Inhibitory Effect of Chlorogenic Acid Analogues Comprising Pyridine and Pyrimidine on α-MSH-Stimulated Melanogenesis and Stability of Acyl Analogues in Methanol

In continuation of studies for α-MSH stimulated melanogenesis inhibitors, we have evaluated the design, synthesis, and activity of a new series of chlorogenic acid (CGA) analogues comprising pyridine, pyrimidine, and diacyl derivatives. Among nineteen synthesized compounds, most of them (fifteen) exhibited better inhibitions of melanin formation in B16 melanoma cells. The results illustrated that a pyridine analogue 6f and a diacyl derivative 13a of CGA showed superior inhibition profiles (IC50: 2.5 ± 0.7 μM and 1.1 ± 0.1 μM, respectively) of α-MSH activities than positive controls, kojic acid and arbutin (IC50: 54 ± 1.5 μM and 380 ± 9.5 μM, respectively). The SAR studies showed that both –CF3 and –Cl groups exhibited better inhibition at the meta position on benzylamine than their ortho and para positions. In addition, the stability of diacyl analogues of CGA in methanol monitored by HPLC for 28 days indicated the steric bulkiness of acyl substituents as a key factor in their stability.

Melanocytes are known to be stimulated by various factors including UV radiation, melanocyte-stimulating hormone (α-MSH), a phosphodiesterase inhibitor, such as theophylline. Mainly three tyrosinase family enzymes are involved in the regulation of melanogenesis in mammalians such as tyrosinase and two tyrosinase-related proteins, TYRP1 and TYRP2. Among these, tyrosinase (EC.1.14.18.1) is the main rate-limiting enzyme that controls the biosynthesis of melanin production by the conversion of L-tyrosine to dopaquinone via L-dopa using monophenolase and diphenolase activities [6][7][8][9]. However, an excess secretion of melanin from melanocytes due to prolonged exposure to sunlight leads to dermatologic disorders such as melasma [10], freckles [11], post-inflammatory melanoderma [12], solar lentigines [13], vitiligo [14], and even cancer [15]. Moreover, a wide number of research revealed that many melanogenesis disorders have been linked to neurodegenerative diseases, including Parkinson's, Alzheimer's, and Huntington's diseases as well [16][17][18]. Therefore, the skincare industry is continuously growing around the globe in recent days. Approximately ≈15% of the world's population are consuming skincare and cosmetic products; among them, Asia lies on the highest rank [19].
In this regard, chemists have been paying more attention to make novel skincare products to reduce the effect of pigmentation such as skin-whitening agents with no side effects [20][21][22][23][24][25][26][27][28][29]. Several small heterocyclic molecules are playing a vital role in the field of medicinal chemistry. A wide variety of synthetic agents such as kojic acid derivatives [30][31][32], thiosemicarbazones [28,33], Morita-Baylis-Hillman adducts [34], quinazolinone benzamides [35,36], and cinnamic acid derivatives [37,38] are reported for the treatment of hyperpigmentation disorders. Pyridine is a common structural motif that is ubiquitous in several pharmaceuticals and natural products including vitamins and alkaloids [39][40][41][42]. Similarly, pyridazine, a 1,3-diazine six membered molecule, has also been received important attention in medicinal chemistry because of its range of biological properties such as antihypertensive, anti-cancer, anti-HIV, and so on [43,44]. Moreover, a few scaffolds of both pyridines [45][46][47] and pyrimidines [48][49][50] are highly effective against the melanogenesis process. Chlorogenic acid (5-O-caffeoylquinic acid, 5-CQA) is a polyphenol ester obtained by the reaction between trans-cinnamic acid and quinic acid [51][52][53]. It is predominantly found in beverages and coffee drinks. In addition, several vegetables and fruits such as eggplant, artichoke, sweet potato, tomato, oilseeds, apples, pears, blueberries, and peanuts are also observed as rich sources [54][55][56]. A number of research studies revealed that CGA was used to improve several disorders including obesity and diabetes in traditional medicinal practice. In addition, it exhibits numerous biological activities such as anti-inflammatory, antioxidant, anti-cancer, anti-hypertensive, antimicrobial, anti-obesity, hypolipidemic, liver protective, etc. [57][58][59]. Growing evidence recognized that CGA significantly inhibits the tumor growth in various cancer cell lines. For example, Anqi Zeng et al. and Sapio Luigi et al. have reported the anti-cancer activity of CGA by induction of the apoptosis pathway in breast cancer and osteosarcoma cell lines [60,61]. Likewise, Alessia Salzillo et al. highlighted its synergistic effect in the inhibition of growth and proliferation of osteosarcoma cells potentially via an apoptosis mechanism when used together with doxorubicin [62].
Considering the importance of skincare products, as a part of our continuous work in the way to the preparation of potent CGA analogues targeting α-MSH-stimulated melanogenesis inhibition [63,64], we have reported the design and synthesis of dimethoxy phenyl analogues of CGA comprising of pyridine, pyrimidine, and cyclohexyl caffeamide skeletons ( Figure 1) along with their inhibitory activities against melanin formation. In addition, the stability of diacyl analogues of cyclohexyl ester (i.e., diacyl caffeates) were examined to see the effect of acyl substituents on their stability profiles.

