Antioxidant activities of Clerodendrum cyrtophyllum Turcz leaf extracts and their major components

This study was designed to investigate the antioxidant properties of the extracts and subfractions of various polarities from Clerodendrum cyrtophyllum Turcz leaves and the related phenolic compound profiles. The ethyl acetate fraction (EAF) showed the most potent radical-scavenging activity for DPPH radicals, ABTS radicals, and superoxide anion (O2·-) radicals as well as the highest reducing power of the fractions tested; the n-butyl alcohol fraction (BAF) was the most effective in scavenging hydroxyl radical (OH·), and the dichloromethane fraction (DMF) exhibited the highest ferrous ion chelating activity. Twelve phenolic components were identified from the EAF of C. cyrtophyllum. Additionally, acteoside (1) was found to be a major component (0.803 g, 0.54%) and show DPPH and ABTS radical scavenging activities with IC50 values of 79.65±3.4 and 23.00±1.5 μg/ml, indicating it is principally responsible for the significant total antioxidant effect of C. cyrtophyllum. Our work offers a theoretical basis for further utilization of C. cyrtophyllum as a potential source of natural, green antioxidants derived from plants.


Introduction
Reactive oxygen species (ROS), which comprise oxygen radicals, nonradical oxidizing species and singlet oxygen ( 1 O 2 ), are inevitable by-products of oxidative metabolism in all living organisms [1]. ROS are particularly unstable and can rapidly react with most biological molecules, including proteins, lipids, lipoproteins and nucleic acids; excess ROS can lead to oxidative stress and induce cellular damage or tissue injury associated with ageing, atherosclerosis, carcinogenesis and mutagenesis [2]. Fortunately, the prominent antioxidant enzymes of endogenous ROS defence can efficiently protect against these harmful free radical attacks, but they are unable to prevent damage completely; thus, exogenous antioxidants are vital to maintaining health [3]. However, the usage of synthetic antioxidants has been increasingly restricted over time because of their potential health risks, such as protein or DNA damage, and toxic effects [4]. Consequently, research has focused on identifying safe, naturally a1111111111 a1111111111 a1111111111 a1111111111 a1111111111

