Silk Cotton Flower: Metabolic, Gut and Skin Research
The flower is both the most-eaten part of Bombax ceiba — a genuine vegetable and tea ingredient across South and Southeast Asia, already described on the hub — and, by a wide margin, the most actively researched part in the current scientific literature. Eleven distinct papers, seven of them published in the last three years, cover flower extracts against type 2 diabetes, constipation, colitis, bacteria, cancer cell lines, and, in a 2026 human-cell study, wound healing. All of it is preclinical: cell culture and rodent work, none of it a clinical trial. This page keeps every finding labelled by its actual mechanism and model, because "the flower has research" is not one claim — it is at least five different ones.
Table of Contents
- Antidiabetic Research: Two Independent 2022 Studies
- How the Flower's Antidiabetic Signal Compares to the Leaf's
- Gut Motility: Relieving Constipation in a Mouse Model
- Gut Barrier: Flower Polysaccharide Against Colitis
- Wound Healing in Human Skin Cells (Paradee 2026)
- Antimicrobial Research: Flower Extract as a Nanoparticle Platform
- Cancer Cell Line Research
- The Five-Flower Tea Tradition, Validated Ethnobotanically
- Reading Eleven Papers Together Without Overclaiming
- Cautions and Practical Guidance
- Key Research Papers
- Connections
Antidiabetic Research: Two Independent 2022 Studies
Two separate groups published flower-specific antidiabetic research in 2022, using different methods and reaching complementary but differently-weighted conclusions.
Yin and colleagues (Frontiers in Nutrition) fractionated Bombax ceiba flower extract by polarity and identified the ethyl acetate fraction as the most active: the strongest antioxidant activity of any fraction tested (DPPH, ABTS and FRAP assays all outperformed the synthetic positive control BHT), moderate α-glucosidase inhibition (IC50 2.85 mg/mL), and cytotoxicity against four cancer cell lines (HepG2, MCF-7, Raw264.7, A549) at sub-2 mg/mL concentrations. UPLC/Q-TOF-MS identified 38 compounds, mostly phenolic acids and flavonoids, and network pharmacology analysis proposed 41 potential human protein targets relevant to type 2 diabetes, concentrated in intracellular receptor signaling and fatty-acid pathways.
Yasien and colleagues (Bioengineering) took a different approach: comparing five extraction solvents (water, ethanol, methanol, chloroform, petroleum ether) for yield of polyphenols, flavonoids, glycosaponins and polysaccharides, then running one functional readout — a yeast-cell glucose uptake assay — on the best-yielding (aqueous) extract, which increased glucose uptake by yeast cells 70.38% over control.
These two results are not equally strong, and should not be cited interchangeably. Yin's α-glucosidase inhibition is a specific, quantified, named-enzyme assay directly relevant to carbohydrate digestion. Yasien's yeast-cell glucose uptake assay is a much cruder, older screening method: yeast cells are not human tissue, do not use the insulin-signaling pathway human muscle and fat cells use, and the assay is essentially a permeability/solubility proxy historically used for rapid low-cost screening, not a validated model of glycemic control. It is real data and it is weak evidence, and both things are true at once.
How the Flower's Antidiabetic Signal Compares to the Leaf's
This site's main hub page for Bombax ceiba already covers a separate, older antidiabetic thread from a different part: shamimin, a flavonol C-glycoside isolated from leaves in 1999 with reported hypoglycemic activity in animals, plus more recent leaf-specific work (Xu and colleagues, 2017, two papers) showing an ethanol extract of leaves improved glucose, lipids and diabetic nephropathy markers in high-fat-diet and streptozotocin-treated rodents, with the effect partly attributed to the leaf's mangiferin content.
