Yes, we’ve covered it in discussions here before:
Vaughan DE, Rai R, Khan SS, Eren M, Ghosh AK. “PAI-1 is a Marker and a Mediator of Senescence.” Arteriosclerosis, Thrombosis, and Vascular Biology (ATVB), 2017;37(8):1446–1452.
A blood-clotting brake called PAI-1 is not just a passive flag of aged, “senescent” cells — it appears to actively cause them to age, and blocking it (with drugs or lifestyle) delays multiple features of aging in mice. This is a review arguing PAI-1 is a druggable master-switch for senescence.
For decades, plasminogen activator inhibitor-1 (PAI-1) was filed under “clotting.” It is the body’s main off-switch for the enzymes that dissolve blood clots, and too much of it leaves people prone to dangerous thrombosis. But a team at Northwestern University in Chicago argues that PAI-1’s real story is far bigger: it may be a central driver of biological aging itself.
The big idea is a shift from marker to mediator. Scientists have long noticed that PAI-1 levels climb as cells and tissues age. Old fibroblasts, aged blood-vessel linings, and patients with premature-aging syndromes such as Werner syndrome all show elevated PAI-1. That made it a convenient biomarker of “cellular senescence” — the state in which cells stop dividing but linger, leaking inflammatory and tissue-damaging signals (the so-called senescence-associated secretory phenotype, or SASP). PAI-1 is one of the most prominent molecules in that toxic secretion.
The provocative claim the authors marshal evidence for is that PAI-1 doesn’t just accompany senescence — it helps switch it on. In landmark cell experiments, forcing PAI-1 expression was enough to push cells into senescence even when the usual master regulator, the tumor-suppressor p53, was absent. Conversely, removing PAI-1 let cells keep dividing.
The most striking support comes from “Klotho” mice, a strain engineered to age rapidly and die young. These animals carry PAI-1 levels roughly 45 times higher than normal. When the researchers genetically deleted PAI-1 — or blocked it with an experimental drug called TM5441 — the mice were protected against multi-organ aging, and in the underlying study their median lifespan increased about fourfold.
The authors then make the leap to humans. They note that the very interventions associated with healthier aging — eating more fruit and vegetables, modest caloric restriction, better sleep, and regular exercise — all lower circulating PAI-1. The implication is seductive: PAI-1 might be a shared chokepoint where lifestyle and future drugs converge to slow aging.
It is a compelling, coherent thesis. But it is a thesis. This is a perspective piece assembling correlative and model-organism evidence into a hypothesis — not proof that lowering PAI-1 extends healthy human lifespan. The headline mouse result comes from an extreme premature-aging model, not normal animals. [Confidence: the biology of PAI-1 in senescence is Medium-High; the human anti-aging payoff is Low and unproven.]
The practical message is narrow but real: the lifestyle levers that lower PAI-1 are the same boring, evidence-based ones — and PAI-1 gives a plausible mechanism for why they help vascular aging.
What the review actually supports, and the magnitude where stated:
The one large effect size in the whole piece is from mice, not people: in the cited Klotho model, PAI-1 was ~45-fold elevated, and PAI-1 blockade produced a ~4-fold (≈300% relative) median-lifespan increase, plus a 98% reduction in the aging hormone FGF23. That number is dramatic but comes from a sick, short-lived engineered strain (see Part 2) and should not be read as a human longevity multiplier.
Bottom line for the reader: there is no validated PAI-1-lowering supplement or drug for healthy people today (TM5441 is experimental). The honest take-home is that diet, CR, sleep and exercise plausibly act partly through PAI-1 — a reason to keep doing them, not a new intervention.
More in-depth look at actionable items…
Source data: Vaughan et al., PAI-1 is a Marker and a Mediator of Senescence, ATVB 2017 — a narrative review. The interventions below are derived from the review’s thesis (lower or normalize PAI-1; clear the senescent cells that secrete it) and then independently validated against external primary literature. Because the source is a review, none of these doses come from a controlled longevity trial in normal humans; translational uncertainty is high throughout and is flagged per intervention.
