Bioevaluation and molecular docking analysis of novel phenylpropanoid derivatives as potent food preservative and anti-microbials

Abstract

Novel derivatives were synthesized using natural scaffold, like phenylpropanoids C₆–C₃ backbone to reduce unfavorable browning of food due to tyrosinase and oxidative spoilage. Most of the compounds displayed mushroom tyrosinase inhibition better than kojic acid. Compound CE48 exhibited better anti-tyrosinase (IC₅₀-29.64 μM) and antioxidant (EC₅₀-12.67 μM) activity than the reference compounds, kojic acid (IC₅₀-50.30 μM) and ascorbic acid (EC₅₀-14.55 μM), respectively. Compounds SAM30, SE78, 11F, and CE48 showed better anti-B. subtilis, anti-S. aureus, and anti-A. niger activity, respectively, compared to their parents. Molecular docking studies between inhibitors and mushroom tyrosinase corroborated the experimental reports, except SAM30 (glide score − 8.117) and SE78 (glide score − 6.151). In silico absorption, distribution, metabolism, excretion/toxicity (ADME/T) and toxicological studies of these newly synthesized compounds exhibited acceptable pharmacokinetic and safety profiles, like good aqueous solubility (− 3.34 to − 7.57), low human oral absorption (e.g., SAM30, SE78, FAM34), low gut–blood barrier permeability [36.67–209.88 nm/s in Cancer coli-2 (Caco-2) cells] and [19.45–91.51 nm/s in Madin-Darby Canine Kidney (MDCK) cells], low blood–brain barrier penetration, non-mutagenicity, and non-carcinogenicity. Interestingly, the synthesized compounds also possessed multifunctional properties, like microbial growth inhibitor, free radicals scavenger, and it also prevented browning of raw fruits and vegetables by inhibiting tyrosinase enzyme.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13205-020-02636-0.

Introduction

Currently, the food industry is in a difficult situation chiefly due to the unfavorable browning and microbial deterioration of their products, especially fruits such as apples, bananas, pears; vegetables such as mushrooms, potatoes; and seafood such as shrimp during postharvest handling, processing, and storage (Chen et al. 2019; Chochkova et al. 2014; Larik et al. 2017; Xing et al. 2016). Enzymatic browning is depicted as the second major cause for the loss of quality in vegetables and fruits (Ioannou and Ghoul 2013). This unpleasant discoloration due to enzymatic browning is mainly caused by tyrosinase (EC 1.14.18.1), a copper-containing metalloenzyme that catalyzes two reactions; first, hydroxylation of tyrosine, and second, oxidation of l-3,4-dihydroxyphenylalanine (l-DOPA) into quinones, which eventually results in melanin formation (Xing et al. 2016). Quinones may react irreversibly with sulfhydryl group and amino group of the proteins, which may result in a decrease in protein digestibility and bioavailability of amino acids such as cysteine and lysine (Sheng et al. 2018). Browning degrades nutritive value and other qualities of food products that ultimately affect its shelf-life and economic value (Dong et al. 2016; Xing et al. 2016). Besides, excessive accumulation of melanin can lead to hyperpigmentation disorders in mammals (Xie et al. 2016). In course of processing and storage of fresh-cut produce (fruits and vegetables) and seafood, fluid oozes from the damaged cells and attract microbes due to the rich source of nutrients in comparison to undamaged produce. This provides an ideal platform for the growth of food spoilage microorganisms, which eventually deteriorates the quality of food products (Chen et al. 2019; Qadri et al. 2015).

Furthermore, the oxidative spoilage of food products during processing and storage results in free radical formation, which is one of the major causes for deterioration of food and that results in unpleasant texture, flavor, and color in food products (Bajpai et al. 2017). Taking into account the above factors, many anti-tyrosinase, antioxidants, and antimicrobial compounds have found their application in preservation of food products, but most of them exhibit adverse side effects including carcinogenicity, toxicity, low stability, or low activity (Bajpai et al. 2017; Chen et al. 2019), which cannot be neglected. Therefore, there is an urgent need for the development of new and safer compounds with multifunctional effects like inhibiting microbial growth, scavenging for free radicals, and preventing browning by inhibiting tyrosinase enzyme.

