Phytochemical-Based Evidence of the Health Benefits of Bidens Pilosa Extracts and Cytotoxicity

Bidens pilosa L. is traditionally used as a flavouring agent in foods, in the treatment of diseases, in agriculture as a biopesticide and herbicide, and in the phytoremediation of soils contaminated with heavy metals. The vast range of uses of B. pilosa for a variety of purposes is questionable, hence motivating the objectives of this study, which are to assess the cytotoxicity, health benefits, and/or risks of B. pilosa using chemical-based evidence. A real-time cell analysis (xCELLigence system), ultra-high-performance liquid chromatography coupled to a quadrupole-time-of-flight mass spectrometer, high-performance liquid chromatography with a diode array detector, and chemical-based spectrophotometric methods were adopted in the study. High concentrations of the ethanol extracts exhibited cytotoxic activity on HepG2 (cancerous), and Vero (non-cancerous) cell lines, whereas the water extracts promoted cell proliferation at selected concentrations. The chemical profiling enabled the separation as well as characterisation of 137 phytochemicals. These were mainly phenolic acids, flavonoids, fatty acids, coumarins, and furanocoumarins. There was no toxic compound identified. The ethanol extracts are generally more potent and exhibit stronger antioxidant activity and cytotoxicity, probably due to the presence of more flavonoids and phenolic acids, validating the uses of B. pilosa and its relevance as a source of functional phytochemicals.

, malaria [9], tumours [10], microbial infections [11], rheumatoid arthritis, and inflammation [12]. Heavy metals such as arsenic and cadmium have been reported to bioaccumulate in B. pilosa, which suggests that the plant has the potential for the phytoremediation of heavy metals [4]. In the study of Salazar et al. (2021) [13], B. pilosa was reported to be effective in the removal of lead from contaminated soils [13]. The plant has also been used to control weeds through its allelopathic properties and to inhibit plant pathogens [6]. It has been reported that B. pilosa extracts improve crop yield by controlling insect pests on legume crops and significantly reducing stored grain pests [14,15]. The essential oil produced by B. pilosa has been found to possess antimicrobial activities and free radical scavenging potential [16].
According to reports, B. pilosa is an effective insecticide, nematicide, and herbicide. This was confirmed by Ahmed et al. (2021) [14] that the methanolic extract of the plant has significant insecticidal activity against the grain beetle (Oryzaephilus surinamensis L.), bean weevil (Acanthoscelides obtectus) and rice weevil (Sitophilus oryzae L.) [14]. The no-choice test conducted by Mersie et al. (2019) [17] on Asteraceae weeds revealed that Parthenium beetle (also Mexican beetle) (Zygogramma bicolorata) avoids laying eggs on B. pilosa as much as they could and also does not feed on it [17]. In a two-day multiple-choice test carried out by Bajracharya et al. (2020) [18], Z. bicolorata avoided a few Asteraceae weeds including B. pilosa except Parthenium hysterophorus L. The extracts of B. pilosa significantly reduced the numbers of the Japanese pine sawyer (Monochamus alternatus) as well as the pine-wood nematode (Bursaphelenchus xylophilus) [19]. Due to the allelopathic characteristics of B. pilosa [6], the water extracts of the plant integrated with water irrigation have been found to effectively suppress weeds in a direct-seeded rice production system, which increases rice yield [20].
In ethnoveterinary medicine, B. pilosa has been reported to be effective in the treatment of several ailments in livestock. For instance, in Trinidad and British Columbia, Canada, B. pilosa is used to improve nutrition in horses [21]. According to Chang et al. (2016) [22], B. pilosa benefits chickens by improving growth through the modulation of gut bacteria, decreasing protozoan infection, and reducing caecal damage. The severity of liver fibrosis and injury in rats was reported to be reduced by B. pilosa when administrated for 6 weeks [23], suggesting that it could be used to treat liver diseases. Fructose-induced hypertension and hypertriglyceridemia in rats were reversed by aqueous and methylene chloride extracts of B. pilosa, indicating that the plant extracts had hypotensive effects unrelated to insulin level modulation [24].
The bioactivities of B. pilosa are associated with its diverse phytochemicals, some of which have been identified as porphyrins, tannins, aliphatics, terpenoids, alkaloids, phenylpropanoids, cardiac glycosides, flavonoids, and aromatics [7]. There is a wide range of bioactive compounds present in the genus Bidens, however, the specific phenolic compounds present have not been fully elucidated. Generally, phytochemicals or plant secondary metabolites are used in the manufacture of food additives, pharmaceutical products, cosmetics, flavours, colognes, biopesticides and a variety of other products. However, most medicinal plants, including B. pilosa, are still not commercially grown, making it difficult to obtain plants which are the major natural source of secondary metabolites [25]. Although secondary metabolites are also produced by bacteria, fungi, and several marine organisms such as corals, sponges, and tunicates [26], plants still remain the most reliable source.
Given the vast range of uses of B. pilosa around the world, including food, traditional medicine, biopesticides, and herbicides, the health risks associated with the consumption of the plant are questionable. This leads to the aims of this study: to determine the cytotoxicity of B. pilosa and to use chemical-based evidence to determine the health benefits and/or risks as well as correlate the identified metabolites with the biological activities. To answer these questions, the phytochemicals in B. pilosa extracts were comprehensively characterized using ultra-high-performance liquid chromatography coupled to a quadrupole-time-of-flight mass spectrometer (UHPLC-ESI-Q-TOF-MS). The antioxidant content and activities were evaluated using spectrophotometric methods. Cytotoxicity of the extract was assessed on cancerous and non-cancerous cell lines utilizing a real-time cell analysis system (xCELLigence).