HED methodology (used consistently below): FDA Body Surface Area normalization. HED (mg/kg) = Animal dose (mg/kg) × (Animal Km ÷ Human Km). Standard Km values: mouse = 3, rat = 6, human = 37. For mouse → human the multiplier is 3 ÷ 37 = 0.081. Human absolute dose assumes a 70 kg adult unless noted. These are scaling estimates, not validated clinical doses.
The Core Strategy. This is the most literal translation of the paper’s thesis: pharmacologically inhibit PAI-1 (SERPINE1) itself. PAI-1 is both a SASP component and, per the review, a causal mediator of senescence acting via the p53→PAI-1→PI3K-AKT-GSK3-cyclin D1 axis and via stabilization of IGFBP-3 (blocking t-PA-mediated IGFBP-3 proteolysis). Mechanistic target: direct antagonism of PAI-1, downstream suppression of p16^INK4a/p21/p53. Intended outcome: reduced vascular/pulmonary/renal senescence, improved fibrinolytic balance, delayed multi-organ aging.
Translational Dosing Protocol.
Literature Validation & Source Verification.
Safety, Toxicity & Interaction Profile. Human trials (CML 180 mg/day × 48 weeks; COVID-19 120–180 mg/day) reported no serious drug-attributable adverse events and, notably, no prolongation of PT/aPTT or bleeding time — distinguishing PAI-1 inhibition from conventional anticoagulants. A formal NOAEL in mg/kg is not published in the accessible literature — Safety Data partially Absent for chronic healthy-population use. Long-term oncology context raises an unresolved theoretical concern: senescence is also tumor-suppressive, so chronically disabling a senescence mediator warrants caution. CYP450 interaction profile is not clearly published — Interaction Data Absent.
The Core Strategy. PAI-1 is a SASP factor; senescent cells are the source. Rather than inhibit PAI-1, remove the cells producing it. Fisetin is a flavonoid senolytic that preferentially triggers apoptosis in senescent cells (acting on senescent-cell anti-apoptotic pathways / SCAPs). Mechanistic target: senescent-cell apoptosis → SASP reduction (incl. SERPINE1/PAI-1, IL-6, p16, p21). Intended outcome: lower systemic senescent burden and inflammatory SASP, improved tissue homeostasis.
Translational Dosing Protocol.
Literature Validation & Source Verification.
Safety, Toxicity & Interaction Profile. Fisetin has a strong preclinical safety margin (food-derived flavonoid, found in strawberries). A formal human NOAEL is not established; published human efficacy data remain limited (trials ongoing). No major liver/kidney toxicity signal at studied doses, but **chronic high-dose human safety data are Absent.**Flavonoids including fisetin can inhibit CYP3A4 and CYP2C and modulate P-glycoprotein in vitro — a plausible interaction route at supratherapeutic doses. [Confidence: Medium on senolytic efficacy in mice; Low on human longevity benefit.]
The Core Strategy. The review explicitly cites caloric restriction (CR) and high fruit/vegetable intake as lowering PAI-1 while raising t-PA, shifting fibrinolytic balance favorably. CR is the most reproducible longevity intervention across species. Mechanistic targets: AMPK activation, mTOR/IGF-1 suppression, reduced adipose-derived PAI-1 (adipose tissue is a major PAI-1 source). Intended outcome: lower circulating PAI-1, improved metabolic and vascular aging markers.
Translational Dosing Protocol.
Literature Validation & Source Verification (with effect sizes).
Effect-size note: the PAI-1 reductions above (−26% to −74%) are large and human-verified, but are largely a function of fat-mass loss in overweight subjects; the absolute reduction in lean, normal-weight individuals is likely smaller. One study found subcutaneous adipose PAI-1 mRNA paradoxically rose during VLCD even as plasma PAI-1 fell — a caution against assuming uniform tissue-level effects.
