Antioxidant Activity and Phenolic Compound Identification and Quantification in Western Australian Honeys

This study reports on the total phenolic content and antioxidant activity as well as the phenolic compounds that are present in Calothamnus spp. (Red Bell), Agonis flexuosa (Coastal Peppermint), Corymbia calophylla (Marri) and Eucalyptus marginata (Jarrah) honeys from Western Australia. The honey’s total phenolic content (TPC) was determined using a modified Folin–Ciocalteu assay, while their total antioxidant activity was determined using FRAP and DPPH assays. Phenolic constituents were identified using a High Performance Thin-Layer Chromatography (HTPLC)-derived phenolic database, and the identified phenolic compounds were quantified using HPTLC. Finally, constituents that contribute to the honeys’ antioxidant activity were identified using a DPPH-HPTLC bioautography assay. Based on the results, Calothamnus spp. honey (n = 8) was found to contain the highest (59.4 ± 7.91 mg GAE/100 g) TPC, followed by Eucalyptus marginata honey (50.58 ± 3.76 mg GAE/100 g), Agonis flexuosa honey (36.08 ± 4.2 mg GAE/100 g) and Corymbia calophylla honey (29.15 ± 5.46 mg GAE/100 g). In the FRAP assay, Calothamnus spp. honey also had the highest activity (9.24 ± 1.68 mmol Fe2+/kg), followed by Eucalyptus marginata honey (mmol Fe2+/kg), whereas Agonis flexuosa (5.45 ± 1.64 mmol Fe2+/kg) and Corymbia calophylla honeys (4.48 ± 0.82 mmol Fe2+/kg) had comparable FRAP activity. In the DPPH assay, when the mean values were compared, it was found that Calothamnus spp. honey again had the highest activity (3.88 ± 0.96 mmol TE/kg) while the mean DPPH antioxidant activity of Eucalyptus marginata, Agonis flexuosa, and Corymbia calophylla honeys were comparable. Kojic acid and epigallocatechin gallate were found in all honeys, whilst other constituents (e.g., m-coumaric acid, lumichrome, gallic acid, taxifolin, luteolin, epicatechin, hesperitin, eudesmic acid, syringic acid, protocatechuic acid, t-cinnamic acid, o-anisic acid) were only identified in some of the honeys. DPPH-HPTLC bioautography demonstrated that most of the identified compounds possess antioxidant activity, except for t-cinnamic acid, eudesmic acid, o-anisic acid, and lumichrome.

Individual honey samples were authenticated based on their HPTLC fingerprints following established protocols (see Figures S1-S4) [37,38,40] and based on this authentication, a pooled sample for each honey was prepared by mixing equal amounts of each individual sample from the same floral source. It was deemed that such a pooled sample would better reflect the typical chemical composition of a honey rather than analysing an individual honey with a chemical profile that specifically mirrors its unique location, time of collection and processing [40]. Therefore, the pooled honey samples were used in this study for constituent identification and quantification.
An artificial honey solution was prepared by mixing 21.625 g of fructose, 18.125 g of glucose, 1.000 g of maltose, 0.750 g of sucrose and 8.500 g of water [41].

Preparation of Honey Samples
For the total phenolic content analysis and antioxidant analyses, individual honeys were prepared in triplicates as 20% w/v aqueous solutions while for phenolic identification and quantification experiments, pooled honeys were extracted using an organic solvent. The extraction process involved adding 1 g of each pooled honey sample to 2 mL deionised water in stoppered glass test tubes followed by vortex mixing. The resulting solution was then extracted three times with 5 mL dichloromethane and acetonitrile (1:1, v/v). The combined organic extracts were dried using anhydrous MgSO 4 , filtered, and evaporated to dryness using a heating block (Stuart SBHCONC/1 Sample Concentrator) set at 35 • C. The organic honey extracts were stored at 4 • C until analysis for which they were reconstituted with 100 µL methanol.

Determination of Total Phenolic Content (TPC)
The TPC assay was performed based on the methodology described by Liberato et al. with minor modifications [42]. This protocol has previously been employed in the analysis of the TPC of some Western Australian bee products [37,38,41,43].
In brief, 200 µL of aqueous honey solution (20%, w/v) or 100 µL of gallic acid standards (0.06 mg/mL to 0.18 mg/mL) spiked with 100 µL of artificial honey solution (40%, w/v) [39] were reacted with 1 mL of diluted Folin-Ciocalteu reagent (1 mL of Folin-Ciocalteu reagent in 30 mL deionised water). After 5 min, 800 µL of 0.75% Na 2 CO 3 was added and allowed to react for 2 h, excluded from light. Sample absorbance at 760 nm was then measured (Carry 60 Bio UV-Vis spectrophotometer) using 100 µL of deionised water spiked with 100 µL of artificial honey solution (40%, w/v), 1 mL of Folin-Ciocalteu reagent and 800 µL of 0.75% Na 2 CO 3 as a blank. The analysis was carried out in triplicate and the mean result for each sample was expressed as mg gallic acid equivalent (GAE) per 100 g of honey.

Determination of Antioxidant Activity Using the Ferric Reducing Antioxidant Power (FRAP) Assay
The FRAP assay, which is based on the reduction of ferric 2,4,6-tris(2-pyridyl)-1,3,5triazine [Fe(III)-TPTZ] to ferrous complex at low pH followed by a spectrophotometric analysis, was performed according to the protocol described by Almeida et al. [44] with minor modifications. This protocol has previously been used in our laboratory to determine the FRAP activity of various bee products [37,38,41,43].
In brief, a 1:1:10 (v/v/v) ratio of the FRAP reagent was prepared by mixing 10 mM TPTZ (dissolved in 40 mM HCl), 20 mM aqueous FeCl 3 ·6H 2 O and 300 mM aqueous acetate buffer (pH 3.6). The reagent mixture was freshly prepared prior to each experiment and incubated at 37 • C prior to use. Ferrous sulphate (FeSO 4 ·7H 2 O) standards ranging from 200 µM to 1200 µM, along with the standard concentration of 600 µM which was used as a positive control, were freshly prepared prior to each experiment and stored on ice.
A total of 20 µL of honey solution or standards were mixed with 180 µL of FRAP reagent in a 96-well microplate (Greiner Bio-One 96-well Microplate Flat Bottom), and the absorbance of the reaction mixture after 30 min of incubation at 37 • C was determined at 620 nm (BMG Labtech POLARstar Optima Microplate Reader). The FRAP antioxidant activity was determined based on the interpolation of the standard curve and expressed as mmol Fe 2+ equivalent (FE)/kg of honey (mean of triplicate results). The DPPH assay in this study was based on the protocol described by Karabagias et al. [45] with minor modifications [37,38,41,43]. The radical 2,2-diphenyl-1-picrylhydrazyl (DPPH) is purple in colour and decays to yellow in the presence of antioxidants. The resulting change can be captured at 520 nm. The DPPH reagent mixture was prepared using a ratio of 19:10 (v/v) of 0.130 mM methanolic DPPH solution and 100 mM pH 5.5 aqueous NaC 2 H 3 O 2 buffer. Aqueous Trolox solutions with concentrations ranging from 100-600 µM (pH adjusted to pH 7.0) were used to derive the calibration curve, with the 400 µM standard also serving as a positive control.
A total of 10 µL of aqueous honey solution or standards were placed in a 96-well microplate, followed by 290 µL of DPPH reagent, and then mixed. The reaction mixture was kept in the dark and the absorbance was measured at 520 nm after 120 min using a microplate reader (Greiner Bio-One 96-well Microplate Flat Bottom). The mean radical scavenging activity of triplicate samples of honey solutions or standards was expressed as Trolox Equivalent (TE), calculated based on the interpolation of the standard curve, and for the honey samples then also expressed as µmol Trolox equivalent per kg of honey.