Chemistry
Several pyridine-containing chlorogenic acid analogues 6a-g were prepared, as shown in Scheme 1. The starting material α,β-unsaturated nitrile 2 was obtained by the Horner-Wadsworth-Emmons (HWE) olefination of 3 ,4 -dimethoxy acetophenone 1 with diethyl cyanomethyl phosphonate. The α,β-unsaturated nitrile was treated with DMF-DMA and NH 4 OAc in DMSO at 120 • C yielded enamino nitrile intermediate 3, which was further reacted with an excess of acetyl bromide in ethyl acetate and H 2 O at 0 • C to form the required bromopyridine intermediate 4 [65]. Furthermore, compound 4 was treated with various type of benzyl amines 5a-g comprising -CF 3 and -Cl groups on the benzene ring in the presence of CsOAc and Cu at 90 • C in DMSO, which yielded desired pyridine ring-containing analogues 6a-g.

Chemistry
Several pyridine-containing chlorogenic acid analogues 6a-g were prepared, as shown in Scheme 1. The starting material α,β-unsaturated nitrile 2 was obtained by the Horner-Wadsworth-Emmons (HWE) olefination of 3′,4′-dimethoxy acetophenone 1 with diethyl cyanomethyl phosphonate. The α,β-unsaturated nitrile was treated with DMF-DMA and NH4OAc in DMSO at 120 °C yielded enamino nitrile intermediate 3, which was further reacted with an excess of acetyl bromide in ethyl acetate and H2O at 0 °C to form the required bromopyridine intermediate 4 [65]. Furthermore, compound 4 was treated with various type of benzyl amines 5a-g comprising -CF3 and -Cl groups on the benzene ring in the presence of CsOAc and Cu at 90 °C in DMSO, which yielded desired pyridine ring-containing analogues 6a-g. Various pyrimidine analogues 10a-f were synthesized using the method shown in Scheme 2. Initially, benzyl guanidine intermediates 8a-f [66] were obtained by treating 7 [67] with different types of benzyl amines 5a-f bearing -CF3 and -Cl groups. Then, (E)-1- Various pyrimidine analogues 10a-f were synthesized using the method shown in Scheme 2. Initially, benzyl guanidine intermediates 8a-f [66] were obtained by treating 7 [67] with different types of benzyl amines 5a-f bearing -CF 3  intermediates 8a-f and 9 were coupled under basic condition at refluxing temperature to afford the target compounds 10a-f. intermediates 8a-f and 9 were coupled under basic condition at refluxing temperature to afford the target compounds 10a-f. Furthermore, we approached the synthesis of substituted diacyl 13a-d and dialkyl CGA cyclohexyl ester analogues 15, 17 as portrayed in Scheme 3. Caffeic acid 11 was Oacylated with acid halide in pyridine in the presence of DMAP catalyst to yield corresponding O-acylated analogues 12a-d. The obtained analogues 12a-d were further reacted with thionyl chloride in toluene under N2 atmosphere at 100 °C, which was followed by esterification using cyclohexanol catalyzed by DMAP at rt, which afforded 13a-d. Similarly, compound 15 was obtained by treating caffeic acid 11 with cyclohexanol in refluxing sulfuric acid affording cyclohexyl ester precursor 14, which was followed by O-alkylation of 14 with isobutyl bromide in basic condition. On the other hand, the compound 17, a dimethoxy caffeic acid cyclohexyl ester analogue, was prepared by reacting dimethoxy caffeic acid 16 with thionyl chloride in refluxing toluene followed by esterification with cyclohexanol at rt, affording good yield. Furthermore, we approached the synthesis of substituted diacyl 13a-d and dialkyl CGA cyclohexyl ester analogues 15, 17 as portrayed in Scheme 3. Caffeic acid 11 was O-acylated with acid halide in pyridine in the presence of DMAP catalyst to yield corresponding O-acylated analogues 12a-d. The obtained analogues 12a-d were further reacted with thionyl chloride in toluene under N 2 atmosphere at 100 • C, which was followed by esterification using cyclohexanol catalyzed by DMAP at rt, which afforded 13a-d. Similarly, compound 15 was obtained by treating caffeic acid 11 with cyclohexanol in refluxing sulfuric acid affording cyclohexyl ester precursor 14, which was followed by O-alkylation of 14 with isobutyl bromide in basic condition. On the other hand, the compound 17, a dimethoxy caffeic acid cyclohexyl ester analogue, was prepared by reacting dimethoxy caffeic acid 16 with thionyl chloride in refluxing toluene followed by esterification with cyclohexanol at rt, affording good yield.

