Antioxidant Properties and Characterization of Heterotrigona itama Honey from Various Botanical Origins according to Their Polyphenol Compounds

Stingless bee honey is a good source of antioxidants, which is attributed to the phenolic compounds. The type and concentration of phenolic compounds in honey can be aﬀected by botanical origin. Therefore, in this study, Heterotrigona itama honey from three botanical origins (gelam, acacia, and starfruit) was evaluated for its antioxidant activity and proﬁle of phenolic compounds. Apis mellifera honey was used as a comparison. Antioxidant activity and proﬁle of phenolic compounds in honey were determined using spectrophotometric and chromatographic methods, respectively. The total phenolic content (TPC), total ﬂavonoids content (TFC), free radical scavenging activity (IC 50 ), and ferric reducing antioxidant power (FRAP) of H. itama were ranged between 52.64 and 74.72mg GAE/100g honey, 10.70–25.71mg QE/100g honey, 11.27–24.09mg/mL, and 77.88–164.88 µ mol FeSO 4 .7H 2 O/100g honey, respectively. The ﬁndings showed that the antioxidant activity and phenolic and ﬂavonoid contents in H. itama honey were signiﬁcantly higher than Apis honey. Benzoic acid and taxifolin were found as the predominant phenolic acid and ﬂavonoid in all samples. However, chrysin was signiﬁcantly highest in Apis honey than stingless bee honey. This result suggested that chrysin can be used as a chemical marker to distinguish Apis honey from stingless bee honey. Gallic acid and ellagic acid were found as the chemical marker for gelam honey, salicylic acid, benzoic acid, and 4-hydroxybenzoic acid for starfruit honey while ferulic acid for acacia honey.


Introduction
Honey is produced by stingless bees and honeybees mainly from the nectar of plants or plant sap [1]. It has been documented that honey is a complex substance and contains over 200 components, which may be either produced during the maturation process of honey, added by bees, or derived from plants [2]. e main components of honey are sugar and water. It also comprises important minor components such as organic acids, amino acids, vitamins, minerals, enzymes, and phenolic compounds. e composition of honey and phenolic compounds primarily depends on the botanical origin [3,4] and external factors such as harvesting season, environment, storage, and processing method. Phenolic compounds are the most important antioxidant in honey that are responsible for the honey's therapeutic properties.
us, honey has been used in traditional and modern treatments to treat human illnesses. In modern treatment, honey has been used to treat diseases associated with oxidative stress such as diabetes mellitus, hypertension, atherosclerosis, cancer, and Alzheimer's [5]. In addition, phenolic compounds also can affect the organoleptic properties (color, taste, or flavor) of honey [2]. In a recent study, phenolic compounds have been used as a chemical marker to differentiate honey from different botanical and geographical origins [6]. For instance, abscisic acid was reported as a potential floral marker for two Polish unifloral honey [7], and quercetin was suggested as a marker for sunflower honey [8].
Phenolic compounds (phenolic acids and flavonoids) are the secondary components of plants [9] that transfer to honey through nectar, pollen, or propolis by bees [10]. ey are responsible for the most antioxidant activity in honey [11]. In recent years, there has been a growing interest in identifying phenolic compounds in stingless bee honey as it has been reported to have higher antioxidant activity and is considered as a potential source of antioxidants compared to Apis mellifera honey, which is good for human health and well-being [12]. Biluca et al. [13], da Silva et al. [14], Sousa et al. [15], and Olivera et al. [16] have identified numerous phenolic compounds in Melipona and Trigona honey. All the studies are from America Latin. However, in Malaysia, there is very limited scientific data on the profile of phenolic compounds in stingless bee honey. Ranneh et al. [17] investigated Malaysian stingless bee honey and Tualang honey. ey found gallic, caffeic, syringic, catechin, cinnamic, p-coumaric, apigenin, and 4-hydroxybenzoic as major compounds. Recently, Majid et al. [18] reported six phenolic compounds (chlorogenic acid, p-coumeric acid, epicatechin, rutin, catechin, and protocatechuic acid in unifloral and multifloral honey produced by Heterotrigona itama from Johor. Moreover, there are no available studies on phenolic compounds in stingless bee honey from acacia, gelam, and starfruit. erefore, more studies need to be conducted to explore and identify more phenolic acids and flavonoids present in stingless bee honey from different botanical origins. In this study, three types of stingless bee honey (acacia, gelam, and starfruit) that are commonly consumed in Malaysia were investigated. e main goal of this present study was to determine the phenolic profiles of stingless bee honey and Apis honey by identifying their phenolic acids and flavonoids, correlate antioxidant activity with phenolic compounds, and verify the botanical origin according to the specific phenolic compounds using multivariate statistical method.

