New Alk(en)ylhydroxycyclohexanes with Tyrosinase Inhibition Potential from Harpephyllum caffrum Bernh. Gum Exudate

This work presents the first report on the phytochemical investigation of Harpephyllum caffrum Bernh. gum exudate. A known cardanol, 3-heptadec-12′-Z-enyl phenol (1) and three new alk(en)ylhydroxycyclohexanes, namely, (1R,3R)-1,3-dihydroxy-3-[heptadec-12′(Z)-enyl]cyclohexane (2) (1S,2S,3S,4S,5R)-1,2,3,4,5-pentahydroxy-5-[octadec-13′(Z)-enyl]cyclohexane (3) and (1R,2S,4R)-1,2,4-trihydroxy-4-[heptadec-12′(Z)-enyl]cyclohexane (4) were isolated from the gum. The structures of the compounds were determined by extensive 1D and 2D NMR spectroscopy and HR-ESI-MS data. The ethanolic extract of the gum was found to be the most potent tyrosinase inhibitor with IC50 of 11.32 µg/mL while compounds 2 and 3, with IC50 values of 24.90 and 26.99 µg/mL, respectively, were found to be potential anti-tyrosinase candidates from the gum. Gum exudate may be a potential source for non-destructive harvesting of selective pharmacologically active compounds from plants. The results also provide evidence that H. caffrum gum may find application in cosmetics as a potential anti-tyrosinase agent.


Introduction
Harpephyllum Bernh. is a monotypic genus of the family Anacardiaceae with Harpephyllum caffrum Bernh. as the only recognised species. Anacardiaceae, also known as the cashew family, is characterised by gum/resin secretion [1,2]. The ethnomedicinal uses of H. caffrum include the treatment of skin diseases such as acne and eczema, control and management of infant convulsions and epilepsy and for blood purification [3,4]. Phytochemical studies of H. caffrum have been documented for its leaves [5,6], edible fruits and stem bark [7]. These studies revealed the presence of polyphenolic compounds and their glycosides, lupane-type triterpenoids, alkyl coumarates and cardanols with saturated and unsaturated alkyl chains. The cardanols have similar alkyl chains with alkenyl cyclohexanols but differ in the ring type, the latter being non-aromatic [5][6][7]. However, no information is available on the natural gum exudate produced by this species.
Plant gum exudates are often harvested for their pharmaceutical [8], food [9] and cosmetic applications [10,11]. They serve as binders due to their rich structurally stable carbohydrate and protein compositions [12] and are preferred over their synthetic polymer counterparts because of their biodegradability, non-toxic nature and economic importance [8]. Although the chemistry of gums has reportedly been characterised by several Molecules 2022, 27, 3839 2 of 10 polysaccharides [12,13], terpenoids have been found occurring in large amounts for some gums [14,15], whereas alkenyl cyclohexanols have not been reported from plant gums. We herein report on the phytochemistry of H. caffrum gum exudate using chromatographic separations and spectrometric techniques. A preliminary study on the anti-tyrosinase potential of the isolated compounds is also evaluated.

Chemistry of Isolated Compounds
Compound 1 (61 mg) was isolated as a yellow oil. The 1 H and 13 C NMR data (Tables 1 and 2) of 1 were consistent with those reported by Okoth et al. (2016) for the cardanol, 3-heptadec-12 -Z-enyl phenol [16]. Its molecular formula was obtained as C 23 Figure 1.  16.0 (C-5), 41.2 (C-2), 36.9 (C-6) and 43.7 (C-4) while the only quaternary carbon resonated at δ C 72.9 (C-3) in the 13 C NMR spectrum. These assignments were supported by COSY correlations between H-1 and H-2, H-4 and H-5, H-5 and H-6, in addition to HMBC the correlations of H-1 with C-2, C-3 and C-5, H-4 with C-2, C-3 and C-5 and H-6 with C-2. NOESY correlations were only observed between H-1 and H-2, and H-2 and H-1 on the cyclohexane ring while H-16 had NOESY correlations with both H-13 and H-17 , observed on the alkenyl chain. The placement of double bond was further justified by HMBC correlations of H-13 with C-14 and C-15 in addition to those of H-17 with C-16 and C-15 . The olefinic protons with a characteristic multiplet at δ H 5.32 and resonances of C-11 and C-14 (adjacent to the double bond) observed at δ C 27.2 and 26.9, respectively, indicated a cis mono-unsaturated alkyl chain [16,17]. All COSY, HMBC and NOESY correlations are shown in Figure 2. The absolute configuration of 2 was deduced from the spectrum of the electronic circular dichroism (ECD) (Figure 3), which revealed a high-amplitude positive cotton effect in the regions 230-250 and 295-325 nm. Compound 2 was therefore named (1R,3R)-1,3-dihydroxy-3-[heptadec-12 (Z)-enyl]cyclohexane.