Safety, Toxicity & Interaction Profile. Excellent safety at moderate restriction; risks are lean-mass loss, bone density loss, and disordered-eating potential at aggressive restriction. No CYP issues. NOAEL framing not applicable.
The Core Strategy. The review notes SIRT-1 suppresses PAI-1 and raises eNOS, while SIRT-1 inhibition induces senescence and raises PAI-1. Resveratrol is the prototypical SIRT-1 activator. Mechanistic target: SIRT-1/NAD+ axis → PAI-1 suppression, endothelial protection. Intended outcome: reduced endothelial senescence and PAI-1.
Translational Dosing Protocol.
Literature Validation & Source Verification.
Effect-size / dissent note: human resveratrol data are mixed and frequently null; the high-profile “SIRT1 direct activator” interpretation has been contested (assay-artifact debate). Direct human PAI-1 reduction by resveratrol is not robustly established. This is the weakest-evidence intervention in the report.
Safety, Toxicity & Interaction Profile. Well tolerated to ~1 g/day; **GI distress and diarrhea at higher doses (2.5–5 g).**No major organ-toxicity signal at supplement doses. Resveratrol inhibits CYP3A4, CYP2C9, and CYP1A2 — a meaningful interaction route. NOAEL effectively high but loosely defined.
| Intervention | Primary target-engagement biomarker | Supporting biomarkers |
|---|---|---|
| PAI-1 inhibitor (TM5614) | Plasma PAI-1 antigen + PAI-1 activity (direct target); fibrinolytic shift (↑ t-PA activity, ↓ D-dimer normalization) | p16^INK4a in tissue/T-cells, IGFBP-3 |
| Fisetin | p16^INK4a (CDKN2A) in CD3+ T-cells / adipose; SA-β-Gal in biopsy | Circulating SASP panel: IL-6, PAI-1/SERPINE1, MMP-9 |
| Caloric restriction | Plasma PAI-1 (human-validated to fall 26–74% with fat loss); fasting insulin, IGF-1 | Epigenetic clock pace (DunedinPACE), adiponectin |
| Resveratrol | Plasma PAI-1 + SIRT-1 activity / target-gene readout | eNOS-dependent flow-mediated dilation; free vs conjugated resveratrol PK |
The common, cheap, directly-relevant readout across all five is a plasma PAI-1 antigen + activity assay — the exact marker this paper validates. p16^INK4a in peripheral T-cells is the best general senescence-burden surrogate but is largely research-grade, not routine clinical.
| Intervention | Procurement status | Approx. monthly cost (HED) | Cost-to-benefit assessment |
|---|---|---|---|
| TM5614 / TM5441 | Research chemical / investigational Rx (TM5441 sold for research use only; TM5614 in trials, not approved for longevity) | Effectively N/A for legitimate consumer use; research-grade material highly variable | Poor/unavailable — strongest mechanistic fit to the paper, but not legitimately obtainable or quality-assured for human use. Highest translational uncertainty. |
| Fisetin | Supplement (OTC) | ~$10–25/month (intermittent pulsing uses little material) | Best compound ROI — cheap, oral, intermittent, decent mouse data. Capped by absent human efficacy data and bioavailability. |
| Caloric restriction | Lifestyle (free) | ~$0 (potentially net savings) | Highest practical ROI — free, human-validated PAI-1 reduction, broad benefits. Cost is adherence/discipline, not money. |
| Resveratrol | Supplement (OTC) | ~$15–40/month | Weak ROI — cheap but efficacy/bioavailability poor; the SIRT-1→PAI-1 human link is the least substantiated. |
Ranked by evidence-weighted, risk-adjusted practicality for the stated longevity goal (lowering PAI-1 / senescent burden):
Disclaimer: This is a research analysis, not medical advice. Several interventions involve research chemicals and off-label use with real toxicity and interaction risks. Consult a qualified physician before acting; do not self-administer research chemicals.
