Abstract
Nattokinase is a fibrinolytic serine protease produced during fermentation of soybeans by Bacillus subtilis var. natto. Its biology makes one quality point unavoidable: milligrams state how much material is present, while fibrinolytic units (FU) state activity under a defined assay. Neither number is a complete specification by itself. A responsible enzyme record also needs an identity method, purity and contaminant limits, assay conditions, batch linkage and evidence that activity remains through manufacture and shelf life.
Human evidence is narrower than many online summaries suggest. Small studies show transient changes in fibrinolysis or coagulation markers and a modest blood-pressure signal in one population. The largest long-term trial found no effect on carotid atherosclerosis, blood pressure or measured laboratory outcomes, while a newer cognitive trial was null on its primary endpoint. PAI-1 biology is scientifically interesting, but no controlled human study shows that oral nattokinase durably lowers PAI-1 or slows aging. Geroprotect Essentials II lists 120 mg per full serving; its FU per serving is not yet publicly verified against a batch-linked certificate, so this review does not invent a conversion.
01 · Identity and specification
An enzyme needs more than a percentage—and more than one activity number
Sumi and colleagues’ 1987 paper described a fibrinolytic enzyme extracted from natto. It digested fibrin, acted on a plasmin substrate and was inhibited by serine-protease inhibitors.1 Later work identified nattokinase as subtilisin NAT, a protease rather than a kinase. The name is historical; its measured function is proteolysis.
For a small molecule, an assay percentage may closely track the amount of one defined chemical. Enzymes are different. Two powders can contain the same mass yet deliver different catalytic activity because of expression, purification, drying, storage or denaturation. FU therefore adds information that milligrams cannot. The Japan Nattokinase Association, an industry body, requires FU on certified product labels and describes 2,000 FU/day as its association standard.2 Its own technical page also acknowledges that a fibrinolysis assay can respond to other enzymes, which is why activity does not establish identity by itself.3
02 · Human pharmacology
A measurable signal, small samples and normal-range changes
The cleanest acute experiment was a double-blind crossover study in 12 healthy young men. A single 2,000-FU dose increased D-dimer at six and eight hours and fibrin or fibrinogen degradation products at four hours; factor VIII activity fell, antithrombin rose and activated partial thromboplastin time lengthened at selected time points. All changes remained within normal ranges.4 This supports a short-lived pharmacodynamic effect. It does not show prevention of thrombosis, stroke or cardiovascular events.
An open-label, self-controlled study gave 45 participants—including healthy adults, people with cardiovascular risk factors and people receiving dialysis—4,000 FU/day for two months. Fibrinogen and factors VII and VIII declined over time, but without a placebo group the study could not separate treatment from time, behavior, regression or concurrent care.5 The population mixture also prevents a simple inference to healthy consumers.
Questions remain about what reaches the circulation after oral use. One 11-person pilot detected nattokinase-reactive material by ELISA after 2,000 FU, with a late peak, but an antibody signal cannot by itself prove that intact, catalytically active enzyme entered blood.6 Peptide fragments, indirect intestinal signaling and assay cross-reactivity remain plausible. The human marker changes are more secure than a claim about the exact absorbed molecular species.
03 · Controlled outcomes
One positive blood-pressure trial and two important nulls
In an eight-week randomized, double-blind trial, 86 adults with untreated prehypertension or stage-1 hypertension received 2,000 FU/day or placebo; 73 completed the protocol. Net between-group changes favored nattokinase by 5.55 mmHg for systolic and 2.84 mmHg for diastolic pressure, with a reduction in plasma renin activity.7 This is a genuine controlled signal, but it arose in a selected blood-pressure population, with attrition and a short follow-up. It does not establish a cardiovascular-event benefit.
The Nattokinase Atherothrombotic Prevention Study provides the strongest long-term counterweight. It randomized 265 adults without clinical cardiovascular disease to 2,000 FU/day or placebo for a median of three years. Nattokinase did not change the progression of carotid intima-media thickness or carotid stiffness, and it had no significant effect on blood pressure or any reported laboratory measure, including coagulation and fibrinolysis factors.8 A null result at this amount and duration belongs in the center of the evidence record.