Phenolic Constituent Identification in Honey
The identification of phenolic honey constituents was performed using a validated HPTLC based database of phenolic compounds. In brief, honeys were first fingerprinted using HPTLC under various conditions and the resulting data (i.e., Rf values, colour hues, UV-Vis and fluorescence λmax and λmin prior to derivatisation, UV-Vis and fluorescence λmax after derivatisation) were matched with standards included in the database [30]. Potential matches were confirmed by spectral overlay analysis [30].
In this study, as additional confirmation of correct identification, a mixture of the identified compounds (7.4 µL, for concentrations see Table 1) in each honey was used to over-spot the respective neat honey extract (7 µL). A corresponding increase in the absorbance of the respective honey extract bands was seen as confirmation of the correct identification.
The CAMAG HPTLC system (Muttenz, Switzerland) used in this study consisted of a CAMAG TLC visualizer 2, Linomat V semi-automatic sample applicator, and ADC2 automated development chamber, a TLC scanner IV, a derivatiser, and a TLC plate heater III. The system was operated by VisionCATS Version 3.1 software, which controls all chromatographic operations and analyses.
In order to perform the phenolic compound identification, honey extracts were subjected to the same HPTLC conditions used to establish the database ( Table 2) using two solvent systems: (a) MPA, consisting of toluene: ethyl acetate: formic acid (2:8:1, v/v/v), and (b) MPB, consisting of toluene: ethyl acetate: formic acid (6:5: [44,45], as well as two different derivatising reagents, natural product-polyethylene glycol reagent (NP-PEG) and vanillin-sulfuric acid reagent (VSA). Naringenin (0.5 mg/mL in methanol), with an application volume of 4 µL, was used as HPTLC reference standard, and for all honey extracts a volume of 7 µL was used. All samples were applied as 8 mm bands, 8 mm from the bottom of the HPTLC plate at a rate of 150 nLs −1 (aided by liquid nitrogen at a pressure of 10,000 mmHg). The chromatographic separation was performed on 20 × 10 cm HPTLC plates (glass-backed silica gel 60 F 254 plates) in an automated twin trough development chamber activated with MgCl 2 ·6H 2 O at 33-38% relative humidity. Saturation pads were used to saturate the system for 15 min and plates were preconditioned with the mobile phase for 5 min, and then developed automatically to a distance of 70 mm at room temperature before being automatically dried for 5 min. Photo-documentations under 254 nm, 366 nm, and white light in transmittance mode (T) were performed on the developed plates in order to detect the separated honey constituents. From this information corresponding peak profiles were generated, and major peaks automatically determined by the software.
The scanning of individual major bands in the honey extracts was carried out using the TLC Scanner 4 in both UV-Vis mode (190-900 nm) and fluorescence mode (190-380 nm) with the following settings: Dimension set at 5 × 0.2 mm (micro), optimisation set for maximum resolution, scanning speed 20 nm/s and use of K400 optical filter. Deuterium (190-380 nm) and tungsten (380-900 nm) were used as lamps and the scans in fluorescence excitation mode were set at 380 < /400 nm and the emissions were observed at 190-270 nm.
Three spectral scans were performed for each sample, prior to and after derivatisation with each of the derivatisation reagents used.
To perform the derivatisation of the plates with NP-PEG reagent, plates were first sprayed with 3 mL of 1% NP reagent using a green nozzle at level 3 and then allowed to dry for 5 min at 40 • C. The plates were then sprayed again, this time with 5% PEG reagent using a blue nozzle at level 2, dried for 5 min at 40 • C and the resulting image was captured at 366 nm [29]. To derivatise using VSA reagent, plates were sprayed with 3 mL of 1% vanillin sulphuric acid reagent using a yellow nozzle at level 3, and then heated for 3 min at 115 • C for 3 min, and after cooling for 2 min, the plates were visualised at 366 nm and T white light.
A system suitability test (SST) was performed for each plate analysis as a quality control step. This was performed by utilising the Rf and the minimum height of the reference sample (naringenin) prior to derivatisation at 254 nm and only those plates that passed the set threshold of ±0.05 for the Rf and the minimum height for MPA (Rf 0.690, minimum height 0.108) and MPB (Rf 0.550, minimum height 0.120) were used in the qualitative and quantitative analysis.

Quantification of Phenolic Compounds in Honey
The same chromatographic instrumentation and parameters as described in Section 2.7 were employed in the quantification of the identified phenolic compounds in the various honey samples. Standard concentrations, application volumes, derivatisation and scanning conditions were optimised. The optimised application volumes for the various standards ranged from 5.0 to 9.8 µL (1.2 µL interval) and each compound was quantified at its specific λmax using the evaluation feature of the VisionCATS software. Table 1 summarises the key parameters for the standards used in the quantification experiments.

HPTLC-DPPH Antioxidant Activity
The same chromatographic instrumentation and parameters as described in Section 2.7 were also employed to perform the HPTLC-DPPH analysis for antioxidant activity in the honey extracts and their respective matched constituents. Seven microliters of each honey extract were used for the analysis alongside the standards in varying volumes. After development, the plates were derivatised with 3 mL of 0.4% DPPH solution (1:1 ratio of methanol and water) using the yellow nozzle and sprayed at level 1 [32,33]. Plate images were obtained at transmittance in white light after 1 h, 2 h and 3 h. Peak profiles at 517 nm were also generated and from these the Rf values of the respective peaks were generated. Each band's colour in the form of RGB values was determined and then converted into corresponding hue values [30]. Compounds that possess antioxidant activity will quench the DPPH radical either by electron transfer or hydrogen atom transfer through radical attack, which is observed as a discoloration at 517 nm due to the formation of 2,2-diphenyl-1-hydrazine or a substituted analogue hydrazine [33]. Gallic acid was used as positive control, its quenching activity resulting in a maximum hue value of 40 • (yellow colour). All obtained hue values were calculated using previously reported formula [30]. The DPPH radical scavenging activity (% DPPH RSA) of a band of interest was calculated as follows: where The respective band's % DPPH RSA was then categorised as indicated in Table 3. The DPPH antioxidant activity of luteolin, epicatechin, epigallocatechin gallate, gallic acid, protocatechuic acid, m-coumaric acid and kojic acid was analysed at varying concentrations to validate the bioautographic analysis. Furthermore, the DPPH antioxidant activity of the matched compounds was determined at low and high concentrations to determine their inherent antioxidant activity.