Plant Materials
The aerial parts of B. pilosa were collected from their natural habitat in Lepelle-Nkumpi Local Municipality, Limpopo (24° 19' 17" S, 29° 11' 1" E) Province, South Africa. Permission to collect the plant sample was granted by the landowner and Limpopo province. The samples of Bidens pilosa were identified by Professor Stefan Siebert, a Professor of Geobotany and the curator of the AP Goossens Herbarium, Faculty of Natural and Agricultural Sciences, North-West University. The voucher specimen of B. pilosa was deposited at the A.P. Goossens Herbarium and assigned the accession number 15,452 and barcode PUC0015452, then stored in the BRAHMS database. The authors abide by the International Union for Conservation of Nature (IUCN) policy, as well as national and international guidelines and legislation, in the collection. The plant material was air-dried at room temperature before being pulverized. The moisture content of the pulverized samples was calculated using water mass to dry mass ratio of the pulverized samples. The initial mass (m) was determined before drying the samples, followed by the final mass (m0) after drying the samples in an oven at 105 o C until a constant mass was achieved. The water mass of the samples (m-m0), and the moisture content (MC) were calculated using the formula: Extraction of phytochemicals was carried out by mixing the pulverized samples in ethanol-water solution (7:3, v/v) on a mechanical shaker for 48 h. The plant-to-solvent ratio was 1:10 (m/v). The extracts were filtered through Whatman® filter paper, Grade 1, (diameter 150 mm), concentrated with a rotary evaporator and dried by lyophilisation. The yield was calculated. Samples were then reconstituted in distilled water before being used but were suspended in 2 mL of 50% methanol (Sigma-Aldrich, ≥ 99.90% purity) and 0.1% formic acid (Sigma-Aldrich, reagent grade, ≥ 95%) prior to the MS analysis.

Total Polyphenol Assessment
The total polyphenol content of the extracts was determined using the Folin-Ciocalteu reagent, with minor adjustments, as described by Daniels et al. (2015) [27]. The samples were prepared by serial dilution in the range of 10-100 mg/ mL, while the standard curve was prepared in the range of 10-500 mg/mL gallic acid (Sigma-Aldrich, ≥ 98.5% purity) in 10% ethanol. An aliquot of 25 µL of the sample was pipetted into the wells of a transparent 96-well microplate, then mixed with 125 µL Folin-Ciocalteu reagent (1:10 v/v in distilled water). After 5 min., 100 µL of 7.5% m/v anhydrous sodium carbonate (Na 2 CO 3 ) was added to the wells making the total volume to be 250 µL. The mixture was incubated for 2 h in the dark at room temperature after which absorbance was recorded at 765 nm using a Multiskan™ plate reader (Thermo Electron Corporation, USA). The assay was done in triplicate and the result was expressed as mg gallic acid-equivalents (GAE) per gram of extract's total polyphenol (mg GAE/g) based on the gallic acid standard curve [28].

Total Flavonol
The flavonol content was determined using the quercetin standard, as previously described [27], also using a spectrophotometric method. The standard linear calibration curve was obtained by preparing quercetin in 95% ethanol at concentrations ranging from 0, 5, 10, 20, 40, and 80 mg/L. In a 96-well microplate, 12.5 µL of the plant extracts was pipetted in a well, then mixed with 12.5 µL hydrochloric acid (0.1% HCl, v/v in 95% ethanol), and 225 µL hydrochloric acid (2% HCl, v/v in distilled water), making the total volume 250 µL. The mixture was then incubated for 30 min at room temperature. Absorbance was measured using a spectrophotometer (Multiskan™ plate reader, Thermo Electron Corporation, USA) in the dark at 25 o C at 360 nm. This was done in triplicate for all of the samples. The total flavonol was expressed as mg quercetin-equivalents per g dry mass of extracts (mg QE/g) using the linearity of the quercetin standard.

2,2-Azino-di-3-ethylbenzthiazoline Sulfonate (ABTS) Assay
The 2,2-Azino-di-3-ethylbenzthiazoline sulfonate (ABTS) assay was used to quantify the radical scavenging activity of the samples adopting the method of Re et al. (1999) [29]. The stock solution was prepared by stoichiometrically reacting 7 mM ABTS with 2.45 mM potassium-persulfate (K 2 S 2 O 8 ) at a ratio of 1:0.5 in the dark at room temperature for 24 h to produce ABTS •+ . Prior to the assay, the ABTS •+ solution was mixed with 96% ethanol at 30 o C until an absorbance of 0.70 (± 0.02) was obtained at 734 nm [30]. Trolox (6-hydrox-2,5,7,8-tetramethylchroman-2-carboxylic acid) was used as the standard, and the standard curve was plotted in the concentration range (0-500 µM). Diluted ABTS •+ solution (A 734nm = 0.700 ± 0.02) of 300 µL was added to 25 µL of the plant extracts and allowed to incubate in the dark at 30 o C for 30 min before measuring absorbance at 734 nm. All tests were performed in triplicate at each concentration of the standard and samples. The results were expressed as µM trolox-equivalents per g dry mass (µM TE/g).