Putting the numbers side by side is informative. Root methanol extract, tested by a different group entirely (covered on the Root page), inhibited α-glucosidase at an IC50 of 21.21 µg/mL. The flower's ethyl acetate fraction (Yin, above) inhibited the same enzyme at an IC50 of 2.85 mg/mL — roughly 130 times weaker by this specific in-vitro measure, though the two studies used different extraction methods and cannot be compared as a controlled head-to-head. What this comparison is useful for is not ranking the parts against each other on thin, non-comparable data — it is illustrating why "α-glucosidase inhibition" as a bare claim needs the part, the extract, and the concentration attached every time, because the same assay on the same species produced wildly different numbers depending on which part supplied the material.
The ceiling that applies to all of it: α-glucosidase inhibition is the mechanism of acarbose, a licensed diabetes drug with a real but modest effect on HbA1c in actual patients. Whatever any part of this plant does through that mechanism in a dish, the mechanism's own best-known clinical translation sets realistic expectations for what a human dose could plausibly achieve, if one existed — which it does not, for any part of this tree.
Gut Motility: Relieving Constipation in a Mouse Model
Wang and colleagues (Pharmaceutical Biology, 2023) tested Bombax ceiba flower aqueous extract against loperamide-induced constipation in sixty male mice, with phenolphthalein (a real, if now largely discontinued as an OTC drug for safety reasons, stimulant laxative) as an active comparator. At the highest dose tested (160 mg/kg), the flower extract increased fecal water content, stool number and gastrointestinal transit rate, and shortened the time to first defecation. Mechanistically, the extract raised serum motilin, gastrin and substance P (three hormones that promote gut motility) while lowering somatostatin (which suppresses it), reduced colonic mucosal damage, restored goblet-cell function, down-regulated the water channel AQP3, and increased c-kit expression (a marker of the interstitial cells of Cajal, the gut's pacemaker cells). Twelve known compounds, including protocatechuic acid, chlorogenic acid and rutin, were identified in the extract.
This is a real, dose-dependent, mechanistically coherent prokinetic finding — and it is worth naming plainly that a prokinetic, motility-promoting effect is the physiological opposite of the astringent, motility-slowing tradition attached to this tree's gum, covered on its own page. Both findings can be true simultaneously because they concern different parts with different chemistry; neither should be allowed to stand in for the other.
Gut Barrier: Flower Polysaccharide Against Colitis
A 2026 study (Zhang and colleagues, Chemistry & Biodiversity) extracted an acidic heteropolysaccharide from Bombax ceiba flower using a green deep-eutectic-solvent method, then tested it in a dextran sulfate sodium (DSS)-induced ulcerative colitis mouse model — a standard, validated model of inflammatory bowel disease. The flower polysaccharide significantly reduced disease activity index, body weight loss, colon shortening and spleen enlargement, suppressed inflammatory cytokines and oxidative stress, and restored intestinal barrier function by increasing mucus secretion and tight-junction protein expression.
This is, again, a gut-protective, barrier-restoring finding from the flower — consistent in direction with the constipation-relief result above, and consistent with the general pattern that this part's gut-related research points toward protection and motility rather than the astringent binding traditionally attributed to the gum. It is one mouse study, unreplicated, and a considerable distance from a treatment for human inflammatory bowel disease, which has its own established medical management.
Wound Healing in Human Skin Cells (Paradee 2026)
The strongest single piece of evidence anywhere in this plant's literature, by methodological standard rather than by effect size, is a 2026 study (Paradee and colleagues, Pharmaceutical Biology) testing ethanolic Bombax ceiba flower extract on two human cell lines: HaCaT keratinocytes and Detroit 551 fibroblasts, both standard, widely used human-derived (though immortalized, not patient-fresh) skin cell models.
LC-QTOF/MS identified gallic acid, epicatechin, rutin, quercetin, hesperetin, apigenin and luteolin in the extract, which showed high phenolic (140.2 mg gallic-acid-equivalent/g) and flavonoid (8.3 mg quercetin-equivalent/g) content. The extract scavenged DPPH and ABTS radicals (IC50 48.0 and 33.4 µg/mL respectively) and reduced hydrogen-peroxide-induced reactive oxygen species inside both cell types. In the actual wound-relevant assays, the extract enhanced wound closure (a scratch-assay measure of cell migration) in both keratinocytes and fibroblasts, and increased collagen production in a dose-dependent manner.
