New kejun ying agents paper - AI Agents Discover Aging Interventions from Millions of Molecular Profiles

#1

Autonomous AI Agents Discover Aging Interventions from Millions of Molecular Profiles

https://www.biorxiv.org/content/10.1101/2023.02.28.530532v4

The table in the post shows the top 20 anti-aging interventions identified by AI agents in a recent preprint.

Each row represents a specific intervention ranked by its final score, which is a composite metric reflecting its potential to reverse biological age based on molecular profile analysis.

The columns are:

GSE ID: The identifier for the Gene Expression Omnibus dataset where the intervention’s effects were analyzed.
Intervention: The name or description of the anti-aging treatment or modification.
Category: The type of intervention (Environment, Genetic, or Drug).
Effect Size: A measure of how strongly the intervention reduces biological age (negative values indicate age reversal).
FDR: False discovery rate, indicating the likelihood the result is a false positive (lower is better).
Final Score: Overall ranking score combining effect size, statistical significance, and other factors (higher is better).
P-value: Statistical significance of the age-reversal effect (lower is better).

Here’s an explanation for each row, including what the intervention involves and its potential anti-aging mechanism (drawn from the paper’s context and general biological knowledge).GSE195707 - Mechanical overload followed by senolytics treatment - Environment - -3.87 - 3.3×10⁻⁴ - 138 - 7.7×10⁻⁷
This involves applying physical stress (like resistance training) followed by senolytics, which are compounds that selectively kill senescent (aged, dysfunctional) cells. The mechanism promotes tissue repair and reduces chronic inflammation from senescent cell buildup, potentially reversing age-related decline in muscle and other tissues.

GSE106161 - KMO inhibitor - Drug - -1.56 - 0.025 - 130 - 2.6×10⁻³
Uses a drug to inhibit kynurenine 3-monooxygenase (KMO), an enzyme in the tryptophan metabolism pathway. It reduces production of harmful metabolites linked to neuroinflammation and oxidative stress, which may protect brain health and slow cognitive aging.

GSE159021 - Nf1 knockout - Genetic - -16.48 - 0.032 - 128 - 3.7×10⁻³
Genetically disables the NF1 gene, which regulates cell growth via the Ras pathway. This could enhance cellular resilience to stress or reduce uncontrolled proliferation, helping mitigate age-related diseases like cancer or neurodegeneration.

GSE184479 - AIM2 knockout - Genetic - -5.48 - 0.011 - 128 - 7.9×10⁻⁴
Knocks out the AIM2 gene, part of the inflammasome that triggers immune responses to DNA damage. Reducing this activity curbs excessive inflammation, a major contributor to aging and related conditions like arthritis or cardiovascular disease.

GSE158817 - Cth knockout with H. pylori infection - Genetic - -2.40 - 0.042 - 128 - 5.4×10⁻³
Disables the Cth gene (cystathionine gamma-lyase) in the presence of H. pylori bacteria. It alters hydrogen sulfide signaling, which has antioxidant effects, potentially balancing gut microbiome-immune interactions to lower inflammation and oxidative damage.

GSE171482 - Suv420h knockout - Genetic - -2.39 - 0.042 - 125 - 5.5×10⁻³
Eliminates the SUV420H histone methyltransferase gene, affecting chromatin structure and gene silencing. This epigenetic change may improve DNA repair or stress resistance, countering age-related genomic instability.

GSE81424 - Nkx3.1 knockout - Genetic - -6.58 - 1.3×10⁻⁴ - 123 - 2.6×10⁻³
Knocks out the NKX3.1 transcription factor gene, involved in tissue differentiation. It may regulate hormone signaling or cellular homeostasis, reducing degeneration in organs like the prostate and slowing systemic aging.

GSE122080 - Ouabain - Drug - -5.74 - 3.6×10⁻⁴ - 123 - 8.6×10⁻⁶
Administers ouabain, a cardiac glycoside with senolytic properties. It targets ion pumps to eliminate senescent cells, improving heart function, reducing frailty, and lowering brain inflammation for overall rejuvenation.