Phenylpropanoids (PPs) are notable natural scaffolds, with C₆–C₃ carbon skeleton as core structure, and are the major category of phenolic acids. Commonly known as hydroxycinnamic acid, it includes p-coumaric acid (p-CoA), caffeic acid (CA), ferulic acid (FA), and sinapic acid (SA); (Natella et al. 1999; Teixeira et al. 2013; Neelam and Sharma 2020). A range of PPs and their derivatives have been illustrated to possess broad-spectrum antimicrobial activities (Engels et al. 2012; Khatkar et al. 2015; Lima et al. 2016) mainly due to their cell membrane damaging features (Hemaiswarya et al. 2011). Also, PPs possess strong antioxidant activity, primarily related to the presence of extended side-chain conjugation, hydroxyl function, and methoxyl group in their structure (Jia et al. 2018). Moreover, natural PPs (CA, FA, and SA) have been reported to have anti-tyrosinase activity, which is due to their structural resemblance to l-tyrosine and l-DOPA, natural substrates of tyrosinase (Takahashi and Miyazawa 2010). N-nonyl caffeate, a caffeic ester analog checks postharvest putrefaction of Chinese Olive by inhibiting bacterial growth and enzymatic browning (Jia et al. 2016). N-hydroxycinnamic acids including CA, FA, SA, and their derivatives have been reported to have both antioxidant and anti-tyrosinase activity (Chochkova et al. 2014; Takahashi and Miyazawa 2010, Neelam and Sharma 2020).

Therefore, we designed and synthesized novel PPs derivatives by modifying the carboxylic group of natural scaffold PPs (CA, FA, and SA) to obtain promising multifunctional compounds displaying anti-tyrosinase, antioxidant, and antimicrobial activities with no adverse effects. Molecular docking and absorption, distribution, metabolism, excretion/toxicity (ADME/T) studies were also conducted to envision the binding interactions of PPs derivatives with tyrosinase enzyme (PDB code: 2Y9W) and to predict the pharmacokinetics and toxicity profiles.

Materials and methods

Chemicals

2, 2-Diphenyl-1-picrylhydrazyl (DPPH) was procured from Sigma-Aldrich, India, and all other chemicals required for this study were procured from commercial suppliers. The progress and completion of each reaction was checked by thin-layer chromatography (TLC) on silica gel pre-coated plates of 0.25 mm thickness, purchased from Merck. Melting points were determined in open capillaries on a Sonar melting point apparatus and were uncorrected. ¹H nuclear magnetic resonance (NMR) spectra were documented in a dimethyl sulfoxide (DMSO-d₆₎ solution on a Bruker advance II 400 NMR spectrometer. Fourier transform infrared (FT-IR) spectra were taken on a PerkinElmer spectrum two FT-IR spectrometer using KBr pellets.

Synthesis

Synthesis of PPs derivatives was carried out with minor modifications (Fig. 1) following an earlier report from our laboratory (Khatkar et al. 2015).

Fig. 1.

Synthesis routes of the phenylpropanoid derivatives. The amide derivatives SAM30 and FAM34 were obtained by treatment of sinapoyl chloride (SC) and feruloyl chloride (FC) with appropriate amine in the presence of diethyl ether (a, b). Compounds SE78, 11F, and CE48 were obtained by the general reaction of SC, FC, and caffeoyl chloride (CC) with appropriate alcohol and diethyl ether (c, d, e)

The general process for the synthesis of ester derivatives of phenylpropanoids (SE78, CE48, and 11F)

First, phenylpropanoyl chloride, that is, sinapoyl chloride (SC), feruloyl chloride (FC), and caffeoyl chloride (CC) were prepared by gradual addition of 0.042 mol thionyl chloride (SOCl₂) to 0.035 mol of PPs, that is, SA, FA, and CA, respectively, each in a separate round-bottom flask. The above reaction mixtures were then refluxed for 6 h in case of FC and 8 h for SC and CC each, as depicted in Fig. 1. Furthermore, excess of SOCl₂ was eliminated by distillation.