Why this ranks above most of the rest of the plant's literature, and what it still is not. Human-derived cell lines are a meaningfully stronger evidence tier than the rat and mouse models that dominate this plant's research, and the endpoints tested (migration, collagen synthesis) are directly, mechanistically relevant to wound repair rather than a proxy assay. It is, however, still cell culture in a dish — no wound, no dressing, no person. "Promotes wound closure in cultured human keratinocytes" and "heals wounds" are different sentences, and only the first one is supported here. No topical formulation, concentration, or application protocol has been tested on an actual wound in an actual organism.
Antimicrobial Research: Flower Extract as a Nanoparticle Platform
Two recent papers used flower extract as a "green" reducing agent to synthesize metal nanoparticles with antibacterial activity: Aziz and colleagues (2025) made silver and gold nanoparticles from crude flavonoids extracted from the flowers, and Safdar and colleagues (2023) made selenium nanoparticles from flower extract, testing both for antibacterial activity and (in the selenium case) urea detection. As with the bark-derived silver nanoparticle work covered on this plant's Stem Bark page, the antibacterial activity being measured belongs to the resulting metal nanoparticle, not to the flower extract taken as a food or tea. These are manufacturing applications, included for completeness, not evidence that eating the flower or drinking flower tea has antibacterial effects in a person.
Cancer Cell Line Research
Three papers report cytotoxic activity of flower extracts against cultured cancer cell lines: Diab and colleagues (2022) found polyphenol-rich flower extracts cytotoxic and genotoxic to Huh7 liver cancer cells while also measuring antioxidant activity; Tundis and colleagues (2014) reported in-vitro cancer cell growth inhibition alongside antioxidant activity in flower extracts; and the Yin 2022 study above separately found the flower's ethyl acetate fraction cytotoxic to four different cancer cell lines. This is standard, early-stage cytotoxicity screening — a cell line dying in a dish when exposed to a plant extract is a common and non-specific finding across enormous numbers of plants, foods and even some non-toxic compounds, and is many steps removed from any anticancer therapeutic claim. No animal tumor model and no human data exist for this plant's anticancer potential; it is mentioned here only because the papers exist and a reader may encounter them.
The Five-Flower Tea Tradition, Validated Ethnobotanically
This plant's hub page already describes dried mu mian hua as a component of the traditional Cantonese "five-flower tea" drunk across Hong Kong and southern China in hot weather. A 2024 ethnopharmacology study (Chan and colleagues, Journal of Ethnobiology and Ethnomedicine) documented this tradition directly — confirming Bombax ceiba flower's place as one of the five standard ingredients and recording the tea's traditional "cooling," damp-heat-clearing use. This is ethnobotanical documentation, not pharmacological evidence: it confirms the tradition is real and current, not that the tea achieves any specific physiological effect. A separate 2025 review (Kumari and colleagues, Chemistry & Biodiversity) surveys the flower's potential as a functional-food ingredient more broadly, drawing on the same preclinical literature covered above.
Reading Eleven Papers Together Without Overclaiming
Eleven real, species-and-part-confirmed papers is a substantial body of research by this plant's standards, and it would be a mistake to read that volume as convergence on one health claim. It is not one claim. It is a scatter of preliminary, mostly single-study, entirely preclinical signals across five distinct pharmacological domains — metabolic, gut motility, gut barrier, skin repair, and cytotoxicity — sharing only the fact that they all used flower material. None of the eleven has been replicated by an independent group on the same endpoint. None has reached an animal disease-treatment trial beyond the single mouse studies described above, let alone a human trial. The single strongest piece by evidence tier — the human-keratinocyte wound-healing work — is still cell culture. "Actively researched" and "clinically proven" remain entirely different statements, and this plant's flower, more than any of its other parts, is at real risk of the first being mistaken for the second simply because there is more of it to point to.
