SW033291

SW033291 is the first-published small-molecule inhibitor of 15-hydroxyprostaglandin dehydrogenase (15-PGDH / HPGD), identified via high-throughput screen by the Jain lab at UT Southwestern (Zhang et al., Science 2015). It is the canonical tool compound used to validate 15-PGDH as a gerozyme target across skeletal muscle, bone marrow, neuromuscular junction, cartilage, heart, liver, and brain. SW033291 itself is not in human trials — the Epirium Bio clinical candidate is a separate, structurally distinct derivative. The compound’s value is as a mechanistic probe: it has established, across a substantial body of preclinical work, that lowering 15-PGDH activity restores regenerative tissue PGE2 signalling that declines with age.

TL;DR: Potent (~low-nM Ki), selective, reversible 15-PGDH inhibitor. Doubles tissue PGE2 back to physiologic levels without supraphysiologic overshoot. Preclinical efficacy across six tissue/disease contexts (bone marrow recovery, aged muscle, NMJ, cartilage, heart failure, liver, brain). No human trials. Clinical translation is pursued by Epirium Bio using a separate compound.

Identity and physicochemical properties

• PubChem CID: 3337839 (confirmed via PubChem REST API 2026-05-23)
• InChIKey: LCYAYKSMOVLVRL-UHFFFAOYSA-N
• IUPAC name: 2-butylsulfinyl-4-phenyl-6-thiophen-2-ylthieno[2,3-b]pyridin-3-amine
• Molecular formula: C₂₁H₂₀N₂OS₃
• Molecular weight: 412.6 Da
• ChEMBL ID: not assigned (research tool compound; no ChEMBL record as of 2026-05-23) needs-canonical-id
• DrugBank ID: not applicable (not a pharmaceutical drug)
• CAS number: not confirmed via PubChem; needs-canonical-id
• Scaffold: thienopyridine core with butylsulfinyl, phenyl, and thiophen-2-yl substituents
• Canonical SMILES: CCCCS(=O)C1=C(C2=C(S1)N=C(C=C2C3=CC=CC=C3)C4=CC=CS4)N

Key structural features: The thienopyridine–amine scaffold places SW033291 in the class of heterocyclic SDR-enzyme inhibitors. The butylsulfinyl group is the primary recognition element for 15-PGDH’s NAD⁺-binding pocket. The compound is a research reagent — not yet GMP-manufactured.

Mechanism of action

15-PGDH (gene: HPGD) is an NAD⁺-dependent short-chain dehydrogenase/reductase (SDR family) that catalyzes the first step in prostaglandin catabolism: oxidation of the C-15 hydroxyl group of prostaglandin E₂ (PGE₂) to yield biologically inert 15-keto-PGE₂. SW033291 competitively and reversibly inhibits this reaction, raising endogenous tissue PGE₂ to physiologic levels (~2-fold increase in most tissues studied) ¹.

Kinetics: The inhibitor’s potency against human 15-PGDH is in the sub-nanomolar range. Per Palla et al. 2021 (verified study page) and Singla et al. 2026 (verified study page), which both cross-reference Zhang 2015: Ki = 0.1 nM against purified 15-PGDH in biochemical assays ¹. Exact IC₅₀ and full kinetics require Zhang 2015 SI — PDF not yet in archive (status: pending). needs-pdf-verification (Ki = 0.1 nM is from secondary citations in verified companion pages; primary source verification against Zhang 2015 SI pending.) needs-replication (independent Ki confirmation from a second lab would strengthen this.)

Selectivity: Selectivity profiling against a panel of related SDR-family enzymes (reported in Zhang 2015 SI) shows preference for 15-PGDH over related enzymes including 11β-HSD1 and other SDR members. Full selectivity data requires SI access.

Downstream signalling: PGE₂ elevation engages EP2 and EP4 GPCRs on stem/progenitor cells in aged tissues. EP2/EP4 signalling activates:

• cAMP → PKA → CREB pathway → transcription of pro-proliferative and anti-apoptotic genes
• β-arrestin → PI3K/AKT → cell survival
• Wnt/β-catenin stabilization → stem cell self-renewal (established in HSC and satellite-cell contexts)

The gerozyme concept positions 15-PGDH as an enzyme whose accumulation with aging actively suppresses regenerative PGE₂ signalling — not merely a bystander marker. SW033291 pharmacologically reverses this age-elevated suppression ².