For me the key questions are:
What is the median and standard deviation of PAI-1 levels at different ages.
How is it measured and how reliable is the measurement?
Probably the paper on the amish population will be most useful.
Then target 50% of the median PAI-1 levels (heterozygous SERPINE mutation).
PAI-1 levels that are too low are also harmful (as mentioned by the video)
The paper:
An analysis of Khan et al., “A null mutation in SERPINE1 protects against biological aging in humans,” Science Advances, 2017.
A rare, naturally occurring loss-of-function mutation in the SERPINE1 gene, found in a small Amish community, halves blood levels of the protein PAI-1 and is linked to longer telomeres, better metabolic health, and roughly a decade of extra lifespan. This is the first human evidence that a single anti-aging molecular target validated in mice may also operate in people.
For decades, anti-aging research has lived mostly in petri dishes and mouse cages. A 2017 study from Northwestern University finally brought one of its most promising targets into humans — and it did so by studying an Amish community in Berne, Indiana, that carries a genetic quirk found almost nowhere else on Earth.
The protein at the center is PAI-1 (plasminogen activator inhibitor-1), encoded by the gene SERPINE1. PAI-1 is best known for regulating blood clotting, but it has a second life: it is a core component of the “senescence-associated secretory phenotype” — the toxic cocktail of signals that aging, “zombie” cells pump out into surrounding tissue. In mice engineered to age rapidly, switching off PAI-1 protects organs, preserves telomeres, and dramatically extends life. The open question was always whether the same is true for us.
The Berne Amish offered a rare natural experiment. Because of their geographic and genetic isolation, a single frameshift mutation in SERPINE1 — traced back to one married couple roughly six generations ago — has spread through the kindred. Carriers make far less PAI-1 for their entire lives, essentially mimicking a lifelong, gentle drug treatment.
The researchers studied 177 community members, 43 of whom carried one copy of the null mutation. The carriers stood out. Their telomeres — the protective caps on chromosomes that fray with age — were about 10 percent longer on average. They had lower fasting insulin and, strikingly, not a single case of diabetes among the 43 carriers, compared with 7 percent of non-carriers. And when the team reconstructed the family tree using birth and death records stretching back generations, carriers lived a median of about 10 years longer.
The “big idea” is that aging may have manipulable molecular levers, and PAI-1 looks like one of them. Crucially, orally active PAI-1 inhibitor drugs already exist and were in early human trials in Japan. Decades of observing these mutation carriers — who show no abnormal bleeding despite low PAI-1 — hint that partially blocking the protein could be safe.
This is an observational study of a tiny, unusual population, so it is a beginning rather than a verdict. But it is a rare moment when a longevity target jumps convincingly from mouse to human.
The honest take-home is that the headline benefit here is genetic, not a lifestyle hack — you cannot acquire this mutation. But the paper points to actionable directions and quantifies what lowering PAI-1 is worth.
The effect sizes are substantial. Carriers averaged a roughly 10-year longer median lifespan (85 vs 75 years; about 13 percent) and a 7-year longer mean age at death (82 vs 75; Cohen’s d about 0.63, a medium-to-large effect). Their telomeres were about 10 percent longer, which — given the cohort lost roughly 9 percent of telomere length per decade — translates to looking biologically about 11 years younger at the chromosomal level. Fasting insulin was 28 percent lower (model-adjusted), and diabetes prevalence was 0 percent versus 7 percent.
What is actionable for the rest of us: PAI-1 is not fixed by genetics alone. It is raised by insulin, glucose, free fatty acids, obesity, and visceral fat, and it is lowered by interventions that already have evidence behind them — caloric restriction, weight loss, exercise, and metabolic drugs such as metformin. The paper explicitly notes these all converge on reducing PAI-1 expression. So the practical message is that strategies that improve insulin sensitivity and reduce adiposity may be partly working through the same pathway this fortunate Amish family inherited. Pharmacological PAI-1 inhibitors are an emerging “watch this space” intervention, not yet available for longevity use.