Chemistry of Isolated Compounds
Compound 1 (61 mg) was isolated as a yellow oil. The 1 H and 13 C NMR data (Tables 1 and 2) of 1 were consistent with those reported by Okoth et al. (2016) for the cardanol, 3-heptadec-12′-Z-enyl phenol [16]. Its molecular formula was obtained as C23H38O based on HR-ESI-MS (m/z 329.2823 [M-H] − ; calcd for C23H37O, 329.2844). This single mass analysis further justified the assignment of alkenyl chain length of the proposed cardanol. The structures of 1 and other compounds isolated from H. caffrum gum are presented in Figure  1.

COSY correlations were between H-1 and H-6, H-2 and H-3 and H-3 and H-4 while NOESY correlations were observed between H-2 and H-6 and H-4 and H-1 which
supported the position of the alkenyl chain on the ring. This was further justified by HMBC correlations of H-1 to C-4, C-5 and C-6. The various NOESY and HMBC correlations between positions 12 and 18 were similar to those observed for 2; thus, the location of the double bond was assigned. The double bond was also assigned cis, similar to 2. Unlike 2, the ECD spectrum of 3 ( Figure 3) showed a high-amplitude positive cotton effect in the regions 230-250 and a weak one around 295-325 nm. Compound 3 was therefore named (1S,2S,3S,4S,5R)-1,2,3,4,5-pentahydroxy-5-[octadec-13 (Z)-enyl]cyclohexane. Compound 4 (1.04 g) was obtained as a brown oil. In comparison to compound 2 (with one oxygenated methine proton), the 1 H NMR spectrum (Table 2) displayed a cyclohexane ring consisting of two deshielded oxygenated methine protons at δ H 3.46 (bs, H-1) and δ H 3.86 (m, H-2), whose multiplicities suggested they are equatorial and axial, respectively [18].  Table 2. Compound 4 had better solubility and good peak resolution in CD 3 OD compared to CDCl 3 . However, the 1 H NMR was run in both solvents (Figure 4) for accurate comparison with 6 which was run in CDCl 3 . The oxygenated methine protons of 4 had chemical shifts that were more upfield (δ H 3.51 and 3.90) in comparison to 5 (δ H 4.31 and 4.35). However, the same methine protons had similar values with those of 6, justifying their similar positions, while the differences in their multiplicities suggested a difference in their configuration. Compound 6 further differed from 4 because of its longer alkenyl chain length. All spectral data ( Figures S1 to S36)       In previous studies, phenolic lipids like such as cardanols, cardols and anacardic acids have been identified from different species of Anacardiaceae like such as Anacardium occidentale [19], Mangifera indica [20] and H. caffrum [7], making them chemotaxonomically significant for this "cashew family" [21] and informing highlighting that some bioactive compounds believed to be localiszed in specific plant parts can be found in the "plant waste", gum. Contrastingly, alkyl and alk(en)ylhydroxycyclohexanes have not gained popularity in Anacardiaceae. Roumy     In previous studies, phenolic lipids such as cardanols, cardols and anacardic acids have been identified from different species of Anacardiaceae such as Anacardium occidentale [19], Mangifera indica [20] and H. caffrum [7], making them chemotaxonomically significant for this "cashew family" [21] and highlighting that some bioactive compounds believed to be localised in specific plant parts can be found in the "plant waste" gum. Contrastingly, alkyl and alk(en)ylhydroxycyclohexanes have not gained popularity in Anacardiaceae. Roumy   In previous studies, phenolic lipids such as cardanols, cardols and anacardic acids have been identified from different species of Anacardiaceae such as Anacardium occidentale [19], Mangifera indica [20] and H. caffrum [7], making them chemotaxonomically significant for this "cashew family" [21] and highlighting that some bioactive compounds believed to be localised in specific plant parts can be found in the "plant waste" gum. Contrastingly, alkyl and alk(en)ylhydroxycyclohexanes have not gained popularity in Anacardiaceae. Roumy et al. (2009) initially reported four cyclic alkyl polyol derivatives from Tapirira guianensis [17] while Okoth et al. (2016) reported an alk(en)ylhydroxycyclohexane alongside close cyclohexanone derivatives and some phenolic lipids [16]. The present study suggests that the alk(en)ylhydroxycyclohexane may be part of yet-to-be-identified phytochemicals present in the H. caffrum tree. Further studies involving other species of Anacardiaceae will be required to establish if this rare class of compounds also has chemotaxonomic importance to the family.