Statistical Analysis
Analysis of variance (ANOVA) was performed using Graphpad Prism 9 (GraphPad Software, San Diego, CA, USA) in order to determine whether there was a significant difference in the total phenolic content, FRAP activity, and DPPH antioxidant activity of different honeys. Tukey's post hoc comparisons were used to identify differences between the groups (p < 0.05).  Table S2 (Supplementary Materials). The average TPC value for Red Bell honey (n = 8) was found to be the highest (59.4 ± 7.91 mg GAE/100 g), followed by Jarrah honey (50.58 ± 3.76 mg GAE/100 g), Coastal Peppermint honey (36.08 ± 4.2 mg GAE/100 g) and Marri honey (29.15 ± 5.46 mg GAE/100 g). The average TPC of the four honeys differed significantly when analysed using One way ANOVA (p < 0.0001). Tukey's post hoc analysis demonstrated that Red Bell honey had higher TPC than the other three honeys (Coastal Peppermint and Marri honey (p = <0.0001), Jarrah honey (p = 0.0407) while Jarrah honey also showed higher TPC compared to Coastal Peppermint (p = 0.0016) and Marri (p = <0.0001) honeys). No difference, however, was observed when the mean TPC values of Coastal Peppermint and Marri honeys were compared (see Figure 2).   Table 4 shows the average FRAP antioxidant activity of the investigated Western Australian honeys, expressed as mmol Fe 2+ equivalent/kg. Based on the analysis of the 32 individual samples tested, mean FRAP activity was 6.26 and ranged from 3.47 (Corymbia calophylla (Marri) honey) to 11.66 (Calothamnus spp. (Red Bell) honey). The FRAP antioxidant activity of individual honeys is shown in Table S2 (Supplementary Materials). When the means of each honey type were analysed, it was found that Calothamnus spp. (Red Bell) honey had the highest activity, followed by Eucalyptus marginata (Jarrah), whereas Agonis flexuosa (Coastal Peppermint) and Corymbia calophylla (Marri) honey had comparable FRAP activity. One way ANOVA analysis demonstrated a significant difference (p = <0.0001) between the means of the honeys and post hoc analysis showed that the average Red Bell honey's FRAP activity was higher than that of Jarrah honey (p = 0.01030), Coastal Peppermint honey (p = 0.0001), and also Marri honey (p = <0.0001). The FRAP activity of Jarrah honey was also found to be higher than that of Marri honey (p = 0.0058), whereas Coastal Peppermint and Marri honeys had comparable average FRAP antioxidant activities (p = 0.5061) (see Figure 2). In line with findings reported by others [46][47][48][49], a high  Table 4 shows the average FRAP antioxidant activity of the investigated Western Australian honeys, expressed as mmol Fe 2+ equivalent/kg. Based on the analysis of the 32 individual samples tested, mean FRAP activity was 6.26 and ranged from 3.47 (Corymbia calophylla (Marri) honey) to 11.66 (Calothamnus spp. (Red Bell) honey). The FRAP antioxidant activity of individual honeys is shown in Table S2 (Supplementary Materials). When the means of each honey type were analysed, it was found that Calothamnus spp. (Red Bell) honey had the highest activity, followed by Eucalyptus marginata (Jarrah), whereas Agonis flexuosa (Coastal Peppermint) and Corymbia calophylla (Marri) honey had comparable FRAP activity. One way ANOVA analysis demonstrated a significant difference (p = <0.0001) between the means of the honeys and post hoc analysis showed that the average Red Bell honey's FRAP activity was higher than that of Jarrah honey (p = 0.01030), Coastal Peppermint honey (p = 0.0001), and also Marri honey (p = <0.0001). The FRAP activity of Jarrah honey was also found to be higher than that of Marri honey (p = 0.0058), whereas Coastal Peppermint and Marri honeys had comparable average FRAP antioxidant activities (p = 0.5061) (see Figure 2). In line with findings reported by others [46][47][48][49], a high correlation (0.912) was observed between FRAP antioxidant activity and TPC, indicating that the antioxidant activity of these honeys is strongly related to their phenolic constituents. Table 4 shows the average DPPH radical scavenging activity of the investigated Western Australian honeys, expressed as mmol TE/kg honey. Based on the results of the analysed 32 samples, a mean radical scavenging activity of 2.44 was found, ranging from 1.01 (Corymbia calophylla (Marri) honey) to 5.41 (Calothamnus spp. (Red Bell) honey). The DPPH radical scavenging activity of individual honey samples is shown in Table S2 (Supplementary Materials). When the mean values were compared, it was found that Red Bell honey had the highest activity. One way ANOVA analysis found a significant difference (p = 0.0001) amongst the means of the different honeys and post hoc analysis demonstrated that the mean DPPH antioxidant activity of Red Bell honey was higher when compared to Eucalyptus marginata (Jarrah) (p = 0.0026), Agonis flexuosa (Coastal Peppermint) (p = 0.0053), and Marri honey (p = <0.0001). Jarrah, Coastal Peppermint and Marri honey, have, however, comparable DPPH radical scavenging activities (p = >0.05) (see Figure 2). A high correlation (0.832) between DPPH antioxidant activity and TPC values of the individual honeys was observed, confirming that phenolic constituents contribute to honey's antioxidant activity. Furthermore, a high correlation (0.948) between DPPH and FRAP antioxidant activity was also observed.

