UHPLC-ESI-Orbitrap-MS Analysis of Biologically Active Extracts from Gynura procumbens (Lour.) Merr. and Cleome gynandra L. Leaves

This study aimed to determine the total phenolic content, DPPH scavenging, α-glucosidase, and nitric oxide (NO) inhibition of Gynura procumbens and Cleome gynandra extracts obtained with five different ethanolic concentrations. The findings showed that the 100% ethanolic extract of G. procumbens had the highest phenolic content and the lowest IC50 values for DPPH scavenging and NO inhibition activity compared to the properties of the other extracts. For C. gynandra, the 20% and 100% ethanolic extracts had comparably high total phenolic contents, and the latter possessed the lowest IC50 value in the NO inhibition assay. In addition, the 20% ethanolic extract of C. gynandra had the lowest IC50 value in the DPPH scavenging assay. However, none of the extracts from either herb had the ability to inhibit α-glucosidase enzyme. Pearson correlation analysis indicated a strong relationship between the phenolic content and DPPH scavenging activity in both herb extracts. A moderately strong relationship was also observed between the phenolic content and NO inhibition in G. procumbens extracts and not in C. gynandra extracts. The UHPLC-ESI-Orbitrap-MS revealed major phenolics from the groups of hydroxycinnamic acids, hydroxybenzoic acids, and flavonoid derivatives from both herbs, which could be the key contributors to their bioactivities. Among the identified metabolites, 24 metabolites were tentatively assigned for the first time from both species of studied herbs. These two herbs could be recommended as prospective natural products with valuable medicinal properties.


Introduction
Plants and herbs have a history of traditional uses and are important parts of cultural heritage. eir appreciation as food and links to health-promoting benefits are also significant. Since ancient times, herbs have been utilised traditionally to cure many illnesses, which has prompted modern science to fully understand their benefits. Gynura procumbens (Asteraceae), locally known as Sambung nyawa, is an annual grown herb that has thick leaves and hardened stems with a slight purple tint during maturation. e young leaves can be eaten raw as salads. Its ethnomedicinal usages are well reported; for example, in Indonesia, G. procumbens is used to treat fever, skin rashes, and ringworm infection [1]. In ailand, it is used to treat inflammation, viral infections, and rheumatism [2]. Scientific investigations on G. procumbens include antidiabetic, antihypertensive [3], anticancer [4], and anti-inflammatory [5] studies. In terms of its phytochemical constituents, phenolic compounds, such as kaempferol, quercetin, astragalin [6], kaempferol 3-O-rutinoside, rutin, and chlorogenic acid [7], were identified as the key metabolites that contributed to the bioactivities in the reported studies.
Inflammation is a coordinated response in the body towards harmful stimuli, such as injuries, pathogenic infection, and allergens. Macrophages are the cells responsible for initiating inflammation by producing inflammatory mediators, such as cytokines, interferons, and nitric oxide (NO) as response to stress [15]. Study has also indicated how the production of excessive inflammatory mediators leads to the onset of diabetes [16]. Among the treatments available for diabetes, the inhibition of α-glucosidase is one of them, which is responsible for glucose breakdown in the intestinal wall [17]. Plants' phenolic compounds have been reported to have the ability in inhibiting this enzyme, yet more often, modern drugs can be prescribed to combat the occurrence of inflammation-related diseases and diabetes [18]. However, prolonged use of these drugs stimulates unwanted side effects towards the liver, kidney, and other organs [19]. In return, naturally occurring phytochemicals from herbs have been ventured as an alternative medicine since they possess reduced or no toxicity when consumed at lower doses. G. procumbens and C. gynandra are two herbs that have the potential to be explored further for their anti-inflammatory and antidiabetic properties based on past studies. An optimum and proper extraction protocol may help researchers study the beneficial health properties of herbs, thus enabling the development of herbal-based products [20]. Extraction protocols that vary based on the type of sample, extraction solvents, temperature, extraction time, and instruments used play important roles in the standardization of herbs [21].
In this study, the efficacy of G. procumbens and C. gynandra extracted with different ethanol concentrations was tested for α-glucosidase and NO inhibition. Nevertheless, detailed metabolite profile and the effect of solvents extractions on distribution of metabolites of both herbs are still lacking. As such, this study proposed to investigate the aforementioned properties of the herbal extracts. e total phenolic content (TPC) and 2,2-diphenyl-1-picrylhydrazyl (DPPH) free radical scavenging activity were also tested as to support the anti-inflammatory and antidiabetic properties of the studied herbs. e relationship between TPC and biological activity was studied using the Pearson correlation model. Active ethanolic extracts from both herbs were then subjected to ultra high-performance liquid chromatography-electron spray ionisation-orbitrap-mass spectrometry (UHPLC-ESI-Orbitrap-MS), and potential metabolites that may contribute to the tested bioactivities were tentatively identified and reported. Only the leaves from both herbs were used in this study. e leaves were washed under running tap water and gently dried using laboratory tissue paper. After that, they were ground with liquid nitrogen using a mortar and pestle and freeze-dried immediately. e dried powder samples were stored at -20°C prior to analysis.