R IC 50 (µM) a,c
Kojic acid b - ->100 a Data are displayed as mean ± SD (n = 2), b Kojic acid and arbutin were used as the positive control, c Cytotoxicity at >30 µM for all the tested analogues (except 6f). All pyridine skeletons containing analogues (Table 1, 6a-g) exhibited good inhibition of melanin formation in B16 melanoma cells compared to two positive controls (i.e., kojic acid). Based on the degree of inhibition ability, compound 6f (IC 50 : 2.5 ± 0.7 µM) containing -CF 3 at the benzylic meta position was found to be excellent among the pyridine series. However, its para and ortho position (6e, IC 50 : 8.5 ± 0.7 µM and 6g, IC 50 : 7.5 ± 0.7, respectively) showed slightly diminished inhibitions. The -Cl substitution at the benzylic para and meta position (6b, IC 50 : 4.5 ± 0.7 µM and 6c, IC 50 : 5.0 ± 1.4, respectively) have similar inhibitions, but its ortho position showed somewhat lower inhibition effect (6d, IC 50 : 8.5 ± 0.7 µM). It seems that pyridine analogues have better activity with electronwithdrawing substituents, since benzyl (i.e., without an electron-withdrawing group) analogue 6a was found to be the least active in the series.
Compared to pyridine analogues, the pyrimidine analogues displayed a lower inhibition power of melanin production. Analogues 10c (IC 50 : 32.0 ± 2.0 µM), 10d (IC 50 : 40.0 ± 8.0 µM), and 10f (IC 50 : 20.0 ± 1.0 µM) were found to be better in inhibition ability compared to positive control kojic acid. However, the analogues 10a without benzyl substitution and 10e with -CF 3 at the benzylic para position were weaker in inhibition power than kojic acid, displaying more than 100.0 µM IC 50 values. Although most analogues were better than the positive control, a clear correlation of structure-activity relationship (SAR) in this series was not observed.
Likewise, the activity against α-MSH stimulated melanin formation in B16 melanoma cells by the acyl 13a-d, and alkyl 15, 17 caffeic acid cyclohexyl ester derivatives were performed. As the result depicted in Table 1, acyl 13a-d showed a highly potent inhibition of α-MSH stimulation compared to kojic acid and most of the pyridine analogues of CGA. All analogues 13a-d exhibited IC 50 values lower than 3.0 µM concentration. The SAR study indicated that the activity of acyl analogues is marginally diminished in increasing length of carbon chains and steric crowds of acyl substituents. For example, analogue 13a (IC 50 : 1.0 ± 0.1 µM) with n-propyl substitution was found to be the most potent, whereas analogue 13d (IC 50 : 2.3 ± 0.2 µM) with the largest substituent (i.e., cyclohexane) unit among the series displayed slight inferior potency. Likewise, alkyl (15 and 17) analogues of caffeic acid cyclohexyl esters were also poorly active to inhibition of melanin formation (i.e., IC 50 : >100 µM).
Our previous report demonstrated that most CGA analogues displayed a wide margin of safety window to their IC 50 value of inhibitory effect when cytotoxic effects were measured by the MTT assay method [63,64]. Herein, the results were similar to the previous report for the new analogues where the preliminary assay showed no significant cytotoxic effects in >30 µM concentration except for analogue 6f. However, their cytotoxic IC 50 value will be measured in our further works.