Science Advances has a 2024 Journal Impact Factor of approximately 12.5 (CiteScore approximately 19.6).
The impact score of this journal is 12.5, evaluated against a typical high-end range of 0–60+ for top general-science journals, therefore this is a High impact journal. (It sits clearly in Q1 multidisciplinary science but below the very top tier occupied by Nature, Science, and Cell, which carry impact factors in the 40–60+ range — hence “High” rather than “Elite.”)
Lifespan
PAI-1 (the mechanistic readout)
Telomere length (primary endpoint)
Metabolic endpoints
Cardiovascular composite scores (e′ velocity, brachial pulse pressure, carotid IMT): all trended favorably in carriers but the individual cardiovascular composite did not reach significance (P = 0.09); only the comprehensive score including LTL was significant (0.53 units lower, P = 0.005). [Confidence: Medium]
The unifying node is cellular senescence and the SASP. PAI-1 is not just a marker of senescence; per the cited mechanistic work it is “necessary and sufficient” for replicative senescence in vitro and is a key downstream effector of p53. Lowering PAI-1 plausibly reduces senescent-cell burden and the inflammatory secretome that drives tissue aging. [Confidence: Medium]
Mapping to canonical longevity pathways:
This is the first human demonstration that a private loss-of-function mutation lowering PAI-1 is associated with longer telomeres, healthier metabolism, lower diabetes, and longer lifespan. Prior PAI-1/longevity causal data came almost entirely from mouse models of accelerated aging (Klotho, BubR1). This study translates that target into a human population whose effect on circulating PAI-1 is large (a genetic “knockdown” rather than a small common-variant nudge), and links it to a near-term druggable target — oral PAI-1 inhibitors already in early human testing. It also introduces composite “biological aging” scores validated against 5-year cardiovascular outcomes in CARDIA. [Confidence: High]
This paper provides the strongest human signal to date that PAI-1 is a real, possibly causal, and druggable aging lever — but the evidence is preliminary, drawn from a small, genetically peculiar, healthier-than-average population, with borderline statistics and an observational design. It strongly motivates trials of PAI-1 inhibitors and supports the broader idea that targeting senescence and the insulin/IGF-1 axis matters, without proving that lowering PAI-1 will extend life in the general population. [Overall Confidence: Medium]
Sources for journal metrics:
Primary source: Khan SS, Shah SJ, Klyachko E, et al. A null mutation in SERPINE1 protects against biological aging in humans. Sci Adv. 2017;3(11):eaao1617.
From this, and a broader research review, identify
- What is the median and standard deviation of PAI-1 levels at different ages.
- How is it measured and how reliable is the measurement?
- What was the median and standard deviation of PAI-1 levels in the Amish group with the mutation in the SPERPINE1 gene?
- What would be the levels to target if the target is 50% of the median PAI-1 levels (heterozygous SERPINE mutation).
A key caveat before the numbers: PAI-1 is one of the least standardized common plasma analytes. Values depend on whether you measure antigen (ng/mL) or activity (IU/mL or AU/mL), the assay vendor, time of day, fasting state, and the 4G/5G genotype. So cross-study “medians by age” should be treated as ballpark, not canonical.