Extraction of Samples.
Six replicates of freeze-dried samples of both G. procumbens and C. gynandra were extracted with different concentrations of ethanol/water ratios (0, 20, 50, 70, and 100% ethanol). In general, the dried samples (4 g) were mixed in 250 mL conical flasks with e mixture was then homogenised using a homogeniser (Ultra Turrax, IKA, Germany) at 6000 rpm for 1 min followed by shaking in a shaker (ES-20, Biosan, Latvia) at 220 rpm for 15 min at room temperature. e supernatant was filtered with125 mm diameter filter paper (Advantec, Japan). e extraction procedures were repeated twice. e collected supernatant was concentrated using a rotary evaporator (RII, Buchi, Switzerland) and further freeze-dried (Freezone 6, Labconco, USA) to eliminate any moisture. Prior to bioassay analysis, the crude extracts of 0 and 20% ethanol/water were dissolved in deionised water, whereas the others were dissolved in dimethyl sulfoxide (DMSO).

Total Phenolic Content (TPC) Assay.
e TPC assay was carried out in accordance with the method described in [22] with minor modifications. Modifications were made in terms of the volume of reagents used in the assay. In general, 20 μL of 350 μg/mL of the extract was mixed with 100 μL of Folin-Ciocalteu's reagent (10-fold dilution) in a 96-well plate. en, 80 μL of 7.5% sodium carbonate was added, and the mixture was left in the dark for 30 min prior to the absorbance reading at 750 nm (SpectraMax PLUS, USA). Gallic acid was serially diluted ranging from 0.78 to 100 μg/ mL and used to make a standard curve in this assay. e results are expressed as mg gallic acid equivalent (GAE) per 100 mg of dried extract.

2,2-Diphenyl-1-picrylhydrazyl (DPPH) Scavenging Assay.
e DPPH scavenging assay was performed based on the method described in [22] with minor modifications to the concentration and volume of DPPH used. In general, 50 μL of extract with serial dilutions ranging from 10.94 to 350 μg/ mL was mixed with 100 μL of DPPH (0.15 mM) in a 96-well plate and incubated in the dark for 30 min. e absorbance of the solution was measured at 515 nm. Quercetin was used as the positive control, and all experiments were performed in six replicates. e results are expressed as the concentration (μg/mL) of extract needed to scavenge 50% of DPPH (IC 50 ).