As we continuously utilized caffeamide and caffeate analogues to investigate the inhibition of α-MSH stimulation on B16 melanoma cells, we found diacetyl catechol derivative (18, IC 50 : 0.02 ± 1.05) as a highly potent analogue in previous work [63]. However, it was found to be highly unstable upon normal storage condition at room temperature because of the reactive acyl moiety. Since the activity of acyl analogues (13a-13d) of caffeates are highly promising in the present work as well, we aimed to perform chemical stability testing of synthesized acyl analogues in methanol solution, and their stability profile is depicted in Figure 2 and Table 2. The detailed procedure is given in the Section 4. ative (18, IC50: 0.02 ± 1.05) as a highly potent analogue in previous work [63]. However, it was found to be highly unstable upon normal storage condition at room temperature because of the reactive acyl moiety. Since the activity of acyl analogues (13a-13d) of caffeates are highly promising in the present work as well, we aimed to perform chemical stability testing of synthesized acyl analogues in methanol solution, and their stability profile is depicted in Figure 2 and Table 2. The detailed procedure is given in the Materials and Method section.   As in our previous observation, the caffeate analogue 18 was found to be the least stable compound as it was decomposed by more than 90% within 10 days, and its amount was dropped to less than 1% in 28 days in methanol solution. In this series, cyclohexyl acyl analogue 13d displayed an excellent stability profile by retaining approximately 87% of its amount in methanol even in 28 days. The substrate having n-propyl 13a and npentyl 13b substituents showed the moderate stability profile, as 39% and 52% remained unreacted in solution for 28 days, respectively. The result suggested that the linear alkyl acyl-substituents are highly reactive; thus, they can be decomposed easily upon exposure to nucleophiles such as methanol. However, the installations of sterically bulky substituents improve stability. For instance, analogues 13c with a bulkier t-butyl substituent showed the higher stability rate as 94% remains unreacted after 28 days.
In general, analogues in the pyrimidine series are less effective than the corresponding caffeamide, thiazole, and pyridine analogues. Although the pyrimidine series are not as effective as the pyridine series in the inhibition of the melanogenesis process, the effect of -CF 3 at the benzylic meta position has a similar pattern, since 10f (IC 50 : 20.0 ± 1.0 µM) displayed higher activity than its ortho (IC 50 : 40.0 ± 8.0 µM) and para (IC 50 : >100.0 µM) derivatives.
By observing the result of our current and previous pyridine/pyrimidine and thiozole analogues of CGA, we found it to be less potent in inhibition activity than parent analogues (i.e., caffeamide and caffeate). Despite the higher inhibition ability of caffeamide and caffeate, the stability of the acyl-substituents on phenol were the most challenging. By substituting with bulkier acyl-substituents, the stability profile was improved in the present work. As we found that the most of our analogues potently block the melanogenesis in B16 melanoma cells with a wide safety profile, thus, the CGA analogous might become a good scaffold for the discovery of anticancer agents and/or cosmeceuticals.