GSE150609 - Roxadustat - Drug - -4.27 - 0.028 - 123 - 3.1×10⁻³
Uses roxadustat to inhibit prolyl hydroxylase, stabilizing hypoxia-inducible factors. This boosts red blood cell production and metabolic adaptation, mimicking low-oxygen benefits like those from exercise or calorie restriction to enhance resilience.

GSE163972 - HG-9-91-01 - Drug - -0.73 - 2.3×10⁻⁵ - 123 - 3.6×10⁻⁷
Involves treatment with HG-9-91-01, a kinase inhibitor. It likely modulates signaling pathways for cell growth or stress, though specifics are limited; it may reduce inflammation or improve cellular repair.

GSE137096 - VEC-Cre endothelial mutagenesis - Genetic - -4.50 - 5.6×10⁻⁴ - 122 - 1.5×10⁻³
Induces mutations in endothelial (blood vessel lining) cells using Cre recombinase. This improves vascular health, enhancing blood flow and nutrient delivery while reducing age-related stiffness and organ damage.

GSE102227 - Mcl1 knockout - Genetic - -9.62 - 1.2×10⁻⁴ - 120 - 1.0×10⁻¹⁰
Disables the MCL1 anti-apoptotic gene. It encourages programmed death of damaged cells, preventing senescence accumulation and promoting healthier tissue turnover.

GSE159795 - B6 vs NZO strain comparison - Genetic - -7.24 - 4.2×10⁻³ - 120 - 3.9×10⁻¹⁰
Compares genetic strains (C57BL/6 vs. New Zealand Obese mice). Differences highlight protective alleles against metabolic disorders, reducing obesity-related inflammation and extending healthspan.

GSE165055 - IP subunit expression - Genetic - -1.72 - 3.1×10⁻³ - 120 - 1.4×10⁻⁴
Increases expression of an immunoproteasome subunit. This enhances protein degradation and immune function, clearing damaged proteins to maintain cellular health against aging.

GSE167301 - SETBP1 dominant-negative - Genetic - -7.76 - 0.012 - 120 - 9.3×10⁻⁴
Uses a dysfunctional form of SETBP1 to inhibit its normal activity in gene regulation. It suppresses harmful signaling, reducing stress and potential oncogenic effects associated with aging.

GSE147830 - Clever-1 deficiency - Environment - -5.69 - 5.1×10⁻³ - 120 - 2.7×10⁻⁴
Creates a deficiency in Clever-1 (a macrophage receptor). This modulates immune cell movement and reduces chronic inflammation, improving tissue environment for longevity.

GSE86500 - Cold exposure in Ybx2 KO - Environment - -6.82 - 0.014 - 120 - 1.1×10⁻³
Applies cold stress to Ybx2 gene knockout models. Cold activates metabolic adaptations like fat burning; the knockout amplifies resilience, similar to hormesis effects that extend lifespan.

GSE128145 - CtBP2 mutant - Genetic - -3.90 - 4.4×10⁻⁴ - 120 - 1.1×10⁻⁵
Mutates CtBP2, a gene repressor. This alters expression of metabolism-related genes, potentially boosting energy efficiency and stress resistance.

GSE148794 - Serpina3n - Drug - -3.12 - 1.0×10⁻⁴ - 120 - 1.9×10⁻⁶
Treats with Serpina3n, a protease inhibitor. It prevents excessive protein breakdown, reducing inflammation and fibrosis in tissues affected by aging.

GSE176174 - BHLHE40 knockout - Genetic - -7.43 - 0.034 - 120 - 4.1×10⁻³ (partial in image, but inferred as last row)
Knocks out BHLHE40, a circadian and stress regulator. This improves adaptation to daily rhythms and stressors, reducing oxidative damage and supporting longevity.

