**3. Results**

#### *3.1. Gallic Acid Equivalence (GAE)*

The gallic acid equivalence for the various crude extracts and fractions is shown in Table 1. GAE values varied in the range of 92.97 ± 0.99 (for mesocarp methanol fraction, MMF) to 1405.41 ± 17.96 (for seed methanol fraction, SMF) μmol GAE/g dry extract, using a calibration curve of gallic acid (*r*2 = 0.9997). The GAE values were further grouped as the following: low GAE (0–500 μmol GAE/g dry extract), moderate (500–1000 μmol GAE/g dry extract), and high (1000–1500 μmol GAE/g dry extract). All tested extracts and fractions were in the category of low GAE, except for SMF.


**Table 1.** Gallic acid equivalence (GAE) and total antioxidant capacity (TAC) of crude extracts and fractions of *D. indum* fruit.

HF: hexane fraction; DF: dichloromethane fraction; MF: methanol fraction; CM: crude methanol extract. \* Indicates the values in the same column are significantly different (*p* < 0.050) in comparison with SMF (marked \*\*) as measured by one-way analysis of variance (ANOVA) of unequal variance (Welch's ANOVA) with Games Howell post hoc test.

#### *3.2. Total Antioxidant Capacity (TAC)*

Neocuproine reagen<sup>t</sup> confers higher stability when compared with other chromogenic reagents such as 2,2-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) (ABTS) and DPPH [20]. TAC for the various *D. indum* fruit extracts (Table 1) varied in the range of 104.52 ± 1.64 (for MMF) to 1515.79 ± 75.86 (for SMF) μmol TE/g dry extract using a calibration curve of trolox (*r*2 = 0.9995). Spearman correlation was run to determine the direction and magnitude of the relationship between the GAE and TAC of all extracts. Interestingly, there was a very strong, positive correlation between GAE and TAC (*<sup>r</sup>*s = 0.929, *n* = 16, *p* < 0.010).

#### *3.3. DPPH Radical Scavenging Activity*

Table 2 shows that only six from the twelve crude extracts and fractions of *D. indum* fruit parts had IC50 values within the tested concentration range (3.91–500.00 μg/mL). None of the mesocarp crude extracts and fractions reached 50% inhibition of DPPH radicals despite there being concentration-dependent DPPH scavenging activity as shown by mesocarp dichloromethane fraction (MDF) and mesocarp hexane fraction (MHF). The IC50 values of various crude extracts and fractions varied in the range of 31.71 ± 0.88 (for SMF) to 497.97 ± 6.43 (for exocarp hexane fraction, EHF) μg/mL and maximum percentage of inhibition varied in the range of 27.27 ± 1.08% (for MHF) to 93.11 ± 0.22% (for SMF). Spearman correlation analysis showed that there was a negative correlation between the IC50 values of DPPH radical scavenging assay and GAE values of the tested extracts (*<sup>r</sup>*s = −0.587, *n* = 6, *p* < 0.050).


**Table 2.** 2,2-Diphenyl-picrylhydrazyl (DPPH) radical scavenging and linoleic acid peroxidation inhibition activities of *D. indum* fruit.

NA: not active; HF: hexane fraction; DF: DCM fraction; MF: methanol fraction; CM: crude methanol extract; QUE: quercetin. \* Indicates that the values in the same column are significantly different (*p* < 0.050) in comparison with QUE (marked \*\*) as measured by Welch's ANOVA with Games Howell post hoc test.

A total of twenty-six (26) subfractions were obtained from the fractionation of exocarp dichloromethane fraction (EDF) using column chromatography. Subfractions Di-6, Di-9, Di-11, Di-17, Di-21, Di-22, Di-23, Di-24, Di-25, and Di-26 were observed to possess clearly different TLC antioxidant bioautograms after visualization with 0.4 mM DPPH in methanol (Figure S1). Hence, DPPH radical scavenging activity of these ten EDF subfractions were determined where all selected subfractions presented concentration-dependent DPPH radical scavenging activities. However, only five subfractions, labelled as Di-21, Di-22, Di-24, Di-25, and Di-26 exhibited appreciable antioxidant activity with IC50 values of 83.73 ± 0.92, 87.96 ± 1.02, 53.42 ± 1.61, 37.66 ± 0.71, and 69.82 ± 1.28 μg/mL, respectively. Comparison with the IC50 value of EDF (260.82 ± 1.31 μg/mL) clearly signified that the five subfractions possessed stronger DPPH radical scavenging activities and thus were considered as antioxidant-enriched EDF subfractions.