The solutions of different alcohol (0.0052 mol) in diethyl ether (50 ml) were refluxed with a solution of phenylpropanoyl chloride (0.0052 mol) in diethyl ether (50 ml) at 80 °C in hot water bath for 4–12 h, as described in Fig. 1. The reaction mixture was cooled down to room temperature, and crude products were formed after evaporation of solvent, which were purified by re-crystallization with lower alcohol.

Butane-1, 3-diyl (2E, 2′E) bis [3-(4-hydroxy-3-methoxyphenyl) prop-2-enoate] (11F)

Rf TLC mobile phase: Chloroform:Ethyl acetate:Acetic acid (50:50:1) = 0.61; (%) Yield = 74.63; M.P. = 116–118 °C; IR (KBr) cm⁻¹: 1113 (C–O str., ester), 1465 (C–H bend, alkane), 1515 (C=C skeletal str., phenyl), 1632 (C=C., alkene), 1714 (C=O str., ester), 2922 (C–H str., alkane); ¹H NMR (400 MHz, DMSO-d₆, δ ppm) δ 7.52–7.37 (m, 2H), 7.24 (s, 1H), 7.12 (S, 1H), 6.97 (dd, J = 12.0, 4.0 Hz, 1H), 6.86 (dd, J = 13.4, 8.3 Hz, 1H), 6.54–6.46 (M, 1H), 4.90 (d, J = 6.4 Hz, 1H), 3.91 (d, J = 6.9 Hz, 1H), 3.93–3.76 (m, 4H), 3.86–3.76 (m, 3H), 1.26–1.09 (m, 3H), 1.06–1.03 (m, 1H).

2-oxo-1, 2-diphenylethyl (2E)-3-(3, 4-dihydroxyphenyl) prop-2-enoate (CE48)

Rf TLC mobile phase: Toluene:Acetone (8:2) = 0.66; (%) Yield = 69.18; M.P = 90–92 °C; IR (KBr) cm⁻¹: 1113 (C–O str., ester), 1450 (C=C skeletal str., phenyl), 1632 (C=C., alkenes), 1793 (C=O str., ester). ¹H NMR (400 MHz, DMSO-d_6, δ ppm) δ 7.92 (d, J = 7.12 Hz, 3H), 7.79 (t, J = 7.56 Hz, 2H), 7.62 (t, J = 7.78 Hz, 3H), 7.41 (d, J = 15.88 Hz, 1H), 7.04 (d, J = 1.96 Hz, 1H), 7.04–6.94 (m, 3H), 6.95 (dd, J = 8.2, 1.92 Hz, 1H), 6.77 (d, J = 8.12 Hz, 1H), 6.19 (d, J = 15.88 Hz, 1H), 4.28 (s, 2H).

2, 2-bis {[(2E)-3-(4-hydroxy-3, 5-dimethoxyphenyl) prop-2-enoyl] methyl} butyl (2E)-3-(4-hydroxy-3, 5-dimethoxyphenyl) prop-2-enoate (SE78)

Rf TLC mobile phase: Chloroform:Methanol (7:3) = 0.82; (%) Yield = 58.87; M.P = 98–99 °C; IR (KBr) cm⁻¹: 1113 (C–O str., ester), 1461 (C–H bend, alkane), 1516 (C=C skeletal str., phenyl), 1631 (C=C., alkene), 1702 (C=O str., ester), 2941 (C–H str., alkane). ¹H NMR (400 MHz, DMSO-d_6, δ ppm) δ 7.38–7.34 (m, 1H), 7.38–6.92 (m, 1H), 4.62 (s, 1H), 4.56 (s, 1H), 3.78 (dd, J = 29.1, 9.1 Hz, 6H), 1.27–1.07 (m, 2H), 0.81 (d, J = 19.0 Hz, 2H).

