Untargeted metabolic analysis using LC-Q-TOF-MS and toxicity assessment of Eryngium foetidum in zebrafish embryos

INTRODUCTION

For centuries, ancient cultures have used bark, seeds, fruits, and other parts of plants to treat diseases and common illnesses (1–4). It is noteworthy that traditional medicine has gradually changed its empiric basis to a complex knowledge based upon the structure-activity of the molecules isolated from medicinal plants (5, 6). Currently, traditional medicine needs to be further assessed for non-clinical research based on pharmacology and toxicology principles by performing in vitro and in vivo studies. In addition, it is also necessary to identify new chemical entities (NCEs) as well as determine their pharmacological or toxicological properties (7, 8).

The chemical diversity of secondary metabolites in the plant kingdom is boundless (9). Chemical analysis and elucidation of the actions attributed to metabolites were previously performed by conventional analytical methods such as UV-Vis or paper chromatography for decades (10, 11). Despite these resources, sophisticated metabolomic techniques such as NMR, GC-MS, or LC-MS were employed to correlate the metabolic composition of plants with the human diet, biological responses, pharmacological treatments, and toxic effects of single molecules (12, 13), as well as the measurement of synergistic mechanisms acting on multiple targets and signaling pathways (9, 14–16).

Eryngium foetidum L. (Apiaceae), commonly named “culantro Mexicano” or “cilantro cimarron", occurs predominantly in tropical and subtropical regions around the world (17). This plant is an aromatic herb with particular features, such as fleshy, branched roots, a straight and cylindrical stem, and oblanceolate leaves with toothed margins. It grows in a basal rosette pattern. The whole plant is glabrous and strongly scented, measuring approximately 45 cm in height (18, 19). In traditional medicine, including that practiced in Mexico, E. foetidum has been widely used for gut disorders, hypertension, asthma, some types of cancer, heart disease, vision problems, inflammation, convulsive neurological disorders, and microbial infections, as well as being used in nutraceutical products (20–24).

The objective of this investigation was to examine the toxic effects of organic extracts derived from E. foetidum utilizing zebrafish (Danio rerio) embryos as a model and determine the metabolic profile of the most toxic extract employing a non-target screening workflow involving liquid chromatography coupled to mass spectrometry (LC-MS).

EXPERIMENTAL

Chemicals

Methanol, acetonitrile, water, and formic acid (LC-MS grade) were purchased from Tedia Company, Inc. (USA); n-hexane, methylene chloride, and methanol for the maceration process (technical grade) and ammonium hydroxide (LC-MS grade) were purchased from Sigma-Aldrich (USA).

Plant material and extraction

E. foetidum was collected in July 2018 in Cunducán, Tabasco (18º05’10.5” N 93º15’29.6 W) and identified by Dr. Nelly Jiménez-Pérez. The voucher specimen (No. 35776) was deposited at the herbarium of the Academic Division of Biological Sciences of the Universidad Juárez Autónoma de Tabasco, Mexico.

Dried and ground aerial parts of E. foetidum (101.90 g) were subjected to maceration with n-hexane (500 mL) three times for 72 hours at room temperature. After filtration, the extract was concentrated in vacuo at 40 °C and 80 rpm to obtain n-hexane extract (HEEf). The process was repeated with the same plant material to obtain dichloromethane extract (DEEf) and methanol extract (MEEf) using dichloromethane and methanol, resp. (25).

Zebrafish and zebrafish embryos

The wild-type zebrafish (Danio rerio) were purchased from a local fish farm. All experiments were reviewed and approved by our Institutional Commission on Research Ethics (CIEL-UJAT; protocol number 0884) and conducted in accordance with the Mexican Federal Regulations for Animal Experimentation and Care (SAGARPA).

Adult fish were separated into male and female groups and acclimatized to a continuous room temperature of 26 °C for at least 7 days and a 14/10 h light/dark cycle. Water was buffered with a pH ranging from 7–8, a hardness of 30–35 mg L^–1 CaCO₃ supply, and bubbled with oxygen saturation. The fish were fed Artemia salina (Artemia Biogrow^®) twice daily. Under these conditions, the night before exposure, adult fish were placed in 6-L breeding tanks in a 2:1 ratio (female-male) at 28 °C to obtain zebrafish embryos. Spawning was induced by light period initiation. The fertilized embryos obtained were transferred into a clean Petri dish and washed with bottled water for debris removal. Then, embryos were observed under an inverted microscope (Primo Vert, Carl Zeiss Iberia, S.L., Spain) to remove any unfertilized or dead embryos. After 2 hours post-fertilization (hpf), embryos were used for all toxicity tests (26, 27).