Comparison with COX inhibitors: NSAIDs (COX-1/2 inhibitors) suppress PGE₂ biosynthesis and would be expected to have opposite effects on tissue regeneration. This distinction is important for translational design — 15-PGDH inhibitors and NSAIDs have opposite PGE₂ directionality. contradictory-evidence — the coincidental use of NSAIDs (common in aged patients with musculoskeletal pain) could pharmacologically antagonize 15-PGDH inhibitor effects in co-administration; no interaction study published as of 2026-05-23.

Preclinical efficacy by indication

Tissue regeneration and hematopoietic recovery — Zhang 2015

The original discovery paper demonstrated SW033291 efficacy in two distinct regenerative contexts ¹:

1. Bone marrow recovery after lethal irradiation: Mice treated with SW033291 showed accelerated hematopoietic recovery after myeloablative irradiation and bone marrow transplantation. Survival was substantially improved versus vehicle control. The mechanism involved enhanced engraftment and proliferation of transplanted HSCs via EP2/EP4 signalling.

2. Colon tissue repair after irradiation: Colonic epithelial recovery was accelerated, with increased proliferation of crypt stem cells.

Both effects were abrogated in PGE₂ receptor knockout contexts, confirming on-target mechanism.

Dimension	Status
Pathway conserved in humans?	yes — 15-PGDH is expressed in human bone marrow and colon
Phenotype conserved in humans?	partial — HSC aging is conserved; irradiation model may not reflect chronic aging
Replicated in humans?	no needs-human-replication

Aged muscle / sarcopenia — Palla et al. 2021

The Blau lab (Stanford) established that 15-PGDH is elevated in aged mouse and human skeletal muscle, suppressing PGE₂ signalling ³. SW033291 was administered i.p. daily for 1 month to aged mice (>24 months old); the specific mg/kg dose is not stated in Palla 2021 — it is referenced only via Zhang 2015. Results in aged mice:

• Restored muscle mass (TA, gastrocnemius, soleus), fiber cross-sectional area, plantar flexor force, and treadmill endurance
• Elevated PGE₂ restored to levels on par with young muscle (LC-MS/MS)
• Reduced atrogene expression (MuRF1, Atrogin-1) and TGF-β signalling (myostatin, Smad3)
• Enhanced mitochondrial biogenesis (PGC1α, citrate synthase, mtDNA/nuclear DNA ratio) and autophagy flux
• Effects abrogated by conditional EP4 knockout in myofibers (confirming PGE₂-EP4 axis; not satellite-cell-depletion controls)

Human translational signal: microarray analysis (Raue 2012 dataset) of vastus lateralis biopsies showed significantly elevated HPGD expression in older adults (mean 78 ± 6 yr) versus young adults (mean 25 ± 3 yr). Specific fold-change and cohort n are not stated in the text of Palla 2021 (figures only; n not reported in the paper). needs-human-replication

Dimension	Status
Pathway conserved in humans?	yes — 15-PGDH elevation in aged human muscle confirmed by biopsy
Phenotype conserved in humans?	partial — muscle wasting is conserved; satellite cell pharmacology not yet tested in vivo in humans
Replicated in humans?	no

Neuromuscular junction regeneration — Bakooshli et al. 2023

Blau lab (Stanford) — note Bhutani is not an author on this paper — demonstrated SW033291 restores NMJ integrity after acute (nerve crush) and chronic (aging) denervation ⁴. Key findings (verified against local PDF 2026-05-23):