Cautions and Practical Guidance
- Culinary and traditional tea use of the flower has a long, informal safety record and is a different question from any of the concentrated-extract findings above, all of which used specific laboratory preparations at specific doses that do not correspond to a cup of tea or a cooked vegetable serving.
- No human trial or human dosing study exists for any concentrated flower extract, for any of the indications discussed on this page.
- Do not use flower preparations in place of prescribed diabetes, inflammatory bowel disease, or wound care. All three have monitored, evidence-based treatments this plant has not been tested against.
- If you take diabetes medication, be aware the flower joins the leaf and root in carrying preclinical hypoglycemic signals — a third independent reason, from a third plant part, for caution about combining concentrated preparations with glucose-lowering drugs.
- Pregnancy and breastfeeding: no safety data exist for concentrated extracts; the culinary flower, eaten as food in the amounts traditionally used, is a different and lower-concern category, but has not been formally studied either.
Key Research Papers
- Paradee N et al. (2026). Bombax ceiba flower extract enhances antioxidant defense and wound healing effects in human keratinocytes and fibroblasts. Pharmaceutical Biology, 64(1):599-614. Human cell lines; the strongest single result on this page. — PubMed
- Yin K et al. (2022). A preliminary study of the chemical composition and bioactivity of Bombax ceiba L. flower and its potential mechanism in treating type 2 diabetes mellitus using UPLC-Q-TOF-MS and network pharmacology analysis. Frontiers in Nutrition, 9:1018733. — PubMed
- Yasien S et al. (2022). Comparative Evaluation of Various Extraction Techniques for Secondary Metabolites from Bombax ceiba L. Flowering Plants along with In Vitro Anti-Diabetic Performance. Bioengineering, 9(10):486. Yeast-cell assay — the weakest evidence tier on this page; read alongside the caveat above. — PubMed
- Wang L et al. (2023). Therapeutic effects of Bombax ceiba flower aqueous extracts against loperamide-induced constipation in mice. Pharmaceutical Biology, 61(1):125-134. Phenolphthalein comparator; prokinetic, not astringent. — PubMed
- Zhang Q et al. (2026). Polysaccharides From Bombax ceiba Flower by Deep Eutectic Solvent Based Three-Phase Partitioning and Their Protective Effects Against DSS-Induced Colitis in Mice. Chemistry & Biodiversity, 23(3):e03434. — PubMed
- Aziz N et al. (2025). Green synthesis and antibacterial activity of silver and gold nanoparticles using crude flavonoids extracted from Bombax ceiba flowers. Cellular and Molecular Biology. Nanoparticle-mediated, not flower-extract-alone activity. — PubMed
- Safdar M et al. (2023). Bombax ceiba flower extract mediated synthesis of Se nanoparticles for antibacterial activity and urea detection. World Journal of Microbiology & Biotechnology. — PubMed
- Diab KA et al. (2022). Polyphenol Content, Antioxidant, Cytotoxic, and Genotoxic Activities of Bombax ceiba Flowers in Liver Cancer Cells Huh7. Asian Pacific Journal of Cancer Prevention, 23(4). Cell-line screening only. — PubMed
- Tundis R et al. (2014). In vitro cancer cell growth inhibition and antioxidant activity of Bombax ceiba (Bombacaceae) flower extracts. Natural Product Communications. — PubMed
- Chan KT et al. (2024). Ethnopharmacology of five flowers herbal tea, a popular traditional beverage in Hong Kong and South China. Journal of Ethnobiology and Ethnomedicine. Documents tradition; not a pharmacology trial. — PubMed
- Kumari N et al. (2025). Exploring the Therapeutic Potential of Bombax ceiba L. Flowers as Source of Functional Food Development: A Review. Chemistry & Biodiversity. — PubMed