Fish embryo acute toxicity (FET) test. – The test was performed following the recommendations described in the Organization for Economic Cooperation and Development (OECD) test guidelines (28, 29). Briefly, the organic extracts were dissolved in DMSO (stock solution, 5 mg mL^–1). Then, a dilution process was carried out to obtain experimental dilutions: HEEf were 33, 3.3, 0.33, 0.033, and 0.0033 μg mL^–1, whereas dilutions of DEEf and MEEf were 50, 5, 0.5, 0.05 and 0.005 μg mL^–1. The vehicle used for each concentration of the extracts was DMSO 0.1 % in potable water. Twelve fertilized zebrafish embryos were incubated for each organic extract dilution in 24-well culture plates at 28 ± 0.5 °C in a final volume of 2 mL. The survival and morphology of the embryos were examined at 6, 12, 24, 48, 72, and 96 hours after exposure to organic extracts under an inverted microscope. Embryo mortality (coagulation) and any anatomical malformation or changes of the caudal and dorsal regions of the head were recorded each day throughout the exposure period (30).

LC-ESI-Q-TOF-MS conditions

Metabolic analysis (fingerprint) was performed using a liquid chromatography system (1290 Infinity II, Agilent Technologies, USA), equipped with a quaternary pump, coupled to an electrospray ionization with quadrupole time-of-flight (Q-TOF) mass spectrometer Agilent 6545 (Agilent Technologies). Based upon earlier reports, the HEEf was dissolved in acetonitrile (1 mg mL^–1), diluted 1:100, and filtered by a 0.45-μm nylon filter (31, 32). Chromatographic separation was performed utilizing a Zorbax Eclipse Plus C18 column RRHD (2.1 × 50 mm, 1.8 μm) with a mobile phase A (0.1 % formic acid-water) and mobile phase B (0.1 % formic acid-acetonitrile) in positive ion mode, or mobile phase A (0.1 % ammonium hydroxide-water) and mobile phase B (0.1 % ammonium hydroxide-acetonitrile) in negative ion mode. A gradient elution method (0:100 V/V, 0–1 min; 25:75 V/V, 1–5 min; 50:50 V/V, 5–10 min; 75:25 V/V, 10–20 min and 100:0 V/V, 20–25 min) was conducted for chemical separation. Further, a 5-μL injection and a flow rate of 0.2 mL min^–1 were used for this analysis. MS/MS experiments were performed to annotate metabolic profiling as well as identify fragmentation patterns by acquisition mode targeted MS2, and the fixed collision energy was 20 eV. High resolution was achieved with a reference solution added to each electrospray ionization (ESI) step: 121.0509 m/z (purine, C₅H₅N₄) and 922.0098 m/z (hexakis(1H,1H,3H-tetrafluoropropoxy)phosphazine, C₁₈H₁₉F₂₄O₆N₃P₃). The instrument parameters were as follows: capillary voltage 3500 V, nozzle voltage 1000 V, nebulizer pressure 241,316.5 Pa, drying gas flow rate 8 L min^–1, gas temperature 325 °C, fragmentor 175 V.

Data analysis

Data were recorded as the mean of survived zebrafish embryos from a basal count of 12 zebrafish embryos per group after exposure to organic extract concentrations. The median lethal concentration (LC₅₀) was calculated by probit analysis after a 96-hour exposure using the USDA-ARS PriProbit analysis program (version 1.63). Survival plots were constructed by the Prism 7 software (GraphPad. Inc., USA) (33).

Data acquisition and analysis of the experiments in LC-ESI-Q-TOF-MS were performed by Agilent MassHunter Workstation software (version B.07.00). The ions were annotated using the Molecular Feature Extraction (MFE) algorithm and Molecular Formula Generation software (Agilent Technologies) to compare with the METLIN Metabolite Database (; , La Jolla, CA, USA). Finally, MS fragmentations at each retention time were reviewed using the fragmentation tool of the ChemDraw Pro 12.0 software (Supplementary data) (31).

