Momordicoside A

• Synonyms:
  (3β,9β,10α,22S,23R,24R)-22,23,24,25-Tetrahydroxy-9-methyl-19-norlanost-5-en-3-yl 6-O-β-D-glucopyranosyl-β-D-glucopyranoside, 19-Norlanostane, β-D-glucopyranoside deriv.
• Molecular Formula: C42H72O15
• Molecular Weight: 817.0 g/mol
• SMILES: CC(C1CCC2(C1(CCC3(C2CC=C4C3CCC(C4(C)C)OC5C(C(C(C(O5)COC6C(C(C(C(O6)CO)O)O)O)O)O)O)C)C)C)C(C(C(C(C)(C)O)O)O)O
• InChI: InChI=1S/C42H72O15/c1-19(27(44)32(49)35(52)39(4,5)53)20-13-14-42(8)25-11-9-21-22(40(25,6)15-16-41(20,42)7)10-12-26(38(21,2)3)57-37-34(51)31(48)29(46)24(56-37)18-54-36-33(50)30(47)28(45)23(17-43)55-36/h9,19-20,22-37,43-53H,10-18H2,1-8H3/t19-,20+,22+,23+,24+,25+,26-,27-,28+,29+,30-,31-,32+,33+,34+,35+,36+,37-,40-,41+,42-/m0/s1
• InChIKey: WBYJYPOPDKQBQJ-WVMSUIAMSA-N

Momordicoside A for Diabetes & Inflammation Research

Momordicoside A (C₄₂H₇₂O₁₅, MW 817.01) is a cucurbitane-type triterpenoid glycoside first isolated from the seeds of Momordica charantia L. (bitter melon, Cucurbitaceae), with its structure definitively established by spectral analysis and X-ray crystallography as the 3-O-β-gentiobioside of cucurbit-5-ene-3β,22(S),23(R),24(R),25-pentaol [1]. This compound belongs to the momordicoside family (A, B, F1, F2, K, L, U, etc.), which share a common cucurbitane skeleton but differ critically in glycosylation patterns and oxidation states [2]. Momordicoside A has been identified as a protein tyrosine phosphatase 1B (PTP1B) inhibitor and α-glucosidase inhibitor, with documented antiproliferative, anti-inflammatory, and glucose metabolism-modulating activities in peer-reviewed studies [3].

PTP1B Inhibition, Not AMPK Activation

Designed for insulin signaling pathway studies via phosphatase inhibition; distinct from AMPK-activating momordicosides.

Weak α-Glucosidase Inhibitor Reference

Provides a low-potency benchmark for enzyme inhibition assays; not a high-potency screening candidate.

In Vivo Model Feasibility

Acute toxicity profile supports use in metabolic disease research models; independent dose-range verification required.

• [1] Okabe H, Miyahara Y, Yamauchi T, Miyahara K, Kawasaki T. Studies on the constituents of Momordica charantia L. I. Isolation and characterization of momordicosides A and B, glycosides of a pentahydroxy-cucurbitane triterpene. Chemical and Pharmaceutical Bulletin. 1980;28(9):2753-2762. doi:10.1248/cpb.28.2753 View Source
• [2] Liu Y, Wang M, Zhao Y, et al. Cucurbitane-type triterpene glycosides from the fruits of Momordica charantia. Planta Medica. 2008;74(10):1291-1294. doi:10.1055/s-2008-1081297 View Source
• [3] PubChem. Momordicoside A (Compound Summary). National Center for Biotechnology Information. CID 71717038. View Source

Momordicoside A vs. Analogs: Substitution Risks

Cucurbitane-type triterpenoid glycosides from Momordica charantia share a common aglycone core but exhibit divergent biological activities driven by variations in glycosylation at the C-3 position and oxidation patterns at C-7, C-19, and C-25 . Momordicoside A bears a 3-O-β-gentiobioside moiety, whereas momordicoside B possesses a 3-O-β-D-xylopyranosyl(1→4)-[β-D-glucopyranosyl(1→6)]-β-D-glucopyranoside, and momordicosides K, L, F1, and F2 carry alternative sugar substitutions that profoundly affect enzyme inhibition potency and target selectivity [1]. In head-to-head enzymatic assays, momordicoside A demonstrates α-glucosidase inhibitory activity that differs markedly from its structural analogs, and its α-amylase inhibition profile and anti-inflammatory gene expression effects are not predictable from chemical similarity alone [2][3]. Consequently, substituting momordicoside A with momordicoside B, K, L, or other in-class compounds without specific validation of the intended assay endpoint will introduce uncontrolled variables that compromise reproducibility and data interpretability.