Ferric Reducing Antioxidant Power (FRAP) Assay
The ferric reducing antioxidant power (FRAP) assay was carried out using the method previously described by Daniels et al. (2015) [27]. FRAP reagent was prepared from a solution of 10 mL of 0.3 M acetate buffer (pH 3.6), 1 mL of 10 mM 2,4,6-tripyridyl-s-triazine (TPTZ) dissolved in 0.1 M HCl, and 1 mL of 20 mM iron (III) chloride rate of 650 L/h, and the rest of the MS settings optimised for best resolution and sensitivity. Data were acquired by scanning from m/z 150 to 1500 in resolution mode as well as in MSE mode. In MSE mode, two channels of MS data were acquired, one at low collision energy (4 V) and the second collision energy ramp at 40-100 V to obtain fragmentation data. Leucine enkephalin was used as a lock mass (reference mass) for accurate mass determination, and the instrument was calibrated with sodium formate. A volume of 2 µL was injected. The mobile phase consisted of water solvent A and acetonitrile (solvent B). Both solvent lines contained 0.1% formic acid. The gradient started at 100% solvent A for 1 min, reaching 28% B over 22 min in a linear way, then 40% B over 50 s and a wash step of 1.5 min at 100% B, followed by re-equilibration to initial conditions for 4 min. The flow rate was set to 0.25 mL/min, and the column temperature was held constant at 55 °C. Separation was achieved on a Waters BEH C18, 2.1 × 100 mm, 1.7 μm column. Based on the following three criteria, compounds were identified and assigned tentative names of identity. (1) Accurate mass match: the masses were automatically matched using XCMS, which was linked to Metlin (http://metlin. scripps.edu/metabo_search.php), MassBank, NIST, and other libraries. All compounds whose accurate mass error (AME) is > 5 ppm was considered unidentified [32]. (2) Mass fragmentation patterns: if available, the mass fragmentation patterns of the compounds are searched in the aforementioned databases. A few phenolic compound standards were spiked under similar LC-MS conditions and fragmentation patterns were compared to identify a given compound based on the retention time, mass fragmentation, and ionisation modes i.e. (by corresponding ions of protonated and deprotonated moieties). Since many compounds could be identified by UPLC-ESI-QTOF-MS, it was not possible to obtain all standards. Therefore, other similar compounds' MS and MS 2 fragment ions in literature and databases were used for the annotation of the compounds. (3) The number of carbon atoms in the peak: the carbon atoms were calculated if isotope abundances were available. The predicted number of carbon atoms in the putatively identified compound was used to reduce false annotations.

Analytical Protocols and Validation Procedure
The developed UHPLC-MS method was validated with a linearity curve, the limit of detection (LOD), the limit of quantification (LOQ), precision, and recovery according to the International Conference on Harmonization (ICH) recommendations [33]. Different concentrations of the standard mixture (3.906, 7.813, 15.63, 31.25, 62.5.0, 125.0, and 250.0 mg/L), which were approximated as 3.9, 7.8, 15.6, 31.3, 62.5, 125.0, and 250.0 mg/L, respectively, were hexahydrate (FeCl 3 ·6H 2 O) dissolved in distilled water. In a 96-well microtiter plate, an aliquot of 10 µL of varying concentrations of plant extracts and standard (0-1000 µM) was pipetted and mixed with 300 µL FRAP reagent. The mixture was incubated at 37 o C for 30 min and absorbance was read at 593 nm. L-Ascorbic Acid was used as the standard, and the results were expressed as µM ascorbic acid-equivalents per g dry mass (µM AAE/g).

Quantification of Phytochemicals and Profiling
The preliminary profiling of B. pilosa extracts was carried out using high-performance liquid chromatography with diode-array detection (HPLC-DAD), an Agilent 1200 Series HPLC device (Agilent, Santa Clara, USA), and a photodiode array detector. The phytochemicals of the plant extracts were identified and quantified using a Waters Synapt G2 qTOF mass spectrometer (Waters, Milford, MA, USA). Before injection, all samples were filtered through a 0.45 μm filter (Nylon Membranes, Supelco, Bellefonte, USA).

Method for HPLC-DAD Analysis
An HPLC-DAD with a reverse-phase RP-18 HS C18 discovery column (150 mm x 4.6 mm i.d., 5 μm, column temperature: room temperature) was used as previously described with minor modifications for preliminary profiling by Zeb (2015) [31]. The mobile phase consisted of water containing 0.1% trifluoroacetic acid (A), and methanol containing 0.1% trifluoroacetic acid (B), and then the flow rate was set at 1 mL/min. The elution program was performed with a gradient starting at 0 min, 100% of A, moving to 100% of B after 25 min. The detected wavelengths were recorded at three different UV channels; 280, 320, and 360 nm. The injection volume was 20 µL. The HPLC-DAD system was allowed to warm up before each run, and the baseline was monitored until it was stable before sample analysis. Peak identification was accomplished by comparing the retention duration and UV absorption spectra obtained with the standards.