• Hpgd mRNA rises ~20-fold by day 90 post-sciatic-nerve-transection in murine skeletal muscle (10-fold at day 14); protein rises ~4-fold; enzymatic activity rises proportionally. Aged EDL has 20.2 ± 2.9% of myofibers denervated vs 4.4 ± 1.5% in young EDL
• Primary cellular source is denervated myofibers (identified by snRNA-seq of DN myonuclei) — NOT Schwann cells. The seeder’s Schwann-cell framing was incorrect
• SW033291 dose: 5 mg/kg i.p. once daily, vehicle 10% ethanol / 5% Cremophor EL / 85% D5W. 14 days for crush cohort; 1 month for aged-mouse cohort. Mice were male C57BL/6 (2–4 mo young from Jackson Labs; 24–26 mo aged from NIA colony)
• After sciatic-nerve crush: 37.2 ± 4.9% increase in plantar flexor tetanic force (PGDHi vs vehicle, day 14); motor axon counts 797 ± 69 (PGDHi) vs 415 ± 53 (vehicle), ~1.9× more; NMJ innervation 97.4 ± 2.1% vs 84.5 ± 2.2%
• In aged chronically-denervated mice: cleaved caspase-3+/ChAT+ apoptotic motor neurons reduced from 11.6 ± 2.5% (vehicle) to 4.2 ± 0.6% (PGDHi); CREB phosphorylation 37 ± 2% (PGDHi) vs 22 ± 3% (vehicle) of ChAT+ neurons
• Mechanism is motor-neuron-centric (EP4 → cAMP → CREB phosphorylation in spinal-cord motor neurons), not Schwann-cell-centric — Schwann cells were not directly tested in this paper
• 15-PGDH protein aggregates also co-localize with “target fibers” (NADH-bullseye histology) in n=10 patients with neurogenic myopathies (axonal neuropathies, myositis, ALS, SMA, lumbar radiculopathy, neurogenic amyloidosis); 9/10 cases positive

Local PDF is available at DOI lookup 10.1126/scitranslmed.adg1485.

Dimension	Status
Pathway conserved in humans?	partial — 15-PGDH aggregates documented in human neurogenic-myopathy biopsies (n=10 patients, 9/10 positive); EP4-CREB motor-neuron mechanism not human-validated
Phenotype conserved in humans?	partial — NMJ denervation and sarcopenia are conserved aging phenotypes
Replicated in humans?	no needs-human-replication

Articular cartilage / osteoarthritis — Singla et al. 2026

Bhutani lab, Science 2026 (online 2025-11-27) ⁵. 15-PGDH is elevated in aged and OA cartilage; PGDHi promotes cartilage regeneration in mouse PTOA + aged-mouse OA + ex-vivo human OA explants. The key mechanistic claim — verified against PMC13127300 — is that the regenerative effect operates via gene-expression reprogramming of preexisting chondrocytes, NOT via stem/progenitor expansion.

Key findings (verified 2026-05-23 against PMC):

• Aged mouse cohort n=9/group; PTOA (DMM) cohort n=7/group; baseline 15-PGDH expression in young (n=3) vs aged (n=5) mice p=0.036
• PGDHi reduces a 15-PGDH-high “hypertrophic-like” chondrocyte cluster (defined by CD200, Ihh, Mef2c, Wnt5b, Spp1, Pth1r, Alpl, Runx2) and expands an ECM-synthesizing articular chondrocyte cluster (defined by Bmp5, Fgf2, Stat5a, Hdac9). Col10a1/Mmp13/Col2a1/Acan/Sox9 are NOT the defining marker sets (the seeder’s guesses on this were wrong; see PMC for full panel)
• OA-associated pain reduced: CatWalk paw-contact p=0.010, mean intensity p=0.028 (vehicle vs PGDHi)
• Ex-vivo human OA cartilage: n=11 patients (ages 55–75, total-knee-replacement source) used for FACS characterization of the CD200+ 15-PGDH+ subpopulation; n=5/group used for GAG-content (DMMB) and stiffness (nanoindentation, Young’s modulus) PGDHi treatment readouts
• scRNA-seq deposited as GEO GSE308009
• PTOA histology evaluated at 6 weeks post-injury; pain readouts at 8 weeks (separate cohorts)
• Bhutani is last/corresponding author; PMID 41308124; PMC13127300 (released 2026-04-29)

Dimension	Status
Pathway conserved in humans?	partial — 15-PGDH elevation confirmed in human OA tissue; CD200+ subpopulation characterized in human explants (n=11)
Phenotype conserved in humans?	yes — osteoarthritis is the direct phenotype
Replicated in humans?	no (ex vivo only) needs-human-replication

Age-related heart failure — Zhang L et al. 2025

Experimental Gerontology 2025 ⁶. 15-PGDH inhibition by SW033291 ameliorated diastolic dysfunction and cardiac fibrosis in naturally aged (18–20 month) C57BL/6 mice. Key findings:

• Cardiac 15-PGDH expression elevated ~2-fold in aged vs young mouse heart
• SW033291 treatment reduced left-ventricular stiffness index (E/e’ ratio by echocardiography), interstitial fibrosis (Masson trichrome), and cardiac hypertrophy markers
• Mechanistically proposed via cardiac fibroblast EP4-cAMP pathway suppressing pro-fibrotic TGF-β signalling
• Gold OA; doi:10.1016/j.exger.2025.112710 confirmed in archive