3 Likes
#2

https://www.biorxiv.org/content/10.1101/2023.02.28.530532v4

Even jure is in the paper

#3

Gemini AI Review of Paper

Review: Autonomous AI Agents Discover Aging Interventions from Millions of Molecular Profiles

Authors: Ying, K., et al. (Gladyshev Lab, Harvard/Brigham and Women’s)

Source: bioRxiv 2023.02.28.530532 1

Executive Summary

This paper represents a “brute force” computational biology breakthrough. Instead of testing one drug at a time, the authors deployed AI agents (“ClockBase Agent”) to systematically re-analyze 2 million publicly available RNA-seq and DNA methylation samples from the Gene Expression Omnibus (GEO)2222. By applying over 40 distinct aging clocks to data originally collected for other purposes (cancer, immune studies, etc.), they uncovered hidden age-modifying effects in 43,602 intervention comparisons 3.

The result is a massive, hypothesis-free catalog of potential geroprotectors. The study validates its computational findings by experimentally testing Ouabain (a cardiac glycoside), demonstrating that intermittent treatment reduces frailty and neuroinflammation in aged mice 4.

Key “Tell It Like It Is” Insights:

  • Destruction > Creation: Loss-of-function interventions (Knockout/Knockdown) are significantly more likely to decelerate aging than Gain-of-function (Overexpression). It is easier to break a pro-aging pathway than to engineer a pro-youth one.
  • Most Drugs Accelerate Aging: A sobering finding—systematic analysis shows that significantly more FDA-approved drugs accelerate biological aging than decelerate it.
  • Context Matters: Many “anti-aging” hits are likely disease-specific rescues (e.g., curing an infection lowers “biological age” by reducing inflammation) rather than true slowing of the baseline rate of aging.

Identified Anti-Aging Interventions

The AI identified >500 significant anti-aging interventions. Below are the high-confidence hits categorized for translation.

1. Top Pharmacological Candidates (Small Molecules)

  • Ouabain: (Cardiac glycoside) Validated in this paper. Reduces frailty and neuroinflammation8.
  • Fenofibrate: (PPAR$\alpha$ agonist) Lipid-lowering agent, showed strong age-reduction scores9.
  • Rapamycin (Sirolimus): Confirmed “gold standard” positive control1010.
  • Tretinoin: (Retinoid) Top-ranked FDA-approved anti-aging drug in their analysis11.
  • Roxadustat: (HIF-PH inhibitor) Mimics hypoxia, stimulates erythropoiesis1212.
  • Serpina3n: (Serine protease inhibitor) Immune modulator13.
  • Metformin: Confirmed concordance with DrugAge database1414.
  • Ascorbic Acid (Vitamin C): Concordant with longevity databases15.
  • Naktide: (CD36 inhibitor)16.

2. Genetic & Biological Targets (Theoretical)

  • IRF4 Knockout: Immune cell differentiation factor; top-ranked genetic anti-aging effect17.
  • Mettl3 Knockout: Disrupts m6A RNA methylation (epigenetic modifier)18.
  • Nf1 Knockout: Neurofibromin 119.
  • Bach2 Expression: Paradoxical finding; transcriptional repressor in T-cells20.
  • miR-155 Overexpression: Pro-inflammatory miRNA, yet reduced biological age in specific contexts21.
  • AIM2 Knockout: Inflammasome component; reducing inflammation lowers aging signal2222.

3. Environmental & Lifestyle

  • Mechanical Overload + Senolytics: Physical stress followed by clearing senescent cells23.
  • Cold Exposure: Specifically effective in Ybx2 KO mice contexts2424.
  • Hypoxia (Context dependent): While general hypoxia accelerated aging, Roxadustat (hypoxia mimetic) reduced it, suggesting intermittent signaling (hormesis) vs. chronic stress25252525.

Part 3: Actionable Intelligence

The Protocol: Theoretical Translation for a Human Biohacker

Warning: The top hit, Ouabain, has a lethal therapeutic window. Proceed with extreme caution.