Method for Liquid Chromatography-mass Spectrometry (LCMS) Analysis
A Waters Synapt G2 Quadrupole time-of-flight (QTOF) mass spectrometer (MS) connected to a Waters Acquity ultra-performance liquid chromatograph (UPLC) (Waters, Milford, MA, USA) was used for high-resolution UPLC-MS analysis. Electrospray ionization was carried out in both negative and positive modes with a cone voltage of 15 V, a desolvation temperature of 275 °C, a desolvation gas flow respectively). The cells were grown in Dulbecco's Modified Eagle's Medium (DMEM) (Sigma, Darmstadt) supplemented with 10% fetal bovine serum (FBS) (Thermo Scientific, USA) in a humidified incubator with 5% CO 2 at 37 °C. Cells were handled in a sterile laminar flow hood, which was cleaned with 70% ethanol [36].

xCELLigence Cell Proliferation Assay
The impedance technology of the xCELLigence system (Real-Time Cell Analyser-RTCA; ACEA Biosciences Inc., San Diego, USA) with RTCA software (version 1.2.1) was used to monitor the cytotoxic effects of B. pilosa on two adherent cell lines (HepG2 and Vero) for 24 h. The system provides proliferation profiles as cells interact with the microelectrodes, allowing the IC 50 /EC 50 values to be calculated. Cell growth is measured in real-time and converted into cell index (CI) values. Others have also documented using this method [37][38][39]. The HepG2 and Vero cells were seeded at a density of 5 × 10 4 cells/mL in the 96-well goldplated E-plate and left to adhere for a period of 24 h. The cells were exposed to a concentration range of 15.6, 31.25, 62.5, 125, 250, 500, and 1000 µg/mL of reconstituted plant material after extraction with two solvents (See Sect. 2.1) in triplicate. The unexposed (control) cells received DMEM culture medium only.

Statistical Analyses
The significant differences in the data obtained in antioxidant content and activity were evaluated by two-way analysis of variance (ANOVA), (p < 0.05), using GraphPad Prism® 8.0.1. The results were presented as means ± standard deviation. In the cytotoxicity, the cell index data obtained after the exposure was normalised with RTCA data analysis software. Normalisation was done at a specific time point (start of the test compound treatment), which was then set as 1.0 by the software and all the other values are represented as a proportion of this value. The dose-response curve and the EC 50 values were obtained using the xCELLigence system's RTCA-integrated software. The EC 50 value was calculated in the RTCA software under the analysis profile by selecting the exposure period. The data are presented as mean [µg/ mL] ± standard deviation (n = 5).

Phytochemical Yields and Moisture Content
The extract yields from the plant correlated with the solvent type, with the ethanol-water (7:3, v/v) extracting injected for quantification. The linearity of the calibration curve was checked by plotting the peak areas against the series of standard solution concentrations (mg/L) and determining the correlation coefficient using a linear regression model. In all cases, the system was linear (r > 0.99). LOD and LOQ were calculated by the parameters of the analytical curves (standard deviation of the response and slope). The standard deviation of the y-intercepts of regression lines was used as the standard deviation of the blank. The LODs and LOQs were estimated as 3.3 and 10 times the standard deviation of the blank/slope ratio of the calibration curve, respectively.
The repeatability of B. pilosa samples was assessed by using intraday and interday variations, and the relative standard deviation (RSD) was taken as a measure of precision. The intraday and interday variations were analysed using six replicates during a single day and by duplicating the experiments on three consecutive days. The percentage relative standard deviations (%RSD) of the peak areas (UV detection) and retention times were determined for each peak detected. The intraday repeatability (expressed as %RSDs) of the retention times was in the range of 0.14 to 3.14%, whereas the interday repeatability was from 1.01 to 2.90%. The intraday repeatability (expressed as %RSDs) of the total peak area was 0.32-0.70%, whereas the interday repeatability was 1.01-1.13%.
The recovery test was done by adding the known amounts of standards at low (50% of the known amounts), medium (same as the known amounts) and high (150% of the known amounts) levels into samples to evaluate the accuracy of the method. This was applied in triplicate.

Maintenance of Cells
Cell lines from the human liver, HepG2 (cancerous), and African monkey kidney, Vero (non-cancerous), were used to measure the cytotoxic potential of B. pilosa extracts (water and 70% ethanol). Since different cell lines originate from different organ tissues and these organs have different sensitives towards compounds, the measured cytotoxic responses may be dependent on the cell type used [34]. The HepG2 liver cell line was chosen because one of the main functions of the liver is detoxification. These cells are immortalised and universally used as a model to assess toxic potency. Although the origin of the Vero, non-cancerous cells are different (kidney), both the liver and kidneys are involved in the excretion of substances from the body [35]. Both cell lines were obtained from the American Type Culture Collection (Manassas, VA, USA) (HB-8065 and CCL-81 while the ethanol activities are 146.02 µmol AAE/g and 144.81 µmol TE/g.