Honey Samples.
Natural stingless bee honey produced by H. itama from acacia (Acacia mangium), starfruit (Averrhoa carambola L.), and gelam (Meleleuca cajaputi Powell) were obtained from three different farms that cultivated acacia, gelam, and starfruit trees located in Johor, Terengganu, and Pahang, respectively ( Figure 1). e samples were collected during the flowering season between August and November 2016. A. mellifera honey (acacia) was used as a comparison in this study. e botanical origin of the honey samples was determined based on the location of beehives and the availability of the floral source on the farm. e samples were collected using a vacuum pump and filtered to remove foreign materials. ey were stored in airtight containers and kept at 4°C until further analysis. e details of honey samples used are summarized in Table 1.

Total Phenolic Content.
e total phenolic content (TPC) was determined using Folin-Ciocalteu assay according to the method described by Garjanovic et al. [19] with some modifications. A 200 µL of honey solution (0.05 g/ mL methanol) was added to 1 mL of Folin-Ciocalteu reagent in a test tube. en, the mixture was allowed to stand at room temperature. After 6 min, 800 µL of sodium carbonate solution (7.5% Na 2 CO 3 ) was added to the mixture. en, the mixture was left to stand for 2 h in a dark room at room temperature. e absorbance of the mixture was measured at 740 nm using a UV-visible spectrophotometer (GENESYS ™ 10S, ermo Fisher Scientific, MA). e calibration curve was prepared using gallic acid at concentrations between 10 and 50 µg/mL (r 2 � 0.9999). Results were expressed as mg gallic acid equivalents (GAE) per 100 g. e analysis was performed in triplicate.   e total flavonoid content (TFC) was measured according to the method suggested by Kamboj et al. [20] with minor modifications. One milliliter of honey solution (0.25 g/mL methanol) was mixed with 0.3 mL of 5% sodium nitrite (NaNO 2 ). e mixture was left to stand for 5 min at room temperature. Afterward, 0.3 mL of 10% aluminum chloride (Al 3 Cl) was added, and the mixture was allowed to stand at room temperature for 6 min. Finally, 2 mL of 1 M of sodium hydroxide (NaOH) was added to the mixture. e absorbance of the mixture was measured at 510 nm using a UV-visible spectrophotometer (GENESYS ™ 10S, ermo Fisher Scientific, MA). Quercetin was used as a standard to plot the calibration curve at the concentrations of 10-50 µg/mL (r 2 � 0.98). e results were expressed as mg quercetin equivalent (QE) per 100 g. e analysis was carried out in triplicate.

Free Radical Scavenging Activity.
e scavenging activities of the honey samples were measured by DPPH assay as described by Meda et al. [21] with slight modifications. A 400 mg of honey sample was mixed with different amounts of methanol (3,6,9,12, and 15 mL) separately. en, 0.75 mL of honey in methanol was mixed with 1.5 mL of DPPH solution. e mixture was left in the dark area for 15 min at room temperature. e absorbance was measured against a blank solution containing water with a UV-visible spectrophotometer (A GENESYS TM 10S, ermo Fisher Scientific, MA) at 517 nm. e antiradical activity of the sample was calculated using the following formula: where A c is the absorbance of the control and A s is the absorbance of the sample. Each honey sample was analyzed in five dilutions. e antiradical activity was expressed as IC 50 (the concentration of honey solution needed to reduce the concentration of DPPH in the solution to 50% of its initial concentration). e concentration of honey sample required to scavenge 50% of DPPH (IC 50 ) was determined from the plotted graph of scavenging activity against the honey dilutions.