Phenolic Compound Identification
The phenolic compound identification was carried out based on a previously reported database filtering approach [30]. The summary of the data (as described in Section 2.7) used to determine the identity of various phenolic constituents in the four investigated pooled honey samples is shown in Tables S3-S18 (Supplementary Materials). In addition, the identified candidate compounds for each significant band in the four different Western Australian honeys are shown in Table 5 along with correlations and percent match data based on the spectral overlays of four different UV-Vis spectra of the unknown and the candidate match compounds (254 nm and 366 nm prior to derivatisation, and 366 nm after derivatisation with VSA and NP-PEG reagents).
Based on the results obtained using database 1A and 1B ( Figure 3A), the compound at Rf 0.570 in Calothamnus spp. (Red Bell) honey was identified as protocatechuic acid (10) as shown by the similarity of the spectral overlays of the unknown band and the standard when analysing their UV-Vis spectra prior to derivatisation ( Figure 4A,B), after derivatisation with NP-PEG ( Figure 4C,D), and also after derivatisation with VSA ( Figure 4E,F). The unknown band at Rf 0.423 in Red Bell honey was identified as epigallocatechin gallate (5), and the unknown band at Rf 0.226 as kojic acid (14). By employing database 2A and 2B ( Figure 3B), which utilised a less polar solvent, the unknown band at Rf 0.550 in Red Bell honey was identified as t-cinnamic acid (13), the band at Rf 0.380 as protocatechuic acid (10), the band at Rf 0.270 as gallic acid (8), and the band at Rf 0.115 as kojic acid (14) (see Figure 5 for structures).  Bell honey was identified as t-cinnamic acid (13), the band at Rf 0.380 as protocatechuic acid (10), the band at Rf 0.270 as gallic acid (8), and the band at Rf 0.115 as kojic acid (14) (see Figure 5 for structures).    All other compounds reported here were identified in the three other honey samples in the same manner. Table 2 summarises the identified honey constituents. Figure 3C,D All other compounds reported here were identified in the three other honey samples in the same manner. Table 2 summarises the identified honey constituents. Figure 3C,D shows the identified compounds in Agonis flexuosa (Coastal Peppermint) honey. Figure 6A,B summarises the identified compounds in Corymbia calophylla (Marri) honey, while Figure 6C,D summarises the identified compounds in Eucalyptus marginata (Jarrah) honey. A comparison between the peak profile of each honey and the respective honey overspotted with a mixture of its identified constituents was also used to further confirm the phenolic compound determination. For confirmation, scans were performed, for example, at each specific λmax of each identified compound in Calothamnus spp. (Red Bell) honey ( Figure 7A-D) using databases 2A and 2B) and based on this analysis, the Rf of the matched compounds were found to be similar to that of the identified bands in the honey. Moreover, an increase in the absorbance confirmed the presence of the compounds in the honey. Profile comparisons for the other investigated honeys are included in the Supplementary Materials (Figures S5-S11). A comparison between the peak profile of each honey and the respective honey overspotted with a mixture of its identified constituents was also used to further confirm the phenolic compound determination. For confirmation, scans were performed, for example, at each specific λmax of each identified compound in Calothamnus spp. (Red Bell) honey ( Figure 7A-D) using databases 2A and 2B and based on this analysis, the Rf of the matched compounds were found to be similar to that of the identified bands in the honey. Moreover, an increase in the absorbance confirmed the presence of the compounds in the honey. Profile comparisons for the other investigated honeys are included in the Supplementary  Materials (Figures S5-S11).

Phenolic Compound Quantification
Optimised parameters, such as standard concentrations, application volumes, mode in obtaining the profile/chromatogram, and derivatisation for quantification of the identified phenolic compounds in the four pooled honey samples, are shown in Table S19 (Supplementary Materials). Based on the findings of the optimisation, it was concluded that standard concentrations ranging from 5 µ g/mL to 50 µ g/mL, application volumes ranging from 5.0 to 9.8 µ L (1.2 µ L interval), peak profiles obtained by scanning the plate at the respective specific λmax, and the absence of any derivatisation constituted the best approach for quick and accurate quantification of phenolic compounds in the honey matrices. Linearity was observed to be greater than 0.99 for each standard and the percent recovery ranged from 95.2 to 102.6%. Table 2 details the standard concentrations, linearity, and % recovery of each identified constituent that was used in the quantification experiment. Furthermore, a sample of an HPTLC plate and its corresponding peak profile used in the quantification of phenolic compounds in Calothamnus spp. (Red Bell) honey is shown in Figure 8A-C.

Phenolic Compound Quantification
Optimised parameters, such as standard concentrations, application volumes, mode in obtaining the profile/chromatogram, and derivatisation for quantification of the identified phenolic compounds in the four pooled honey samples, are shown in Table S19 (Supplementary Materials). Based on the findings of the optimisation, it was concluded that standard concentrations ranging from 5 µg/mL to 50 µg/mL, application volumes ranging from 5.0 to 9.8 µL (1.2 µL interval), peak profiles obtained by scanning the plate at the respective specific λmax, and the absence of any derivatisation constituted the best approach for quick and accurate quantification of phenolic compounds in the honey matrices. Linearity was observed to be greater than 0.99 for each standard and the percent recovery ranged from 95.2 to 102.6%. Table 2 details the standard concentrations, linearity, and % recovery of each identified constituent that was used in the quantification experiment. Furthermore, a sample of an HPTLC plate and its corresponding peak profile used in the quantification of phenolic compounds in Calothamnus spp. (Red Bell) honey is shown in Figure 8A  By utilising the optimised conditions, the concentration of the compounds identified in honey ranged from 0.003 µg/g (t-cinnamic acid (13)) to 13.49 µg/g (2,3,4-trihydroxy benzoic acid (6)) (see Table 6 for the specific quantities). Protocatechuic acid (10) was found to be the most abundant constituent in Calothamnus spp. (Red Bell) honey, followed by epigallocatechin gallate (5), kojic acid (14), gallic acid (8), and t-cinnamic acid (13). In the case of Agonis flexuosa (Coastal Peppermint) honey, epicatechin (4) was found to be the most abundant, followed by kojic acid, epigallocatechin gallate, lumichrome (15), syringic acid (11), luteolin (1), and m-coumaric acid (12). For Corymbia calophylla (Marri) honey, gallic acid was found to be the most abundant compound, followed by epigallocatechin gallate, eudesmic acid (7), epicatechin, luteolin, taxifolin (3), kojic acid, and m-coumaric acid (12). 2,3,4-Trihydroxy benzoic acid (6) was found to be the most abundant compound in Eucalyptus marginata (Jarrah) honey, followed by epicatechin, o-anisic acid (9), taxofolin, lumichrome, m-coumaric acid, kojic acid, and hesperitin (2) (see Figure 5 for structures).