There is no single clean median±SD-by-decade table in the literature; reported reference ranges overlap heavily. The consensus picture:
| Group | Typical PAI-1 antigen | Notes |
|---|---|---|
| Healthy adults (broad) | ~5–40 ng/mL; mean ~18 ± 10 | Wide fasting reference range cited as 3–72 ng/mL |
| Young adults | lower end | Rises after middle age |
| Middle-aged (esp. males) | 50–60 ng/mL “not uncommon” | Driven largely by insulin resistance/adiposity |
| Elderly | Higher still | Steep rise after midlife, paralleling senescent-cell load |
The honest summary: PAI-1 rises progressively with age, with the steepest climb after middle age. But the dominant driver is not chronological age per se — it is insulin resistance, BMI, triglycerides, and the 4G/5G polymorphism. One large analysis found t-PA/PAI-1 complex rises clearly with age while free PAI-1 age-trends were weaker, underscoring how method-dependent the “age effect” is. I would not quote a precise median±SD for, say, a 40-year-old, because the literature doesn’t support that level of precision. [Confidence: Medium on the trend; Low on exact per-decade values]
Two assay families:
The reliability problem is biological/pre-analytical, not analytical:
Bottom line: the assay is precise, but a single PAI-1 draw is a noisy estimate of a person’s “true” level. Standardize timing (early-morning fasting), and ideally repeat-measure, before acting on a number. [Confidence: High]
One clarification: the paper reported PAI-1 as mean ± SD, not median:
Note the SD (6.6) exceeds half the mean — the distribution is heavily right-skewed, so the median is almost certainly lower than 5.9 (likely ~4–5 ng/mL), though the paper doesn’t publish it. Treat 5.9 ± 6.6 as the best available figure.
Target calculation (50% of the heterozygous level):
Two ways to read your goal, worth separating:
The paper’s own safety argument supports the second framing: heterozygotes (partial reduction, ~6 ng/mL) had 20+ years with no bleeding, whereas homozygotes (near-zero) have a bleeding disorder. Driving toward ~3 ng/mL moves you toward that homozygous deficiency zone. [Confidence: Medium — the “optimal” target for longevity is not established by any trial.]
A practical note: because of the diurnal and biological variability above, hitting “3” vs “6” ng/mL on a single morning draw is within the noise band of the measurement itself, so any targeting should be based on repeated, time-standardized sampling.
Sources:
Identify and provide details on all the orally active PAI-1 inhibitor drugs that already exist and have at least phase 1 clinical trials completed.
The honest bottom line: despite decades of interest, only two orally active PAI-1 inhibitors have ever reached human clinical trials. One is essentially the entire clinical-stage field today; the other failed years ago. Everything else (TM5275, TM5441, TM5484, TM5509, S35225, MDI-2268, etc.) is preclinical or research-tool only.
This is the drug the Amish paper was referring to (“orally active PAI-1 inhibitors… early-phase clinical testing in humans in Japan”).
| Drug | Developer | Oral | Highest stage | Status |
|---|---|---|---|---|
| TM5614 / RS5614 | Tohoku Univ. / Renascience (Japan) | Yes | Phase III (ongoing) | Active; multiple completed Phase I/II trials, POC obtained |
| Tiplaxtinin (PAI-039) | Wyeth/Pfizer (USA) | Yes | Phase I | Discontinued (bleeding risk / risk-benefit) |
Sources:
Ashitaba exudate is an herb that will work. A standardized dose is from a company called Chalcurb. You don’t want the leaves. I actually can’t find this product. Does anybody know somebody in Japan?
https://www.amazon.com/Swanson-Chalcurb-Ashitaba-Milligrams-Capsules/dp/B07C4XWHLL
This is what is needed but it’s out of stock
Maybe…
Short answer: yes, plausibly — but only as an indirect PAI-1 suppressor, and the evidence is entirely preclinical (cell + mouse). No human study has ever directly measured a PAI-1 reduction from Ashitaba. It’s a fundamentally different mechanism from the direct inhibitors (TM5614, tiplaxtinin) discussed earlier.
Ashitaba (Angelica keiskei Koidzumi) is a Japanese plant whose yellow stem exudate (sap) is rich in two prenylated chalcones — xanthoangelol (XA) and 4-hydroxyderricin (4-HD) — which together make up ≥90% of the chalcone content. Critically for PAI-1, the activity is structure-dependent and not shared by both compounds.
