Method Article

Integrated UHPLC-MS/MS and HRMS Strategy for Screening Potentially Hepatotoxic Furan-Containing Compounds in Dioscorea bulbifera L.

DOI:

10.3791/70865

May 19th, 2026

In This Article

Summary

Loading...
$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

This study developed an integrated strategy combining UHPLC–MS/MS and HRMS to screen for potentially hepatotoxic furan-containing compounds in Dioscorea bulbifera L.

Abstract

Loading...
$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

Dioscorea bulbifera L. (DBL), a traditional herbal medicine used to treat thyroid disorders and tumors, has raised considerable safety concerns due to its potential hepatotoxic effects. Previous research suggests that this liver injury may be associated with the metabolic activation of furan-containing compounds (FCCs) present in DBL. Here, we systematically characterized FCCs and their reactive metabolites in DBL extract through an integrated analytical approach combining ultra-high performance liquid chromatography coupled with tandem mass spectrometry (UHPLC-MS/MS) and ultra-high performance liquid chromatography coupled with tandem high-resolution mass spectrometry (UHPLC-HRMS). Employing both targeted and untargeted mass spectrometric analyses, we identified a total of 849 constituents in the aqueous extract of DBL, including 10 furanoditerpenoids and 17 additional FCCs. To elucidate the metabolic activation pathways, we used glutathione (GSH) and 4-bromobenzylamine (BBA) as dual trapping agents within a mouse liver microsomal (MLM) incubation system. Through P450-mediated metabolism, seven corresponding adducts derived from diosbulbin B (DSB), C (DSC), D (DSD), E (DSE), L (DSL), N (DSN), and 2-pentylfuran were successfully trapped and characterized. This study establishes a sensitive and specific approach for the comprehensive profiling of potentially hepatotoxic furan compounds in DBL, and provides valuable insights into the formation of reactive metabolites and toxicity mechanisms related to furan-containing herbal medicine.

Introduction

Loading...
$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

Safety concerns related to traditional Chinese medicine (TCM), especially drug-induced liver injury (DILI), hinder their widespread clinical utility and modernization1,2,3. Growing evidence suggests that most constituents of TCM are not intrinsically toxic; instead, their hepatotoxicity results from metabolic activation in vivo1,4,5. Furan-containing compounds (FCCs) are well recognized as prominent hepatotoxic precursors, undergoing P450–mediated metabolic activation to produce highly reactive electrophilic intermediates, specifically cis-enedials, which are crucial in initiating liver toxicity. These electrophilic metabolites form covalent adducts with cysteine thiols and amino groups of essential proteins, thereby facilitating the progression of toxic effects6,7,8.

In TCM, Dioscorea bulbifera L. (DBL) has been widely used for the treatment of thyroid disorders, tumors, and skin diseases9,10. A total of 152 chemical constituents have been isolated and characterized from DBL, including 41 furanoditerpenoids11. Among these, diosbulbin B (DSB), a major bioactive component, has been demonstrated to induce hepatic injury in murine models, an effect closely linked to the formation of the DSB-derived cis-enedial (DDE)12,13. Driven by advances in analytical instrumentation, mass spectrometry has become an important tool for analyzing complex chemical components and metabolites of TCM14,15. Conventionally, cis-enedials produced by DSB metabolism are detected by trapping electrophilic species with N-acetyl-L-lysine in combination with glutathione or N-acetyl-L-cysteine, followed by LC-MS/MS analysis of the resulting chemically stable pyrrole derivatives16,17. Building on this approach, we previously developed an improved strategy employing simultaneous trapping of furan-derived metabolic intermediates with GSH and BBA, coupled with LC-MS/MS analysis using a neutral loss scan of 129 Da and a precursor ion scan of m/z 272 for GSH adducts, together with a 1:1 precursor ion scan of m/z 169 and 171 derived from the BBA moiety for pyrrole detection (Figure 1)18,19. While effective for known furanoids, this approach exhibits notable limitations when applied to complex mixtures, particularly for FCCs lacking prior structural information. Therefore, there remains a pressing need for a comprehensive analytical strategy capable of enabling thorough chemical characterization of complex matrices of TCM while simultaneously supporting efficient screening of hepatotoxic precursors.