  • The “ClockBase” Stack Hypothesis:
    • Core Driver (The “Dirty” Senolytic): Ouabain.
      • Mouse Protocol: 0.5 mg/kg, twice/week, intermittent (3 months)26.
      • Human Extrapolation: The mouse dose is massive. 0.5 mg/kg mouse approx 0.04 mg/kg human approx 2.8 mg for a 70kg human.
      • Safety Caveat: STOP. Clinical Ouabain/Digoxin doses for heart failure are ~0.125–0.25 mg. A 2.8 mg dose is likely cardiotoxic or fatal.
      • Biohacker Translation: Micro-dosing or “Hormetic Dosing.” Using standardized Strophanthusextracts (source of Ouabain) at clinical heart-support doses (e.g., 3–6 mg of extract, not pure alkaloid, or strictly adhering to <0.25 mg pure equivalent) in a “Pulse” protocol (e.g., 2 weeks on, 2 weeks off) to mimic the intermittent senolytic effect without inducing arrhythmia.
    • Metabolic Support: Fenofibrate (145 mg/day).
      • Rationale: Identified as a top FDA-approved anti-aging agent27. Improves lipid profile and mitochondrial oxidation.
    • Regenerative Signal: Tretinoin (Topical).
      • Rationale: Systemic toxicity is high, but topical application reduces skin age (a major organ) and lowers systemic inflammatory signaling via skin barrier improvement.
    • Background: Rapamycin (Weekly pulse).
      • Rationale: Validated positive control in the study28.
  • Timing:
    • AM: Fenofibrate (requires food).
    • Pulsed (Intermittent): Ouabain/Strophanthus (Mid-day, monitor heart rate). Do not take daily forever; the study used specific treatment windows29.
    • Evening: Tretinoin (Topical).

Biomarkers to Track (N=1 Experiment)

  • Safety (Critical for Ouabain/Digoxin analogs):
    • ECG/EKG: Mandatory. Watch for ST-depression or arrhythmia (pro-arrhythmic risk).
    • Serum Electrolytes: Potassium (K+) and Magnesium. Low K+ potentiates Ouabain toxicity.
    • Creatinine/eGFR: Renal clearance is critical for cardiac glycosides.
  • Efficacy (Verification):
    • DunedinPACE (Blood): The most sensitive clock for short-term intervention changes30.
    • hs-CRP: The paper notes Ouabain reduced neuroinflammation31. Systemic CRP is a rough proxy.
    • HRV (Heart Rate Variability): Ouabain improves cardiac output32. improved HRV would be a functional surrogate.
    • Frailty Proxy: Grip Strength (dynamometer) – The mice showed preserved strength and coat condition33.

Feasibility & ROI

  • Cost: Low. Ouabain (as Strophanthus) and Fenofibrate are generic and cheap.
  • Metabolic Effort: High. The safety monitoring required for cardiac glycosides (Ouabain) makes this a high-effort protocol. You are playing with sodium-potassium pumps in the heart.
  • ROI Analysis:
    • Ouabain: Speculative/Risky. High risk of toxicity for a theoretical senolytic benefit that might be achieved more safely with Dasatinib/Quercetin or Fisetin.
    • Fenofibrate: High. Low risk, established lipid benefits, and now potential aging-clock benefits.

Population Applicability

  • Scalability: Niche.
  • Sex Differences: The mouse validation for tAge (transcriptomic age) was in females 34, while the frailty/functional validation was in males35. This suggests broad applicability, but the specific molecular pathways might differ by sex.
  • Target: People with early signs of frailty or cardiac output decline. Healthy, young biohackers should likely avoid Ouabain due to the “if it ain’t broke, don’t fix it” rule of cardiac pharmacology.