Ferric Reducing Antioxidant
Power. e reducing ability of the sample was carried out according to the procedure described by Khalil et al. [22] with slight modifications. e method was based on the reduction of Fe 3+ -TPTZ to a bluecolored Fe 2+ FRAP reagent. It was prepared by mixing 0.3 M of acetate buffer (pH 3.6), 10 mM of TPTZ, and 20 mM FeCl 3 .6H 2 O in a ratio of 10:1:1 at 37°C. A 200 µL of the honey sample (0.05 g/mL) in methanol was added to 1.5 mL of FRAP reagent in a test tube. After 10 min of incubation at 37°C in a water bath, the absorbance was measured using the UV-visible spectrophotometer (GENESYS ™ 10S, ermo Fisher Scientific, MA) at 593 nm. e antioxidant potential of the sample was determined from a standard curve using ferrous sulfate heptahydrate FeSO 4 .7H 2 O at concentrations between 10 and 100 µg/mL (r 2 � 0.98). Triplication of the test was performed, and the results were expressed as the mean average.

Extraction of Phenolic Compounds Using XAD-2 Resin.
e extraction of phenolic compounds was prepared according to the method described by Kassim et al. [23] and Ferreira et al. [24] with some modifications. A 10 g of honey sample was mixed with 100 mL of acidified water (pH 2; 0.02 M HCl) and 150 g of Amberlite XAD-2 resin (Fluka Chemei; pore size � 9 nm; particle size � 0.3-1.2 mm). e mixture was then homogenized for 10 min. e mixture was poured into a column (35 × 2 cm) and washed with 100 mL of acid solution (0.02 M HCl) to remove all sugars and other polar constituents in honey. Subsequently, 300 mL of deionized water was used to rinse the mixture in the column. e phenolic fractions were eluted with 100 mL of methanol and evaporated using a rotary evaporator until dry at 40°C. Methanol was used based on its ability to extract more phenolic compounds as reported by Kassim et al. [23]. e residues were dissolved in 250 µL methanol for HPLC analysis.

HPLC Analysis.
e detection of phenolic compounds was performed using high-performance liquid chromatography equipped with a diode array detector (HPLC-DAD). e column used was a reversed-phase C18 column, ACE (4.6 × 250 mm, particle size 5 µM) from Altmann Analytik GmbH & Co. KG (Munich, Germany). Approximately 10 µL of the phenolic extract was injected into the HPLC, and the separation was performed using 0.25% formic acid and 2% methanol in water (solvent A) and methanol (solvent B). e gradients used were: 10% methanol (B) for 15 min, 40% methanol (B) for 20 min, 45% methanol (B) for 30 min, 60% methanol (B) for 50 min, 80% methanol (B) for 52 min, 90% methanol (B) for 60 min, and followed by isocratic elution with 90% methanol (B) for 65 min. en, the gradient was changed to 10% methanol for 68 min until 73 min. e flow rate used was 1 mL/min. Phenolic compounds were detected using UV absorption spectra monitored at 290 nm and 340 nm. e phenolic compounds were identified by comparing the chromatographic retention times with the