α-Glucosidase Inhibition Assay.
e inhibition of the α-glucosidase enzyme by G. procumbens and C. gynandra extracts was evaluated based on the method described in [23] with minor modifications to the used volume of the enzyme and substrate. Prior to the experiment, α-glucosidase and the substrate PNPG were dissolved in 50 mM phosphate buffer at pH 6.5. In brief, 10 μL of plant extract with a serial dilution ranging from 10.94 to 350 μg/mL was mixed with 130 μL of 30 mM phosphate buffer and 10 μL of the enzyme (2 U/mL) in a 96-well plate. After 5 min of incubation, 50 μL of the PNPG substrate was added followed by and reincubation at room temperature for another 15 min. e reaction was ceased by the addition of 50 μL of glycine (pH 10) before the absorbance was measured at 405 nm. Quercetin was used as the positive control in this study. e results are expressed as the concentration (μg/mL) of extract needed to inhibit 50% of α-glucosidase (IC 50 ).

Nitric Oxide (NO) Inhibition Activity.
e NO inhibition assay was performed in accordance with the method described in [24]. e RAW 264.7 cells were grown in culture flasks using phenol-red DMEM under 5% CO 2 at 37°C. Once confluency reached 80%, cells were detached using 2.5 mL TrypLE ™ Express enzyme. Prior to cells seeding in 96-well plates, cells were counted using the standard Trypan blue counting technique, where the cell concentration was set to 1 × 10 4 cells/mL in all wells. Seeded cells (50 μL/well) were left in an incubator for 24 h before proceeding with induction and treatment. In all, 50 μL (1 μL IFN-c + 1μL LPS + 48 μL DMEM) triggering agent was added into the designated wells, followed by 50 μL of plant extract (serially diluted from 15.63 to 500 μg/mL). Curcumin was used as the positive control in this assay. All analyses were performed in six replicates, and the cells were incubated 17-24 h at 37°C under a 5% CO 2 atmosphere.

Measurement of Nitrite.
A Griess assay was performed to measure the accumulation of nitrite ions (NO 2 − ), a conversion product from NO in a simple manner. An incubated 96-well plate was removed, and 50 μL of media from the plate was transferred attentively into a new 96-well plate.
en, 50 μL of Griess reagent (1% sulfanilamide, 0.1% N-(1naphtyl)-ethylene diamine dihydrochloride, and 2.5% phosphoric acid) was added, and the plate was left in the dark for 15 min at room temperature. Sodium nitrite (NaNO 2 ) at 200 μM was used as a positive control in this assay. Absorbance at 550 nm was measured after incubation.
2.9. Cell Viability Test. Cell viability was assessed by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay to determine the cytotoxicity of the plant extract. Fresh phenol-red DMEM (100 μL) was added to the wells containing cells, followed by 20 μL of MTT (dissolved in 1× PBS buffer). e plate was then incubated for 4 h at 37°C under a 5% CO 2 atmosphere. Next, all the media were discarded, and 100 μL of DMSO was added. e plate was left for 15 min in the dark at room temperature before the absorbance was measured at 570 nm. e percent viability of the cells was calculated by comparing the absorbance of the treated cells with the control group (untreated cells).

Metabolite Profiling Using the UHPLC-ESI-Orbitrap-MS.
e separation and identification of metabolites from G. procumbens and C. gynandra were achieved using a UHPLC system (Ultimate 3000 ™ , ermo Scientific) coupled with mass spectrometry (Q Exactive ™ Hybrid Quadrupole-Orbitrap, ermo Scientific). Separation of the metabolites was performed using a ermo C18 column (2.1 mm × 100 mm, 1.9 μm) with mobile phases of LCMSgrade acetonitrile with 0.1% formic acid as buffer B and deionised water with 0.1% formic acid as buffer A. e flow rate was set at 274 μL/min, and UV detection was set at Evidence-Based Complementary and Alternative Medicine 3 280 nm. e gradient setting of the mobile phases was set for a total run time of 30 min and divided as follows: equilibration of column for 5 min at 95% of solvent A, a steady decrease in solvent A for 20 min until reaching 5%, maintenance of 5 % solvent A for another 5 min, and a steep increase in solvent A to 95% in one minute. e column was re-equilibrated for 4 min at 95% solvent A. All samples were prepared at a concentration of 2.0 mg/mL in 30% methanol and filtered by a nylon syringe filter (0.45 μM, 13 mm diameter). Ten microliters of each sample was injected into the system. Ionization of the metabolites was performed using an ESI probe in negative mode. e capillary temperature was set at 300°C with a scanning range from 50-1500 amu. All analyses were performed and monitored using Xcalibur 2.2 software ( ermo Scientific Inc, Waltham, MA, USA). e identification and characterization of metabolites were performed by relative comparison from previously reported data and from online databases [25,26]. e mass error in ppm was calculated by comparing the theoretical monoisotopic mass from the online databases to the observed mass.