HPTLC DPPH Assay
The DPPH-HPTLC assay was carried out to determine which constituents contributed to the respective honey's overall antioxidant activity. Previously, the HPTLC-DPPH assay was developed to determine the DPPH antioxidant activity after 1 h of exposure to the reagent, the corresponding peak profiles were obtained using white light, and the antioxidant activity was expressed as mg GAE/100 g of honey [32]. In this experiment, incubation time as well as the mode for peak profile generation were optimised ( Figure  S12, Supplementary Materials). It was found that the colour of the unreacted DPPH on the plate degraded by 14.7% after 2 h and by 19.3% after 3 h. It was also observed that a significant decrease in the absorbances of the test compounds was observed after 2 and 3 h of incubation time as compared to 1 h ( Figure S8, Supplementary Materials). Furthermore, naringenin (used as reference standard) showed a DPPH radical scavenging activity (% DPPH RSA) of 66.5% after 1 h, 81.4% after 2 h, and 81.9% after 3 h of incubation. DPPH scavenging activity is generally evaluated at the point when the absorbance remains constant [50]. Because of this, in this study the photo-documentations and the recording of the corresponding peak profiles were carried out after 2 h in order to allow sufficient time for compounds to react with the DPPH reagent but not too long that the reagent would autodegrade and produce false positive results. Furthermore, it was also observed that peak profiles generated using a scan at 517 nm were more sensitive in determining the reaction of the individual bands as compared to profiles generated using white light, and therefore scanning at 517 nm was adopted for all analyses in this study ( Figure S12, Supplementary Materials).
The use of a change in hue to determine the DPPH antioxidant activity was also validated. Based on the findings (Table 7), all standards showed an increase in % DPPH RSA which correlated with increases in sample concentration, indicating that hue values can be a very useful tool in describing the antioxidant activity of a particular compound. The DPPH antioxidant activity of the four pooled and extracted Western Australian honeys along with the phenolic compounds that were previously identified in each honey were analysed in the HPTLC-DPPH assay using two solvent systems (MPA and MPB).  (Table 8). The DPPH antioxidant activity of the compounds identified in each pooled honey sample was also determined at a low concentration to mimic the concentration of the compounds in each honey and also at a high concentration in order to determine whether the activity is based on its concentration in the honey or an inherent antioxidant activity of the constituent (see Table 9). Based on the data generated, it was found that most compounds were antioxidant with the exception of eudesmic acid, o-anisic acid, t-cinnamic acid, and lumichrome, which remained inactive even when analysed at a higher concentration (see Figure 5 for structures). For Calothamnus spp. (Red Bell) honey, one of the identified constituents, t-cinnamic acid at Rf 0.562, was observed to be inactive, which was consistent with the finding that the t-cinnamic acid (13) standard did not possess any DPPH antioxidant activity. The other compounds in Red Bell honey, identified as protocatechuic acid (10) at Rf 0.390, gallic acid (8) at Rf 0.281, and kojic acid (14) at Rf 0.115, were found to be the dominant antioxidants in the honey ( Figure 9B).
For Agonis flexuosa (Coastal Peppermint) honey, the respective quantities present for the bands at Rf 0.473 (m-coumaric acid (12)), at Rf 0.420 (syringic acid (11)), at Rf 0.395 (luteolin (1)), at Rf 0.187 (epicatechin (4)), at Rf 0.100 (kojic acid (14)), and at Rf 0.090 (epigallocatechin gallate (5)), showed the expected antioxidant activity based on the calibrated antioxidant activity of the standards, except for the constituent at Rf 0.279 (identified as lumichrome (15)) which showed an unexpected result since lumichrome standard itself was found to be inactive, indicating that there might be a constituent that was co-eluting with lumichrome at this Rf value which might cause the honey band at this Rf to show antioxidant activity ( Figure S13, Supplementary Materials).
In the case of Corymbia calophylla (Marri) honey, the bands at Rf 0.484 (eudesmic acid (7) For Eucalyptus marginata (Jarrah) honey, the band at Rf 0.530, identified as hesperitin (2), was found to be consistent in its behaviour with the analysed activity of the corresponding standard, which was found to be inactive at low concentration. The band at Rf 0.455 (o-anisisc acid (9)) was found to be active, whereas the corresponding standard showed no activity, even at higher concentration, implying that this honey band has a co-eluting constituent which causes a low level of antioxidant activity. The band at Rf 0.282 was identified as lumichrome (15), which showed a very low level of activity similar to the standard, which was found to be inactive in the investigated concentrations. The activity of the compounds at Rf 0.420 (m-coumaric acid (12)), Rf 0.322 (taxifolin (3)), Rf 0.100 (kojic acid (14)), and Rf 0.083 (epigallocatechin gallate (5)) were found to be consistent with the activity of the respective standards ( Figure S15, Supplementary Materials).
For Calothamnus spp. (Red Bell) honey, one of the identified constituents, t-cinnamic acid at Rf 0.562, was observed to be inactive, which was consistent with the finding that the t-cinnamic acid (13) standard did not possess any DPPH antioxidant activity. The other compounds in Red Bell honey, identified as protocatechuic acid (10) at Rf 0.390, gallic acid (8) at Rf 0.281, and kojic acid (14) at Rf 0.115, were found to be the dominant antioxidants in the honey ( Figure 9B). For Agonis flexuosa (Coastal Peppermint) honey, the respective quantities present for the bands at Rf 0.473 (m-coumaric acid (12)), at Rf 0.420 (syringic acid (11)), at Rf 0.395 (luteolin (1)), at Rf 0.187 (epicatechin (4)), at Rf 0.100 (kojic acid (14)), and at Rf 0.090 (epigallocatechin gallate (5)), showed the expected antioxidant activity based on the calibrated antioxidant activity of the standards, except for the constituent at Rf 0.279 (identified as lumichrome (15)) which showed an unexpected result since lumichrome standard itself was found to be inactive, indicating that there might be a constituent that was co-eluting with lumichrome at this Rf value which might cause the honey band at this Rf to show antioxidant activity ( Figure S13, Supplementary Materials).
In the case of Corymbia calophylla (Marri) honey, the bands at Rf 0.484 (eudesmic acid (7) For Eucalyptus marginata (Jarrah) honey, the band at Rf 0.530, identified as hesperitin (2), was found to be consistent in its behaviour with the analysed activity of the corresponding standard, which was found to be inactive at low concentration. The band at Rf