Here, we report an integrated strategy for the systematic characterization of potentially hepatotoxic FCCs in DBL (Figure 2). The strategy involves three key steps: (i) comprehensive chemical profiling of the aqueous extract of DBL using widely targeted metabolomics (WTM) based on UHPLC–MS/MS; (ii) trapping of reactive intermediates generated from FCCs with GSH and BBA; and (iii) detection and structural elucidation of GSH/BBA adducts formed via metabolic activation of potentially toxic constituents using HPLC–MS/MS and UHPLC–MS/MS. This integrated approach enables unambiguous identification of key toxic precursors responsible for DBL-induced hepatotoxicity from complex TCM matrices and provides a highly sensitive and specific framework for the safety assessment of TCM.

Access restricted. Please log in or start a trial to view this content.

Protocol

Loading...
$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

NOTE: The reagents and the equipment used in this study are listed in the Table of Materials.

1. Preparation of the DBL test solution

  1. Collect and authenticate three batches of the tubers of Dioscorea bulbifera L. Following the National Specification for Processing of TCM, remove fibrous roots from the tubers, wash them to remove dirt, slice them into 1 cm thick circular slices while fresh, dry the slices in a forced air oven at 60 °C, then remove and cool them to room temperature20.
  2. Weigh 200 g of DBL from each batch and grind them into a coarse powder (particle size 250-850 µm) using a high-speed crusher.
  3. Soak 20 g of crude powder in 200 mL of deionized water (1:10, w/v) for 30 min. Concurrently, process a mixture of the three batches of coarse powder prepared in step 1.2 (mixed at a ratio of 1:1:1, w/w).
  4. Extract the soaked material by heating under reflux for 60 min. Filter the extract through gauze or a sieve to remove insoluble residues. Add 200 mL of deionized water to the remaining residue and repeat the reflux extraction 2x, each for 60 min. Combine all three filtrates and concentrate the pooled extract by boiling to a final volume of 100 mL.
  5. Transfer the concentrated extract to a vacuum freeze dryer and lyophilize to obtain powder.
  6. Accurately weigh 18 mg of each batch of extract powder. Dissolve each in 1.5 mL of 50% acetonitrile, and vortex the mixture for 30 s to ensure homogeneity.
  7. Centrifuge the solution at 13,800 x g for 10 min at 4°C and collect the supernatant as the MLM incubation stock solution.
  8. Pipette 20 µL of the stock solution, dilute it to 1:50 with 50% acetonitrile, and vortex for 30 s. Then filter the solution through a 0.22 µm microporous membrane to obtain the sample solution ready for mass spectrometric analysis.

2. Chemical characterization of DBL

  1. Chromatographic conditions
    1. Use a C18 column (2.1 mm x 100 mm, 1.8 µm) and set the column oven temperature to 40 °C. Prepare mobile phase A as 0.1% (v/v) formic acid in water and mobile phase B as acetonitrile.
    2. Program the gradient elution as follows: 0–10 min, 2%–50% B; 10–11 min, 50%–95% B; and 11–13 min, 95% B. Set the flow rate to 0.4 mL/min and inject 2 µL of the sample.
  2. Perform mass spectrometric analysis using the following parameters: Ion source: IonDrive Turbo V electrospray ionization (ESI); Acquisition mode: Multiple reaction monitoring (MRM) with simultaneous acquisition in positive and negative ion modes; Ion spray voltage: +5.5 kV (positive) / −4.5 kV (negative); Curtain gas: 35 psi; Source temperature: 400 °C; Ion source gas 1: 60 psi; Ion source gas 2: 60 psi; Declustering potential (DP): ±100 V.
  3. Data processing
    1. Acquire and process MRM data using SCIEX Analyst Software (v1.6.3). Convert the raw data files (.wiff) to TXT format using MSConvert. Annotate the detected compounds by querying the proprietary BiotreeDB (V2.1) MS library. The annotation was based on the matching of retention times (RT), mass-to-charge ratios (m/z), and fragmentation patterns with authentic standards.