General Information
1 H NMR spectra were recorded on a Jeol RESONANCE ECZ 400S (400 MHz). Chemical shifts are reported in ppm from tetramethylsilane (TMS) with the solvent resonance resulting from incomplete deuteration as the internal reference (CDCl 3 : 7.26 ppm) or relative to TMS (δ 0.0). Data are reported as follows: chemical shift (ppm), multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, br = broad, m = multiplet, dd = doublet of doublet, td = triplet of doublet), coupling constants (Hz), number of protons. 13 C NMR spectra were recorded on a Jeol RESONANCE ECZ 400S (100 MHz) with complete proton decoupling. Chemical shifts are reported in ppm from tetramethylsilane with the solvent as the internal reference (CDCl 3 : 77.16 ppm). High-resolution mass spectrometry was performed on LCQ Fleet-Thermo Scientifics (Waltham, MA, USA). All reactants or reagents were purchased from Aldrich TCI, alfa aesar, and acros were directly used without further purifications. Silica gel column chromatography was performed with Silica Gel of Kieselgel 60 (Merck, KGaA, 64271 Darmstadt, Germany).

General Procedure for the Synthesis of Pyridine Analogues (6a-g) of CGA
2-Bromo-4-(3,4-dimethoxyphenyl) pyridine 4 (0.13 mmol), cesium acetate (0.05 g, 0.28 mmol), and copper powder (1 mg, 0.013 mmol) was taken in anhydrous DMSO (0.13 mL). Then, benzylamine derivatives 5a-5g (0.19 mmol) were added under nitrogen atmosphere, and the reaction mixture was stirred at 90 • C for 24 h. After the completion of the reaction, the reaction mixture was cooled to room temperature, the precipitate was filtered with ethyl acetate, the organic layer was washed with water, dried over anhydrous MgSO 4 , concentrated under reduced pressure, and purified by column chromatography. All the synthesized compounds were identified by 1 H NMR, 13

Synthesis of O-Acylated Derivatives of Acrylic Acid (12a-12d)
To synthesize acrylic acid derivatives 12a-d, the starting material caffeic acid 11 (11.10 mmol) was dissolved in pyridine (10.0 mL) under nitrogen atm, DMAP (0.110 mmol) was added, and it was stirred at 0 • C for 10 min. Then, alkyl acid chlorides (554.14 mmol) were added dropwise and stirred at room temperature for 24 h. Once the reaction was complete, the organic layer was washed three times with ethyl acetate and 1N HCl; then, it was dried with MgSO 4 , filtered and concentrated. The pure product was obtained using silica gel column chromatography (EA:Hex = 1:1).

Synthesis of Cyclohexyl Ester Analogues 13a-d of CGA
To a solution of 12a-d (1.0 equiv.) in toluene (1.0 mL), SOCl 2 (5.0 equiv.) was added dropwise and stirred at rt for 3 h. Once the starting materials were consumed, the excess SOCl 2 and solvent were removed. Then, the residue was dissolved in DCM (2.0 mL) and DMAP (0.2 equiv.) was added; and stirring was continued for 30 more minutes. Next, cyclohexanol (1.0 equiv.) was added to the reaction mixture and stirred for additional 12 at rt. After the reaction was completed, the solvent was removed under reduced pressure, and the obtained crude product was purified by using silica gel column chromatography to afford 13a-d. (

Synthesis of (E)-Cyclohexyl 3-(3,4-dihydroxyphenyl)acrylate (14)
To the caffeic acid (11) (0.10 g, 0.55 mmol) in cyclohexanol (20 mL), sulfuric acid (0.2 mL, 3.7 mmol) was added dropwise and stirred at reflux for 1 h. Reaction progress was monitored by using TLC. After that, it was diluted with ethyl acetate and washed three times with NaHCO 3 solution and three times with aqueous NaCl solution. The organic layers were collected and dried by using MgSO 4 , and solvent was evaporated under reduced pressure. The obtained crude reaction mixture was purified by silica gel column chromatography (EA:Hx = 1:3) to give compound 14 as a brown solid.