RESULTS AND DISCUSSION

Organic extract yields and fish embryo acute toxicity

The maceration process obtained a yield of 0.11, 0.10, and 0.78 % for HEEf, DEEf, and MEEf, resp. Fig. 1a shows the percentage of fish embryos that survived after exposure to HEEf. At 24 h, a concentration- and time-dependent survival was observed, whereas a minimal survival % was observed after 96 h (HEEf 33 μg mL^–1). The calculated LC₅₀ was 2.63 μg mL^–1 (95 % CI 0.58–28.5 μg mL^–1). Interestingly, a high survival rate was noted when zebrafish embryos were exposed to DEEf (Fig. 1b). Treatment with MEEf did not produce any mortality in the zebrafish embryo population exposed to this extract.

Fig. 1. Effect of dilutions of: a) HEEf; b) DEEf and c) MEEf, on survival of zebrafish embryos. Data plotted are means calculated from twelve replicates.

It is noteworthy that embryos exposed to the vehicle (DMSO 0.1 %), DEEf, or MEEf presented normal embryonic development. Figs. 2a and 2b illustrate representative micrographs of the zebrafish zygote and normal developing zebrafish embryo; Figs. 2c–g shows the phenotypic defects induced by HEEf exposure, such as somite formation and head malformation, yolk sac edema, and caudal region; and Fig. 2h shows a coagulated embryo (dead state). Fish embryos showed morphological malformations when exposed to HEEf 33 μg mL^–1 after 72 h.

Fig. 2. Representative micrographs of: a) zygote in vehicle; b) embryonic development without malformations; c) somite formation; d) caudal fin hypertrophy malformation; e) head malformation; f) yolk sac edema malformation; g) caudal region malformation; h) coagulated embryo (dead).

Metabolic profiling by LC-ESI-QTOF-MS

As HEEf was the most toxic extract in the zebrafish embryo preparations, it was of interest to determine its chemical composition by LC-Q-TOF-MS via the application of gradient elution. The total compound chromatograms (TCC) of the total HEEf content and acetonitrile in a positive and negative ion mode were then obtained (Fig. 3).

Fig. 3. The total compound chromatograms (TCC) of the following samples: a) acetonitrile and b) n-hexane extract of E. foetidum in positive ion mode; c) acetonitrile and d) n-hexane extract of E. foetidum in negative ion mode.

Thirteen compounds from the base peak chromatograms were annotated. The peak number was assigned by increasing order regarding retention time. The identification, molecular formula, ion type, experimental m/z, and MS/MS fragments of 9 compounds shown in positive ion mode are listed in Table I: 1 – absindiol, 2 – 5,8,12-trihydroxy-9-octadecenoic acid, 3 – diglycidyl resorcinol ether, 4 – eplerenone, 5 – diisopropyl adipate, 6 – anethole, 7 – 3''-hydroxy-geranylhydroquinone, 8 – 2,4,8-eicosatrienoic acid isobutylamide, and 9 – 25-azacholesterol. Their chemical structures are illustrated in Fig. 4. Ion fragments in positive ion mode were included in Supplementary data 1.

Table I. Compounds annotated by LC-ESI-QTOF-MS in HEEf by positive ion mode

AUC – area under curve; RIA – relative ion abundance.

Fig. 4. Chemical structures of compounds annotated in HEEf by positive ion mode.

Four compounds were also annotated by negative ion mode and are listed in Table II: 10 – 5,8,12-trihydroxy-9-octadecenoic acid, 11 – lauryl sulfate 12 – 6-octadecenyl valerate and 13 – (R)-3,4-dihydro-2-(4,8,12-trimethyl-3,7,11-tridecatrienyl)-2H-1-benzopyran-6-ol. Their chemical structures are presented in Fig. 5. Ion fragments in negative ion mode are included in Supplementary data 2.

Table II. Compounds annotated by LC/QTOF-MS in HEEf by negative ion mode

AUC – area under curve; RIA – relative ion abundance.

Fig. 5. Chemical structures of compounds annotated in HEEf by negative ion mode.