Antioxidant activity
The antioxidant activities of samples were determined using standard methods. VC and BHT were used as positive standards in the radical-scavenging assays. Gallic acid was used as a positive standard in the ferric reducing power assay. Ethylene diamine tetra acetic acid (EDTA) was used as a positive standard for the ferrous ion-chelating activity assay.
2.3.1. DPPH radical-scavenging activity. The DPPH radical scavenging activities were estimated [11] by mixing 0.1 ml of the extract with 3.9 ml of 60 μM solution of DPPH in ethanol. After 30 min of reaction, the absorbance was measured at 517 nm. The inhibition percent and 50% inhibition (IC 50 ) values of DPPH radicals were calculated. http://dx.doi.org/10. 17504/protocols.io.jiqckdw [PROTOCOL DOI] 2.3.2. ABTS radical-scavenging activity. The method described by [11] was used to determine the ABTS radical-scavenging capacity. An aliquot of extract (0.1 ml) was added to 3.9 ml of ABTS radical solution. The mixture was reacted for 30 min, and the absorbance at 734 nm was measured. The inhibition percent and IC 50 values of the extracts for ABTS radical were calculated. http://dx.doi.org/10.17504/protocols.io.jirckd6 [PROTOCOL DOI] 2.3.3. Superoxide radical-scavenging activity. The superoxide radical scavenging effects were examined [12]. Briefly, 1 ml of the extract was added to 1 ml of 50 μM NBT solution, 1 ml of 468 μM NADH, and a 1 ml aliquot of 60 μM PMS reaction mixture. After 5 min, the absorbance was read at 560 nm. The inhibition percent and IC 50 values were calculated. http://dx. doi.org/10.17504/protocols.io.bdaxi2fn [PROTOCOL DOI] 2.3.4. Hydroxyl radical-scavenging activity. The scavenging of hydroxyl radicals was determined following the method of Guo et al. [12]. The reactions were performed with 0.3 ml of 20 mM sodium salicylate, 2.0 ml of 1.5 nM FeSO 4 , 1.0 ml of sample, and 1.0 ml of 6 mM H 2 O 2 . The reaction mixture was incubated for 1 h at 37˚C. The absorbance was measured at 510 nm. The inhibition percent and 50% of absorbance (EC 50 ) were calculated. http://dx.doi. org/10.17504/protocols.io.bdazi2f6 [PROTOCOL DOI] 2.3.5. Reducing power. The reducing power of the samples were assayed using the method of Guo et al. [12] Briefly, 1 ml of extract was added to 2.5 ml of phosphate buffer (0.2 M, pH 6.6) and 2.5 ml of 1% potassium ferricyanide. After 20 min, 2.5 ml of 10% trichloroacetic acid (TCA) was added, and then the mixture was centrifuged at 3000 rpm for 10 min. The upper layer (2.5 ml) was mixed with 2.5 ml of distilled water and 0.5 ml of 0.1% ferric chloride, and after 10 min, the absorbance was measured at 700 nm. The EC 50 values were calculated from the graph of inhibition percentage against extract concentration. http://dx.doi.org/10. 17504/protocols.io.bda2i2ge [PROTOCOL DOI] 2.3.6. Ferrous ion-chelating activity. The ferrous ion-chelating activities were determined according to Guo et al. [12] A 1 ml aliquot of extract was added to a solution of 100 μL of FeCl 3 (2.0 mM), 3.7 ml of distilled water and 200 μL of ferrozine (5.0 mM). After 20 min, the absorbance was recorded at 562 nm. The inhibition percent and IC 50 values were calculated. http://dx.doi.org/10.17504/protocols.io.bda4i2gw [PROTOCOL DOI] 2.3.7. Total phenolic content (TPC). The TPCs in the samples were determined by a colorimetric method based on the procedure described by Zhou et al. [11] Folin-Ciocalteu (FC) reagent (2 ml) was added to 2 ml of diluted extract. After 3 min, 750 μL of sodium carbonate anhydrous solution (7.5%, w/v) was added, and the mixture was adjusted to 10 ml with distilled water. After 2 h, the absorbance was recorded at 765 nm. Calibration curves were constructed with gallic acid as the standard at concentrations ranging from 0-100 μg/ml. http:// dx.doi.org/10.17504/protocols.io.bda5i2g6 [PROTOCOL DOI] 2.3.8. Total flavonoid content (TFC). The amounts of total flavonoids were quantified [11]. The reaction mixture consisted of 1.0 ml of extract, 0.3 ml of 5% sodium nitrite and 4 ml of 60% ethanol. After 6 min, 0.3 ml of 10% aluminium nitrite was added. After 6 mins, 4 ml of 1 M sodium hydroxide solution was added. Then, the volume was brought to 10 ml, and the absorbance was measured at 510 nm. The TFC was calculated and is expressed as rutin equivalents (RE). A calibration curve was constructed with different concentrations of rutin (15-

Statistical analysis
Triplicate analyses were performed, and the results are presented as the mean ± standard deviation. Each experiment was performed three times. Statistical analyses were performed using ANOVA; p < 0.05 indicated significance. This analysis was carried out using Sigmaplot (Version 13.0).

Scavenging effects on DPPH free radical
DPPH radical quenching assays are commonly used for the determination of antioxidant activities, and an antioxidant candidate that proves promising in the scavenging of DPPH radical may inhibit one of the many mechanisms by which oxidative stress is caused by lipid peroxidation [13]. To evaluate the radical-scavenging abilities of the various extracts of C. cyrtophyllum in our specific experimental setup, we found a significant (p<0.05) dose-dependent decrease in the concentration of DPPH due to the scavenging activities of the extracts (Fig 2A). As illustrated in Table 1, EAF and BAF showed the lowest IC 50 values (0.36 mg/ml), corresponding to the greatest DPPH radical-scavenging capacities, followed by RF (IC 50 was 1.33 mg/ml). DMF was found to exert the weakest radical-scavenging effect and show the highest IC 50 value (4.04 mg/ml). The DPPH radical-scavenging activity tended to decrease in the following order: VC>BHT>EAF> BAF>ECE>PEF>RF>DMF. However, none of the C. cyrtophyllum extracts were more effective than the positive standards, VC and BHT (IC 50 values of 0.07 and 0.08 mg/ml, respectively).
Consistent with previous reports, DPPH radicals could be effectively inhibited by the ethanol extract of C. infortunatum leaves with an inhibition of 92.6% at the tested concentration (250 μg/ml), whereas the chloroform and petroleum ether extracts exhibited low DPPH scavenging activities, and their percentage inhibitions were 52.2% and 16.7% at the same concentration [14]. The scavenging ability of C. cyrtophyllum is not very remarkable when compared to those of other medicinal Clerodendrum plants.