General process for the synthesis of amide derivatives of phenylpropanoids (SAM30 and FAM34)

The reaction mixture of appropriate phenylpropanoyl chloride (0.0052 mol) in 50 ml diethyl ether maintained at 0–10 °C was treated by adding drop-wise solution of appropriate amine (0.0052 mol) dissolved in 50 ml diethyl ether (Fig. 1). The above reaction mixture was continuously stirred for 30 min, followed by filtration of the precipitated amide. The resultant amide was re-crystallized with lower alcohol.

N, N′-bis [(2E)-3-(4-hydroxy-3, 5-dimethoxylphenyl) prop-2-enoyl] benzene-1, 2 dicarboxamide (SAM30)

Rf TLC mobile phase: Chloroform:Ethyl acetate:Acetic acid (50:50:1) = 0.56; (%) Yield = 62.28; M.P = 96–97 °C; IR (KBr) cm⁻¹: 1515 (C=C skeletal str., phenyl), 1600 (N–H bend, 2° amide), 1668 (C=O., amide), 3209 (N–H str., 2° amide). ¹H NMR (400 MHz, DMSO-d_6, δ ppm) δ 7.84–7.70 (m, 4H), 7.75 (d, J = 5.68 Hz, 2H), 7.59 (d, J = 15.8 Hz, 1H), 7.49–7.45 (m, 1H),7.36–7.26 (m, 1H), 7.11 (d, J = 3.2 Hz, 2H), 6.99 (dd, J = 14.12, 7.32 Hz, 1H), 6.85 (s, 2H), 6.59–6.51 (m, 2H), 3.63 (s, 6H).

(2E)-3-(3-methoxy-4-hydroxyphenyl)-N-({2-[(2E)-3-(3-methoxy-4-hydroxyphenyl) prop-2-enoyl]hydrazinyl} carbonothioyl)prop-2-enamide (FAM34)

Rf TLC mobile phase: Chloroform:Ethyl acetate:Acetic acid (50:50:1) = 0.71; (%) Yield = 52.39; M.P = 198–199 °C; IR (KBr) cm⁻¹: 1401 (C=C skeletal str., phenyl), 1631 (N–H bend, 2° amide), 1696 (C=O., amide), 3152 (N–H str., 2° amide). ¹H NMR (400 MHz, DMSO-d_6, δ ppm) δ 7.93–7.71 (m, 2H), 7.60 (d, J = 17.5 Hz, 1H), 7.93–7.13 (m, 9H), 7.53–7.17 (m, 6H), 7.22 (s, 1H), 7.22 (s, 1H), 6.99 (d, J = 11.1 Hz, 1H), 6.83–6.08 (m, 1H), 6.39–6.10 (m, 1H), 3.81 (dd, J = 18.1, 12.8 Hz, 7H), 3.35 (s, 8H), 2.53 (d, J = 15.5 Hz, 1H), 1.86 (d, J = 5.0 Hz, 1H).

Biological studies

Extraction of tyrosinase enzyme from mushroom

The common mushroom, Agaricus species was procured from our laboratory and used for the extraction of tyrosinase enzyme, as per the method described earlier (Zaidi et al. 2014), with minor modifications. Pre-washed mushrooms were sliced and homogenized by blender. The mixture was treated with 50 mM citrate buffer (pH 6.0) for 40 min at 4 °C with constant shaking. The suspension was then centrifuged at 5000 rpm for 30 min. The resulting supernatant was collected and used as a crude enzyme source for further studies.