Part 4: The Strategic FAQ

  1. Safety of Translation: “You used 0.5 mg/kg of Ouabain in mice. That converts to ~2.8 mg in humans, which is roughly 10x the therapeutic dose for heart failure patients. Did you observe cardiotoxicity in the mice, and do you genuinely believe a non-toxic dose in humans will retain the senolytic efficacy?”
  2. Pro-Aging Drugs: “Your data shows Morphine and Fluorouracil (chemotherapy) as ‘anti-aging’ in Figure 4f36. Is this a genuine rejuvenation signal, or is the clock misinterpreting cellular suspended animation/toxicity as ‘youth’? This casts doubt on the ‘anti-aging’ classification of toxic stressors.”
  3. Destruction vs. Creation: “You found that loss-of-function (Knockout) interventions vastly outperform overexpression for anti-aging37. Does this confirm the theory that aging is driven by ‘hyper-function’ (Blagosklonny’s hypothesis), and we should focus on breaking pathways rather than boosting them?”
  4. Transcriptomic Noise: “You analyzed RNA-seq data from millions of samples with vastly different batch effects and sequencing depths. Even with your standardization, how much of the ‘anti-aging’ signal in the Other category is simply technical noise or batch artifacts masquerading as biological signal?”
  5. Disease Masquerading: “Many of your top anti-aging hits are in disease models (e.g., Cth KO with H. pylori). Are these interventions actually ‘anti-aging’, or are they simply ‘anti-disease’—restoring a sick animal to baseline, which the clock interprets as ‘younger’ compared to the dying control?”
  6. Senolytic Specificity: “Ouabain is identified as a senolytic. Did you verify if it clears senescent cells in healthy aged tissue, or only in the specific context of the pathologies present in your 26-month-old mice?”
  7. Clock Discordance: “In Figure 4a, you show a massive scatter of P-values. Did you find interventions that accelerated GrimAge but decelerated Transcriptomic Age? How do you reconcile conflicting outputs from different clocks on the same sample?”
  8. Tretinoin Systemic Use: “Tretinoin is listed as a top FDA anti-aging drug38. The samples were likely skin/epithelial. Do you have any evidence from your multi-tissue analysis that systemic retinoids benefit internal organs (liver/heart), or is this strictly a skin-deep finding?”
  9. Fenofibrate Mechanism: “Fenofibrate is a PPAR$\alpha$ agonist. Do your transcriptomic maps show that its anti-aging effect is purely metabolic (lipid oxidation), or did it activate distinct longevity/autophagy pathways independent of lipid handling?”
  10. The “Agent” Black Box: “Your AI agent achieved 99% accuracy on technical tasks39, but did it hallucinate biological context? For example, did it correctly distinguish between a ‘control’ that was a sham surgery vs. a ‘control’ that was a wild-type young mouse in all 43,000 comparisons?”
#4

KMO inhibitor is interesting.

When researching IPAM I learned a lot about the tryptophan/kynurenine pathway and I had hypothesized that inhibiting the kynurenine pathway to increase tryptophan metabolites and limit kynurenine toxic metabolites and then supplementing what kynurenine metabolites were actually necessary could be a viable longevity intervention.

1 Like
#5

I was looking at this pathway and it appears that one of the enzymes in it becomes less available over time. Its probably a splicing thing, but I have not put the effort into finding out what it might be.

#6

If you find out which one I would be interested to hear.

#7

Can you post more about this?
Could Linus Pauling (my gateway to biohacking in the early 1970s) have had it right?

#8

The average person gets plenty of vitamin C.

1 Like
#9

That’s not what Linus Pauling believed. Without supplements, we get enough vitamin C to avoid scurvy.
So… what’s an optimum amount. One way to look at that is relative to the vitamin C levels in animals that make their own vitamin C. Humans of course don’t. I remember reading somewhere that to get a comparable level of vitamin C to animals that make their own we would need to take approx 10g of vitamin C per day. Not many people do that. I take 6 grams daily - from your point of view, a waste, but from mine, maybe not.

#10

But animals that do make their own vitamin C don’t have a particulary high lifespan either.
I do understand the idea of keeping your extracellular matrix intact in order to suppress cancer, but collagen formation is not the limiting factor here as we have multiple interventions that can increase collagen synthesis (isotretinoin, collagen peptides, …).

#11

What a fantastic thread! Thank you!