Discussion
The data obtained in the TPC assay were consistent with previous studies where Red Bell honey had shown higher phenolic content than nine other monofloral honeys from Western Australia [37]. By using the same conditions for the assay, Manuka honey from Australia and New Zealand [43] was found to have a TPC of 35.08 mg GAE/100 g (minimum 22.6, maximum 66.3) indicating that Calothamnus spp. (Red Bell) honey and also Eucalyptus marginata (Jarrah) honey have higher TPC than Manuka honey, which is generally seen as a honey with high antioxidant activity [51]. By comparing the findings of this study with TPC data for other monofloral honeys across the globe, TPC values of 18.9 ± 3.82 to 23.7 ± 4.37 GAE/100 g [52] were reported for some Romanian monofloral honeys, while Mexican monofloral honeys had TPC range of 18.02 ± 0.49 to 102.77 ± 1.29 GAE/100 g [53], Czech and Slovak honeys had TPC between 54.0 ± 1.7 and 254.2 ± 1.4 GAE/100 g [54], and Brazilian honeys were reported to have TPC between 13.3 and 100 GAE/100 g [55]. The TPC values obtained in this study are lower in comparison. The assay used in this study was, however, a modified Folin-Ciocalteu assay in which the concentration of the sodium carbonate solution was optimised in such a way that sugar interference was muted, as sugars were also observed to react with the reagent leading to an overestimation of TPC without this modification [41].
Similar to the generated TPC values, the FRAP activity of Calothamnus spp. (Red Bell) honey was also observed to be higher compared to the other investigated Western Australian honeys [37] and also when compared to that of Manuka honey (2.88 to 10.72 mmol Fe 2+ /kg) [43]. By comparing the FRAP activity with that of other monofloral honeys from across the globe, Bangladeshi monofloral honeys were reported to have FRAP activity of 1.00-8.00 mmol Fe 2+ /kg [56], the FRAP activity of Oak honeydew honey from Croatia was reported to be 4.8 mmol Fe 2+ /kg [57], Polish monofloral honeys were reported to have between 1.00 and 7.00 mmol Fe 2+ /kg FRAP activity [58], and Thai monofloral honeys 0.61 to 4.34 mmol Fe 2+ /kg [59]. Compared to these findings, the honeys from Western Australia investigated in this study showed a higher FRAP activity.
The DPPH radical scavenging activity of Calothamnus spp. (Red Bell) honey was also observed to be higher compared to the other investigated Western Australian honeys [37] and when compared to that of Manuka honey (mean = 1.98, range of 0.56 to 4.35 mmol TE/kg) [43]. Polish monofloral honeys were reported to have 0.20 to 1.20 mmol TE/kg DPPH activity [58], Oak honeydew honey from Croatia was reported to have a DPPH activity of 4.5 mmol TE/kg [57], and Thai monofloral honeys of 0.107 to 1.224 mmol TE/kg [59]. These values were lower compared to the DPPH radical scavenging activity of the investigated WA honeys.
High correlations between TPC values, FRAP and DPPH antioxidant activity were observed in this study, consistent with other reports [46][47][48][49]. DPPH and FRAP assays were chosen to express the total antioxidant activities of honey because the application of multiple assays can be helpful in reflecting the antioxidant properties of honeys more accurately than a single assay can do [60]. DPPH and FRAP assays have been widely used to determine the antioxidant activity of various plant extracts and food products since they use stable free radicals and the determination of antioxidant capacity is simple, quick and easy to perform, results are readily validated, accurate, and highly reproducible and the reagents are inexpensive and easy to prepare [61,62]. The TPC assay was employed to confirm that the antioxidant assay can be attributed to the phenolic compounds present in honey as it has been found that high antioxidant potential in FRAP and DPPH assays is usually observed for samples with high phenolic and flavonoid content [11].
By employing the HPTLC database to identify the phenolic constituents in the honeys, kojic acid (14) and epigallocatechin gallate (5) were found in all investigated honeys. m-Coumaric acid (12) was present in most honeys except Calothamnus spp. (Red Bell) honey. Lumichrome (15) was identified in Agonis flexuosa (Coastal Peppermint) and Eucalyptus marginata (Jarrah) honey, gallic acid (8) was found in both Red Bell and Corymbia calophylla (Marri) honey, taxifolin (3) was only found in Marri and Jarrah honey, while luteolin (1) and epicatechin (4) were only found in Coastal Peppermint and Marri honey. Hesperitin (2) was only identified in Jarrah honey, eudesmic acid (7) only in Marri honey, syringic acid (11) only in Coastal Peppermint honey, and protocatechuic acid (10) and t-cinnamic acid (13) only in Red Bell honey. Compounds that were only identified in a specific honey might in the future potentially be used as biomarkers for that honey.
The HPTLC-based database for phenolic compound identification was previously employed in the analysis of Manuka honey where kojic acid, gallic acid, epigallocatechin gallate, lumichrome, 2,3,4-trihydroxy benzoic acid, and o-anisic acid were also identified. However, leptosperine, mandelic acid, lepteridine, methyl syringate, salicylic acid, and benzoic acid were only found in Manuka honey [30]. This implies that there are some compounds that are ubiquitous in honeys while others are unique and can only be found in a certain honey. Since all the honeys investigated in this study along with Manuka honey belong to the plant family Myrtaceae, it can be speculated that this might explain some of the overlaps in the compounds identified in the four honey types.
To date, reports on the presence and concentration of phenolic compounds in honeys originating from Western Australia has been very scant. Prior to this study, only for Jarrah honey had some compounds been reported. Using HPLC-ESI-MS/MS analysis, Anand et al. in 2019 were able to quantify quercetin, hesperitin, cinnamic acid, methyl syringate, rutin, sinapic acid, ferulic acid, p-coumaric acid, phenyllactic acid, syringic acid, caffeic acid, vanillic acid, chlorogenic acid, p-hydroxybenzoic acid, protocatechuic acid, and gallic acid [39]. By employing the HPTLC-based database in this study, only hesperitin was identified from the compounds reported by Anand et al., which can be attributed to a number of reasons: Firstly, Anand et al. (2019) utilised a Strata-X cartridge solid phase extraction which was eluted with acidified water (pH-2), and then with methanol prior to their analysis [39]. This study, however, employed a liquid-liquid solvent extraction using dichloromethane and methanol (1:1) as a solvent system.
The solvent system used in the development of the HPTLC plates was toluene: ethyl acetate: formic acid 6:5:1 (MPB) which has frequently been used to fingerprint honeys [32,33,40,63,64]. A more polar solvent system, toluene: ethyl acetate: formic acid 2:8:1 (MPA), was also utilised in order to identify compounds of higher polarity. A better separation in the bands in honey was observed with MPB, however, it was found to be unable to fully develop all honey constituents as seen by dark bands on the baseline of the plate prior to derivatisation, and after derivatisation with NP-PEG and VSA reagents (Figures 2 and 5), as well as after derivatisation with DPPH reagent (Figure 9). It is a recommendation that another solvent system with higher polarity is also used in the future in order to identify those more polar compounds that were not fully captured by the solvent systems used in the current study.
This research utilised pooled honey samples, as the composition of such a pooled sample will be more representative of the typical chemical composition of the respective honey compared to the analysis of a randomly chosen single sample. Specifically, eight samples were pooled to represent the Red Bell honey used in this study, six samples each were pooled for Coastal Peppermint and Jarrah honey, and 13 samples were used to represent Marri honey.
An HPTLC-DPPH assay was previously employed in the qualitative and quantitative analysis of the antioxidant fingerprints of honeys [32][33][34]. Islam et al. in 2020 and 2021 utilised the method for the quantification of antioxidant band activities for various Australian honeys. However, the analysis was performed using dichloromethane as an extraction solvent, the incubation time was set to only 1 h, and peak profiles were obtained with white light [32,33]. In this study, however, a more polar extraction solvent was used (dichloromethane: methanol 1:1 v/v) which led to the observation of more antioxidant bands. The incubation time was also optimised as it was found that 1 h was not enough for some phenolic compounds to fully react with the DPPH reagent ( Figure S8, Supplementary Materials). Longer incubation times of 2 h and 3 h were also tested and it was found that 2 h is the optimum time for the multiple types of polyphenols present in honey to react with the reagent but not long enough for the DPPH reagent to autodegrade.