In vitro (human endothelial cells):
In mice:
In humans (no direct PAI-1 data):
| Ashitaba (XA) | TM5614 / Tiplaxtinin | |
|---|---|---|
| Action on PAI-1 | Lowers production (↓ mRNA expression, esp. inflammation-driven) | Directly inhibits the PAI-1 protein’s activity |
| Driver targeted | Obesity + chronic inflammation (TNFα), via AMPK activation, anti-adipogenesis | The molecule itself, regardless of source |
| Evidence level | Cell + mouse only | Human Phase I–III |
XA also activates AMPK (the same nutrient-sensing longevity pathway flagged in the original paper) and suppresses adipocyte differentiation — so it plausibly reduces the upstream metabolic-inflammatory drivers that raise PAI-1, rather than neutralizing PAI-1 directly.
So: a reasonable, low-risk dietary candidate for modestly dampening elevated PAI-1 in a metabolic-syndrome context, but it is nowhere near the evidence base — or the direct, potent mechanism — of the clinical PAI-1 inhibitors. It complements rather than replaces the insulin-sensitizing/weight-loss strategies that lower PAI-1.
Sources:
I’m convinced haha. This is going on my list of low risk things to try.
I am taking TM5614 for about 2 years now. Cycle and limit dosing with the aim to limit bleeding risk. Never had any issues but have no objective way to judge if it is making a positive difference. My hair quality is very good for my age (57). Virtually no grey hair or hair loss, but lifespan is more important to me.
Vendors related to this:
One important distinction up front: **most retail “Ashitaba” products are leaf/stem powder, which is not the exudate.**The exudate is the thick yellow sap squeezed from cut stems; when freeze-dried it concentrates to roughly 8% chalcone, versus a fraction of that in leaf powder. Since the PAI-1 research used the exudate/sap, the standardized-chalcone products below are the closest match. I can’t verify live stock, pricing, or that every label’s “chalcone” truly comes from sap vs. leaf — supplement marketing routinely blurs this, so confirm chalcone standardization before buying.
If your goal is the PAI-1/chalcone effect specifically, look for a label that states sap- or exudate-derived and a standardized chalcone percentage (≈8–10%) — Tier 1. Tier 2 leaf powders are fine as a general superfood but deliver far less chalcone per gram and aren’t what the studies tested. Also note the EU classifies **ashitaba sap as a “Novel Food”**with assessed safety limits, which is a useful quality/dose reference point.
A standard caution: dietary supplements aren’t tightly regulated for content accuracy, and ashitaba contains furanocoumarins (potential drug-interaction/photosensitivity concerns), so verify third-party testing and check interactions if you’re on medication. This is information, not a purchase recommendation.
Sources:
Interesting… how did you get it? What dose are you taking?
I source it from a supplier in China without independent testing. I appreciate that some people will find that too risky. We will all have to make our own trade offs in terms of risks and benefits. I take about 120mg daily in my coffee. Happy to share contact details of supplier to anyone interested. It helps to pool demand to get a lower price.
Have you measured PAI-1 and seen it has gone down? If so what assay?
Unfortunately I have no measurements of PAI-1
Where do you buy it?
Xiamen Jenny Chemical Technology Company in Xiamen City. Best to contact Olivia Wang. Number (also on WhatsApp) is +86 166 2100 6118. They also have TM5441, but the scientific evidence in favour of TM5614 is slightly more advanced. I would go for TM5614. Happy to join an order as well to get a better rate. You have to negotiate with them for a competitive price.
Another new study on PIA-1:
Analysis of: Ferraresso et al. “Age-associated increases in PAI-1 silenced with siRNA-lipid nanoparticles reduces thrombosis and prolongs lifespan.” Blood (2026), 147(18):2053–2063. doi:10.1182/blood.2025029834 , Published 2026 Apr 30
A single intravenous dose of a liver-targeted siRNA nanoparticle knocks down the pro-clotting, pro-aging protein PAI-1 by ~90% for about 10 days, shrinking blood clots and improving survival in old mice — and repeated dosing extended lifespan by 20% in a fast-aging mouse model.