#### *3.4. Linoleic Acid Peroxidation Inhibition*

Table 2 demonstrates that despite the percentage of linoleic acid peroxidation inhibition was significantly lower (*p* < 0.050) than quercetin, both EDF and EHF exhibited the highest percentage of inhibition (51.08 ± 0.84 and 51.46 ± 0.62 μg/mL, respectively) between all samples for the tested concentration range (0.98–125.00 μg/mL). Spearman correlation indicated a very weak but not statistically significant negative correlation (*<sup>r</sup>*s = −0.197, *n* = 8, *p* > 0.050) between the maximum percentage of linoleic acid peroxidation inhibition and maximum percentage of DPPH radical scavenging of the various *D. indum* crude extracts and fractions.

#### *3.5. GC-MS of SMF, EDF and EDF Subfractions*

Based on the assays employed, SMF and EDF were considered as fractions with prominent antioxidant activities. The difference in the metabolites of both fractions were investigated using GC-MS after derivatization using MSTFA. The total ion current (TIC) chromatograms of SMF and EDF are shown in Figure 1 and the metabolites identified in both fractions are listed in Tables 3 and 4. The mass spectra data of the metabolites identified in SMF and EDF are listed in Table S2 [24–36]. Phenolics, fatty acids, and dicarboxylic acids were detected in both fractions. The total area percentage of phenolics in the TIC chromatogram of EDF was 53 times more than that of SMF. On the other hand, SMF contained amino acids, saccharides, polyol, and sesquiterpene, which were not detected in EDF.


**Table 3.** Metabolites identified in the *D. indum* seed methanol fraction (SMF) through Gas Chromatography-Mass Spectrometry (GC-MS) analysis.

RT = Retention Time, C = Carbon, H = Hydrogen, O = Oxygen, N = Nitrogen, M+ = molecular ion, *m/z*.


**Table 4.** Metabolites identified in the *D. indum* exocarp dichloromethane (DCM) fraction (EDF) through GC-MS analysis.

RT = Retention Time, C = Carbon, H = Hydrogen, O = Oxygen, N = Nitrogen, M+ = molecular ion, *m/z*.

A total of nine phenolics have been identified in EDF, namely vanillic acid, syringic acid, ferulic acid, isoferulic acid, sinapic acid, vanillin, syringic aldehyde, *p*-hydroxybenzaldehyde, and coniferyl aldehyde. Numerous studies correlated the in vitro antioxidant activities of plant extracts with their phenolic contents. GC-MS of five antioxidant-enriched subfractions of EDF added another four phenolics: *p*-hydroxybenzoic acid, homovanillic acid, *p*-coumaric acid, and sinapic aldehyde, to the list of phenolics detected in EDF. The distribution of phenolic antioxidants in subfractions Di-21, Di-22, Di-24, Di-25, Di-26, and EDF is shown in Table 5.


**Table 5.** Area percentage (%) of phenolics in exocarp DCM fraction (EDF) and subfractions through GC-MS analysis.

ND: not detected. \* Indicates the phenolic was detected only in subfractions.

The total area percentage of phenolic antioxidants in the subfractions were lower than EDF, in the following decreasing order: EDF > Di-21 > Di-25 > Di-22 > Di-24 > Di-26. EDF contained the highest number of phenolic antioxidants (9 phenolic antioxidants), followed by Di-22 (8), Di-24 (8), Di-25 (8), Di-21 (4), and Di-26 (3). Vanillin was the major phenolic in EDF. Syringic acid and vanillic acid presented as the major phenolic in subfractions Di-21 and Di-22, respectively, while sinapic acid was the major phenolic in subfractions Di-24, Di-25, and Di-26.