The findings of the HPTLC-DPPH assay for honeys are often expressed in a qualitative manner by presenting active bands that showed a discoloration of the DPPH reagent [34]. In some instances, the antioxidant band activity was also quantified, expressed as mg GAE/100 g of honey [32,33]. Quantification of individual bands is, however, challenging given that some antioxidant bands are very low in absorbance and might thus be below the limit of detection of this quantification method.
In this study, the colours of the unreacted DPPH reagent and the colours of the active bands in the analysed samples, converted into hue values, were compared and from this, their DPPH radical scavenging activity was calculated. The HPTLC software usually provides colour information in the form of RGB values which can be converted into hue values (based on the hue, saturation, and brightness (HSB) colour space) [30]. The use of hues in expressing the colour of a particular band was found to be very helpful in the early stage of identification of an unknown sample using the HPTLC-derived database where, upon the use of a suitable derivatisation agent, a discrimination of one compound group from another based on colour was possible [30]. It was found in this study that colour captured in the form of hue values can also be used to express the results of the HPTLC-DPPH assay (Tables 6-8). Various antioxidant compounds were tested, and the findings demonstrated that hues varied according to the sample concentration that was applied. However, linear regression did not reach 0.99 indicating that the current parameters used in this study are only able to describe the antioxidant results in a semi-quantitative manner, expressed here in inferences ranging from + to +++. More optimisation is required in order to use the method for full quantification of the antioxidant activity of individual bands in an unknown sample.
The DPPH antioxidant activities of the individual constituents that were identified in each honey were also determined as a mixture at low concentration to mimic the concentrations that were quantified in honey and also at higher concentration (Table 7) to determine whether its activity is based on its concentration in the honey or inherent antioxidant activity of the constituent. All identified compounds except eudesmic acid (7), o-anisic acid (9), t-cinnamic acid (13), and lumichrome (15) showed activity towards the DPPH reagent. The inactive compounds (Table 7) lack a hydroxyl group in the phenolic ring that can react with the DPPH reagent [65,66] indicating that not all phenolic compounds are antioxidant.
The reaction of compounds with DPPH is governed by the reagent's steric accessibility indicating that smaller molecules have greater access to the radical site as compared to larger molecules [50]. This explains why flavonoids tend to react slower compared to smaller molecules like simple phenolic acids. The reactivity of flavonoids with DPPH on the other hand is dictated by the so-called Bors criteria ( Figure 10). The first criterion is the presence of a catechol group on Ring B (Bor creases the stability of the resulting antioxidant radical. The second is the pr double bond combined with a 4-oxo group on Ring C (Bors 2), which facil delocalization. The third is the presence of OH groups at positions 3 and 5 i with a 4-oxo group, which enables electron delocalization via hydrogen b [65,66]. Among the flavonoids identified in this study, taxifolin (3) possesse criteria, confirmed by very high radical scavenging activity even when analy concentration. Luteolin (1), epicatechin (4), and epigallocatechin gallate (5) Bors 1 criterion, while hesperetin (2) does not possess any, which explai shown only a very weak radical scavenging activity. The trends seen in th activity of the different phenolic compounds investigated in this study w with the trends that were previously reported [65][66][67]. The HPTLC-DPPH been demonstrated to be a very powerful tool in the identification of antio uents. However, DPPH or a similarly structured radical does not exist in food system [50] and it is therefore suggested that a more biochemically re dant model should be used in future studies. The first criterion is the presence of a catechol group on Ring B (Bors 1), which increases the stability of the resulting antioxidant radical. The second is the presence of a 2,3 double bond combined with a 4-oxo group on Ring C (Bors 2), which facilitates electron delocalization. The third is the presence of OH groups at positions 3 and 5 in combination with a 4-oxo group, which enables electron delocalization via hydrogen bonds (Bors 3) [65,66]. Among the flavonoids identified in this study, taxifolin (3) possesses Bors 1 and 2 criteria, confirmed by very high radical scavenging activity even when analysed at a lower concentration. Luteolin (1), epicatechin (4), and epigallocatechin gallate (5) all possess the Bors 1 criterion, while hesperetin (2) does not possess any, which explains why it has shown only a very weak radical scavenging activity. The trends seen in the antioxidant activity of the different phenolic compounds investigated in this study were consistent with the trends that were previously reported [65][66][67]. The HPTLC-DPPH assay has thus been demonstrated to be a very powerful tool in the identification of antioxidant constituents. However, DPPH or a similarly structured radical does not exist in a biological or food system [50] and it is therefore suggested that a more biochemically relevant antioxidant model should be used in future studies. This is the first report on the use of band colours as a basis of expressing antioxidant activity in samples, which demonstrates that colour values derived from HPTLC analysis can also be used to (semi-quantitatively) express antioxidant activity in addition to more traditional quantification (using generated standard curves) that HPTLC can also perform. While this presents a novel analytical angle to HPTLC-DPPH analysis, some limitations need to be acknowledged. Given the very general nature of the DPPH assay and its common use in natural product research as a screening tool for antioxidant activity, a qualitative (i.e., active or inactive) or semi-quantitative (i.e., activity ranges from + to +++) approach might suffice in many instances. This can be achieved, as illustrated in this study, by expressing antioxidant activity of individual honey bands as % RSA, which is a widely accepted way of expressing antioxidant activity. However, should a fully quantitative result be the aim, more optimisation is needed, specifically to determine the concentration range of each match compound that yields linear regression equal to or greater than 0.99.
In recent years High Performance Thin Layer Chromatography (HPTLC) has emerged as a very versatile tool for various aspects of honey analysis. It can, for example, be used to identify and quantify various sugars in honey [68,69] and with this can also be used to identify post-harvest sugar adulterations [70]. It is also applied to identify and quantify the presence of hydroxymethyl furfural (HMF) in honey, which is a marker for excessive heat treatment-associated degradation and thus reduced honey quality [71][72][73]. HPTLC in combination with DPPH derivatisation has also been successfully used to visualise and quantify (as gallic acid equivalents) antioxidant honey constituents [32,33]. Moreover, the HPTLC analysis of organic honey extracts has been demonstrated to yield unique signatures that are reflective of a honey's floral origin and can thus be used for honey authentication [40,63,64]. This study contributes to the growing body of literature that demonstrates the versatility of HPTLC in the analysis of honey. The identity of some phenolic constituents in the four investigated Western Australian honeys was revealed using a HPTLC-based database along with their quantification, also using HPTLC. Moreover, compounds that contribute to these honeys' antioxidant activity could be identified and semi-quantified using a modification of the previously published HPTLC-DPPH analysis protocol.