As we age, a single protein quietly tips our blood toward clotting. Plasminogen activator inhibitor-1 (PAI-1) blocks the body’s own clot-dissolving machinery, and its levels climb steadily across the lifespan — from roughly 21 ng/mL in healthy young adults toward 50 ng/mL in older people. That climb tracks with the diseases of aging: thrombosis, obesity, diabetes, atherosclerosis, and Alzheimer’s. Tellingly, the rare humans born with partial PAI-1 deficiency don’t bleed dangerously, but they do enjoy less cardiovascular disease and, on average, live about 10% longer. PAI-1 has therefore become one of the more intriguing “druggable” levers in geroscience.
The problem has been hitting it. PAI-1 shape-shifts between active, latent, and cleaved forms, frustrating small-molecule and antibody inhibitors, none of which has yet won FDA approval. A team led by Christian Kastrup at the Versiti Blood Research Institute in Milwaukee took a different route: rather than blocking the protein, switch off its production. Borrowing the lipid-nanoparticle (LNP) technology behind the COVID-19 mRNA vaccines and the approved siRNA drug Onpattro, they packaged a small interfering RNA (siPAI-1) that homes to the liver — a primary source of circulating PAI-1 — and degrades its messenger RNA.
The results are striking in their consistency. One shot cut circulating PAI-1 by about 90% in mice, an effect lasting roughly 10 days before fading. Treated blood dissolved clots faster in the test tube, with no sign of impaired normal clotting. In a brutal surgical model of deep-vein thrombosis, treated young mice formed almost no clot, and — most notably — old mice given siPAI-1 survived at 83% versus just 33% of controls. The drug also normalized the sky-high PAI-1 of obese mice, trimmed cholesterol by 14% in atherosclerosis-prone mice, and, given weekly, extended lifespan by 20% in Klotho-deficient “fast-aging” mice.
The big idea is a shift in strategy: instead of chasing a slippery protein with a drug that lasts hours, you silence the gene that makes it for days to weeks, using a delivery platform already proven safe in humans. Six months of weekly dosing produced no detectable toxicity, organ damage, or excess bleeding. The caveats are equally important — these are mouse studies, the lifespan win came in a progeroid model rather than normally aging animals, and the human payoff remains hypothetical. But as a proof of concept that a clinically translatable RNA therapy can durably lower an aging-associated risk factor, it is a notable step.
This is a preclinical mouse study; there is no consumer intervention to adopt today. The take-home value is directional — it sharpens why PAI-1 is worth watching as a longevity target — and the effect sizes show why interest is warranted.
The headline magnitudes: a single dose produced ~86–90% knockdown of circulating PAI-1 lasting ~10 days. In old mice subjected to total vena cava ligation, survival rose from 33% to 83% — an absolute risk reduction of 50 percentage points, a relative risk of death of 0.25 (a 75% reduction), and a number-needed-to-treat of just 2. Clot burden in young mice fell ~93% (Cohen’s d ≈ 2.4). In atherosclerosis-prone mice, total cholesterol dropped 14% (287 → 246 mg/dL; d ≈ 1.6) — clinically meaningful, since a ~15% cholesterol reduction is associated with materially lower cardiovascular risk. Weekly dosing extended lifespan 20% in fast-aging mice.
The practical message for the longevity-minded reader: PAI-1 is a measurable blood biomarker that rises with age, obesity, and metabolic disease. Lifestyle factors that lower PAI-1 — weight loss, exercise, and improved insulin sensitivity — are already established and freely available. Asking a physician to include PAI-1 (or its driver, metabolic/visceral-fat status) in a risk panel is reasonable. But siPAI-1 itself is not available, not human-tested, and should not be inferred to be safe or effective in people from this work. [Confidence: High that PAI-1 is a valid aging biomarker; Low that the drug benefits translate to humans.]
Blood carries a 2024 Journal Impact Factor of approximately 23.1 (CiteScore ~23, ~96th percentile in hematology). The impact score of this journal is 23.1, evaluated against a typical high-end range of 0–60+ for top general-science and clinical journals; therefore this is a High impact journal