Total Phenolic Content and Total Flavonoids Content.
e antioxidant capacity of honey is usually measured by evaluating the contents of phenolic and flavonoids as they are the key compounds responsible for the antioxidant activity of honey [13]. TPC values in the honey samples ranged between 52.64 and 74.62 mg GAE/100 g honey ( Table 2). A significant difference (p < 0.05) was observed in TPC values between H. itama (gelam and starfruit honey) and A. mellifera honey. Gelam honey showed the significantly highest value of TPC, while acacia honey obtained the lowest value. However, no significant differences (p > 0.05) were observed among H. itama honey from different botanical origins. In contrast, Abu Bakar et al. [25] demonstrated the TPC value of H. itama from Malaysia was higher than our results. e TPC values varied between 435.69 and 516.07 mg GAE/100 g honey. Recently, Imtiazah et al. [26] reported 368.11 mg GAE/100 g honey total phenolic content. However, Tufail Ahmad et al. [27] reported 5.86 mg GAE/ 100 g honey total phenolic content in Malaysian stingless bee honey produced by Geniotrigona thoracica, which was lower than our results. Ranneh et al. [17] also reported lower TPC in stingless bee honey collected from the forest in Kedah and Johor. e values varied between 228.09 and 235.28 mg GAE/kg honey. Ismail et al. [28] reported 33.2-60.2 mg GAE/100 g honey of total phenolic content, which agreed with our study. In comparison to other countries, the results found in this study were comparable to those found by Silva et al. [29] and da Silva et al. [14], where the TPC values were ranged between 1.30 and 66 mg GAE/100 g honey for stingless bee honey from Paraiba and Amazona, Brazil. Furthermore, Silva et al. [29] also reported the higher TPC values in nine Melipona subnitida honey from two semiarid regions, Paraiba, Brazil.
e TFC values of H. itama honey and Apis honey are shown in Table 2. e highest TFC value (25.71 mg QE/100 g honey) was found in starfruit honey, followed by gelam (20.67 mg QE/100 g honey), acacia (10.70 mg QE/100 g honey), and Apis (7.02 mg QE/100 g honey) honey. A significant difference (p < 0.05) was observed in all honey samples investigated. In comparison with Malaysian stingless bee honey, Ranneh et al. [17] and Selvaraju et al. [30] reported lower TFC where the values ranged between 97.88 and 101.5 mg CE/kg and 10.18 and 12.68 mg CE/kg, respectively. In a recent study, Imtiazah et al. [26] also demonstrated a lower TFC in Malaysian stingless bee honey (64.25 mg CEQ/kg honey). However, higher TFC was found in multifloral stingless bee honey collected in Sabah and Kelantan with values between 43.2 and 65.9 mg QE/100 g honey [28]. e results from this study were consistent with those reported for stingless bee honey from Alagoas, Brazil, 11.69-49.50 mg QE/100 g honey [31]. Nevertheless, a higher TFC value was found in six species of Melipona honey from Brazil [32] and a lower TFC value in Melipona beecheii honey from Cuba [11].
Overall, stingless bee honey had significantly higher TPC and TFC as compared to Apis honey. A low concentration of phenolic acids and flavonoids may contribute to the low TPC and TFC values of Apis honey as shown in Table 3. In addition to that, the difference can be explained by the different bee species and the way honey is produced by two different bees [33]. While the variation observed between stingless bee honey is due to the different botanical origins of the honey as the botanical origin has a direct association with the phenolic compounds present in honey [34]. Comparison of TPC and TFC data with similar honey from Malaysia and other regions could not be done as there is no related publication available.

e Free Radical Scavenging Activity and Ferric Reducing Antioxidant
Power. Antioxidant activities of honey samples were estimated using 2,2-diphenyl-1-picrylhydrazyl (DPPH) and ferric reducing antioxidant power (FRAP) assays. e free radical scavenging activity of honey samples was expressed as IC 50 , indicating the amount of antioxidant needed to reduce the initial concentration of DPPH solution by 50%. Honey with a low value of IC 50 has greater antioxidant activity than honey with a high value of IC 50 [35]. Results of the free radical scavenging activity of honey samples are presented in Table 2. e IC 50 value of H. itama honey varied between 11.27 and 24.09 mg/mL with the gelam honey having the lowest value. us, gelam honey had greater antioxidant activity than starfruit and acacia honey, but no significant difference (p > 0.05) was observed among them. Aljadi and Kamaruddin [36] also reported that gelam honey (honeybee honey) had higher antioxidant activity than coconut honey. Kek et al. [37] reported 26.63 mL/g in Malaysian stingless bee honey produced by H. itama. While Ismail et al. [28] reported a range of 10.6-19.7 mg/mL, which was consistent with our results. When compared to the stingless bee honey from Brazil, the IC 50 value in this study was lower than the results reported (25.39-51.55 mg/mL). While the IC 50 value of H. itama honey (11.27-24.09 mg/ mL) was lower than Apis honey (53.65 mg/mL). A similar finding was reported by Duarte et al. [31]. However, there was no significant difference found among the honey samples except between gelam and Apis honey. In contrast, Alvarez-Suarez et al. [12] reported a higher IC 50 in M. beecheii honey than Apis honey. e ferric reducing antioxidant power (FRAP) analysis was performed to measure the ability of phenolic compounds as an antioxidant to reduce ferric ions (Fe 3+ ) to ferrous ions (Fe 2+ ). From the results in Table 2 [40] presented in the honey samples as the samples used in this study belonged to different botanical origins. erefore, honey from different botanical origins possesses distinct antioxidant activity [14].