3. Characterization of GSH/BBA conjugates generated in MLM Incubations

  1. MLM incubation
    1. Slowly thaw the mouse liver microsomal suspension on ice. On ice, prepare the incubation mixture in a 2 mL microcentrifuge tube by sequentially adding the components listed below. Add the NADPH-regenerating system last. For the control group, replace the NADPH-regenerating system with an equal volume of ice-cold PBS. Vortex briefly to mix.
      1. Prepare an incubation mixture consisting of the following components: Mouse liver microsomal suspension (MLM, 4 mg/mL protein): 125 µL; magnesium chloride (MgCl₂) solution (32 mM): 10 µL; DBE (12 mg/mL) or DSB standard solution (35 mM): 10 µL; glutathione (GSH) solution (50 mM): 100 µL; BBA solution (250 mM): 10 µL; PBS buffer (pH 7.4, containing 1.0 mol/L EDTA and 0.1 mol/L K2HPO4): 145 µL; NADPH-regenerating system solution (5.0 mM): 100 µL.
    2. Immediately upon completion of the incubation, remove all tubes from the incubator. Terminate the enzymatic reaction by adding 1 mL of ice-cold acetonitrile (-20 °C) to each tube.
    3. Vortex for 3 min to precipitate proteins. Centrifuge at 13,800 x g for 10 min at 4 °C.
    4. Collect the supernatant, filter through a 0.22 µm membrane, and transfer the filtrate to an autosampler vial.
  2. Preliminary screening of MLM incubation of DBL extracts by HPLC-MS/MS
    1. Chromatographic conditions
      1. Use a C18 column (150 mm x 4.6 mm, 3 µm) and set the column temperature to 25 °C. Prepare mobile phase A as 0.1% (v/v) formic acid in water and mobile phase B as acetonitrile.
      2. Run the following gradient: 0–2 min, 10% B; 2–15 min, linear increase from 10% to 90% B. Set the flow rate to 0.8 mL/min and inject 5 µL of the prepared sample.
    2. Mass spectrometric conditions. Set the full-scan range to m/z 500–1000. Precursor ion (PI) scans at m/z 169 and 171, and a neutral loss (NL) scan of 129 Da in positive ion mode, and a PI scan at m/z 272 in negative ion mode. Set key parameters as follows: ion spray voltage (IS), +5,500 V (positive) / −4,500 V (negative); declustering potential (DP), ±70 V; entrance potential (EP), ±10 V; collision energy (CE), +35 V (positive) / −40 V (negative); collision cell exit potential (CXP), ±10 V; ion source gas 1 and 2 (GS1/GS2), 50 psi each; curtain gas (CUR), 40 psi; and source temperature (TEM), 650 °C.
    3. Acquire and process all mass spectrometric data using SCIEX Analyst Software (v1.6.3).
  3. Identification of pyrrole derivatives and FCCs by UHPLC-HRMS
    1. Chromatographic conditions
      1. Use a C18 column (2.1 mm x 100 mm, 1.9 µm) and set the column oven to 30 °C. Prepare mobile phase A as 0.1% (v/v) formic acid in water and mobile phase B as acetonitrile.
      2. Program the gradient as follows: 0-20 min, 5%-95% B; 20-21 min, 95% B. Set the flow rate to 0.3 mL/min and inject 2 µL of the sample.
    2. Conduct mass spectrometric analysis with the following parameters: Ion source: ESI; ion spray voltage: +3.5 KV (positive ions)/-2.5 KV (negative ions); capillary temperature: 320 °C; probe heater temperature: 350 °C; sheath gas: 10 arb; maximum spray current: 100 µA; S-lens RF level: 50%; collision energy: 35 V (positive ions) and 40 V (negative ions); scan range: 100-1000 m/z.
    3. Data preprocessing
      1. Based on the identified FCCs from DBL and the trapping mechanism (Figure 1), draw the predicted GSH/BBA conjugates using ChemBioDraw Ultra 14.0 (Supplementary Figure 1).
      2. Import the acquired HRMS raw files (.raw) into the Qual Browser module of Xcalibur (V4.1). Create an identification workflow in the software to screen for GSH/BBA-conjugates.
      3. Compare the experimental MS1 and MS2 spectra with the predicted structures of GSH/BBA conjugates derived from FCCs (predicted using ChemBioDraw Ultra 14.0) based on the known adduction mechanisms (Figure 1 and Supplementary Figure 1). Use a minimum peak intensity threshold of 1 x 105 and a mass accuracy tolerance of ± 10 ppm for all monitored ions. Confirm conjugates based on the characteristic bromine isotope pattern (approximately 1:1) observed in both MS1 and MS2 spectra, including diagnostic ions at m/z 169 and 171.
      4. Determine the molecular weights of the parent FCCs by subtracting the mass of the GSH/BBA tag from the observed molecular ions. Use 472/474 Da (C17H21O5N4BrS) for the subtraction.
      5. Characterize the parent FCCs in DBL by UHPLC–HRMS based on the results obtained above.

Access restricted. Please log in or start a trial to view this content.