Conclusions
This study investigated the antioxidant activity of four Western Australian honeys, Calothamnus spp. (Red Bell), Agonis flexuosa (Coastal Peppermint), Corymbia calophylla (Marri) and Eucalyptus marginata (Jarrah) honey. It was found that Red Bell honey has the highest total phenolic content, followed by Jarrah, Coastal Peppermint, and Marri honey. The same trends were observed for their respective FRAP and DPPH antioxidant activities.
HPTLC-DPPH bioautography was also carried out to determine which honey constituents contribute to the respective honey's antioxidant activity using a novel method of analysis based on the changes of hues on reaction with the DPPH reagent. This change in hue was used to determine the % RSA of each active band. The method was able to identify the individual bands that contribute to the honeys' overall antioxidant activity. Based on the findings of this analysis, most identified compounds showed antioxidant activity except for t-cinnamic acid, lumichrome, o-anisic acid, and eudesmic acid due to the absence of hydroxyl groups in their benzene ring.
As most analyses were carried out using HPTLC, the study was also able to demonstrate the versatility of this instrumentation in the analysis of various aspects of honey chemistry and bioactivity.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/antiox12010189/s1, Table S1: Table S1 summarises the identity, botanical origin, and families of honeys collected as part of a study on Western Australia honeys; Table S2: Honey Sample Collection and Floral Information, TPC, FRAP, and DPPH Antioxidant Activity, Table S3: Summary of the data used to determine the identity of the unknown bands in Calothamnus spp. (Red Bell) honey (Database 1A), Table S4: Summary of the data used to determine the identity of the unknown bands in Calothamnus spp. (Red Bell) honey (Database 1B), Table S5: Summary of the data used to determine the identity of the unknown bands in Calothamnus spp. (Red Bell) honey (Database 2A), Table S6: Summary of the data used to determine the identity of the unknown bands in Calothamnus spp. (Red Bell) honey (Database 2B), Table S7: Summary of the data used to determine the identity of the unknown bands in Agonis flexuosa (Coastal Peppermint) honey (Database 1A), Table S8: Summary of the data used to determine the identity of the unknown bands in Agonis flexuosa (Coastal Peppermint) honey (Database 1B), Table S9: Summary of the data used to determine the identity of the unknown bands in Agonis flexuosa (Coastal Peppermint) honey (Database 2A), Table S10: Summary of the data used to determine the identity of the unknown bands in Agonis flexuosa (Coastal Peppermint) honey (Database 2B), Table S11: Summary of the data used to determine the identity of the unknown bands in Corymbia calophylla (Marri) honey (Database 1A), Table S12: Summary of the data used to determine the identity of the unknown bands in Corymbia calophylla (Marri) honey (Database 1B), Table S13: Summary of the data used to determine the identity of the unknown bands in Corymbia calophylla (Marri) honey (Database 2A), Table S14: Summary of the data used to determine the identity of the unknown bands in Corymbia calophylla (Marri) honey (Database 2B), Table S15: Summary of the data used to determine the identity of the unknown bands in Eucalyptus marginata (Jarrah) honey (Database 1A), Table S16: Summary of the data used to determine the identity of the unknown bands in Eucalyptus marginata (Jarrah) honey (Database 1B), Table S17: Summary of the data used to determine the identity of the unknown bands in Eucalyptus marginata (Jarrah) honey (Database 2A), Table S18: Summary of the data used to determine the identity of the unknown bands in Eucalyptus marginata (Jarrah) honey (Database 2B); Table S19: Parameters used in optimising the quantification of phenolic compounds in honey, Figure S1. HPTLC fingerprint patterns for various samples of Calothamnus spp. (Red bell, n = 8), Figure S2. HPTLC fingerprint patterns for various samples of Agonis flexuosa (Coastal Peppermint, n = 5), Figure S3. HPTLC fingerprint patterns for various samples of Corymbia calophylla (Marri, n = 13), Figure S4. HPTLC fingerprint patterns for various samples of Eucalyptus marginata (Jarrah, n = 6), Figure  and Corymbia calophylla (Marri) spiked with the identified compounds based on database 1A and 1B (blue) scanned at the λmax of each specific compounds prior to derivatization, Figure S9A-G. Profile comparison of Corymbia calophylla (Marri) honey (green) and Corymbia calophylla (Marri) spiked with the identified compounds based on database 2A and 2B (blue) scanned at the λmax of each specific compounds prior to derivatization, Figure S10A-D. Profile comparison of Eucalyptus marginata (Jarrah) honey (green) and Eucalyptus marginata (Jarrah) spiked with the identified compounds based on database 1A and 1B (blue) scanned at the λmax of each specific compounds prior to derivatization, Figure S11A-G. Profile comparison of Eucalyptus marginata (Jarrah) honey (green) and Eucalyptus marginata (Jarrah) spiked with the identified compounds based on database 2A and 2B (blue) scanned at the λmax of each specific compounds prior to derivatization, Figure S12A,B. Comparison of the profiles of compounds identified in Corymbia calophylla (Marri) honey after derivatised with DPPH reagent and obtained after 1 h with transmittance in white light (green) vs. scanned at 517 nm (A) and comparison of the profiles of compounds identified in Corymbia calophylla (Marri) honey after being derivatised with DPPH reagent scanned at 517 nm and taken at 1 h (green), 2 h (blue), 3 h (grey) (B) developed using mobile phase 1B, Figure S13A-D. HPTLC plate image (a) of Agonis flexuosa (Coastal Peppermint) honey after derivatised with DPPH reagent and developed using mobile phase A (A,B) and developed using mobile phase B (C,D) obtained with transmission in white light, and comparison of the profiles of Agonis flexuosa (Coastal Peppermint) honey (green) and Agonis flexuosa (Coastal Peppermint) honey spiked with the identified compounds (blue) after being derivatised with DPPH reagent obtained at 517 nm (b-left) and comparison of the profiles of Agonis flexuosa (Coastal Peppermint) honey obtained at 254 nm (green) and 366 nm (blue) prior to derivatisation and the profile of Agonis flexuosa (Coastal Peppermint) honey spiked with the identified compounds (grey) obtained at 277 nm prior to derivatisation (b right), Figure S14A-C. HPTLC plate image (a) of Corymbia calophylla (Marri) honey after being derivatised with DPPH reagent and developed using mobile phase A (A) and developed using mobile phase B (B,C) obtained with transmission in white light, and comparison of the profiles of Corymbia calophylla (Marri) honey (green) and Corymbia calophylla (Marri) honey spiked with the identified compounds (blue) after being derivatised with DPPH reagent obtained at 517 nm (b-left) and comparison of the profiles of Corymbia calophylla (Marri) honey obtained at 254 nm (green) and 366 nm (blue) prior to derivatisation and the profile of Corymbia calophylla (Marri) honey spiked with the identified compounds (grey) obtained at 277 nm prior to derivatisation (b right), Figure S15A-D. HPTLC plate image (a) of Eucalyptus marginata (Jarrah) honey after being derivatised with DPPH reagent and developed using mobile phase A (A and B) and developed using mobile phase B (C,D) obtained with transmission in white light, and comparison of the profiles of Eucalyptus marginata (Jarrah) honey (green) and Eucalyptus marginata (Jarrah) honey spiked with the identified compounds (blue) after being derivatised with DPPH reagent obtained at 517 nm (b-left) and comparison of the profiles of Eucalyptus marginata (Jarrah) honey obtained at 254 nm (green) and 366 nm (blue) prior to derivatisation and the profile of Eucalyptus marginata (Jarrah) honey spiked with the identified compounds (grey) obtained at 277 nm prior to derivatisation (b right).

Data Availability Statement:
No new data were created or analysed in this study. Data sharing is not applicable to this article.