Correlation Analysis.
e Pearson's correlation showed a significant correlation between FRAP and TFC (0.991).
is indicates that flavonoids are the greatest contributor to the reducing power of honey. Nevertheless, no significant correlation was observed between TPC and TFC (0.922), between DPPH (IC 50 ) and TPC (−0.813) and TFC (−0.658), and between FRAP and TPC (0.919) and DPPH (IC 50 ) (−0.588). e findings from this study also suggest that the antioxidant activity of honey is not only influenced by the phenolic and flavonoids content but also other antioxidant compounds that partially contribute to the antioxidant activity of honey. As known, honey contains a wide variety of significant antioxidant compounds such as organic acids, amino acids, enzymes, carotenoids, Maillard reaction products, and proteins [34,41,42]. Duarte et al. [30] also reported a significant correlation between the FRAP and flavonoids content in Africanized and stingless bee honey from Alagoas, Brazil. Other studies also demonstrated a significant correlation between the FRAP and flavonoids content, suggesting flavonoids as one of the key components responsible for the antioxidant activity of honey [22,40,43,44].

Limit of Detection and Limit of Quantitation of Phenolic
Compounds.
e limit of detection (LOD) and limit of quantitation (LOQ) values were calculated based on the 3 * standard deviation of blank response/slope and 10 * standard deviation of blank response/slope, respectively.

Phenolic Profile.
Phenolic compounds identification was performed using HPLC analysis. e chromatograms of the standard of phenolic compounds and honey samples are shown in Figures 2(a) and 2(b), respectively. Table 4 shows the phenolic compounds present in all honey samples. Sixteen compounds were found in starfruit honey, while 15 compounds were identified in acacia and gelam honey.
However, stingless bee honey (248.69-1,212.70 µg/100 g honey) exhibited significantly higher taxifolin content than Apis honey (76.50 µg/100 g honey). Our result was in accordance with those previously found in Amazon honey from M. (Michmelia) seminigra merrillae [14]. On the other hand, luteolin was found as the predominant flavonoid in Melipona asilvai and Melipona quadrifascita honey; kaempferol in Melipona anthidioides, Melipona scutellaris, and M. subnitida honey; and apigenin in Melipona mandacaia honey from Sergipe state, Brazil [32]. Different types and concentrations of phenolic compounds in honey were reported from different studies, which might be attributed to the different botanical origins, bee species, and geographical locations of honey.
In this present study, ferulic acid, salicylic acid, ellagic acid, benzoic acid, 4-hydroxybenzoic acid, taxifolin, myricetin, quercetin, and naringenin were detected in stingless bee honey. To the best of our knowledge, this is the first report of the presence of these compounds in Malaysian stingless bee honey. e findings revealed that stingless bee honey contains a wide range of phenolic compounds. is finding could have a significant impact on the stingless bee industry sustainability as well as promote Malaysian stingless bee honey globally.
Overall, stingless bee honey exhibited more types and higher concentrations of phenolic compounds compared to Apis honey. is suggests that stingless bee honey is a good source of natural phenolic acids and flavonoids. ese findings are consistent with those reported by Alvarez-Suarez et al. [12] and Guerrini et al. [46]. Phenolic acids and flavonoids have been proven to have antioxidant activity, which has a significant correlation with honey's medicinal properties, and each of them has specific health benefits [47]. Furthermore, stingless bee honey has the potential in treating eye illnesses, wounds, cancer, diabetes mellitus, Table 4: e profile and concentration of phenolic compounds (µg/100 g honey) in Heterotrigona itama honey from different botanical origins (acacia, gelam, and starfruit) and Apis mellifera (acacia) honey.
hypertension, microbial infection, fertility problems, and dysregulated lipid profiles [47]. is indicates that stingless bee has a great potential to be commercialized and consumed by a large population. Acacia, gelam, and starfruit honey can be a good choice as an antioxidant source for humans. Since they have high antioxidant content than Apis honey, we believe that stingless bee honey has a better effect on human health.