Qualitative Analysis of Phytochemicals Using UHPLC-MS and HPLC-DAD
Due to the overlapping of numerous compounds, particularly in the phenolic fraction, HPLC-DAD produces complex profiles of the compound fraction. Nevertheless, UPLC-MS could exclude interferences using MS modes and product ion scan thus verifying the structure of the compound. The examination of base peak chromatograms of B. pilosa extracts shown in Fig. 2 enabled phytochemicals to be screened. In this study, UHPLC-ESI-MS enabled the detection of a total of 137 peaks (Table 1). [M-H-CO 2 ] − were formed, enabling tentative identification (dihydroxy benzoic acid) using the NIST database and literature [40]. Peak 22 has [M-H] − at m/z 167 and produced a significantly more than the water. The ethanol-water extract yield was 14.55 ± 0.86%, whereas the water yield was 8.49 ± 0.28%. The yields from both solvents are notably different, which could be attributable to the characteristics of these solvents. The residual moisture of the pulverized plant sample was determined after drying. The moisture content of the pulverised samples was evaluated to be 5.14% dry matter.

Antioxidant Content and Activity
The antioxidant content concentration (total polyphenol and total flavonol) varies depending on the extraction (water extract and 70% ethanol extract). Similarly, there was a strong relationship between the phytochemicals and antioxidant activity in the test samples. Figure 1 show that the concentration of the phytochemicals in the ethanol samples is higher compared to the water samples, and this has an obvious ripple effect on the antioxidant activities. It indicates that the ethanol extracts were more potent than the water extracts. The concentrations of the total polyphenols ranged from 15.66 to 20.19 mg GAE/g in the water extract and from 30.86 to 32.52 mg GAE/g for the ethanol extract, with mean values of 18.65 mg GAE/g and 31.56 mg GAE/g respectively. The mean value of the flavonol content of the water extract is 2.92 mg QE/g while the ethanol extract is 18.31 mg QE/g. The antioxidant activities of the ethanol extracts are higher, indicating a higher potency, and they correlate with the phytochemical concentrations. The antioxidant activities of water extracts of ABTS and FRAP are 60.02 µmol AAE/g and 116.29 µmol TE/g, respectively,       form (34)  In general, those chlorogenic acids (CGAs) with a greater number of free equatorial hydroxyl groups in quinic acid are more hydrophilic than those with a greater number of free axial hydroxyl groups. On the other hand, the ease of removal of the caffeoyl residue during fragmentation is 1 > 5 > 3 > 4 [44] since hydroxyl groups in the quinic acid are axial in positions 1 and 3, and equatorial in positions 4 and 5 [44]. For di-O-CQAs, the more the additional caffeoyl groups are attached to free equatorial hydroxyl groups (owing to steric interactions), the stronger the retention [45]. That is, the loss of the caffeoyl group (C) is likely to be in the order of 1-C > 5-C > 4-C > 3-C [44,45]. This enabled peaks 77 and 78, 79, 87, and 89 to be identified as 1,5 di-O-CQA, 3,5 di-O-CQA, 3,4 di-O-CQA, and 4,5 di-O-CQA respectively since the elution order is 1,3-diCQA <<< 1,4-diCQA < < 3,4-diCQA < 1,5-diCQA < 3,5-diCQA < < 4,5-diCQA [44]. Some      [49]. CFA was initially detected in B. pilosa [48]. Derivatives of CFAs were shown at 23, 39, 40, and 41 and individual adducts of 2 CFAs were shown at m/z 623 (peak 38,42) in agreement with previous work [49]. Chlorogenic acids (CGAs) glycosides showed a typical fragmentation pattern and UV absorbance. All CGAs glycosides were more hydrophilic than their isobaric chlorogenic acids. For instance, mono and diacyl CGAs glycosides were eluted before di-and tri-acyl CGAs, respectively. These glycosides showed fragments that originated from the cinnamoyl glycoside part, e.g., m/z 341 and 323 for caffeoyl glucoside; m/z 355 and 337 for

Flavan-3-ol
Peaks 113 and 114 were identified as procyanidin A-type dimer having a sequence of two stereoisomers: catechin and epicatechin flavan-3-ols, respectively. Thus, peak 113 was identified to be catechin-catechin while peak 114 was identified as (epi)catechin-(epi)catechin because the order of hydrophobicity was (

Flavanones
Five chromatographic peaks (45,52,62,65, and 71) from B. pilosa ethanol and water extracts were considered with flavanone skeletons. Flavanones usually peak 129 and its fragmentation pattern, which is similar to previous work [54]. Its hydroxylated form was represented at peaks 121 and 128, while the acetylated derivative conjugate was formed at peak 130. Linoleic acid was found in both extracts of B. pilosa analysed in this study. It is an essential fatty acid that possesses anti-inflammatory properties by providing the building blocks for prostaglandins. Several other hydroxylated fatty acids were also detected as major peaks, and they showed extra loss of water molecules [54]. They are known to possess antimicrobial and cytotoxic activities [55]. The polyunsaturated fatty acids included oxo-dihydroxydecanoic acid at peaks of 117 and 124, trihydroxy-fatty acid, and trihydroxyoctadecadienoic acid (peaks of 122 and 126), all found in the water extract. Trihydroxyoctadecanoic acid showed main MS/MS fragments at m/z 311 and 293 due to the subsequent loss of two water molecules and the main fragment at m/z 211 due to the C15\ C16 bond cleavage [54]. Additionally, dihydroxy derivatives of octadecanoic acid and dihydroxyoctadecanoic acid were shown at peak 127. Besides, LC-MS revealed several monohydroxy-fatty acid peaks, including hydroxyoctadecanoic acid (peak 131) and its dimeric form, at peaks 123 and 135. It also reveals its methyl dimer derivative, detected at peak 136. Lastly, the saturated fatty acid, amino hexadecanediol, was identified at peak 119, in agreement with previous work [54].

