THE SCIENCE BEHIND WINNOW

The strains make the difference. That's why we publish the data.

Plastic binding is strain specific, wildly variable, and better when the right strains work together. Winnow is built with the best strains for gut health + microplastic binding.

CLINICAL FOUNDATION

The clinically-studied strains inside Winnow.

These strains form the clinical backbone of Winnow. Their evidence comes from randomized controlled trials in human populations.

L. rhamnosus

ATCC 53103

1,977

publications

204

RCTs

Plus 32 meta-analyses and 50 systematic reviews — among the most rigorously evaluated probiotic strains ever studied.

CLINICAL EVIDENCE COVERS

  • Supports digestive comfort during and after antibiotic use

  • Helps your stomach bounce back when things feel off

  • Supports healthy skin from the inside out

  • Daily immune support to help you feel your best through the seasons

  • Promotes gut comfort and easy, predictable regularity

L. reuteri

DSM 17938

301

publications

104

RCTs

Plus 8 meta-analyses and 20 systematic reviews. Among the most clinically documented strains for pediatric GI health.

CLINICAL EVIDENCE COVERS

  • Supports a smooth, balanced digestive rhythm

  • Helps your gut recover quickly when it's working overtime

  • Reinforces a strong, resilient gut lining

  • Supports a balanced, well-regulated immune response

L. acidophilus

ATCC SD5212

96

publications

5

RCTs

Plus 1 systematic review. Studied for urogenital and digestive health endpoints in combination with L. acidophilus isolates.

CLINICAL EVIDENCE COVERS

  • Makes dairy easier to digest

  • Everyday digestive comfort and regularity

  • Supports your body's natural defenses

  • Supports a healthy urogenital microbiome

Counts pulled from PubMed publication-type filters · April 2026. RCT count reflects studies tagged "Randomized Controlled Trial" specifically; total clinical trials including phased trials are higher. Clinical evidence is for gut, immune, and urogenital endpoints — not microplastic binding, which is preclinical.

THE BINDING HYPOTHESIS

How probiotics interact with plastic particles.

This is the preclinical layer of Winnow's formulation. The mechanism is surface adsorption — the bacteria stick to plastic particles. The mechanism operates entirely within the intestinal tract.

1Encounter

Probiotic and particle meet

Microplastic particles drift freely through the gut lumen alongside our probiotic strains.

2Bind

Probiotics bind plastics

Plastic particles cling to the probiotic's sugary outer coating (called EPS) and the textured protein layer on its cell wall (called S-layer).

3Pass

Bound particles pass

Bound plastic particles are escorted through normal intestinal functions.

Exopolysaccharides (EPS)

Thick polysaccharide capsules on the bacterial cell surface adsorb particles via electrostatic and hydrophobic interactions.

S-layer proteins

A crystalline lattice of surface proteins forms an additional structured layer with binding sites for particle adsorption.

Passage

Bound particles cluster on bacterial surfaces and pass through normal intestinal transit, reducing free-floating particles.

TRY THE MECHANISM

See it play out before you dig in to the data.

Drag any probiotic up or down to bind microplastic particles before they reach the gut barrier. Each capture is the same surface adsorption shown above — just played out at hand speed.

Bind & pass

See if you can bind the plastic particles before they reach the gut barrier. Drag any probiotic up or down to intercept.

Bound

0

Escaped

0
Plastic
Probiotic
Bound
Epithelium
Mucus

Illustrative only. Capture rates, particle counts, gut transit, and timing shown here are simplified for visualization and do not represent in vivo performance, binding efficiency, or product efficacy.

OUR DATA

Not all probiotics bind plastic. Strain matters.

Below are highlights from our internal lab assay across 8 strains and 5 common plastic polymers.

Plastic binding of two L. plantarum strains
% of particles captured (in vitro assay)
Winnow strain Poor binder
100%
50%
0%
52%
0%
50%
15%
72%
17%
79%
21%
70%
19%
PVC PS PP PE PET
PVC
52%
0%
PS
50%
15%
PP
72%
17%
PE
79%
21%
PET
70%
19%

Binding values show percent of particles adsorbed. Each strain was tested at 1 × 109 CFU/mL. PVC = polyvinyl chloride, PS = polystyrene, PP = polypropylene, PE = polyethylene, PET = polyethylene terephthalate. Results are not peer reviewed and should be interpreted as preclinical data, not evidence of plastic removal in humans. Winnow Labs, 2025.

Explore the full data

BROAD COVERAGE

5

of the most common polymers tested (PVC, PS, PP, PE, PET).

WIDE VARIABILITY

0-88%

binding observed across strains.

STRONG PERFORMANCE

87%+

What strong binding can look like when you find the right strain.

BLEND SYNERGY

1.8x

more binding by Winnow's full consortium to a blended plastic mix.

Why the bottom row matters.

L. plantarum is one of the best-studied species for plastic-binding in the published literature. But the table shows an important point: species name does not predict performance on its own. One L. plantarum strain showed near-zero binding. Another showed strong binding. Two bacteria can share the same species label and still behave very differently. That is why strain designation matters. Binding is strain-specific, not species-specific.

OUR DATA

Strain-level binding by polymer.

Percent of particles adsorbed in our internal lab assay. PVC = polyvinyl chloride, PS = polystyrene, PP = polypropylene, PE = polyethylene, PET = polyethylene terephthalate.

Winnow strain

Lactobacillus reuteri

Lactobacillus plantarum

Lactobacillus brevis

Lactobacillus acidophilus

Lactobacillus rhamnosus

Lactococcus lactis

Lactobacillus paracasei

Bifidobacterium longum

Lactobacillus plantarum (not in Winnow)

PVC

33%

52%

41%

30%

5%

23%

9%

7%

0%

PS

57%

50%

51%

48%

59%

33%

39%

0%

15%

PP

84%

72%

79%

87%

70%

63%

35%

76%

17%

PE

85%

79%

82%

88%

73%

68%

41%

34%

21%

PET

72%

70%

51%

41%

73%

71%

67%

62%

19%

Binding values show percent of particles adsorbed. Each strain was tested at 1 × 109 CFU/mL. Results are not peer reviewed and should be interpreted as preclinical data, not evidence of plastic removal in humans. Winnow Labs, 2025.

Why the bottom row matters. L. plantarum is one of the best-studied species for plastic-binding in the published literature. But the table shows an important point: species name does not predict performance on its own. One L. plantarum strain showed near-zero binding. Another showed strong binding. Two bacteria can share the same species label and still behave very differently. That is why strain designation matters. Binding is strain-specific, not species-specific.

BETTER TOGETHER

The right multi-strain consortia can exceed individual performance.

When multiple binding strains are combined, synergistic surface interactions can improve binding rates. Winnow's full consortia bound 1.8x more of a blended plastic mix than any single strain alone.

Observed binding

Single strain average Winnow formula
PE
44% 84%
PP
61% 72%
PS
47% 69%
PET
53% 69%
PVC
23% 26%
Mix
50% 68%

YOU'VE READ THE DATA

Now try the formula.

8 strains, 40 billion CFU. Clinically-studied gut health strains backed by human trials. Science-backed microplastic binding you can trust.

Try Winnow — $1.33/day

Want to learn more, check out the following resources:

Exposure Calculator

Estimates personal microplastic exposure based on 34 questions across seven categories.

Research Atlas

A curated and searchable database of over 60,000 microplastic research articles.

Journal

Deep dives and meta-analyses on various topics in microplastics and nanoplastics