Tyrosinase Inhibitory Activity
The anti-tyrosinase inhibitory activity of the isolated compounds (1)(2)(3)(4) and Harpephyllum caffrum gum-ethanolic extract HCG-EtOH was evaluated using L -tyrosine as substrate. The test solutions for each compound and extract were prepared with concentrations varying from 1.56 to 200 µg/mL and results are presented as half maximum inhibitory concentration (IC 50 ) shown in Table 3. The ethanolic extract, HCG-EtOH, was identified as potent tyrosinase inhibitors with IC 50 value (11.32 µg/mL) comparable with that of the control, arbutin (9.85 µg/mL) although lower than kojic acid (4.34 µg/mL). Compounds 2 and 3 showed good inhibition with IC 50 values of 24.90 and 26.99 µg/mL, respectively, while 1 and 4 were moderately active with IC 50 values of 41.77 and 34.90 µg/mL, respectively. The results suggest that HCG-EtOH's potent anti-tyrosinase activity may be due to the synergistic effects of its constituents, of which 2 and 3 possibly play a significant role in the enzyme-substrate interactions. However, this proposed synergy warrants further investigation. We opined that the basic cyclohexanol moiety of 2-4 (IC 50 between 24.90 and 34.90 µg/mL) could be a better tyrosinase inhibition pharmacophore than phenolic in 1 (IC 50 ; 41.77 µg/mL), since all compounds had similar alkenyl side chains. Tyrosinase inhibitory potential of alk(en)ylhydroxycyclohexanes has not been previously reported. Yu et al. (2016) reported earlier that cardanols exhibited their anti-tyrosinase activity by altering tyrosinase conformation and significantly decreasing the steady state rate of tyrosinase diphenolase activity [22]. A previous study showed that the stem bark of H. caffrum could be considered as an anti-tyrosinase agent with an IC 50 of 40 µg/mL observed for its ethanolic extract [23]. Therefore, it is most likely that gum exudate from H. caffrum is a better anti-tyrosinase agent compared to the stem bark as revealed in the current study (IC 50 of HCG-EtOH, 11.32 µg/mL). Table 3. Tyrosinase inhibitory activity of different extracts and compounds from Harpephyllum caffrum (HCG-EtOH) compared with the positive control: arbutin and kojic acid.

General Experimental Procedures
The infrared (IR) spectra were obtained on a Perkin Elmer Spectrum 100 Fourier transform infrared spectrophotometer (FT-IR) with universal attenuated total reflectance (ATR) sampling accessory. Electronic Circular Dichroism (ECD) data were acquired using an Applied Photophysics Chirascan spectrometer (Applied Photophysics Ltd., Leatherhead, UK). 1 H, 13 C and 2D nuclear magnetic resonance (NMR) spectra were recorded using deuterated methanol (CD 3 OD) at room temperature on a Bruker Avance III 400 MHz spectrometer. The high-resolution electrospray ionization mass spectra (HR-ESI-MS) were obtained on a Waters SYNAPT G1 High-Definition Mass Spectrometer (Waters, MA, USA). All column chromatography (CC) was carried out using Merck silica gel 60 (0.040-0.063 mm) while Merck 20 cm × 20 cm silica gel 60 F 254 aluminium sheets were used for thin-layer chromatography (TLC). The TLC plates were analysed under UV (254 and 366 nm) before further visualisation by spraying with 10% sulfuric acid in methanol (MeOH) solution followed by heating. Analytical grade solvents and other chemicals used were supplied by either Merck (Darmstadt, Germany) or Sigma (St. Louis, MO, USA) chemical companies.