Other Compounds (Carboxylic Acids, Amino Acids or Alkaloids)
The fragmentation of the carboxylic acids occurred through the release of one or two water (18 Da)

Characterisation of Fatty Acids
Next to flavonoids and phenolic acids, fatty acids are the third most abundant class of compounds in B. pilosa ethanol and water extracts. There are fourteen known fatty acid peaks (117, 119, 121, 122, 124, 126, 127, 128, 129, 130,  131, 132, 135, and 136) and three unknown fatty acid peaks (133, 134, and 137). Polyunsaturated and hydroxylated fatty acids were the predominant forms. Peak 132 was identified as linoleic acid. This was confirmed by its dimeric form at profiling phenolic compounds from the samples analyzed. The amount has been expressed as caffeic acid, ferulic acid, coumaric acid, rutin catechin, epicatechin, and hesperidin equivalents, and the names of other phenolic compounds are used alternatively and reported in many papers.

Groups of Phytochemicals Identified in the Extracts of B. pilosa
A total of 137 metabolites were annotated in the positive and negative ESI modes of both solvent extracts (Figs. 4 and  5). The predominating classes were phenolic acids, flavonoids, fatty acids, coumarins and furanocoumarins, as well as other hydroxyl compounds (Figs. 4 and 5). Among the polyphenolic compounds, phenolic acids and their derivatives were found to be the most abundant compounds in both the water and 70% ethanoic extracts. However, far more phenolic acids, both in bond and free forms, were detected in the ethanol extracts than in the water extract. The phenolic acids determined in the study were 1855.4 ± 4.8 and 384.2 ± 3.2 mg/L CAE/FAE/CoAE in the ethanol extract and water extract, respectively. Flavonoids were the second ions (− 46 Da) [54]. Compound 16 was identified as phenylalanine consistent with previous studies [54].

HPLC DAD and UPLC-QTOF-MS Quantitation of Phenolic Compounds
The UV/vis absorptions, capable of distinguishing phenolic subclasses, were considered a starting point for compound quantification. Many aglycone standards have been used in the process, even for the glycosylated forms, especially for O-linked compounds, because their absorption coefficients and the wavelengths of maximum absorption do not change significantly with glycosylation, owing to the fact that glycosidic and alkyl residues are poor chromophores [57]. For C-linked glycosides, however, a new series of conjugated bonds significantly alters the absorption spectra, making the compound's absorption different from the aglycone's [57]. In this study, however, C-linked glycosides were not very prevalent. Based on the qualitative analysis, five chemical markers, namely phenolic acids (caffeic acid (peak 2), ferulic acid (peak 5), and coumaric acid (peak 4)), flavonol (peak 6), dihydrochalcone (peak 8), flavanone (peak 7), and flavano-3-ols (peak 1/3) were selected for simultaneous quantitative determination. They were the most abundant in the extracts. Based on the UV spectrum of the markers, UV detection wavelengths were chosen. The phenolic acids had the strongest UV absorption at 300, 309, and 322 nm, whereas flavanones had UV absorption at 284, 330 nm, flavonols at 254, 255, and 354 nm, dihydrochalcones at 284 nm, and finally flavan-3-ol at 278 nm.
Phenolic compounds were detectable and quantifiable in the aqueous and ethanol extracts of B. pilosa (Table 2) and sampled with a high degree of sensitivity, as indicated by their limits of quantification and detection ( Table 3). The limits of detection and quantification observed, along with the calculated recoveries, indicate this method's suitability for   6A and 7A). Interestingly, the lowest and highest concentrations of the water extract stimulated cell growth of the Vero (non-cancerous cells). However, the ethanol extract caused statistically significant toxicity to both cell lines (Figs. 6B and 7B). The two highest concentrations (500 µg/ mL and 1000 µg/mL) were toxic to the HepG2 cells (Fig.  6B), whereas only the highest concentration (1000 µg/mL) of the ethanol extract was cytotoxic to the Vero cells. However, at concentrations ranging from 31.25 µg/mL-500 µg/ mL, the Vero cell lines did not show progressive proliferation (Fig. 7B) compared to the control and 15.6 µg/mL. The lowest concentration (15.6 µg/mL) of the water and ethanol extracts thus stimulated cell growth of the Vero (non-cancerous) cells. An EC 50 was calculated based on the cytotoxicity caused by the ethanol extract to both cell lines (