Glycosylation Determines Enzyme Profile

The 3-O-β-gentiobioside moiety of momordicoside A produces weak α-glucosidase inhibition; analogs B, K, L carry different sugar patterns that markedly shift potency. Class membership alone does not predict assay outcome.

Mechanism Divergence: PTP1B vs. AMPK

Momordicoside A inhibits PTP1B, while momordicosides Q–T activate AMPK. Substituting one for the other changes the signaling cascade under investigation and requires revalidation of endpoints.

Anti-Inflammatory and Antiproliferative Profiles Are Compound-Specific

Gene expression modulation and antiproliferative ranking vary across cucurbitane glycosides. Data from one analog cannot be assumed to transfer to momordicoside A without direct evidence.

• [1] Okabe H, Miyahara Y, Yamauchi T, Miyahara K, Kawasaki T. Studies on the constituents of Momordica charantia L. I. Isolation and characterization of momordicosides A and B, glycosides of a pentahydroxy-cucurbitane triterpene. Chemical and Pharmaceutical Bulletin. 1980;28(9):2753-2762. doi:10.1248/cpb.28.2753 View Source
• [2] Perera WH, Shivanagoudra SR, Perez JL, et al. Anti-inflammatory, antidiabetic properties and in silico modeling of cucurbitane-type triterpene glycosides from fruits of an Indian cultivar of Momordica charantia L. Molecules. 2021;26(4):1038. doi:10.3390/molecules26041038 View Source
• [3] Nhiem NX, Kiem PV, Minh CV, et al. α-Glucosidase inhibition properties of cucurbitane-type triterpene glycosides from the fruits of Momordica charantia. Chemical and Pharmaceutical Bulletin. 2010;58(5):720-724. doi:10.1248/cpb.58.720 View Source

Momordicoside A Evidence Guide

α-Glucosidase Inhibition vs. Analogs

In a standardized α-glucosidase inhibitory assay, momordicoside A exhibited an IC₅₀ of 0.47 mM (470 μM) [1]. This value positions momordicoside A as having weak-to-moderate inhibitory activity relative to other cucurbitane-type glycosides isolated from the same plant material. In a separate study evaluating 14 cucurbitane-type triterpene glycosides, compounds with alternative glycosylation patterns (e.g., compounds 12 and 13 in that study) showed moderate inhibitory activity, whereas momordicoside A (identified as compound 3 in that work) was categorized as exhibiting weak inhibitory activity and was not among the most potent inhibitors [2]. This differential potency, attributable to the 3-O-β-gentiobioside moiety of momordicoside A, underscores that not all momordicosides are interchangeable for α-glucosidase inhibition assays.

α-Glucosidase IC₅₀

Reported

0.47 mM (470 μM)

Weak inhibitor; 5.3‑fold less potent than acarbose

Ranked among the weaker cucurbitane glycosides in the tested panel; not a high‑potency lead

Evidence Dimension	α-Glucosidase inhibitory activity (IC₅₀)
Target Compound Data	0.47 mM (470 μM)
Comparator Or Baseline	Most potent compounds in class (e.g., compounds 12 and 13 in Nhiem et al. 2010) demonstrated moderate inhibitory activity; acarbose positive control: IC₅₀ = 87.65 ± 6.51 μM (0.088 mM)
Quantified Difference	Momordicoside A is approximately 5.3-fold less potent than acarbose and categorized as weak inhibitor relative to the most active compounds in the series
Conditions	In vitro α-glucosidase enzyme inhibition assay; enzyme source not specified

Why This Matters

Researchers selecting a cucurbitane glycoside for α-glucosidase studies must recognize that momordicoside A is a weak inhibitor (IC₅₀ 0.47 mM), not a high-potency candidate, and other momordicosides (e.g., K, L) may exhibit different inhibitory profiles requiring independent validation.

