Benefits
Body fat over six months in adults with overweight
In a 6-month placebo-controlled trial, 10 billion CFU of B420 a day did not lower body fat measured by DXA more than placebo in the main comparison (-3.0%, P = 0.28, per-protocol), and no group differed in the intention-to-treat analysis. Body fat fell versus placebo only in the B420 plus polydextrose group (-4.5%, P = 0.02, per-protocol).
Daily energy intake over six months
In the same 6-month trial, people taking B420 alone recorded about 320 fewer calories a day at the end than at baseline in 5-day food diaries, compared with about 23 fewer on placebo, a significant difference after correction for multiple groups. Only 22 people in the B420 group were analyzed, and food diaries are self-reported.
Waist circumference over six months
B420 alone did not change waist size significantly versus placebo in the trial's main comparison. Waist shrank by 2.6 cm versus placebo in the B420 plus polydextrose group (P = 0.047). A post-hoc analysis pooling both B420 groups found a 2.4 cm difference, but it was not the planned analysis.
Gut lining during a course of diclofenac
In a placebo-controlled trial in 50 healthy adults, 200 billion CFU of B420 a day for 2 weeks before and 2 weeks during diclofenac use did not reduce the rise in fecal calprotectin, a marker of gut inflammation. Blood and stool hemoglobin also did not differ between groups.
Gut bacteria and stool fatty acids (microbiota markers)
In an exploratory analysis of the 6-month trial, Lactobacillus and Akkermansia were more abundant in stool with B420 at the end of the intervention, and stool valeric acid rose more than on placebo. These are lab markers, and they do not show a health effect on their own.
Immune cell activity with B420 in multi-strain yogurt and drink (blend data)
In two small crossover trials, yogurts or drinks combining B420 with one or two Lactobacillus strains raised the share of blood immune cells able to engulf particles in healthy adults. B420 was not given alone, so these immune marker results cannot be credited to B420 itself.
Mechanism of action
Gut barrier signals (lab and animal work)
In cell culture, fluid from B420 cultures strengthened tight junctions between gut lining cells and protected them from damage by a harmful E. coli. In mice on a high-fat diet, B420 lowered plasma endotoxin and E. coli adhesion in the gut. In the human trial, serum zonulin did not differ significantly between B420 alone and placebo.
Fat mass in diet-induced obese mice (animal work)
In mice fed a high-fat diet for 12 weeks, daily B420 reduced fat mass, weight gain and glucose intolerance compared with vehicle, and lowered liver inflammation. These are animal results from the maker's research group and were the basis for the human trial.
Survival through the gut
In the diclofenac trial, stool levels of B. lactis DNA were far higher with B420 than with placebo. In the body fat trial, B420 was detected in the stool of 82% of people in the B420 groups, although the test also picked up some related B. lactis strains.
Strain identity at the genome level
A genome comparison found that B420 and Bi-07 differ from the type strain of their subspecies by only 40 to 55 single-letter DNA changes and a CRISPR region change. Such strains are hard to tell apart by standard methods, which is why the strain name on a label matters.
Clinical trials
Randomized, double-blind, placebo-controlled four-arm trial at four Finnish centers (NCT01978691): B420 10 billion CFU a day in 12 g microcrystalline cellulose, 12 g Litesse® Ultra (Polydextrose), both together, or 12 g microcrystalline cellulose, stirred into a smoothie for 6 months. Primary outcome: relative change in DXA body fat mass versus placebo. Fully funded by DuPont Nutrition & Health; five authors were DuPont employees. A later exploratory microbiota analysis of the 134-person per-protocol subset came from the same trial (Stenman et al. 2016, EBioMedicine)
225 healthy adults aged 18 to 65 with BMI 28 to 34.9; 209 in the intention-to-treat and 134 in the per-protocol population (only 24 to 25 per-protocol in the B420 alone group).
Null for B420 alone on the primary outcome: no group differed in body fat in the intention-to-treat analysis, and in the per-protocol analysis B420 alone was -3.0% versus placebo (P = 0.28) while B420 plus polydextrose was -4.5% (P = 0.02); polydextrose alone was +0.3% (P = 1.00). A post-hoc factorial analysis that pooled both B420 groups against the two groups without B420 was significant (-4.0%, P = 0.002). B420 alone lowered food-diary energy intake versus placebo. Adverse events did not differ between groups.
Randomized, double-blind, placebo-controlled parallel trial run in 2006 (EudraCT 2005-005796-15): B420 200 billion CFU a day (two capsules) or maltodextrin capsules for 14 days, then with diclofenac 75 mg twice daily for 14 more days, then 2 weeks of follow-up. The trial site was contracted by Danisco Sweeteners Oy; four authors were employees of IFF, which makes B420 (Mäkelä et al. 2021, Br J Clin Pharmacol)
50 healthy men and women aged 20 to 40 in Finland; 47 completed the diclofenac phase (25 B420, 22 placebo).
Null primary result: fecal calprotectin rose by 48.2 μg/g with B420 and 31.3 μg/g with placebo (difference 16.9, 95% CI -14.0 to 47.8, P = 0.276). Stool and blood hemoglobin changes did not differ. Stool B. lactis DNA was much higher with B420. Two episodes of severe abdominal pain occurred in the B420 group and resolved without treatment; there were no serious adverse events.