyme does not represent the many bile salt hydrolases in a human gut. Amir Zarrinpar, the corresponding author, is a co-founder and equity holder in Endure Biotherapeutics. The University of California, San Diego reviewed that arrangement. The funders, including the National Institutes of Health and the Department of Veterans Affairs, had no role in study design, data collection, interpretation, or the decision to submit the work for publication.

For a healthy adult who wants more healthy years, the finding is a lead, not a lever. A bile-acid pattern seen in centenarians can be pushed further in old male mice, and insulin sensitivity moves with it. Nothing in the paper shows that copying the pattern extends human life.

Sources

  • Zarrinpar and colleagues. Microbial bile acid deconjugation counters metabolic decline in aging. Cell Metabolism, online 16 November 2026. DOI 10.1016/j.cmet.2026.10.016. PMID 43038987.

  • Sato and colleagues. Novel bile acid biosynthetic pathways are enriched in the microbiome of centenarians. Nature, November 2021. PMID 34325466.

Articles in this section are educational and are not medical advice, a diagnosis, or a prescription. Consult a qualified professional before acting on anything you read here.

Read next