#12

My Prompt:

What is the scientific and clinical research that supports the idea that Ascorbic Acid (Vitamin C) is a longevity supplement (increases healthspan or lifespan in humans)?

From Gemini:

The scientific consensus on Ascorbic Acid (Vitamin C) as a “longevity” agent is nuanced: it is not a confirmed lifespan extender in healthy humans in the same vein as caloric restriction or rapamycin (in animal models), but it is a critical healthspan preserver.

The strongest argument for Vitamin C in longevity is not that taking more makes you live forever, but that modern humans often operate at “sub-clinical” deficiency levels that accelerate epigenetic aging and structural decay.

1. The Core Scientific Verdict

Does it increase maximum Lifespan? No human data supports this. Does it reduce All-Cause Mortality? Yes, but with a caveat. Epidemiological studies consistently show that higher plasma vitamin C levels are associated with lower risks of mortality from cardiovascular disease and cancer. However, this often correlates with fruit/vegetable intake rather than supplementation alone.

  • The Curve: The relationship is likely asymptotic or U-shaped. Going from “deficient” to “optimal” yields massive benefits. Going from “optimal” to “mega-dose” (via oral supplements) yields diminishing returns due to absorption saturation.

Full Gemini Response: https://gemini.google.com/share/4be467630f47

#13

1. The Pharmacokinetic Evidence (The “Levine” Limit)

(Gemini)
The definitive evidence for the saturation curve comes from a series of NIH studies led by Dr. Mark Levine in the 1990s and 2000s. These studies changed the RDA and proved that oral Vitamin C has a strict biological ceiling.

  • The Study: Levine et al. (Proc Natl Acad Sci USA, 1996 & 2001) conducted depletion-repletion studies in healthy volunteers.
  • The Curve: They found a steep sigmoidal relationship between dose and plasma concentration.
    • Steep Rise: Going from 0mg to 100mg/day causes a massive spike in plasma levels (moving from scurvy/deficiency to ~60 µM).
    • The Plateau: At 200mg/day, the curve flattens. Plasma levels reach ~70–80 µmol/L.
    • The Ceiling: At doses above 400mg/day (even up to 2500mg), plasma levels barely budge, capping out at roughly 80 µmol/L.
  • Why this happens (The Mechanism): This is due to the SVCT1 transporter (Sodium-Dependent Vitamin C Transporter 1) in the gut and kidneys.
    • Absorption Saturation: At a dose of 200mg, bioavailability is near 100%. At a dose of 1250mg, bioavailability drops to <50%. The transporter physically cannot pump faster.
    • Renal Threshold: Once plasma levels hit ~70 µM, the kidneys stop reabsorbing Vitamin C and dump the rest into the urine.

Confirming once again that Americans have the most expensive urine in the world. :sweat_smile:

3 Likes
#14

Google Gemini Deep Search analysis of these compounds (excluding Rap, Metformin and Vitamin C):

Pharmacological Modulation of the Aging Continuum: A Comprehensive Analysis of Emerging Small Molecule Candidates for Healthspan Extension

Executive Summary

The transition of geroscience from theoretical exploration to translational reality hinges on the identification of pharmacological agents capable of modulating the fundamental hallmarks of aging. This report provides an exhaustive, expert-level analysis of five high-priority small molecule candidates—Ouabain, Fenofibrate, Roxadustat, Serpina3n, and Naktide—identified through a convergence of deep transcriptomic profiling, autonomous AI-driven discovery, and rigorous preclinical validation in mammalian models.

Collectively, these agents challenge the traditional “one drug, one disease” paradigm. Instead, they exhibit pleiotropic effects that intervene in upstream drivers of organismal decline: cellular senescence, metabolic rigidity, hypoxic compromise, proteostatic collapse, and oxidative amplification.