Results

Loading...
$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

Identified FCCs in DBL
To characterize the chemical profile of the DBL, a combination of triple quadrupole-based WTM and HRMS profiling was employed. This dual approach allowed for the comprehensive analysis of extracted ion pairs, generating highly informative extracted ion chromatograms (EICs, Figure 3A) and total ion chromatograms (TICs, Figure 3B). Under the optimized chromatographic and detection conditions, a total of 849 compounds spa...

Access restricted. Please log in or start a trial to view this content.

Discussion

Loading...
$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

Previous investigations into the hepatotoxicity of DBL have primarily focused on the metabolic intermediates of individual FCCs using targeted capture strategies9,22. Although these approaches provide valuable mechanistic insights, their effectiveness in complex herbal matrices is often limited by the poor detectability of low-abundance bioactivated intermediates. In the present study, we developed an upgraded analytical workflow for the systematic identification...

Access restricted. Please log in or start a trial to view this content.

Disclosures

Loading...
$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

The authors have nothing to disclose.

Acknowledgements

Loading...
$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,

This work was supported by the National Natural Science Foundation of China (No. 82360759), the Natural Science Foundation of Guizhou Province (No. ZK [2024]406, ZK [2024] 397, Postdoctoral Station [2021]007, BSH [2024]005), and Guizhou University of Traditional Chinese Medicine TD Hopes [2023]005.

Access restricted. Please log in or start a trial to view this content.

Materials

List of materials used in this article
NameCompanyCatalog NumberComments
4-BromobenzylamineJiuding Chemical (Shanghai) Technology Co., LtdX114214A
AcetonitrileSigma-Aldrich (Shanghai) Trading Co., Ltd1000294000
CentrifugeWuxi BaiTaKe Biotechnology Co., LtdCR3180
Chromatographic sample bottlesZhejiang ALWSCI Technology Co., LtdC0000008
diosbulbin BChengdu PufeiDe Biotech Co., Ltd18022603
Dioscorea bulbifera L.-DBL-1; DBL-2; DBL-3The plant materials were authenticated by Prof. Weike Jiang (Guizhou University of Traditional Chinese Medicine, Guiyang, China), with voucher specimens deposited at the Key Laboratory of Basic Pharmacology of Guizhou Province.
Electronic BalanceShanghai Anting Electronic Instrument FactoryFA1004B
Formic acidThermo Fisher Scientific Co., Ltd195715
GlutathioneSigma-Aldrich (Shanghai) Trading Co., LtdSLBX5462
High speed  crusherYongkang Biao Hardware Products Co., LtdXC-2000Y
Kunming mice Liver Microsomes (male)Dalian Meilun Biotechnology Co., Ltd.MB12496-1
Magnesium ChlorideBeijing Solarbio Science & Technology Co., LtdM8161
MethanolSigma-Aldrich (Shanghai) Trading Co., Ltd646377
millipore filterTianjin Jinteng Experimental Equipment Co., LtdNylon66
NADPH Na4Beijing Solarbio Science & Technology Co., Ltd102X022
OvenTianjin Laboratorial Instrument & Equipment Co., LtdGFL125
Phosphate-Buffered SalineBeyotime Biotechnology  Co., LtdC0221B
Thermostatic water bath oscillatorShanghai Zijia Instrument Co., LtdTHZ-82A
Ultra-freezerThermo Fisher Scientific Co., LtdTDE60086FV-ULTS
Ultrasonic CleanerShanghai Zijia Instrument Co., LtdL10-250A
Vortex MixerJiangsu Haimen Medical Instrument FactoryVM-D