Dimension	Status
Pathway conserved in humans?	partial — PGE₂/EP4 axis in cardiac fibrosis is conserved; 15-PGDH cardiac aging not yet confirmed in human tissue
Phenotype conserved in humans?	yes — diastolic dysfunction / HFpEF is a dominant aging cardiac phenotype
Replicated in humans?	no needs-human-replication

Expanding contexts (2024–2025)

Recency search (PubMed 2023–2026, query “SW033291” + “15-PGDH inhibitor” + “HPGD aging”; 2026-05-23) revealed additional preclinical applications:

• Blood-brain barrier / Alzheimer’s disease (Koh et al. 2025, PNAS, PMID 40397680): 15-PGDH inhibition mitigated BBB deterioration, suppressed neuroinflammation, and preserved cognitive function in mouse AD and TBI models. Neuroprotection occurred independent of amyloid pathology. This represents a new mechanistic branch — the PGE₂/EP4 → cAMP → PKA axis in cerebrovascular endothelial cells and microglia.

• Liver regeneration after acetaminophen injury (Li et al. 2025, BBRC, PMID 39855045): SW033291 reduced hepatic necrosis area, lowered ALT/AST, raised glutathione, and promoted hepatocyte proliferation in APAP-overdose mouse model. Mechanism: elevated PGE₂ → EP2/EP4 → hepatocyte proliferation.

• Type 2 diabetes metabolomics/transcriptomics (Metabolites 2024, PMID 39330516): metabolomic and transcriptomic characterization of SW033291 effects in a T2DM mouse model — expanded the compound’s documented downstream pathway reach but does not constitute aging-specific evidence.

• MASH-associated fibrosis (Cells 2025, PMID 40643509): 15-PGDH inhibition ameliorated MASH-associated apoptosis and fibrosis in an NASH mouse model.

None of these additional contexts have human trial data; all are preclinical. needs-human-replication

Pharmacokinetics and dosing

Typical preclinical doses: 5 mg/kg/day, intraperitoneal injection, in mice (confirmed in Bakooshli 2023; dose not explicitly stated in Palla 2021). Human-equivalent dose calculation is not available (no PK data in humans).

Route of administration: All published aging-context studies use i.p. injection in mice. Oral bioavailability in rodents has been characterized to some degree (Zhang 2015 mentions it) but the compound is not optimized for oral delivery. The Epirium Bio clinical candidate is presumably an orally bioavailable derivative — the properties of SW033291 itself do not predict clinical route ².

PK parameters in mice: No plasma half-life data is publicly reported in the primary papers reviewed. Given the i.p. route and the preclinical-tool status, PK optimization was not the primary focus of the published work. dose-response-unclear

Selectivity caveat: Although selectivity against related SDR enzymes was assessed (Zhang 2015 SI), broader kinome or off-target profiling has not been publicly reported. Butylsulfinyl-thienopyridine scaffolds have not been widely studied for kinome activity.

Safety and oncology tradeoff

15-PGDH functions as a tumor suppressor in multiple cancer types (colon, lung, breast). Loss-of-function mutations in HPGD are associated with increased colon cancer risk in epidemiological data. Pharmacological inhibition raises PGE₂, which at supraphysiologic concentrations can be pro-tumorigenic via EP2/EP4 → cAMP → PKA → immunosuppression pathways in the tumor microenvironment.

This creates an explicit cancer-aging tradeoff (see cancer-aging-tradeoffs): the intervention that restores youthful regenerative signalling may, with chronic use, lower the threshold for cancer initiation or progression — particularly in colonic epithelium (where 15-PGDH is most potently tumor-suppressive).

Key caveats mitigating this concern:

• SW033291 raises PGE₂ ~2-fold, toward physiologic levels — not to supraphysiologic levels. The distinction between physiologic PGE₂ restoration and pharmacologic PGE₂ overload may be safety-relevant.
• The mouse studies (including colon-regeneration data in Zhang 2015) did not report increased tumor burden at the doses and durations studied — but these were short-term experiments, not chronic oncogenicity studies.
• No long-term carcinogenicity data exists for any 15-PGDH inhibitor in any species. long-term-unknown

Immunosuppression concern: PGE₂ at high levels suppresses T-cell and NK-cell cytotoxic activity (EP2/EP4 on immune cells → cAMP → PKA → suppressed TCR signalling). Chronic elevation might impair anti-tumor immune surveillance.