The 2026 ICC-PACS trial randomized 120 people with asymptomatic intracranial or carotid stenosis to 8,000 FU/day or placebo for six months; 88 completed per protocol. The primary MoCA cognitive outcome was null (between-group mean difference 0.038; 95% CI −1.109 to 1.163). A visuospatial subscore favored nattokinase in an exploratory analysis, but that cannot convert a null primary trial into a demonstrated cognitive benefit.9
04 · PAI-1 and longevity
Compelling biology with a missing human bridge
In a purified biochemical system, subtilisin NAT cleaved active recombinant plasminogen activator inhibitor-1 (PAI-1) and enhanced tissue-plasminogen-activator-induced clot lysis.10 That experiment establishes enzyme–substrate chemistry under controlled conditions. It does not establish that swallowing nattokinase lowers circulating PAI-1 in a person.
The longevity interest comes from a different line of evidence. In 177 members of the Berne Amish community, including 43 carriers of a rare loss-of-function SERPINE1 variant, carriers had longer leukocyte telomeres, lower fasting insulin and lower diabetes prevalence. In an extended pedigree analysis, mean age at death was 82 years in carriers and 75 years in noncarriers.11 This is important human genetics, but it represents lifelong partial PAI-1 deficiency in a founder population—not a trial of an oral enzyme.
No controlled human trial cited here shows durable PAI-1 lowering after oral nattokinase. NAPS, the best long-duration study, found no overall effect on measured fibrinolysis and coagulation factors. The defensible conclusion is that PAI-1 supplies a research rationale, not a verified mechanism of Geroprotect or a human longevity claim.
05 · Safety in context
Standard haemostasis precautions without manufacturing a routine hazard
Controlled studies at 2,000 FU include a 12-person acute study in which marker changes stayed within normal ranges and a 265-person trial with median three-year exposure that did not report a clinically meaningful coagulation signal in its abstracted outcomes. A product-specific toxicology programme reported no mutagenicity or clastogenicity, no adverse effects at the highest tested 90-day rat dose and tolerability in a small four-week human study; the work evaluated NSK-SD and should not be generalized automatically to every nattokinase material.12
A published case described cerebellar haemorrhage after 400 mg/day in a person taking aspirin for previous stroke who had pre-existing cerebral microbleeds.13 A case report cannot estimate incidence or prove what happens in a healthy population at another potency. It does support the ordinary precaution already used for fibrinolytic supplements: anyone using anticoagulant or antiplatelet medicines, living with a bleeding disorder or preparing for a procedure should have the full formula reviewed by a clinician or pharmacist.
No human trial has established a clinically meaningful or additive bleeding rate for 120 mg of nattokinase with 360 mg of fucoidan, and no evidence justifies calling the formula an unstudied haemostatic “collision.” Conversely, absence of a measured rate is not proof of no interaction. The proportionate position is medicine- and procedure-focused review, not an alarmist prediction. A documented nattokinase allergy has also been reported in people with natto allergy, so relevant allergy history belongs in individual review.14
06 · Formula and COA context
What 120 mg in Essentials II means—and what remains to be matched
A full six-tablet labelled serving of Geroprotect Essentials II contains 120 mg of Nattokinase. That is a valid mass statement. It is not yet a public potency statement because the current FU per serving has not been verified against a certificate demonstrably linked to the raw-material lot used in the finished product and to retained activity after tablet manufacture.
The source dossier refers to a supplier certificate reporting 24,074 FU/g for one powder batch. Multiplying that figure by the labelled mass would be inappropriate until the new certificate set is matched to purchasing records, manufacturing batch records and a finished-product or justified activity-retention specification. The same review should confirm the identity method, FU assay method and acceptance range, vitamin K2 result, allergen statement, contaminants, microbiology, storage condition and shelf-life activity. Until then, no FU figure should appear on the product page or be used to compare the formula with trials.
No component-controlled trial has tested whether nattokinase adds to, duplicates or synergises with the other Essentials II ingredients. The evidence can support a transparent formulation rationale around fibrinolysis-related research. It cannot support prevention of thrombosis or stroke, cognitive enhancement, PAI-1 reduction, cardiovascular-event reduction, slowed aging or extended life.
Evidence verdict
Real enzyme activity; mixed human outcomes; product potency still to be verified
| Statement | Evidence boundary |
|---|---|
| Nattokinase is fibrinolytically active | Supported in biochemical assays; oral human studies also show small, transient marker changes. |
| Milligrams alone establish enzyme potency | No. Mass should be paired with FU, identity, contaminants and stability. |
| Nattokinase lowers blood pressure | One short RCT was positive; the larger, longer NAPS trial was null. A general consumer outcome is not established. |
| Nattokinase lowers PAI-1 and slows aging | Unproven. The cited cleavage experiment is cell-free, while the human longevity evidence is lifelong SERPINE1 genetics. |
| The labelled 120 mg has a known FU value | Not yet publicly verified against a current, batch-linked activity certificate and finished-product retention record. |
| Essentials II is clinically validated | Unproven. Ingredient studies are not trials of the finished formula. |
Questions
Frequently asked questions
Is nattokinase a kinase?