• [1] Pham HY et al. Isolation process research of momordicoside A from the fruits of Momordica charantia L. and its anti-diabetes activities in in vivo model. Vietnam Academy of Science and Technology Project Report. Institute of Marine Biochemistry; 2012-2013. View Source
• [2] Nhiem NX, Kiem PV, Minh CV, et al. α-Glucosidase inhibition properties of cucurbitane-type triterpene glycosides from the fruits of Momordica charantia. Chemical and Pharmaceutical Bulletin. 2010;58(5):720-724. doi:10.1248/cpb.58.720 View Source

α-Amylase & α-Glucosidase Inhibition vs. Momordicoside L

In a direct comparative study evaluating purified cucurbitane glycosides from an Indian cultivar of M. charantia, momordicoside A and momordicoside L were assayed side-by-side for α-amylase and α-glucosidase inhibition. At a fixed concentration of 0.87 mM for α-amylase inhibition, momordicoside A demonstrated 76.6% inhibition, comparable to momordicoside L (76.2%) and acarbose (68.0–76.6%) [1]. For α-glucosidase inhibition at 1.33 mM, momordicoside A exhibited 33.5% inhibition, notably higher than momordicoside L (23.7%) but substantially lower than the most active compound in the panel, karaviloside VIII (56.5%) [1]. These data reveal that while momordicosides A and L share similar α-amylase inhibition profiles, their α-glucosidase inhibitory activities diverge by approximately 1.4-fold (33.5% vs. 23.7%), confirming that minor structural differences translate into measurable functional divergence.

vs. Momordicoside L

Head‑to‑head

α‑Amylase: 76.6% vs 76.2% (indistinguishable); α‑Glucosidase: 33.5% vs 23.7% (1.41× higher)

α‑Glucosidase activity diverges; compounds are not interchangeable for this endpoint

At fixed 1.33 mM; α‑amylase profiles similar but α‑glucosidase difference materially affects assay interpretation

Evidence Dimension	α-Amylase inhibition (%) at 0.87 mM; α-Glucosidase inhibition (%) at 1.33 mM
Target Compound Data	α-Amylase: 76.6% inhibition; α-Glucosidase: 33.5% inhibition
Comparator Or Baseline	Momordicoside L: α-amylase 76.2% inhibition, α-glucosidase 23.7% inhibition; karaviloside VIII: α-glucosidase 56.5% inhibition (most active in panel)
Quantified Difference	α-Amylase: difference ≤0.5% (statistically indistinguishable); α-Glucosidase: momordicoside A is 1.41-fold more active than momordicoside L, but only 59% as active as karaviloside VIII
Conditions	In vitro enzymatic assays; compounds tested at 0.87 mM (α-amylase) and 1.33 mM (α-glucosidase); acarbose used as positive control

Why This Matters

For α-glucosidase inhibition studies, momordicoside A and momordicoside L are not interchangeable; momordicoside A provides 41% greater inhibition at equivalent concentration, a difference that would materially affect experimental outcomes and data interpretation.

• [1] Perera WH, Shivanagoudra SR, Perez JL, et al. Anti-inflammatory, antidiabetic properties and in silico modeling of cucurbitane-type triterpene glycosides from fruits of an Indian cultivar of Momordica charantia L. Molecules. 2021;26(4):1038. doi:10.3390/molecules26041038 View Source

PTP1B Inhibition vs. AMPK Activation in Analogs

Momordicoside A has been identified as a protein tyrosine phosphatase 1B (PTP1B) inhibitor, a mechanism distinct from the AMPK-activating properties reported for other momordicosides (e.g., momordicosides Q, R, S, and T) [1][2]. PTP1B is a validated negative regulator of insulin signaling; its inhibition enhances insulin receptor phosphorylation and downstream glucose uptake [1]. While quantitative IC₅₀ data for PTP1B inhibition by momordicoside A are not currently published in the peer-reviewed literature, the compound is commercially categorized as a phosphatase inhibitor . In contrast, momordicosides Q, R, S, and T have been shown to stimulate GLUT4 translocation in L6 myotubes and 3T3-L1 adipocytes via AMPK activation, with in vivo glucose tolerance enhancement demonstrated in both insulin-sensitive and insulin-resistant mice [2]. These divergent mechanisms—PTP1B inhibition (momordicoside A) vs. AMPK activation (momordicosides Q–T)—represent distinct pharmacological entry points for diabetes research and are not functionally interchangeable.

