Dr. Axe Wire · Study decoded · gut

Konjac Fiber Fed Good Gut Bacteria in Lab Study

A modified konjac fiber encouraged beneficial gut bacteria and short-chain fatty acids while suppressing harmful microbes in fecal fermentation samples.

By Dr. Josh Axe · November 7, 2026 · 1 source · Watch

Photo: Pexels

A modified fiber from konjac root fed beneficial gut bacteria and suppressed harmful strains, according to a study published November 2, 2026 in Food & Function.

Researchers prepared a modified fiber, konjac oligo-glucomannan (KOGM), by breaking down konjac glucomannan with an enzyme from the probiotic bacterium Pediococcus acidilactici. The fragments averaged 13.4 kilodaltons, produced at 50 degrees Celsius, pH 6.0, and 180 minutes of enzyme exposure. In growth tests, KOGM selectively fed beneficial bacteria, including Lacticaseibacillus paracasei and Lactiplantibacillus plantarum, while suppressing harmful bacteria such as Klebsiella pneumoniae and Enterobacter species.

The team then tested six mixtures in human fecal fermentation samples over 24 hours: a blank control, a commercial comparator fiber called mannan oligosaccharides, unmodified konjac fiber, the new KOGM fiber, and each combined with the bacterial enzyme. The fiber-plus-enzyme combination reached the lowest pH, 3.89, with the highest measured acetate (2,062.72 micrograms per milliliter), propionate (1,356.81 micrograms per milliliter) and valerate (16.89 micrograms per milliliter) — all short-chain fatty acids gut bacteria produce from fiber. A separate unmodified-fiber-plus-enzyme mixture produced the highest acetate among all groups, while unmodified konjac fiber alone produced the most butyrate, at 206.7 micrograms per milliliter.

Short-chain fatty acids feed the cells lining the gut wall and are linked to a healthier gut barrier. The researchers reported that both konjac fibers increased microbial diversity and sped up community stabilization during fermentation. KOGM specifically enriched beneficial groups, including Bacteroidetes and Actinobacteria, boosted helpful genera like Bifidobacterium and Prevotella, and reduced potentially harmful bacteria including Escherichia-Shigella.

Short-Chain Fatty Acids Produced After 24-Hour Fermentation

Fecal fermentation samples, lab dish model

Acetate

2062.72 µg/mL

Propionate

1356.81 µg/mL

Butyrate

206.7 µg/mL

Valerate

16.89 µg/mL

Food & Function, 2026, Yang et al.

This is a lab dish fermentation model using fecal samples, not a feeding study in living animals or people, so it does not show how this fiber behaves once it passes through a digestive tract.

Dr. Axe’s Take

How a fiber gets broken down changes what it feeds in the gut, and that matters more than just calling something fiber. This enzymatic processing concentrated konjac's effect on favorable bacteria like Bifidobacterium while crowding out less helpful strains. For a reader at home, the practical move isn't this lab-made compound — it's eating fiber-rich plant foods that feed the same beneficial bacteria: root vegetables, oats, beans, onions and garlic. I'd rotate those into meals daily rather than wait on a prepared konjac extract, since whole-food fiber already does similar work for your gut lining.

— Dr. Axe

Take the Gut Health Quiz →

Study summary

Who: Fecal fermentation samples tested in a lab dish, not a human or animal feeding trial.
Design: In vitro fecal fermentation model comparing six treatments: blank control, commercial mannan oligosaccharide fiber, unmodified konjac fiber, modified konjac fiber (KOGM), and each konjac fiber combined with a bacterial enzyme.
Dose: KOGM produced via enzymatic hydrolysis at 50°C, pH 6.0, 80 U/mL enzyme, for 180 minutes; fermentation ran 24 hours.
Primary result: Fiber-plus-enzyme mixture reached pH 3.89 and produced 2,062.72 µg/mL acetate, 1,356.81 µg/mL propionate, 16.89 µg/mL valerate after 24 hours.
Secondary: Unmodified fiber alone produced the highest butyrate (206.7 µg/mL); modified fiber enriched Bifidobacterium, Prevotella and Bacteroidetes while reducing Escherichia-Shigella and Proteobacteria.
Funding / conflicts: Funder listed in the journal record: National Natural Science Foundation of China. Conflicts not reported in the abstract.
Date: Nov 2, 2026