Scavenging effects on ABTS free radical
The relatively stable ABTS radical is recommended for use in the determination of antioxidant activity of plant extracts, as the colour of plant extracts do not interfere in this determination [15]. The efficiency of various C. cyrtophyllum extracts to scavenge ABTS radicals was increased significantly (p<0.05) with increasing concentration of the extract (Fig 2B). At 0.8 mg/ml, the scavenging activities of the extracts for ABTS radicals decreased in the following order: EAF (86.1%) > BAF (72.9%) > RF (42.6%) > ECE (41.3%) > DMF (17.7%) > PEF (16.7%). At 1.6 mg/ml, the rank of the antioxidant activities was as follows: EAF (93.5%) > BAF (93.9%) > ECE (82.3%) > RF (72.8%) > DMF (33.7%) > PEF (29.9%). RF achieved 93.1% inhibition of ABTS radicals, which was more than PEF (63.1%), at the highest tested concentration (4.0 mg/ml). As indicated in Table 1, the IC 50 values of all the examined fractions indicated that the ABTS radical-scavenging ability of EAF was greatest, as it showed the lowest IC 50 , 0.10 mg/ml, which was almost the same as that of the positive standards, VC and BHT (IC 50 values of 0.06 and 0.10 mg/ml, respectively).
With respect to other extracts, the aqueous and hydroalcoholic extracts of Verbena officinalis demonstrated ABTS radical-scavenging capacities with IC 50 values of 99.27 and 301.11 μg/ ml, respectively [16], indicating that C. cyrtophyllum is less effective than other Verbenaceae plants, but EAF of C. cyrtophyllum is more effective in comparison.

Superoxide radical-scavenging activity
Although the superoxide radical is not a highly reactive oxidative species that is toxic to cellular components, its dismutation can result in dangerous hydroxyl radical formation via Fenton-

PLOS ONE
type chemistry and induce lipid peroxidation [17]. The ECE and five subfractions of C. cyrtophyllum exhibited linear concentration-dependent inhibition of superoxide radicals ( Fig 2C). As shown in Table 1, EAF exhibit the strongest superoxide radical-scavenging ability with the lowest IC 50 value, 13.06 μg/ml, which is better than those of positive standards VC and BHT (IC 50 values of 22.22 and 26.73 μg/ml, respectively). The IC 50 values were in the following order: EAF > VC > BHT > BAF > ECE > RF > DMF > PEF. The PEF subfractions of various plants generally show the weakest free radical scavenging activities, which could be because few lipophilic extracts are capable of showing antioxidant activity. EAF was the most effective superoxide radical scavenger, indicating that the potential antioxidant compounds in C. cyrtophyllum were of medium polarity.
With respect to other extracts, the ethanolic extracts of Stachytarpheta angustifolia showed superoxide radical inhibitions in a range of 73.3%-80.8% at concentrations of 125-250 μg/ml, and the IC 50 value was 64.68 μg/ml [18]. C. cyrtophyllum extracts, especially the EAF subfraction, might exhibit better superoxide radical scavenging ability than S. angustifolia.

Hydroxyl radical-scavenging activity
Among oxygen-centred radicals, hydroxyl radicals are the most chemically reactive, and they therefore easily react with biomolecules such as proteins, lipids, and nucleic acids in almost every biological membrane, causing cell damage and hence resulting in ageing, cancer and several other diseases [19]. In this study, the Fe 2+ /H 2 O 2 system was used to generate hydroxyl radicals and measure the scavenging activities of ECE and five subfractions. The scavenging activities of the various extracts on hydroxyl radicals increased quickly with increasing concentration (Fig 2D). In Table 1, we present the IC 50 values of the various C. cyrtophyllum extracts and the positive control (VC) for the scavenging of hydroxyl radicals, and the values were 1.99 mg/ml, 0.90 mg/ml, 0.84 mg/ml, 1.07 mg/ml, 0.65 mg/ml, 3.41 mg/ml and 0.17 mg/ml for ECE, PEF, DMF, EAF, BAF, RF, and VC, respectively. The scavenging activities for hydroxyl radicals in descending order were VC > BAF > DMF > PEF >EAF > ECE >RF.
The ethanolic extracts of Vitex negundo exhibited dose-dependent scavenging and quenched approximately 40.2% and 56.7% of hydroxyl radicals at concentrations of 0.5 mg/ml and 1.2 mg/ml, respectively [20]. Apparently, C. cyrtophyllu extracts could be used to obtain effective hydroxyl radical scavengers that can help prevent oxidative damage; their effect is weaker than that of S. angustifolia but comparable to that of V. negundo.