Statistical Analysis.
All bioassay results were reported as the mean of six biological replicates with the standard deviation. One-way analysis of variance (ANOVA) was performed with a significant difference between the collected data set at a confidence interval of 95%. e Pearson correlation test (r value) was calculated to evaluate the relationship between the bioactivities. All calculations were performed using GraphPad PRISM version 5.01 for Windows (San Diego, CA, USA).

Total Phenolic Content of G. procumbens and C. gynandra Extracts.
e TPC results of the G. procumbens and C. gynandra extracts are presented in Table 1. e TPC assay was performed to measure the relative amounts of phenolic compounds from the herbal extracts since most phenolic compounds were found to possess health benefits [27]. G. procumbens extracted with 100% ethanol was observed to have the highest phenolic content with 5.91 mg GAE/100 mg dried extract (de). e TPC trend was 100% > 70% > 50% > 20% > 0% ethanolic extract as the ratio of water increased. On the other hand, the TPC of C. gynandra extracts showed mixed results, with 0%, 20%, and 100% ethanolic extracts showing significantly high phenolic content with 3.38, 3.48, and 3.71 mg GAE/100 mg de, respectively. e TPC values increased as the ratio of ethanol increased and started to drop in the 50% and 70% ethanolic extracts before rising again at the 100% ethanolic concentration. e difference in the TPC trend shown by the two herbs could be due to the type of metabolites being extracted. Previous study has indicated how polar metabolites in nature have the tendency to be extracted with polar solvents and vice versa [28]. Secondary metabolites, such as tannins, hydroxycinnamic acids, and flavonoids are regarded as polar in nature, whereas sterols, terpenoids, and lipids are semipolar and nonpolar [29]. Water, being a universal solvent, is also more polar than ethanol. Increasing the water ratio in ethanol makes the final concentration of the extraction solvent more polar. In this study, different water/ethanol concentrations were thought to extract metabolites of different polarities, thus giving different clusters of metabolites in each extraction solvent.

DPPH Scavenging Activity of the G. procumbens and C. gynandra Extracts.
e ability of the ethanolic extracts of G. procumbens and C. gynandra to scavenge DPPH free radicals was tested, and the results are presented in Table 1

α-Glucosidase Inhibition Activity of the G. procumbens and C. gynandra Extracts.
Unlike the rest of the assays, neither extract (G. procumbens or C. gynandra) showed any inhibition against the α-glucosidase enzyme. In this study, extracts from both plants at 10 mg/mL were tested against α-glucosidase, and no activity was detected except for the positive control, quercetin. Quercetin showed 92.3% inhibition at 50 μg/mL. is finding contradicts findings from other researchers, for example, Ngwe and colleagues [30] reported that they obtained IC 50 values of the 95% ethanolic and water extracts of G. procumbens to be 1.05 μg/mL and 1.06 μg/mL, respectively. Another study reported that the water extract of G. procumbens yielded an IC 50 of 0.092 mg/ mL [31]. Further investigation from the reported studies leads to the observation of certain parameters that were not applied in our studies. G. procumbens was extracted thrice with water for 12 h, which was different from our extraction protocol. e incubation temperature during the assay also seemed to play an important role in the enzyme activity [32]. Khatib et al. [29] reported low inhibition of the α-glucosidase enzyme using a Momordica charantia ethanolic extract and suggested different origin, maturity, postharvest, and processing conditions as possible factors for their results. On the other hand, this is the first ever reported study on the inhibition of α-glucosidase of C. gynandra ethanolic extracts. e inhibitory pattern of both herbs was similar in that the inhibitory activity increased gradually from 0% > 20% > 50% > 70% and 100% ethanolic extracts. e MTT assay further confirmed that none of the G. procumbens and C. gynandra extracts possessed cytotoxicity towards the cells, which strengthened the possibility for the development of anti-inflammatory drugs.