Our study showed that HEEf had a higher yield compared with dichloromethane and methanol extracts, indicating that aerial parts contain more non-polar metabolites than hydrophilic molecules (34). In fact, data suggest that the medicinal properties of the plant may be generated predominantly by these types of molecules compared to others with different chemical properties. However, this suggestion remains to be reconsidered and elaborated.

In accordance with OEDC, an acute zebrafish embryotoxicity test is a required component in the environmental hazard assessment of industrial chemicals, plant products, biocides, pharmaceuticals, and feedstuffs (29, 35, 36). In the FET test performed here, exposure of zebrafish embryos to HEEf decreased survival in a concentration- and time-dependent manner. Unfortunately, it was not possible to determine the non-observed adverse effect level (NOAEL), a measure that indicates the greatest concentration or amount of a substance that initiates no detectable adverse alterations in morphology, growth, development, or lifespan of an organism, as the lower concentration tested (3.3 × 10^–3 μg mL^–1) still produced a 10 % mortality in zebrafish embryos. Interestingly, the embryos exposed to HEEf at 33 µg mL^–1 (Fig. 1a) showed a non-concentration- and time-dependent behavior, probably due to the biological variability of the embryos, although they are from the same progenitors; genetic variability and the high fecundity of the fishes can influence their phenotype and therefore the biological responses to xenobiotics and the environment (37).

The compounds annotated by the LC-ESI-QTOF-MS analysis that might be associated with mortality and altered morphogenesis of some organs are the diglycidyl resorcinol ether, diisopropyl adipate, and lauryl sulfate shown in Tables I and II. In this context, diglycidyl resorcinol ether is an additive that demonstrated in vitro genotoxic effects in mammalian cells, mutagenicity by the Ames test (38), and carcinogenic effects (39), and due to its higher relative abundance than the other compounds, it may be mainly responsible for the toxic effects observed in the embryos. In addition, diisopropyl adipate, an emollient and permeation enhancer ingredient for pharmaceuticals, has not been examined comprehensively regarding toxicity yet (40, 41). Further, lauryl sulfate, a surfactant used in pharmaceuticals, cosmetics, and foods, is a previously reported recognized human and environmental toxic agent (42–44).

Environmental pollutants interfere not only with plant metabolism by modifying enzymatic and genetic activity, cellular detoxification, and biological interaction with the environment, producing biotic stress and susceptibility (45–48), but these pollutants might also be part of a biomagnification chain that consequently results in toxic effects on human health in different ways (47, 49, 50). More investigations on this topic are recommended since it is not known whether all edible plants that consumers are exposed to may produce toxic responses.

Similar to this investigation, Castro and colleagues (51) reported that the toxicity of E. foetidum on zebrafish embryos and larvae was dependent upon the extraction solvent, the extract concentration used, and exposure duration. They found that E. foetidum polar extracts (aqueous, ethanol, and methanol) increased mortality rates, delayed hatching time, and produced changes in enzymatic activities. In addition, these organic extracts delayed the development of zebrafish embryos and elevated glutathione-S-transferase activity, free radical production, and oxidative damage. They concluded that concentrations above 0.625 μg mL^–1 might exert adverse effects in the early stages of zebrafish embryonic development. The above findings are in agreement with our data showing that MEEf adversely altered fish embryo development; however, our highest concentration of 33 μg mL^–1 was still much lower. More investigation is required to examine this issue.

Finally, complementary investigations were published on animal models. Janwitthayanuchit and colleagues (52) reported a 24-week chronic toxicity study where mice were fed with a rodent diet supplemented with ground freeze-dried E. foetidum leaves at 0.8, 1.6, or 3.2 %. Data demonstrated that the modified diet produced tubulonephrosis and chronic interstitial nephritis in animals fed with 1.6 or 3.2 % preparations. These toxic effects showed that the malformations observed in the zebrafish embryos can also induce damage in rodents and, probably, higher animals.

CONCLUSIONS

The current study provided a metabolic profile of E. foetidum by LC-ESI-QTOF-MS. The environmental pollutants found were diglycidyl resorcinol ether, diisopropyl adipate, and lauryl sulfate, possibly related to the observed anatomic malformations and elevated mortality rates noted in zebrafish embryos after exposure to HEEf. Therefore, further investigation needs to be performed on edible plant species to examine various exposure scenarios to xenobiotic pollutants, since these may represent a source of toxic substances that might exert adverse effects on human health.