  • Ouabain, a cardiac glycoside, is repurposed here as a potent senolytic, leveraging the electrochemical vulnerability of senescent cells to reduce frailty and neuroinflammation.
  • Fenofibrate, a PPAR$\alpha$ agonist, is revealed as a “metabolic rejuvenator” with unexpectedly high age-reduction scores in transcriptomic clocks, driving autophagy and reversing tissue fibrosis.
  • Roxadustat, a hypoxia-inducible factor prolyl hydroxylase (HIF-PH) inhibitor, mimics the adaptive benefits of hypoxia to rejuvenate erythropoiesis and intestinal stem cell function.
  • Serpina3n, a serine protease inhibitor, emerges from AI screens as a top-tier immune modulator, paradoxically acting as both a biomarker of stress and a powerful therapeutic for neuroprotection.
  • Naktide, a peptide inhibitor of the Na+/K±ATPase/Src complex, offers a novel mechanism to uncouple cellular work from oxidative stress, silencing the “oxidant amplification loop” that drives metabolic syndrome and uremic cardiomyopathy.

This document synthesizes mechanism of action, preclinical efficacy data, transcriptomic validation, and clinical translational potential into a cohesive narrative, providing a definitive reference on these emerging geroprotectors.

The pharmacological candidates analyzed in this report were selected not merely for their ability to treat a specific pathology, but for their capacity to reset systemic biological clocks. This selection process increasingly relies on “Autonomous AI Agents”—sophisticated computational systems capable of scouring millions of molecular profiles to match drug-induced gene expression signatures with “youthful” transcriptomic profiles. By evaluating compounds against multi-tissue transcriptomic clocks (tAge), researchers can identify non-obvious geroprotectors that reverse the molecular signature of aging across diverse organ systems.

Conclusion

The pharmacological extension of healthspan is no longer a futuristic concept but a tangible scientific pursuit. Ouabain, Fenofibrate, Roxadustat, Serpina3n, and Naktide represent a new class of geroprotectors that do not merely suppress symptoms but actively intervene in the aging process. Whether by eliminating senescent cells, rewiring metabolism, mimicking regenerative hypoxia, shielding tissues from proteolysis, or breaking the cycle of oxidative stress, these small molecules offer a diverse toolkit for delaying the onset of age-related disease and extending the period of healthy life. Future translational efforts must focus on optimizing dosing regimens to balance efficacy with safety, paving the way for clinical trials targeting aging itself as a treatable condition.

Full Gemini Analysis and report: https://gemini.google.com/share/6c418485d7d5

1 Like
#15

The following table and detailed breakdown identify the generic names, commercial/brand names, and sources for the requested compounds.

Compound Identification Summary

Compound Generic / Scientific Name Commercial / Brand Names Source / Class
Ouabain Ouabain (g-Strophanthin) Strodival (historical), Purostrophan Natural: Derived from the seeds of Strophanthus gratus and the bark of Acokanthera plants.
Fenofibrate Fenofibrate Tricor, Lipofen, Antara, Fenoglide, Triglide, Lofibra Synthetic: A fibric acid derivative (fibrate).
Roxadustat Roxadustat (FG-4592) Evrenzo Synthetic: An oral hypoxia-inducible factor prolyl hydroxylase (HIF-PH) inhibitor.
Serpina3n Serine protease inhibitor A3N None (Research Reagent only) Endogenous: A protein encoded by the Serpina3n gene (primarily murine ortholog of human SERPINA3).
Naktide Naktide (Na/K-ATPase-derived peptide) None (Research Reagent only) Synthetic Peptide: Derived from the N-domain of the Na/K-ATPase -1 subunit.

Detailed Compound Profiles

1. Ouabain

  • Classification: Cardiac glycoside (cardenolide).
  • Status: Historically used for heart failure (IV administration) due to its rapid onset compared to Digoxin. It is largely obsolete in modern Western clinical practice but remains a potent tool in research for Na+/K±ATPase inhibition.
  • Source: Extracted from Strophanthus gratus (climbing oleander) seeds.