Reducing power
Several investigations have indicated that the reducing powers of certain plant extracts are related to their antioxidant activity. The presence of reducers (i.e., antioxidants) can reduce the Fe 3+ /ferricyanide complex to the ferrous form by donating an electron, concomitantly decreasing the oxidized intermediates from lipid peroxidation processes [21]. The ECE and its five subfractions showed dose-dependent reducing powers (0-8 mg/ml) ( Fig 2E). As shown in Table 1, the EC 50 values of reducing power were found to be 0.203 mg/ml and 0.216 mg/ml for EAF and BAF, respectively, while the other subfractions all exhibited lower reducing powers varying from 0.473 to 1.562 mg/ml. The EC 50 value of gallic acid was 0.045 mg/ml. The obtained results suggest that gallic acid has better reducing ability than antioxidants from C. cyrtophyllum extracts. The ethanolic extracts of C. inerme and Lantana camara were examined previously and compared to C. cyrtophyllum. The extracts of C. inerme presented reducing powers of 0.79 (2.5 mg/ml) and 0.86 (5.0 mg/ml) [22]. The ECE and subfractions BAF and EAF obtained from C. cyrtophyllum had better antioxidant properties than those of the other Verbenaceae plants described previously, meaning that it might have a higher content of reductants bearing aromatic hydroxyl groups that can interrupt free radical chain reactions by hydrogen donation.

Ferrous ion-chelating activity
Ferrous ions are regarded as precursors of reactive oxidants, and these ions can catalyse lipid peroxidation via Fenton and Haber-Weiss reactions, resulting in the generation of hydroxyl radicals. Chelating ability is regarded as a significant indicator of potential antioxidant activity, and this parameter can be quantified based on the absorbance of the red colour generated by the reduction of Fe 2+ in the ferrozine complex, as the co-existing chelator may capture the ferrous ion before complex formation [23]. In this assay, both the extracts and EDTA decreased the absorbance in a dose-dependent manner (Fig 2F). The IC 50 values of the ECE and the various subfractions for ferrous binding ranged from 0.43 to 1.65 mg/ml, which were lower than that of the positive standard (EDTA, IC 50 of 0.16 μg/ml), suggesting that EDTA had the strongest chelating ability, as shown in Table 1.
With regard to methanolic extracts, the percentages of inhibition achieved by the leaf, stem and root extracts of C. viscosum were 9.6%, 15.8% and 5.0% at a concentration of 120 μg/ml, corresponding to IC 50 values were between 0.68 mg/ml and 1.10 mg/ml [24]. It seems that the ferrous chelating ability of the C. cyrtophyllu extracts were greater than that of C. viscosum, and these extracts were better able to stabilize the metal ion and oxidize it.