Supplemental material for this article is available online. Additional data will be made available on request.

Abbreviations, acronyms, codes. – DEEf – dichloromethane extract of E. foetidum, FET – fish embryo acute toxicity, HEEf – n-hexane extract of E. foetidum, LC₅₀ – median lethal concentration, LC-ESI-QTOF-MS – liquid chromatography coupled with electrospray ionization-quadrupole time of flight mass spectrometry, MEEf – methanol extract of E. foetidum, NOAEL – non-observed adverse effect level, OECD – Organization for Economic Co-operation and Development (Paris, France), TCC – total compound chromatogram.

Acknowledgments. – R.V.C. thanks CONAHCyT for scholarship 775708 during his Master's studies.

Conflict of interest. – The authors have no conflicts of interest to declare.

Funding. – The current research did not receive any funds.

Authors contributions. – Conceptualization, S.R.M., I.S.L. and O.H.A.; experiments, R.V.C., O.A.P.M., A.Y.H. and L.C.R.; analysis, N.J.P., C.A.S., N.R.C. and O.H.A.; writing, original draft preparation, review and editing, S.R.M., R.V.C., O.A.P.M., L.C.R. and O.H.A. All authors have read and agreed to the published version of the manuscript.

REFERENCES

6. B. B. Petrovska, Historical review of medicinal plants' usage, Pharmacogn. Rev. 6(11) (2012) 1−5;

12. D. G. Cox, J. Oh, A. Keasling, K. L. Colson and M. T. Hamann, The utility of metabolomics in natural product and biomarker characterization, Biochim. Biophys. Acta 1840(12) (2014) 3460−3474;

14. G. S. de Jesus, D. Silva Trentin, T. F. Barros, A. M. T. Ferreira, B. C. de Barros, P. de Oliveira Figueiredo, F. Rodrigues Garcez, É. L. Dos Santos, A. C. Micheletti and N. C. Yoshida, Medicinal plant Miconia albicans synergizes with ampicillin and ciprofloxacin against multi-drug resistant Acinetobacter baumannii and Staphylococcus aureus, BMC Complement. Med. Ther. 23(1) (2023) Article ID 374 (18 pages);

17. B. K. Singh, Y. Ramakrishna and S. V. Ngachan, Spiny coriander (Eryngium foetidum L.): A commonly used, neglected spicing-culinary herb of Mizoram, India, Genet Resour. Crop Evol. 61(6) 1085−1090;

20. M. D. García, M. T. Sáenz, M. A. Gómez and M. A. Fernández, Topical antiinflammatory activity of phytosterols isolated from Eryngium foetidum on chronic and acute inflammation models, Phytother. Res. 13(1) (1999) 78−80;

31. S. Rodríguez-Morales, B. Ocampo-Medina, N. Romero-Ceronio, C. Alvarado-Sánchez, M. Á. Vilchis-Reyes, L. F. Roa de la Fuente, R. Ortiz-Andrade and O. Hernández-Abreu, Metabolic profiling of vasorelaxant extract from Malvaviscus arboreus by LC/QTOF-MS, Chem. Biodivers. 18(4) (2021) e2000820 (8 pages);

34. A. R. Abubakar and M. Haque, Preparation of medicinal plants: Basic extraction and fractionation procedures for experimental purposes, J. Pharm. Bioallied Sci. 12(1) (2020) 1−10;

44. H. Löffler and R. Happle, Profile of irritant patch testing with detergents: sodium lauryl sulfate, sodium laureth sulfate and alkyl polyglucoside, Contact Dermatitis 48(1) (2003) 26−32;

51. T. F. D. Castro, W. F. Carneiro, T. F. Reichel, S. Lacerda, M. R. F. Machado, K. K. C. de Souza, L. V. Resende and L. D. S. Murgas, The toxicological effects of Eryngium foetidum extracts on zebrafish embryos and larvae depend on the type of extract, dose, and exposure time, Toxicol. Res. 11(5) (2004) 891−899;