Mechanism of Shaofu Zhuyu Decoction in Improving Diabetes Mellitus Erectile Dysfunction Inhibition of Ferroptosis Based on Network Pharmacology and Experimental Validation

TCM formula SFZYD reduced ferroptosis markers and improved erectile function in diabetic rats via the Nrf2/HO-1/GPX4 pathway — no human data

Journal: Journal of Traditional Chinese Medicine | Published: 2026-04 | Type: Journal Article | PMID: 42015776 Authors: Wang Z, Mao Y, Zang Y, He S, Sun J, Wei Z, Wang M, Yang Y — all Changchun University of Chinese Medicine and its affiliated hospital, Changchun, China Funding/COI: Multiple Jilin Province and National Natural Science Foundation of China grants (none directly named for SFZYD); no conflicts of interest declared

Summary

Shaofu Zhuyu decoction (SFZYD) is a classical TCM formula traditionally used for blood stasis conditions. This study used network pharmacology to identify ferroptosis — a form of iron-dependent programmed cell death — as a predicted mechanism target, then tested SFZYD in high-glucose-exposed corpus cavernosum endothelial cells (CCECs) and streptozotocin-induced diabetic rats. The researchers report that SFZYD reduced oxidative stress, restored GPX4 and HMOX1 expression, and improved erectile function in the rat model via the Nrf2/HO-1/GPX4 signaling pathway. This is a preclinical mechanistic paper with no human data.

Claims

• Network pharmacology analysis identified 48 ferroptosis-related molecular targets for SFZYD, with HMOX1 and GPX4 as the primary hits
• SFZYD restored HMOX1 and GPX4 mRNA levels in CCECs damaged by high-glucose conditions
• In diabetic rats, SFZYD "effectively alleviated" erectile dysfunction and reduced fibrosis and blood sinus damage in penile tissue (no quantified intracavernosal pressure ratios or ICP/MAP values reported in the abstract)
• After Nrf2 inhibition, SFZYD reversed the resulting decreases in Nrf2, HMOX1, and GPX4 expression and improved ROS levels and mitochondrial membrane potential

Study Quality

This is a multi-method preclinical study combining data mining, network pharmacology, in vitro cell work, and in vivo rat experiments. The use of a specific Nrf2 inhibitor to confirm pathway dependency is methodologically sound — it moves beyond correlation toward causal mechanism testing. Flow cytometry for ROS and mitochondrial membrane potential adds quantitative rigor to the cellular work.

However, the foundational layer — network pharmacology — is increasingly criticized as hypothesis-generating at best and cherry-picking at worst. The method maps drug components to protein targets using databases and compound-target prediction algorithms, which regularly produce large lists of "hits" that do not survive wet-lab validation. That 48 targets were identified and two were selected as "primary" without reporting selection criteria is a structural weakness. No sample sizes are reported anywhere in the abstract, and no quantitative effect sizes (e.g., ICP/MAP ratios for erectile function) are provided for the animal experiments.

Red Flags

• No sample sizes reported — the abstract gives no n for cell experiments, rat groups, or any comparison
• No quantified erectile function outcome — "effectively alleviated erectile dysfunction" with no ICP/MAP data or erectile index reported in the abstract
• Network pharmacology is hypothesis-generating, not causal — the 48-target hit list is a computational prediction, not a validated target screen
• All authors from a single TCM institution — no independent replication, inherent confirmation bias risk for a study validating that institution's traditional formula
• COI not disclosed — standard Chinese government funding declarations do not rule out undisclosed relationships
• No human data of any kind — preclinical findings in rats have historically poor translation to ED treatment
• Journal specialization — the Journal of Traditional Chinese Medicine has an editorial interest in publishing positive TCM findings; no comment on peer review rigor
• "Data mining" methodology underdescribed — what databases, what inclusion criteria, what statistical thresholds

Strengths

• Mechanistic design using pathway inhibition (Nrf2 inhibitor) to test causal claims, not just correlation
• Multi-level validation: computational → cellular → animal → pathway knockdown
• Histological endpoints (H&E, Masson staining) provide structural tissue data, not just biomarker proxies
• Flow cytometry adds quantitative cellular readouts (ROS, mitochondrial membrane potential)
• Ferroptosis is a genuinely novel and biologically plausible mechanism in diabetic vascular injury

Verdict

This paper is mechanistically interesting as a hypothesis-generator: ferroptosis in corpus cavernosum endothelium is a real and understudied target, and the Nrf2/GPX4 axis is a legitimate pathway. The multi-method design is more rigorous than most TCM papers. But it answers exactly one question — does SFZYD affect these markers in rats — and does so without reporting the effect sizes that would let anyone evaluate clinical relevance. No human data, no dose-response characterization, no pharmacokinetic work, and no comparison to existing diabetes-ED treatments. File it under "interesting rat data that may never get followed up with a clinical trial."