Clinical program implication: Epirium Bio’s Phase 1/2 design for the clinical candidate will likely include colon cancer surveillance endpoints (e.g., colonoscopy, fecal occult blood) as part of safety monitoring.

Clinical translation status

SW033291 is a research tool compound — it is not in, nor being pursued for, human clinical trials. The compound was not designed for clinical use and lacks the pharmaceutical properties (formulation, manufacturing, bioavailability) required for a clinical candidate.

Epirium Bio (formerly Emmyon; spun out from Stanford / Blau lab) is pursuing a separate, undisclosed 15-PGDH inhibitor derivative as a clinical candidate, reportedly targeting sarcopenia and neuromuscular indications. As of 2026-05-23, the Epirium structure is not publicly disclosed and no NCT-registered trial targeting 15-PGDH with a pharmacological inhibitor appears in ClinicalTrials.gov (query confirmed 2026-05-23; 0 results for “SW033291”; 0 results for “15-PGDH inhibitor”). The Epirium compound should be seeded as a separate page when structural or trial registration data becomes public.

ClinicalTrials.gov active trials for SW033291: 0 (confirmed 2026-05-23)

The Blau & Porpiglia 2026 Annual Review of Pharmacology and Toxicology review ² summarizes the gerozyme concept and translational pipeline. It is the most current synthesis of the target-to-clinic rationale.

Hallmark linkage

Hallmark	Directionality	Mechanism
stem-cell-exhaustion	Reverses	Restores EP2/EP4 → Wnt/β-catenin → satellite cell, HSC, chondrocyte activation
altered-intercellular-communication	Restores	PGE₂ is a paracrine regenerative signal; age-elevated 15-PGDH suppresses tissue PGE₂ communication

Secondary (context-specific, not in frontmatter):

• chronic-inflammation — PGE₂ has complex, context-dependent pro- and anti-inflammatory effects; net impact of 15-PGDH inhibition on chronic inflammaging is not established
• mitochondrial-dysfunction — cardiac PGE₂/EP4 signalling may influence mitochondrial biogenesis; speculative

Limitations and gaps

1. No human data. Every efficacy claim derives from mouse (or ex vivo human tissue) experiments. The gerozyme framing is compelling but unvalidated in humans. needs-human-replication
2. Tool compound only. SW033291’s pharmaceutical properties are not optimized. Doses, route, and systemic exposure used in mouse experiments cannot be directly extrapolated to humans.
3. Cancer risk unknown. Long-term carcinogenicity of 15-PGDH inhibition has not been studied. The tumor-suppressor loss-of-function framing is the dominant safety concern and must be resolved in the Epirium clinical program. long-term-unknown
4. Selectivity not fully characterized. Off-target profiling against the broader proteome has not been published. needs-replication
5. PGE₂ immunosuppression at scale. If systemic PGE₂ rises substantially (even toward physiologic levels) in immunocompromised aged hosts, the net immunological effect is uncertain.
6. NSAID interaction. Concurrent NSAID use (highly prevalent in elderly patients with musculoskeletal pain) would pharmacologically oppose 15-PGDH inhibitor effects; no interaction study published.
7. Sex-specific effects. Most published mouse work uses male mice; female-specific PGE₂ biology (menstrual cycle, sex-hormone–prostaglandin interactions) has not been characterized for this compound.
8. Epirium clinical compound undisclosed. The clinical-stage compound’s structure, PK, and target selectivity profile are not public, making independent evaluation of the class translation gap impossible. unsourced

Footnotes

Footnotes

1. doi:10.1126/science.aaa2340 · Zhang Y, Desai A, Yang SY, Bae KB, Antczak MI, Nowak SJ, … Jain S · Science 2015 Jun 5;348(6240) · in-vivo (mice) + biochemical · model: C57BL/6 mice; irradiation + BMT model; colonic-injury model; biochemical 15-PGDH assay · first identification of SW033291 as 15-PGDH inhibitor; HTS-derived; confirmed via DOI lookup (277 citations; FWCI 18.1; citation percentile 100) · archive status: download pending ↩ ↩² ↩³