Discussion
The use of medicinal plants in various capacities, including disease treatment, pest and weed control in agriculture, and consumption for nutritional benefit, is a common practice among indigenous peoples and those living below the poverty line. The shelf life of the plant product, however, is crucial to its effectiveness because it correlates to the bioactivity of the phytochemicals [60]. Moisture, temperature, and other environmental conditions influence the shelf life of medicinal herbs. According to the study of Müller and Heindl (2006) [61], medicinal plant specimens are best stored at a final moisture content (MC) not higher than 13%, relative humidity of less than 70%, and at 25 °C to reduce the development of microorganisms such as bacteria, fungi, and yeasts, as well as pest infestation. In this study, the moisture content of B. pilosa was found to be 5.14% when air-dried at room temperature. This indicates that the plant could be air-dried until a constant mass to attain a safe condition, especially if intended for consumption as a herbal remedy or diet purposes. The potency of the phytochemicals will also be at its optimal under the conditions stated above.
Plants are known to contain different metabolites, including natural antioxidants, that help to protect and preserve their metabolism as well as their physical integrity. A number of plant phytochemicals with antioxidant characteristics have been identified, with examples including hydrosoluble antioxidants such as phenolics, flavonoids, anthocyanins, stilbene, and lignan, as well as liposoluble antioxidants such as β-carotene, α-carotene, lycopene, lutein, zeaxanthin, minerals, and vitamins [62]. Recent studies suggest that the most abundant compounds detected in the 70% ethanol extract. Given the wide range of uses of B. pilosa around the world, the abundance of flavonoids and phenolic acids in the plant might be the basis of the health benefits, biological activities, and other uses of B. pilosa.
Phenolic acids, including ester forms of hydroxybenzoic and hydroxycinnamic acids, are major classes identified and are estimated at 52% and 42% in water and 70%-ethanol extracts respectively. However, this percentile was estimated with a total of 54 compounds identified in the water and 101 identified in the ethanol extract. Hydroxybenzoic acids account for 28.57%, while hydroxycinnamic acids make up 71.42% of the phenolic acids in the water extract. The ethanol extract, on the other hand, has 11.62% of hydroxybenzoic acids and 88.37% of hydroxycinnamic acids. Phenolic acids have been reported to have quite a number of beneficiary biological activities, such as antidiabetic, antiulcer, antioxidant, anticancer, antimicrobial, antiaging, neuroprotection, hepatoprotection, and cardioprotection [58]. The high content of phenolic acids in the extracts of B. pilosa could be attributed to its medicinal usage in traditional medicine.
Flavonoids are a class of polyphenolic compounds found primarily in plants and provide health benefits through cell signalling pathways. It could be used as an antioxidant agent, phytoalexin, allopathic compound, and antimicrobial agent [59]. In this study, a total of 3 peaks (6%) of flavonoids in the water extract and 33 peaks (33%) of flavonoids in the 70% ethanol extract, belonging to the main subclasses dihydroflavonol, flavonols, flavanones, flavones, and flavan-3-ol, were tentatively characterized through their characteristic UV spectra.

Cytotoxicity Using RTCA
The HepG2 (cancerous) and Vero (non-cancerous) cell lines were used to measure the cytotoxic potential of the plant Spectrophotometric methods revealed the flavonol content, a subclass of flavonoids is 2.92 mg QE/g in the water extract and 18.31 mg QE/g in the ethanol extract in the present study. According to the developed UPLC-ESI-QTOF-MS overproduction of reactive nitrogen species (RNS) and reactive oxygen species (ROS), as well as hydroxyl radicals, nitrosonium, hydrogen peroxide, nitroxyl anion, peroxynitrite, and superoxide, could be the leading causes of oxidative stress. This could result in other pathogenic processes, such as DNA damage and inflammatory diseases [63]. From this point of view, compounds with molecules capable of scavenging or reducing high ROS/RNS and free radicals would be of health benefit in managing illnesses such as cancer, diabetes, heart disease, neurodegenerative diseases, and ageing [58]. Bidens pilosa could offer these benefits because of its high polyphenols (an antioxidant) ranging from 15.66 to 20.19 mg GAE/g in the water extract and from 30.86 to 32.52 mg GAE/g for the ethanol extract, with a subclass of 54% and 42% phenolic acid, respectively. The total amount of total polyphenol estimated in this study is similar to the study of Cortés-Rojas et al. (2013) [64] who estimated the total polyphenol content of B. pilosa to be about 20 mg methods, flavonols were not detected in the water extracts, but other subgroups of flavonoids such as dihydroflavonol, flavanones, and flavones were found at 33% each. However, the flavonol in the ethanol extract was 78.78% of the flavonoid content. Flavonoids and phenols are important components of diets and medications due to their inherent health-promoting properties, which include anti-oxidative, anti-inflammatory, anti-mutagenic, anti-carcinogenic, antimicrobial, and larvicidal properties [59]. Bidens pilosa has been reported to be used in the treatment of various diseases, including hyperglycemia [8], malaria [9], tumours [10], and inflammation [12], which could be linked to the high possession of flavonoids and phenols. The antioxidant activities of the ethanol extracts are higher compared to the water extract. According to this study, the peaks in the ethanol extract are higher than in the water extract, with a total of 101 peaks belonging to four main classes: phenolic  [58], phenolic acids help in disease prevention and therapy through cell growth promotion using their antioxidants and anti-inflammatory properties [58] which is contrary to the activity of flavonoids, particularly against cancerous cells. This implies that flavonoids do not promote cell proliferation, unlike phenolic acids. The simulated growth observed in the Vero cells after exposure to both types of extracts (water and ethanol), could be ascribed to the presence of phenolic compounds in the extracts. Although this might be the case for the Vero cells, in contrast, the HepG2 cells did not show simulated growth patterns. This observation in differences in vulnerabilities between the types of cells is due to genetic differences between the two cell lines [74].
The EC 50 values determined in this study using the xCELLigence technology are substantially lower compared to previous studies where cytotoxicity of B. pilosa methanolic extracts was determined using the MTT assay with HepG2 cell lines [66]. This difference could be attributed to the increased sensitivity of the xCELLigence compared to the MTT assay or the use of different solvents, i.e., we used ethanol instead of methanol. In contrast to the absence of EC 50 values (no cytotoxicity observed) for the water extracts in the current study, Abdel-Ghany et al. (2016) [75] published EC 50 of 1000 µg/ml after exposure of aqueous extracts of B. pilosa to primary cell cultures (rat hepatocytes). These authors used primary cell cultures (short life span) in contrast to the immortal cell lines used in the current study. One should, however, keep in mind that the cells are cultured in tissue culture dishes without the benefit of an immune system and are merely indicative of toxicity potential. The results indicate the potential toxicity exhibited by the ethanol extracts. These effects might be different when whole organisms are used or exposed due to their defence mechanisms.