Antioxidant components
3.7.1 Total flavonoids and total phenolic content. The antioxidant activity of plant extracts is mainly linked to the active phytochemicals ubiquitous in plants [21]. The efficiency of phenol-based antioxidants is based on their redox properties, which allow them to quench oxygen-derived free radicals and are associated with their structural characteristics, such as the number and position of hydrogen-donating hydroxyl groups and alkyl degree of the phenolic moieties. Among these compounds, flavonoids constitute a special class of phenolic compounds and are reported to scavenge or delay oxidation by oxidizing molecules by transferring a single electron to O 2 �and OH � radicals [25].
In this study, TPC and TFC were used to quantify the antioxidant compounds in the C. cyrtophyllum crude extract and fractions (Fig 3). It could be seen that the TPC results were not entirely consistent with those of TFC; BAF showed the highest phenolic and flavonoid contents, followed by EAF. These findings were consistent with a higher efficiency in radical scavenging. As Table 1 shows, the TPCs and TFCs of the test extracts varied in the ranges of 0.07-12.42 mg GAE/g DW and 0.41-93.63 mg RE/g DW. The TPCs were in the following order: ECE > BAF > RF > EAF > DMF > PEF. The TFCs were in the following order: ECE > BAF > EAF > RF > DMF > PEF. These results indicated that TPCs and TFCs of fractions obtained with solvents of various polarities were different.
The strong correlations between the total antioxidant capacities determined based on the DPPH, ABTS, OH � , O 2 �-, reducing powers and ferrous ion chelating activities with the TPCs and TFCs were observed as show in Table 1. The data indicated that the total flavonoids and phenolic contents in BAF and EAF were significantly higher than those of other fractions. EAF showed the most potent radical-scavenging activity for DPPH radicals, ABTS radicals, and superoxide anion (O 2 �-) radicals as well as the highest reducing power of the fractions tested; the n-butyl alcohol fraction (BAF) was the most effective in scavenging hydroxyl radical (OH � ). The results are consistent with reports, which had previously suggested that the antioxidant activities of medicinal plants were mainly contributed by the phenolic compounds and flavonoids in the extracts and therefore could play an important role in the beneficial effects of corresponding important medicinal plants [26,27]. However, DMF exhibited the highest ferrous ion chelating activity, suggesting that other components with non-phenolic hydroxyl groups in the extracts were more effective chelators of ferrous ions than phenolic compounds  https://doi.org/10.1371/journal.pone.0234435.g003 [28,29]. Therefore, the phenolic and flavonoid constituents present in spices corresponding the antioxidant activities require further investigation.

Isolation and identification of compounds 1-12.
The EAF subfraction was subjected to successive separations and purifications using silica gel, polyacrylamide gel and Sephadex LH-20 gel column chromatography and semipreparative HPLC to yield compounds 1-12. The structures (Fig 4) of all these compounds were unequivocally determined by extensive NMR spectroscopic experiments as well as mass spectrometry and comparison with data reported in the literature [30][31][32].
Acteoside (1) [11]. It is interesting to investigate the structure-activity relationships for phenylethanoid glycosides 1-4, which have similar structures, and the main differences are the substituents at C-3 (R1), C-4 (R2) and C-4 0 (R3). It is inferred that the antioxidant activity of phenolics increases when there are free hydroxy groups in the molecule, which is consistent with the reported results [33]. Of the six flavonoids analysed (5-10), only luteolin (5) displayed strong antioxidant activities. In general, the antioxidant activities of flavonoids depend on the structure and substitution pattern of the hydroxy groups. The essential requirement for effective radical scavenging by flavonoids is a 3 0 ,4 0 -O-dihydroxy B-ring structure, which confers higher stability in the radical form and participates in electron delocalization [34]; hence, luteolin (5) has a higher antioxidant capacity. The high radical scavenging capacity of esculetin (12) is probably due to the superior stability of radicals derived from catechol moieties compared to that of phenoxyl radicals [35]. The finding that acteoside (1) is a major component (0.803 g, 0.54%) of C. cyrtophyllum and the EAF subfraction demonstrated that compound 1 is a main active ingredient and is principally responsible for the significant antioxidant effect of C. cyrtophyllum.

Conclusions
Continuing our ongoing research into the antioxidant activity of the components of C. cyrtophyllum, we first demonstrated that its crude extracts and fractions of various polarity possess potential antioxidant and radical-scavenging activities through multiple mechanisms. EAF and BAF exerted the highest antioxidant effects. Fractionation of the EAF led to the isolation and identification of phenolic compounds 1-12. Compounds 2-5 and 7-12 were obtained from C. cyrtophyllum for the first time. Compounds 1-5 and 12 exhibited considerable radical scavenging effects. Considering the yield, acteoside (1) is a main effective ingredient responsible for the DPPH and ABTS radical scavenging activities and could explain the significant antioxidant activities of the C. cyrtophyllum extracts, indicating it may be suitable as a natural antioxidant or alternative to toxic synthetic antioxidants in the food and pharmaceutical industries. Further work on the isolation and elucidation of other specific metabolites in C. cyrtophyllum responsible for the antioxidant activity is in progress in our laboratory, and in vivo biological tests should be conducted.

Ethics statement
This research did not include any human subjects or animal experiments.