No. Despite its name, nattokinase is a subtilisin-family serine protease. It breaks peptide bonds and is studied for fibrinolytic activity.
Are milligrams meaningless for nattokinase?
No. Milligrams report mass, but they do not report catalytic potency. A useful specification needs both mass and FU, plus an identity method, contaminant limits and stability data.
How many FU are in the Geroprotect serving?
The current label states 120 mg, but a FU figure should not be published until the incoming activity certificate is matched to the supplied lot and retained activity in the finished product. This review therefore makes no conversion.
Does nattokinase cause bleeding?
A routine clinical bleeding rate has not been established at the labelled amount. Because nattokinase affects fibrinolysis-related markers and a high-risk aspirin-associated case has been reported, anticoagulant or antiplatelet medicines, bleeding disorders and procedures warrant clinician or pharmacist review.
Does nattokinase remove plaques or improve cognition?
Not on current controlled evidence. NAPS found no effect on carotid atherosclerosis progression, and ICC-PACS was null on its primary cognitive endpoint.
Does nattokinase work synergistically with fucoidan or ginsenosides?
That has not been tested in a component-controlled human trial. A pathway diagram or overlapping laboratory mechanism is not a synergy experiment.
Primary and authoritative sources
References
- Sumi H, et al. A novel fibrinolytic enzyme (nattokinase) in the vegetable cheese Natto; a typical and popular soybean food in the Japanese diet. Experientia. 1987;43:1110–1111. DOI 10.1007/BF01956052.
- Japan Nattokinase Association. JNKA Mark: association criteria for nattokinase products. Industry standard; accessed 14 July 2026.
- Japan Nattokinase Association. How to identify nattokinase: fibrinolysis method and IBox test. Industry technical statement; accessed 14 July 2026.
- Kurosawa Y, et al. A single-dose of oral nattokinase potentiates thrombolysis and anti-coagulation profiles. Scientific Reports. 2015;5:11601. DOI 10.1038/srep11601.
- Hsia CH, et al. Nattokinase decreases plasma levels of fibrinogen, factor VII, and factor VIII in human subjects. Nutrition Research. 2009;29:190–196. DOI 10.1016/j.nutres.2009.01.009.
- Ero MP, et al. A pilot study on the serum pharmacokinetics of nattokinase in humans following a single, oral, daily dose. Alternative Therapies in Health and Medicine. 2013;19(3):16–19. PMID 23709455.
- Kim JY, et al. Effects of nattokinase on blood pressure: a randomized, controlled trial. Hypertension Research. 2008;31:1583–1588. DOI 10.1291/hypres.31.1583.
- Hodis HN, et al. Nattokinase atherothrombotic prevention study: a randomized controlled trial. Clinical Hemorheology and Microcirculation. 2021;78:339–353. DOI 10.3233/CH-211147.
- Zhang K, et al. Nattokinase supplementation for cognitive enhancement in asymptomatic intracranial/carotid stenosis: a randomized controlled trial. Journal of Stroke and Cerebrovascular Diseases. 2026;35:108511. DOI 10.1016/j.jstrokecerebrovasdis.2025.108511.
- Urano T, et al. The profibrinolytic enzyme subtilisin NAT purified from Bacillus subtilis cleaves and inactivates plasminogen activator inhibitor type 1. Journal of Biological Chemistry. 2001;276:24690–24696. DOI 10.1074/jbc.M101751200.
- Khan SS, et al. A null mutation in SERPINE1 protects against biological aging in humans. Science Advances. 2017;3:eaao1617. DOI 10.1126/sciadv.aao1617.
- Lampe BJ, English JC. Toxicological assessment of nattokinase derived from Bacillus subtilis var. natto. Food and Chemical Toxicology. 2016;88:87–99. DOI 10.1016/j.fct.2015.12.025. Product-specific, industry-associated evidence.
- Chang YY, et al. Cerebellar hemorrhage provoked by combined use of nattokinase and aspirin in a patient with cerebral microbleeds. Internal Medicine. 2008;47:467–469. DOI 10.2169/internalmedicine.47.0620. Single case report.
- Awatani-Yoshidome K, et al. Anaphylaxis from nattokinase in a patient with fermented soybean (natto) allergy. Allergology International. 2022;71:153–154. DOI 10.1016/j.alit.2021.08.004. Single case report.