Multivariate Analysis.
Multivariate analysis was performed to determine potential phenolic compounds, which can be used as a chemical marker to distinguish stingless bee honey from Apis honey. e principal component analysis (PCA) was used to identify honey based on the concentration of phenolic compounds. To establish the best parameters for classifying honey samples, two principal components with eigenvalues greater than one were extracted. e first principal component (PC1) accounted for 58.6% of the variance, while the second principal component (PC2) accounted for 26.3%. ese two components explained 84.9% of the variation in the data. Based on the score plot (Figure 3), all honey samples were scattered away from each other, indicating that all honey samples contained different types and concentrations of phenolic compounds. From the score plot, Apis honey was located far away from the stingless bee honey, proving that Apis honey was significantly different from the stingless bee honey in terms of the phenolic compounds. is finding suggests that phenolic compounds can distinguish Apis honey from stingless bee honey. A biplot was further created to display the relationship between honey samples and phenolic compounds (Figure 4). en, a loading plot was performed to assess which phenolic compounds contributed significantly to  the separation ( Figure 5). According to the PC2, chrysin was the only phenolic compound responsible to separate Apis honey from stingless bee honey. As a result, chrysin can be used as a chemical marker to distinguish Apis honey from stingless bee honey.
To study the potential to differentiate acacia, gelam, and starfruit honey, data of Apis honey were removed, and principal component analysis (PCA) was constructed. e first principal component (PC1) accounted for 71.2% of the variance, while the second principal component (PC2) accounted for 27.0%. ese two components explained 98.2% of the variation in the data. Based on the score plot ( Figure 6), all stingless bee honey were well separated, indicating that all honey samples were significantly different in phenolic compounds content. e most influential phenolic compounds responsible for clustering to be identified as a potential marker for all honey samples were determined using biplot (Figure 7(a)) and loading plot (Figure 7(b)). Based on the loading plot, gallic acid and ellagic acid have a strong impact on the  separation of gelam honey. Salicylic acid, benzoic acid, and 4-hydroxybenzoic acid are responsible for the separation of starfruit honey, while ferulic acid is the only one that separated acacia from other honey samples.

Conclusion
is study discovered that botanical origin significantly influences antioxidant activity and the composition of phenolic compounds in honey. Flavonoids were shown to have a significant relationship with the antioxidant activity (FRAP) of honey. In addition, HPLC analysis indicated and validated that stingless bee honey is rich in phenolic components as compared to Apis honey, showing that stingless bee honey is a valuable source of natural phenolic acids and flavonoids. Furthermore, the findings suggest that the botanical origin has a significant impact on the content of phenolic compounds in honey. A multivariate data analysis was used to find a potential marker that might be used to distinguish honey samples based on their phenolic compounds. e results showed that chrysin successfully distinguished the stingless bee honey from Apis honey. Gelam honey was discriminated by gallic acid and ellagic acid, starfruit honey by salicylic acid, benzoic acid, and 4-hydroxybenzoic acid, while acacia honey by ferulic acid. However, the number of samples used in this study was limited.
us, a further investigation with a large number of samples is required in order to confirm the present results.

Data Availability
Data used to support the findings of this study are included in the article.

Conflicts of Interest
e authors declare that there are no potential conflicts of interest.