PTP1B vs. AMPK Mechanism

Class‑level

PTP1B inhibition (momordicoside A) vs. AMPK activation (momordicosides Q–T)

Divergent signaling entry points; assay selection must match mechanism

Quantitative PTP1B IC₅₀ not reported; dose‑response pilot recommended

Evidence Dimension	Primary antidiabetic mechanism of action
Target Compound Data	PTP1B inhibition (inhibitory effect documented; quantitative IC₅₀ not reported in peer-reviewed literature)
Comparator Or Baseline	Momordicosides Q, R, S, T: AMPK pathway activation, stimulation of GLUT4 translocation in L6 myotubes and 3T3-L1 adipocytes; in vivo glucose disposal enhancement in mice
Quantified Difference	Qualitative mechanistic divergence: phosphatase inhibition vs. kinase activation
Conditions	In vitro PTP1B enzyme assay (momordicoside A); in vitro GLUT4 translocation assays and in vivo glucose tolerance tests (momordicosides Q–T)

Why This Matters

Investigators designing experiments targeting insulin signaling pathways must distinguish between PTP1B inhibitors (momordicoside A) and AMPK activators (momordicosides Q–T); the choice of compound dictates the signaling cascade under interrogation and the appropriate assay endpoints (e.g., phosphatase activity assays vs. GLUT4 translocation/AMPK phosphorylation).

• [1] MedChemExpress. Momordicoside A Product Datasheet: Phosphatase Inhibitor. CAS 75801-95-5. View Source
• [2] Tan MJ, Ye JM, Turner N, et al. Antidiabetic activities of triterpenoids isolated from bitter melon associated with activation of the AMPK pathway. Chemistry & Biology. 2008;15(3):263-273. doi:10.1016/j.chembiol.2008.01.013 View Source

Anti-Inflammatory Gene Expression Profile

In a panel of cucurbitane glycosides (compounds 2–6) tested at 50 μM for anti-inflammatory gene expression effects, momordicoside A (compound 5 in the study) differentially modulated pro-inflammatory markers [1]. The study evaluated effects on IL-6, TNF-α, iNOS, and COX-2 expression, demonstrating that individual compounds within this structural class produce distinct patterns of inflammatory marker modulation [1]. While the specific quantitative expression values for each marker across all compounds are not disaggregated in the abstract, the study established that cucurbitane glycosides from M. charantia differentially affect inflammatory gene expression, supporting the conclusion that momordicoside A is not functionally interchangeable with other in-class compounds for anti-inflammatory research applications.

Inflammatory Gene Expression

Context‑dependent

Differential modulation of IL‑6, TNF‑α, iNOS, COX‑2 at 50 μM

Compound‑specific anti‑inflammatory profile; not a class effect

Quantitative per‑marker comparisons not disaggregated; source review advised

Evidence Dimension	Pro-inflammatory gene expression modulation
Target Compound Data	At 50 μM, momordicoside A differentially affected expression of IL-6, TNF-α, iNOS, and COX-2
Comparator Or Baseline	Karaviloside VI, karaviloside VIII, momordicoside L, charantoside XV (all evaluated at 50 μM)
Quantified Difference	Differential effects observed; quantitative comparisons per marker not available in open-access abstract
Conditions	In vitro gene expression analysis in relevant cell models; compounds tested at 50 μM

Why This Matters

This evidence, though categorized as supporting, reinforces that cucurbitane glycosides exhibit compound-specific anti-inflammatory profiles; procurement of momordicoside A for inflammation studies should be based on its specific activity profile rather than assuming class-wide equivalence.

• [1] Perera WH, Shivanagoudra SR, Perez JL, et al. Anti-inflammatory, antidiabetic properties and in silico modeling of cucurbitane-type triterpene glycosides from fruits of an Indian cultivar of Momordica charantia L. Molecules. 2021;26(4):1038. doi:10.3390/molecules26041038 View Source