Conclusion
This study is a comprehensive secondary metabolite profiling and cytotoxicity that aims to provide chemical-based evidence for the health benefits of B. Pilosa. The analytical methods used were HPLC-DAD and acids, flavonoids, fatty acids, coumarins and furanocoumarins, as well as other hydroxyl compounds. The number of flavonoids was relatively high, about six-fold more than that which is present in the water extract. The antioxidant activity is both the individual and joint effect of the metabolites in the plant sample. Our study reveals the antioxidant activities of the ethanol extracts are higher, indicating a higher potency, and this correlates with the number of peaks or phytochemicals. The antioxidant capabilities of different solvent extractions of Severinia buxifolia were studied by Truong et al. (2019) [67], and water extract was shown to have the least activity, while ethanol was found to be the second most active, after methanol.
Other classes of compounds with health benefits such as coumarins and related compounds were also annotated in the extract of B. pilosa. These compounds have become relevant in recent years due to their various biological activities. Coumarins and related compounds in their natural or synthetic forms have been reported to be used as antibiotics [68], anti-inflammatory and antinociceptive [69], photochemotherapy and anti-HIV therapy [70], cancer and antioxidant agents [71]. Aside from their medical use, coumarins are also utilized in various industries, such as the cosmetic, and agrochemical industries, as optical brightening agents, laser dyes and dispersed fluorescent [70]. This study suggests that B. pilosa could serve as a natural source of coumarins and furanocoumarins, with water solvent being a more effective extractant than ethanol.
The cytotoxicity of the extract followed a similar trend to the results of other assays in this study. Cytotoxicity is a non-specific endpoint of toxicity and can be applied as an early warning sign since it is a reflection of the toxic response elicited by chemicals present in mixtures as in the case of plant extracts [72] Cell death is reflected by a slowdown in metabolic activity [34] and can be measured using various viability assays. In this study, the cytotoxic potential of B. pilosa water and ethanol extracts were assessed using different cell lines: HepG2 and Vero. Concentrationdependent cytotoxicity was observed after exposure to the ethanol extracts. The high concentrations of the ethanol extracts were cytotoxic to both cell lines compared to the water extracts rather than promoted cell proliferation -specifically of the Vero (non-cancerous) cells. The cytotoxicity caused by the ethanol extract could be attributed to more phytochemicals present in the ethanol extract, particularly flavonoids when compared to the water extract. An extensive investigation has been carried out with flavonoids against various cancer cell lines such as breast, prostate, liver, lung, melanoma, pancreas, and stomach., which proved effective against cancer cells, making flavonoids an anticancer agent It has, however, been used as a chemoprevention agent as well as a cancer treatment agent [73]. The same UHPLC-ESI-Q-TOF-MS, which enabled the separation as well as annotation of 137 secondary metabolites, mainly predominated by phenolic acids, flavonoids, and fatty acids. The plant (ethanol extract) is cytotoxic but not poisonous, as no potentially poisonous compound was identified. These findings provide further evidence for the value of the plant as a source of functional phytoconstituents and also explain the impact of solvents of extraction (ethanol and water) on the bioactivity of the plant.
Acknowledgements The authors would like to thank Hein Hendrik Smith for his assistance in locating and collecting plant samples (Bidens pilosa).

Funding Statement
The authors do not receive any funding from an external source. Open access funding provided by North-West University.

Availability of Data and Material
All data generated or analysed during this study are included in this article.

Declarations
Competing interests The authors declare that there is no conflict of interest associated with this publication. All authors have read and agreed to the publication of the manuscript.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons. org/licenses/by/4.0/.