Sample Collection and Identification
Gum was collected from H. caffrum bark growing in Reservoir Hills, Durban. The H. caffrum tree has been identified and a voucher specimen (R. Moodley 2) was deposited in the Ward Herbarium, School of Life Sciences, University of KwaZulu-Natal, Westville, Durban, South Africa [7].

Extraction and Purification
Four hundred grams of H. caffrum gum (HCG) was crushed (using a mortar and pestle) without further drying. The crushed gum was then extracted with ethanol (EtOH) and the resulting extract was concentrated using a rotary evaporator to obtain HCG crude EtOH extract (HCG-EtOH). The HCG-EtOH (2 g) was subjected to CC using varying ratios of hexane:EtOAc (from 1:0 to 0:1, v/v) and EtOAc:MeOH (from 1:0 to 0.9:0.1, v/v) as mobile phases. A total of 30 aliquots were collected from which aliquots 5 and 11 yielded compounds 1 and 2, respectively. The remaining aliquots were combined based on TLC profiles to give A (aliquots 13-16), B (17)(18)(19)(20) and C (21-30). Fraction C (520 mg) was re-chromatographed using a gradient elution similar to the one described for the CC of HCG-EtOH, resulting in compounds 3 and 4.

Tyrosinase Inhibitory Assay
Tyrosinase inhibitory activity was determined spectrophotometrically with slight modifications [24]. Briefly, tested samples were dissolved in DMSO to a final concentration of 20 mg/mL. In a 24-well plate, 30 µL of the samples was added to 970 µL of potassium phosphate buffer (0.1 M, pH 6.5) and serially diluted two-fold. The mixture (70 µL) was transferred to a 96-well plate containing 30 µL of potassium phosphate buffer (0.1 M, pH 6.5) to obtain increasing concentration range of (1.56-200 µg/mL). Then, 30 µL of mushroom tyrosinase solution (333 Units/mL in phosphate buffer, pH 6.5) was added to each mixture, which was then incubated at 25 • C for 5 min. Thereafter, 110 µL of 2 mM L-tyrosine was added in all wells of a 96-well microtiter plate and further incubated for 30 min at 25 • C. Arbutin and kojic acid were included as positive controls tested at concertation range (1.56-200 µg/mL) and 1% DMSO was included as a negative control. The absorbance of the mixture was read at 492 nm using an ELISA microplate reader (VarioSkan Flash, Thermo Fisher Scientific, Vantaa, Finland). All experiments were performed in triplicates. The percent inhibition of tyrosinase was calculated as follows: where A 0 is the absorbance of the negative control with the enzyme without test samples, A 1 is the absorbance of the test samples and enzyme.

Statistical Analysis
Data were expressed as mean ± SD from three independent experiments. The IC 50 values, corresponding to the concentration required to inhibit 50% of tyrosinase activity, were calculated from a sigmoidal dose-response of a non-linear regression and R-square values representing the best fit of the model were determined by Pearson nonlinear regression using the GraphPad Prism 5 software (GraphPad Software, San Diego, CA, USA). A one-way ANOVA was performed and differences between means were separated using the Tukey multiple comparison test at p < 0.05. Experiments were performed in triplicate.

Conclusions
Three new alkenyl cyclohexanols of the order, di, tri and pent-ol, were isolated from the gum of Harpephyllum caffrum, alongside a known cardanol. The alkenyl side chain of compound 3 differs from other compounds with an additional methylene unit which was deduced from HR-ESI-MS report. All compounds show moderate inhibition of tyrosinase while the ethanolic extract of the gum displayed potent activity, comparable to arbutin. Gum exudate of H. caffrum may be a potential anti-tyrosinase candidate with cosmetic applications. However, further studies which evaluate the toxicity of the isolates and extracts are required to establish this preliminary report.