2. doi:10.1146/annurev-pharmtox-071724-100856 · Blau HM, Porpiglia E · “From Cell Reprogramming to Tissue Rejuvenation: Countering Aging by Targeting a Gerozyme” · Annu Rev Pharmacol Toxicol 2026;66:7–26 · PMID 40971573 · review · consolidates gerozyme concept and 15-PGDH inhibitor translational rationale across all tissue contexts; introduces the gerozyme framework (enzyme accumulating with aging that mechanistically suppresses regeneration); mentions Epirium Bio clinical development · archive status: hybrid OA; download pending · confirmed via PubMed efetch + DOI lookup (1 citation; citation percentile 95) ↩ ↩² ↩³

3. palla-2021-15pgdh-muscle-rejuvenation · doi:10.1126/science.abc8059 · Palla AR, Ravichandran M, Wang YX, Alexandrova L, Yang AV, Kraft P, Holbrook CA, Schürch CM, Ho ATV, Blau HM · Science 2021 Jan 29;371(6528):eabc8059 · in-vivo (aged mice, >24 mo) + human microarray (Raue 2012 dataset; vastus lateralis; mean aged 78 ± 6 yr vs young 25 ± 3 yr) · n=3–15/group depending on assay (mass/force n=7–12); human cohort n not stated in paper · model: aged C57BL/6 mice (>24 mo vs young 2–4 mo); SW033291 i.p. daily 1 month; exact mg/kg dose NOT stated in paper (referenced via Zhang 2015) · muscle mass, CSA, plantar flexor force, treadmill endurance all improved vs vehicle (p<0.05 to p<0.0001); elevated HPGD in aged human vastus lateralis microarray (fold-change not stated in paper text; figures only) · sex of mice not stated · archive status: download failed; PMC7938328 (OA) · confirmed via DOI lookup (198 citations; FWCI 13.7; citation percentile 100) · verified 2026-05-23 against verified companion study page ↩

4. bakooshli-2023-15pgdh-nmj-regeneration · doi:10.1126/scitranslmed.adg1485 · PMID 37820010 · PMC10763629 · Bakooshli MA, Wang YX, Monti E, Su S, Kraft P, Nalbandian M, Alexandrova L, Wheeler JR, Vogel H, Blau HM (last author — NOT Bhutani; the seeder author list was fabricated and has been corrected) · Sci Transl Med 15(717):eadg1485 (2023) · in-vivo · model: male C57BL/6 (2–4 mo young Jackson + 24–26 mo aged NIA colony); SW033291 5 mg/kg i.p. once daily, 14 d for crush / 1 mo for aged · plantar flexor force +37.2 ± 4.9% at 14 dpi; motor axon counts 1.9× higher with PGDHi; aged motor-neuron apoptosis 11.6%→4.2% with PGDHi; n=10 human neurogenic-myopathy biopsies (9/10 positive for 15-PGDH-target-fiber co-localization) · local PDF available · verified 2026-05-23 against full PDF ↩

5. singla-2025-15pgdh-cartilage-regeneration · doi:10.1126/science.adx6649 · PMID 41308124 · PMC13127300 · Singla M, Wang YX,…, Blau HM, Bhutani N (last/corresponding) · Science 391(6789):1053–1062, in-print 2026-03-05 (online 2025-11-27) · in-vivo (aged C57BL/6 n=9/group + PTOA C57BL/6 n=7/group) + ex-vivo human OA cartilage explants (n=11 patients aged 55–75 for FACS; n=5/group for GAG/stiffness treatment readouts) · scRNA-seq GEO GSE308009 + multiplexed IF · mechanism: gene-expression reprogramming of preexisting chondrocytes (NOT progenitor expansion); CatWalk pain p=0.010/p=0.028 vehicle vs PGDHi · journal PDF closed-access; PMC release 2026-04-29 used for verification · verified 2026-05-23 ↩

6. doi:10.1016/j.exger.2025.112710 · Zhang L (first), …, Guan W (last) · Exp Gerontol 2025;203:112710 · PMID 39952309 · in-vivo · n TBC · model: naturally aged 18–20 month C57BL/6 mice; SW033291 doses in primary source · diastolic dysfunction (E/e’ ratio), cardiac fibrosis, and hypertrophy markers improved vs vehicle · archive status: gold OA; download pending · confirmed via DOI lookup (2 citations; FWCI 8.2; citation percentile 96) ↩