Tyrosinase inhibition assay

The tyrosinase inhibition in mushroom was performed using l-tyrosine as substrate using synthesized PPs derivatives with few modifications in previously described method (Karkouch et al. 2017). Briefly, assay mixture containing 60 μL of potassium phosphate buffer (50 mM, pH 6.8), 105 μL of mushroom tyrosinase, and 15 μL of test sample/standard inhibitor were incubated for 10 min at 25 °C and then 120 μL of l-tyrosine (2.5 mM) was added and further incubated for 30 min at 25 °C. The absorbance of reaction mixture at 475 nm was measured using microplate reader (Molecular Devices, Sunnyvale, USA). Kojic acid was used as a standard inhibitor. The extent of tyrosinase inhibition by the synthesized compounds was represented as IC₅₀, the concentration of the synthesized compounds needed to achieve 50% inhibition of mushroom tyrosinase activity. The IC₅₀ value was calculated by the online IC₅₀ calculator tool of AAT Bioquest, (https://www.aatbio.com/tools/ic50-calculator).

Antioxidant activity by DPPH assay

The antioxidant activity of synthesized compounds was investigated by modifying already reported DPPH assay (Akroum et al. 2010). Briefly, 1 ml methanolic solution of test compound (final concentration, 5–25 µM) mixed with 1 ml of 100 µM methanolic DPPH solution was allowed to stand in dark at room temperature for 30 min. The absorbance of reaction mixture was then measured at 517 nm using a UV-1800, UV–Vis spectrophotometer (Shimadzu, Japan). Ascorbic acid was employed as a standard inhibitor.

The antioxidant capacity of the compounds was represented by EC₅₀ (50% Effective Concentration). EC₅₀ value is defined as the amount of antioxidant required to reduce the concentration of DPPH radicals by 50% (Villano et al. 2007).

Antimicrobial activity of phenylpropanoids and their derivatives

The minimal inhibitory concentration (MIC) of each compound was evaluated against Escherichia coli MTCC 443, Bacillus subtilis (from laboratory of Enzymology and Recombinant DNA Technology, Department of Microbiology, M.D. University, Rohtak, India), Staphylococcus aureus MTCC 737, Aspergillus niger MTCC 282, and Candida albicans MTCC 227 using broth macro-dilution method (CLSI 2012; Zuzarte et al. 2011). Briefly, two-fold serial dilutions of test samples were made in nutrient broth for bacteria or Sabouraud dextrose broth for fungi. Each tube containing test sample was inoculated and incubated aerobically at 37 °C, for 24 h and at 35 °C, for 48 h for bacteria and fungi, respectively. Control tubes containing growth medium and inoculums were maintained for each test batch. Two commercially available antibiotics, amoxicillin and fluconazole, were procured from the local pharmacy and were used as standards.

Computational studies

Molecular docking protocol

In silico molecular docking studies of the designed PPs derivatives was done using Glide module of Schrodinger software (Friesner et al. 2004), (Schrodinger Release 2018-2, Schrodinger, LLC, New York, NY, 2018). The crystal structure of tyrosinase from Agaricus bisporus (PDB Code: 2Y9W) was obtained from Protein Data Bank (https://www.rcsb.org/) and processed using protein preparation wizard. Initially, all chains (B, C, and D), hetatms, and water molecules were omitted except for chain A and an H₂O molecule coordinated with Cu²⁺ followed by generation of their metal-binding state. Using SiteMap module, the binding pocket of 2Y9W (tyrosinase) was defined, and 3-D receptor grid was crafted by selecting the binding site residues. The proposed PPs derivatives were sketched using ACD/chem sketch freeware version 12.0 and converted to 3-D structure using ACD/3-D viewer, and the optimized structure was saved in mol format. Prior to enzyme–ligand docking, the 3-D structure of ligands was processed and optimized using Ligprep module. Finally, after protein and ligand preparation, and characterizing the grid corresponding to the binding pocket (active site) of tyrosinase, the ligands were docked to tyrosinase using flexible docking protocol of GLIDE module in extra precision (XP) mode. Based on glide scoring function, the docked compounds were evaluated.