Antiproliferative Activity in Bioassay-Guided Fractions

Bioassay-guided fractionation of M. charantia fruit ethanol extract led to the isolation of 15 cucurbitane-type triterpene glycosides, with antiproliferative evaluation performed against MCF-7 (breast), WiDr (colon), HEp-2 (laryngeal), and Doay (medulloblastoma) human tumor cell lines [1]. Several isolated cucurbitane-type triterpene glycosides demonstrated antiproliferative activities in this panel [1]. However, momordicoside A was not among the compounds specifically highlighted for potent antiproliferative activity in the published results, in contrast to compound 14 (7β,25-dihydroxycucurbita-5,23(E)-dien-19-al 3-O-β-D-allopyranoside) which was noted for potent hypoglycemic activity in the same study [1]. Separately, a study using momordica extract (not isolated momordicoside A) in K562/AO2 multidrug-resistant leukemia cells identified a non-cytotoxic concentration of 5 μg/mL, with multidrug resistance reversal effects observed (5.4-, 6.5-, and 4.0-fold reversal for adriamycin, vincristine, and daunomycin, respectively) and apoptosis rates of 19.38 ± 1.06% when combined with vincristine [2]. It is critical to note that this MDR reversal study used a crude momordica extract, not purified momordicoside A, and therefore these quantitative data cannot be attributed to momordicoside A alone.

Antiproliferative Ranking

Context‑dependent

Not among highlighted antiproliferative compounds in bioassay‑guided fractionation; crude‑extract MDR data not attributable to pure momordicoside A

Not a lead antiproliferative; requires purified compound for reproducible studies

Crude extract reversal values (5.4–6.5×) cannot be assigned to momordicoside A alone

Evidence Dimension	Antiproliferative activity and multidrug resistance reversal
Target Compound Data	Included in bioassay-guided fractionation but not among the highlighted antiproliferative compounds; MDR reversal data from crude momordica extract (5 μg/mL non-cytotoxic concentration, 5.4–6.5× reversal) cannot be attributed to purified momordicoside A
Comparator Or Baseline	Compound 14 (7β,25-dihydroxycucurbita-5,23(E)-dien-19-al 3-O-β-D-allopyranoside) identified with potent hypoglycemic activity
Quantified Difference	Momordicoside A not among the most antiproliferative compounds in the isolated panel
Conditions	MCF-7, WiDr, HEp-2, and Doay human tumor cell lines (antiproliferative); K562/AO2 multidrug-resistant leukemia cell line (MDR reversal with crude extract)

Why This Matters

Researchers should not select momordicoside A as a lead antiproliferative candidate based on class membership; other cucurbitane glycosides or compounds in the extract may drive the observed activities. This evidence highlights the necessity of using purified, authenticated momordicoside A rather than crude extracts for reproducible mechanistic studies.

• [1] Hsiao PC, Liaw CC, Hwang SY, et al. Antiproliferative and hypoglycemic cucurbitane-type glycosides from the fruits of Momordica charantia. Journal of Agricultural and Food Chemistry. 2013;61(12):2979-2986. doi:10.1021/jf3041116 View Source
• [2] Ye AF, Yin LH, Xiong SD, Han YX, Zhang SH, Wu JB. Study on reversal of multidrug resistance in K562/AO2 cells by momordica. Chinese Journal of Clinical Pharmacology and Therapeutics. 2009;14(7):766-770. View Source

In Vivo Acute Toxicity Assessment

Acute toxicity evaluation of momordicoside A in an in vivo model resulted in classification as non-toxic under the tested conditions [1]. This safety-relevant data point was generated as part of a project that also established a 20 kg-per-batch isolation process for momordicoside A from M. charantia dried fruits and confirmed α-glucosidase inhibitory activity (IC₅₀ = 0.47 mM) [1]. While acute toxicity data for comparator momordicosides (e.g., momordicoside B, K, or L) are not currently available in the peer-reviewed literature for direct cross-study comparison, this dataset establishes a baseline safety profile for momordicoside A that supports its use in in vivo diabetes research models.

Acute Toxicity Assessment

Context‑dependent

Non‑toxic classification in in vivo model

Supports in vivo feasibility; no comparator data for other momordicosides

Species and dose not specified; independent verification before long‑term studies

Evidence Dimension	Acute toxicity classification
Target Compound Data	Non-toxic (in vivo acute toxicity assessment)
Comparator Or Baseline	Comparative toxicity data for other momordicosides (B, K, L, F1, F2) not available in peer-reviewed literature
Quantified Difference	Not calculable due to absence of comparator data
Conditions	In vivo acute toxicity model; project conducted at Institute of Marine Biochemistry, VAST; species and dose not specified in project summary

Why This Matters

The non-toxic classification supports the feasibility of in vivo studies with momordicoside A, a critical consideration for researchers planning animal experiments. Absence of comparator toxicity data underscores that safety extrapolation across momordicoside analogs is not scientifically justified.