2. Fenofibrate

  • Classification: Lipid-regulating agent (Fibrate).
  • Status: FDA-approved prescription medication.
  • Mechanism: It activates peroxisome proliferator-activated receptor alpha (PPAR alpha) to lower triglycerides and increase HDL cholesterol.
  • Commercial Note: Formulations vary significantly in bioavailability (micronized vs. non-micronized), which affects whether they must be taken with food.

3. Roxadustat

  • Classification: HIF-PH Inhibitor.
  • Status: Approved in the EU, Japan, and China for anemia of chronic kidney disease (CKD). It received a Complete Response Letter (rejection) from the US FDA in 2021 due to safety concerns regarding thrombotic events.
  • Mechanism: It mimics the body’s response to hypoxia, stimulating erythropoietin production.

4. Serpina3n

  • Classification: Serine Protease Inhibitor (Serpin).
  • Status: Not a drug. This is a specific gene/protein often studied in mouse models. It is the murine ortholog to the human protein Alpha-1-antichymotrypsin (SERPINA3).
  • Research Context: It is involved in inflammation and neurological repair. There are no commercial pharmaceutical formulations for human use; it is available only as a laboratory reagent (e.g., antibodies, recombinant proteins) for experimental purposes.

5. Naktide

  • Classification: Synthetic Peptide Inhibitor.
  • Status: Experimental Compound. It is not an approved drug for human use.
  • Mechanism: Designed to specifically decouple the interaction between the Na+/K±ATPase pump and Src kinase. This inhibition blocks the signal transduction function of the pump without necessarily stopping its ion-pumping function. It is utilized in longevity and metabolic research to study oxidative stress and adipogenesis.
#16

Before anyone gets too excited about fenofibrate:

I asked ChatGPT 5 about fenofibrate

"There is real preclinical work showing fenofibrate (a PPAR-α agonist) can push cells/tissues in directions that look “rejuvenating” in certain models—more autophagy, less fibrotic signaling, fewer senescent cells—and there are papers that explicitly frame fibrates (including fenofibrate) as having senolytic/autophagic activity in disease models. PMC+2PubMed+2
 But translating that into “unexpectedly high age-reduction scores on transcriptomic clocks” and “metabolic rejuvenator” is mostly omics-clock rhetoric + animal/cell data, not proven human geroprotection.

What’s solid vs what’s hype

1) Autophagy and anti-fibrosis signals: supported (mostly preclinical)

  • Cardiac autophagy / anti-fibrosis in mice: fenofibrate increased autophagy (via FGF21/SIRT1 signaling) and reduced fibrosis/inflammation in diabetic mouse hearts. PubMed
  • Fibrosis pathway suppression in retina models: fenofibrate inhibited subretinal fibrosis pathways (TGF-β/Smad and Wnt signaling) in an experimental setting. Frontiers
  • Senescence/autophagy framing: a widely cited paper argues “fibrates” show senolytic and autophagic activity in a specific disease context (osteoarthritis model), including reduction of senescent cells and increased autophagic flux. PMC

:white_check_mark: Bottom line: “fenofibrate can promote autophagy and reduce fibrotic signaling in some models” is fair.

2) “High age-reduction scores” on transcriptomic clocks: possible, but not equivalent to slowed aging

Transcriptomic clocks can move a lot when you activate a master metabolic regulator like PPAR-α, because you’re rewriting large swaths of gene expression (fat oxidation, peroxisomal genes, lipid handling, inflammation-adjacent pathways). We do have data showing PPAR-α activation produces broad transcriptome shifts in liver models closer to human biology. PMC

:warning: The key caveat: a drug can make an omics clock “younger” without reliably improving lifespan/healthspan across tissues in humans. Clock shifts are signals, not outcomes.

3) “Metabolic rejuvenator” claim: marketing language

It’s an interpretive label, not a clinical conclusion. Fenofibrate’s established role remains triglyceride lowering/remnant risk management, with mixed/neutral CV outcome results overall except certain subgroups (high TG/low HDL, often in diabetes trials). PubMed+1

1 Like
#17

Fenofibrate can increase homocysteine…