In silico ADME and toxicity prediction

In silico ADME studies of the target compounds were performed using QikProp tool of Schrodinger suite. It predicted various pharmacokinetic parameters listed in Table 4.

Table 4.

Results of ADME parameters of phenylpropanoid derivatives

Descriptors	SAM30	SE78	FAM34	11F	CE48	Standard range
DonorHB	4	3	4	2	2	0.0–6.0
AcceptorHB	10.5	12.75	7.5	7	5.5	2.0–20.0
Lipinski’s rule of five	2	3	0	0	0	Max. 4
CNS (Central nervous system)	− 2	− 2	− 2	− 2	− 2	2 active, -2 inactive
QPPCaco (apparent Caco-2 cell permeability, nm/s)	36.67	78.10	42.56	178.13	209.88	> 500 great, < 25 poor
QPlogBB (Brain–blood partition coefficient)	− 3.44	− 3.52	− 3.01	− 1.88	− 1.86	− 3.0–1.2
QPPMDCK (Apparent MDCK cell permeability, nm/s)	19.45	31.44	36.63	76.65	91.51	> 500 great, < 25 poor
QPlogS (Aqueous solubility, mol dm−3)	− 7.57	− 6.5	− 6.41	− 3.34	− 5.38	− 6.5–0.5
HOA (Human oral absorption)	1	1	1	3	3	1 (low), 2 (medium), 3 (high)
QPlogPo/w (Octanol/water partition coefficient)	3.95	5.43	3.08	3.24	3.61	− 2.0–6.5

In silico toxicity profiling has been done by using an online-server, toxicity prediction module of PreADMET ver 2.0 (https://preadmet.bmdrc.kr/toxicity/). It predicts mutagenicity from Ames test (in vitro) and carcinogenicity from mouse (in vivo).

Results and discussion

Chemistry

Compounds (SE78, CE48, 11F, SAM30, and FAM34) were synthesized according to the method described previously (Khatkar et al. 2015), with minor modifications. The synthetic route of esters and amides of PPs were depicted in Fig. 1. In order to obtain ester and amide derivatives of PPs, we converted parent compounds (SA, FA, and CA) into phenylpropanoyl chloride, that is, SC, FC, and CC, respectively, by modifying their carboxylic group by addition of SOCl₂ and refluxed for 6–8 h, instead of stirring and heating the solution as reported earlier (Khatkar et al. 2015). Finally, the target compounds SE78, 11F, and CE48 were obtained by general reaction of SC, FC, and CC with appropriate alcohol and diethyl ether as their common solvent, refluxed for 4–12 h. The amide derivatives SAM30 and FAM34 were obtained by treatment of SC and FC with appropriate amine in presence of diethyl ether at 0–10 °C. To best of our knowledge, these ester and amide derivatives of PPs have previously never been reported in the literature. The structures of the newly synthesized PPs derivatives were confirmed by FT-IR and ¹H NMR spectra that were found to be in full harmony with their structures.

Biological evaluation

Tyrosinase inhibition

The effect of synthesized PPs derivatives in repressing tyrosinase activity was determined using tyrosinase enzyme extracted from Agaricus species and l-tyrosine was taken as a substrate for tyrosinase inhibition assay. The IC₅₀ values of all test samples and standard inhibitor, kojic acid is presented in Table 1. The lower IC₅₀ values represent the greater enzyme inhibitory activity. All of the synthesized compounds exhibit inhibitory effects on mushroom tyrosinase, and their IC₅₀ values ranged from 12.12 ± 0.50 to 111.35 ± 0.75 μM. The mushroom tyrosinase inhibition by the studied compounds followed the order: FAM34 > CE48 > 11F > Kojic acid > SAM30 > SE78 with IC₅₀ values 12.12 ± 0.50 μM, 29.41 ± 0.31 μM, 38.13 ± 0.14 μM, 48.74 ± 2.28 μM, 64.48 ± 0.65 μM, and 111.35 ± 0.75 μM, respectively.