Correlation between TPC, DPPH Scavenging, and NO Inhibition of G. procumbens and C. gynandra Extracts.
To justify whether phenolic compounds from the studied herbs correlate with DPPH scavenging and NO inhibitory activity, a Pearson correlation test was performed at the 95% confidence interval. No correlation data were obtained for the α-glucosidase assay since none of the extracts showed any inhibition towards the enzyme. A strong positive or negative correlation has a scored r value between +0.5 and +1.0 or − 0.5 and − 1.0. If the score is between +0.1 and +0.3 or − 0.1 and − 0.3, then the association is considered weak [33]. e correlation between TPC and the DPPH scavenging assay showed that the G. procumbens extracts scored a strong r value of 0.8443, whereas C. gynandra scored lower with r � 0.5888. is indicates that phenolic compounds could be one of the key contributors to the DPPH scavenging activity in both herbs. DPPH is a stable free radical and is frequently used to test the antioxidant capacity of herbal extracts. Plants' phenolic acids have been proven to be the main source of DPPH scavenging activity [28] in which DPPH free radicals receive electrons from phenolic acids and are reduced to the stable DPPH-H complex [34]. Our finding is also in line with the data presented for G. procumbens extracts [35]. e Pearson correlation between TPC and the NO inhibition activity of G. procumbens extract showed a moderately strong r value at 0.6418. is indicates a possible contribution from phenolic compounds in the inhibition of NO production via RAW 264.7 cell induction. By contrast, the C. gynandra extracts showed a weak negative correlation with an r value of 0.2432. A weak negative correlation may be due to the TPC assay providing an estimation of the total phenolic compounds presents in an extract. However, nonphenolic compounds, such as ascorbic acid and tocopherol, are also able to reduce Folin-Ciocalteu's reagent [36]. e presence of nonphenolic metabolites in C. gynandra extracts could contribute to a higher TPC and may not inhibit NO production.

Tentative Identification and Characterization of Phenolic
Compounds from the G. procumbens and C. gynandra Extracts.
e extracts with the best representation of their total phenolic contents, DPPH scavenging activities, α-glucosidase inhibition, and nitric oxide (NO) inhibition from both herbs were chosen for analysis by UHPLC-ESI-Orbitrap-MS. Ergo, the 100% ethanolic extract of G. procumbens and the 20% and 100% ethanolic extracts of C. gynandra were chosen for this purpose. Table 2 shows the list of tentatively identified phenolic acids grouped according to their class. Among the 58 phenolic acids identified, 27 were found only in the G. procumbens extract, and 11 and 3 were found only in the 20% and 100% ethanolic extracts of C. gynandra, respectively. e remaining 17 phenolic acids can be found in at least two of the extracts analysed. e theoretical monoisotopic mass and mass error of phenolic acids, which were identified as derivative or dimer, were not determined since information of their actual structure were lacking.

Conclusions
e DPPH scavenging, α-glucosidase inhibition, and NO inhibition activities as well as TPC were tested for G. procumbens and C. gynandra extracted with different ethanolic concentrations.
e results showed a preliminary understanding of the potential of both herbs to serve as antioxidant and anti-inflammatory agents. In total, 58 metabolites were identified in both herbs with 24 metabolites were identified for the first time. Tentatively identified metabolites help to reduce the gap in unknown metabolites from both herbs and are important for future reference. Future studies using both herbs as herbal formulations should investigate the synergistic effects of the metabolites on herbal product development.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare no conflicts of interest.