References

Loading...
$$\rightleftharpoonup{xx}$$ $$\longleftharp{xx}$$, $$\longrightharp{xx}$$,
  1. Ma, Z., Shi, Z., Xiao, X., Wang, J. New insights into herb-induced liver injury. Antioxid Redox Signal. 38 (16-18), 1138-1149 (2023).
  2. Yang, Y., et al. A review of herb-induced liver injury in mainland China. Front Pharmacol. 13, 813073(2022).
  3. Shen, T., et al. Incidence and etiology of drug-induced liver injury in mainland China. Gastroenterology. 156 (8), 2230-2241 (2019).
  4. Zhou, Y., et al. Mechanism of drug-induced liver injury and hepatoprotective effects of natural drugs. Chin Med. 16 (1), 135(2021).
  5. Wang, Y., et al. Metabolic activation of toxic natural products from herbal and dietary supplements leading to toxicities. Front Pharmacol. 12, 758468(2021).
  6. Pan, J., et al. Difference in hepatotoxicity of furan-containing components in Cortex Dictamni correlates with the efficiency of their metabolic activation. Phytomedicine. 114, 154778(2023).
  7. Liao, Y., et al. DNA damage and up-regulation of PARP-1 induced by columbin in vitro and in vivo. Toxicol Lett. 379, 20-34 (2023).
  8. Tian, M., Peng, Y., Zheng, J. Metabolic activation and hepatotoxicity of furan-containing compounds. Drug Metab Dispos. 50, 2458-2471 (2022).
  9. Hu, Z., et al. Development of a mechanism-based biomarker for Dioscorea bulbifera L. exposure and hepatotoxicity in rats. Phytomedicine. 102, 154172(2022).
  10. Narzary, C., Sarkar, D., Das, P., Papi, D. Ethnobotany phytochemistry, and pharmacological activity of Dioscorea bulbifera: A comprehensive review. Chem Biodivers. 22 (1), e202401408(2025).
  11. Guan, X., et al. Bioactivity,toxicity and detoxification assessment of Dioscorea bulbifera L.: A comprehensive review. Phytochem Rev. 16 (3), 573-601 (2017).
  12. Jiang, Y., et al. Disruption of acetyl-CoA acyltransferase 2 resulting from exposure to diosbulbin B. Chem Res Toxicol. 38 (7), 1203-1214 (2025).
  13. Yang, Y., et al. Histones H2B/H3 selectively modified by acis-enedial metabolite resulting from metabolic action of diosbulbin B: Potential target proteins for antitumor agent design. J Med Chem. 68 (13), 13408-13420 (2025).
  14. Shao, L., et al. Pharmacokinetics and UPLC–MS/MS analysis of delsoline in mouse whole blood. J Anal Methods Chem. 2018, 9412708(2018).
  15. Shi, H., et al. RefMetaPlant: A reference metabolome database for plants across five major phyla. Nucleic Acids Res. 52 (D1), D1614-D1628 (2024).
  16. Tan, R., et al. Diosbulbin B: An important component responsible for hepatotoxicity and protein covalent binding induced by Dioscorea bulbifera L. Phytomedicine. 102, 154174(2022).
  17. Li, W., et al. Metabolic activation of furan moiety makes diosbulbin B hepatotoxic. Arch Toxicol. 90 (4), 863-872 (2016).
  18. Wang, K., Li, W., Chen, J., Peng, Y., Zheng, J. Detection of cysteine- and lysine-based protein adductions by reactive metabolites of 2,5-dimethylfuran. Anal Chim Acta. 896, 93-101 (2015).
  19. Zhang, Z., et al. Sensitive bromine-based screening of potential toxic furanoids in Dioscorea bulbifera L. J Chromatogr B. 1057, 1-14 (2017).
  20. National Pharmacopoeia Committee. National standard for processing of traditional Chinese medicines. , People’s Medical Publishing House. , Beijing. (1988).
  21. Wang, K., et al. Chemical identity of interaction of protein with reactive metabolite of diosbulbin B. in vitro and in vivo. Toxins. 9 (8), 249(2017).
  22. Wang, K., et al. Selective and sensitive platform for function-based screening of potentially harmful furans. Anal Chem. 86 (21), 10755-10762 (2014).
  23. Sawada, Y., et al. Widely targeted metabolomics based on large-scale MS/MS data for elucidating metabolite accumulation patterns in plants. Plant Cell Physiol. 50 (1), 37-47 (2009).
  24. Zhang, Q., et al. Formation of furan from linoleic acid thermal oxidation: (E,E)-2,4-decadienal as a critical intermediate product. J Agric Food Chem. 72 (8), 4384-4392 (2024).
  25. Fan, X. Furan formation from fatty acids as a result of storage, gamma irradiation, UV-C and heat treatments. Food Chem. 175, 439-444 (2015).
  26. Zhang, S., et al. 8-Epidiosbulbin E acetate triggers apoptosis via metabolic activation-mediated crosstalk between mitochondria and endoplasmic reticulum. Ecotoxicol Environ Saf. 302, 118614(2025).
  27. Li, W., et al. A metabolic activation-based chemoproteomic platform to profile adducted proteins derived from furan-containing compounds. ACS Chem Biol. 17 (4), 873-882 (2022).

Access restricted. Please log in or start a trial to view this content.

Reprints and Permissions

Request permission to reuse the text or figures of this JoVE article

Request Permission

Tags

Hepatotoxicity ScreeningUHPLC HRMSReactive MetabolitesLiver MicrosomesP450 MetabolismGlutathione TrappingFuranoditerpenoids
Video Coming Soon

Related Articles