• [1] Pham HY et al. Isolation process research of momordicoside A from the fruits of Momordica charantia L. and its anti-diabetes activities in in vivo model. Vietnam Academy of Science and Technology Project Report. Institute of Marine Biochemistry; 2012-2013. View Source

Momordicoside A Application Scenarios

PTP1B Inhibition in Type 2 Diabetes Research

Momordicoside A is appropriately deployed as a PTP1B inhibitor in insulin signaling and type 2 diabetes research [1]. Researchers investigating phosphatase-mediated regulation of insulin receptor phosphorylation should select momordicoside A over AMPK-activating momordicosides (Q, R, S, T) when the experimental objective is to interrogate PTP1B-specific mechanisms rather than AMPK-dependent GLUT4 translocation [2]. Experimental design must account for the absence of published PTP1B IC₅₀ data for momordicoside A; dose-response pilot studies are recommended to establish effective working concentrations in the investigator's specific assay system.

α-Glucosidase Assays: Weak Inhibitor Reference

Given its IC₅₀ of 0.47 mM (470 μM) for α-glucosidase inhibition [1] and its categorization as a weak inhibitor relative to other cucurbitane glycosides [2], momordicoside A is best utilized as a weak inhibitor reference compound in α-glucosidase assay development or as a comparator for validating more potent inhibitors (e.g., acarbose, IC₅₀ 87.65 ± 6.51 μM) [3]. Researchers should not select momordicoside A as a lead candidate for potent α-glucosidase inhibition; momordicoside L (23.7% inhibition at 1.33 mM) or karaviloside VIII (56.5%) may be more appropriate depending on the desired potency window [4].

In Vivo Diabetes Studies with Toxicity Baseline

Momordicoside A has an established acute toxicity classification of non-toxic in an in vivo model [1], providing a safety foundation for animal studies in diabetes and metabolic disease research. The availability of a scalable isolation process (20 kg dried fruit per batch) further supports procurement for in vivo work [1]. Researchers should verify species-specific toxicity profiles independently before initiating long-term or high-dose studies.

QC and Analytical Methods for M. charantia

Momordicoside A has been baseline-separated from momordicoside L, momordicoside K, and momordicoside F₂ in a validated 14-minute UHPLC-UV-ELSD/MS method [1], and an SPE-HPLC method with UV detection at 208 nm has been established for its rapid quantification [2]. These analytical benchmarks support the use of authenticated momordicoside A reference standard for QC of bitter melon dietary supplements, botanical extracts, and natural product research materials, with calibration linearity from 10–1,000 mg/L (r² = 0.9992) and recovery >90% [2].

Application

Selection Property

Validation Focus

Insulin signaling pathway studies

PTP1B inhibition mechanism

Phosphatase activity assays; dose‑response pilot (IC₅₀ not published)

Carbohydrate metabolism enzyme assays

Weak α‑glucosidase inhibitor reference

Comparative inhibition vs. acarbose or validated inhibitors

In vivo metabolic disease models

Baseline acute toxicity profile

Species‑specific toxicity verification; dose‑range establishment

Botanical reference standard QC

Validated UHPLC separation from analogs

Calibration linearity, recovery, and method transfer

• [1] MedChemExpress. Momordicoside A Product Datasheet: Phosphatase Inhibitor. CAS 75801-95-5. View Source
• [2] Tan MJ, Ye JM, Turner N, et al. Antidiabetic activities of triterpenoids isolated from bitter melon associated with activation of the AMPK pathway. Chemistry & Biology. 2008;15(3):263-273. doi:10.1016/j.chembiol.2008.01.013 View Source
• [3] Nhiem NX, Kiem PV, Minh CV, et al. α-Glucosidase inhibition properties of cucurbitane-type triterpene glycosides from the fruits of Momordica charantia. Chemical and Pharmaceutical Bulletin. 2010;58(5):720-724. doi:10.1248/cpb.58.720 View Source
• [4] Perera WH, Shivanagoudra SR, Perez JL, et al. Anti-inflammatory, antidiabetic properties and in silico modeling of cucurbitane-type triterpene glycosides from fruits of an Indian cultivar of Momordica charantia L. Molecules. 2021;26(4):1038. doi:10.3390/molecules26041038 View Source
• [5] Wang S, et al. Rapid analysis of momordicoside A in bitter melon by high performance liquid chromatography following solid phase extraction. Se Pu (Chinese Journal of Chromatography). 2001;19(2):128-131. PMID:12541654. View Source