Synthesis of (E)-Cyclohexyl 3-(3,4-dimethoxyphenyl)acrylate (17)
To the solution of 3,4-dimethoxycinnamic acid 16 (0.48 mmol) in toluene (1 mL), thionyl chloride (0.17 mL, 2.4 mmol) was added under nitrogen atm, and the mixture was stirred under reflux for 1.5 h. Then, after removing thionyl chloride, it was dissolved in anhydrous CH 2 Cl 2 under nitrogen atm, and then DMAP (8 mg, 0.049 mmol) was added, and stirring continued for 10 min. Then, cyclohexanol (0.05 mL, 0.48 mmol) was added dropwise to the reaction mixture and stirred at room temperature for 24 h. The reaction mixture was poured into 10 mL of water and extracted with DCM. Dried with MgSO 4 , solvent was evaporated to obtain crude reaction mixture. This was purified by silica gel column chromatography (EA:Hx = 1:6) to obtain compound 17.
Yield: 71%, 1  Mouse melanoma B16 cells were cultured in a 96-well culture plate (2.5 × 10 3 cells/ well) [63]. Culture media were washed from each culture plates after 24 h and treated with freshly prepared 10% FBS-DMEM medium, various concentrations (such as 0.5, 1.0, 3.0, 10.0, 30.0, and 100.0 µM) of compound samples, and stimulants (α-MSH, 10.0 nM). Then, it was incubated at 37 • C for 72 h, and the amount of melanin released into the medium was determined by measuring the absorbance at wavelength of 405 nm using spectrophotometer (Molecular Device, San Jose, CA, USA). The absorbance of stimulated cells using α-MSH was used as a control, and the inhibition level was calculated by subtracting the absorbance values of cells that were not stimulated. The absorbance values were compared with standard curves obtained with synthetic melanin. The synthetic melanin (Sigma-Aldrich, St. Louis, MO, USA) was dissolved in 0.85 M KOH-10% DMSO solution; then, it was serially diluted with 10% FBS-DMEM medium and then aliquoted into a 96-well culture plate. Then, the absorbance of melanin was measured by a spectrophotometer at 405 nm, and a standard curve was prepared. Using this value, the amount of melanin released out of the cell was directly calculated. Results obtained by three or more sets of experiments were expressed as mean ± standard error.

MTT Assay
Melanoma B16 cells were seeded into a 96-well culture plate at a seeding density of 2.5 × 10 3 cells/well and incubated for 24 h at 37 • C supplied with 5% CO 2 [68]. The culture medium of each well was replaced with a fresh 10% FBS-DMEM medium and treated with serially diluted samples and α-MSH. After 72 h of incubation, cells were incubated with fresh 10% FBS-DMEM medium. After 2 h of stabilization, 100 µL of 1 mg/mL MTT solution was added per well to lyse the cells; then, the absorbance at 590 nm was measured using a spectrophotometer. The experiments were carried out three or more times, and the results were expressed as mean ± standard error.

Stability of CGA Acyl Analogues in Methanol Solution
The stability of the synthesized acyl analogues was measured in methanol, since the acetyl analogue of CGA was highly unstable in methanol solution. First, 10 mg of the sample was dissolved in 2 mL of methanol and stored at 25 • C. Then, purity was analyzed using HPLC (Shimadzu LC solution, Kyoto, Japan) for 28 days. The samples in methanol solution were diluted 10 times in the HPLC mobile phase ((A) 0.5% formic acid in H 2 O (B) 0.5% formic acid in acetonitrile (ACN), A/B = 20/80) and injected in HPLC column (inertsil ODS-3V C18 column (250 × 4.00 mm i.d; 5 µm)). Finally, the amount of acyl compound remaining in the solution was measured by calculating the area under the curve of the HPLC chromatogram, and the results were expressed in percentage units.

Conclusions
We have developed the synthesis method of CGA analogues comprising pyridine, pyrimidine motifs, as well as cyclohexyl ester analogues in moderate to good yields. Among the nineteen synthesized compounds, fifteen compounds displayed potent inhibition activity against melanin production in B16 melanoma cells upon α-MSH stimulation. Results showed that pyridine containing 6b, 6c, 6f analogues, and diacyl CGA cyclohexyl ester analogues 13a-d have greatly superior inhibition potency than the positive control, kojic acid. In addition, analogues 13c and 13d showed high stability profile in methanol solution compared to analogue 18 in 28 days of stability monitoring.