Table 1.

Tyrosinase inhibition and DPPH radical scavenging activity of phenylpropanoid derivatives

Compound	Tyrosinase Inhibition IC50 (μM)	DPPH radical scavenging activity EC50 (μM)
SAM30	64.48 ± 0.65	21.09 ± 0.85
SE78	111.35 ± 0.75	15.75 ± 0.18
11F	38.13 ± 0.14	43.12 ± 1.47
FAM34	12.12 ± 0.50	96.99 ± 6.56
CE48	29.41 ± 0.31	12.67 ± 0.01
Kojic acid	48.74 ± 2.28	–
Ascorbic acid	–	14.55 ± 0.19

Results are represented as means ± standard deviation (SD); (n = 3)

Interestingly, N-feruloyl amide (b₄), phenylpropanoid amide (compound 5), and caffeic ester (compound 4) were reported to have higher inhibitory activities than most commonly used reference inhibitor, kojic acid (Jo et al. 2017; Takahashi and Miyazawa 2010). Some of the compounds exhibited lower inhibitory effects, which may be due to the presence of methoxy group at 3rd and 5th position on phenyl ring that may lead to steric hindrance (Tang et al. 2016). These reports suggested that the above-reported compounds were more felicitous than kojic acid when accounted as the mushroom tyrosinase inhibitor and fully support our findings in case of CE48 and FAM34 anti-tyrosinase activity.

In vitro antioxidant activity

The antioxidant activity of the synthesized compounds was assessed in vitro by DPPH radical scavenging assay (Akroum et al. 2010). Results are shown in Table 1. Caffeic acid ester, CE48 (EC₅₀ − 12.67 ± 0.01 μM) exhibited excellent radical scavenging activity compared to ascorbic acid (EC₅₀ − 14.55 ± 0.19 μM), a well-known inhibitor for DPPH assay. It is interesting to note that the higher antioxidant activity of CA ester in comparison to ascorbic acid was in good agreement with earlier reports (Wang et al. 2014). The order of antioxidant activity among the PPs esters was CE48 (CA ester) > SE78 (SA ester) > 11F (FA ester). This relative order of DPPH radical scavenging ability among the synthesized ester derivatives of PPs was found to be in accordance with those obtained by other investigators (Menezes et al. 2011). Furthermore, obtained reports indicated that the amide of SA was more potent antioxidant than the amide of FA. This was in accordance with the finding of Georgiev and co-workers, where they have demonstrated the antioxidant potential of N-hydroxycinnamoyl-amides, using a DPPH assay (Georgiev et al. 2012).

Antimicrobial activity evaluation

The free PPs (CA, FA, and SA), amoxicillin and fluconazole were used as reference compounds. The lowest concentration of each derivative that inhibited the visible microbial growth after the respective incubation period was considered as MIC. The results obtained as MIC are summarized in Table 2. The MIC values of the tested esters and amides varied from 3.75 to 30 mM and 3.75 to 15 mM, respectively. The compound 11F demonstrated better anti-S. aureus activity than the parent (FA) and compounds SAM30 and CE48 were found to be equivalent to their corresponding parent compound, SA and CA, respectively. Among all the investigated compounds, only SA derivatives (SE78 and SAM30) exhibited superior anti-B. subtilis activity than SA, the parent compound (Fig. S1). Except for compound CE48, all the other derivatives showed moderate activity against Gram-negative bacteria, E. coli. In the case of antifungal activity, compound CE48 (3.75 mM) was found to be potent anti-A. niger compound compared to the parent, CA (15 mM). Two synthesized compounds, FAM34 and SE78, displayed similar activity against A. niger and C. albicans, respectively, with MIC value of 7.5 mM. All other derivatives exhibited reduced antifungal activity compared to the parent compound and fluconazole.

Table 2.

Comparison of the antimicrobial activity of phenylpropanoid derivatives and parent compounds, expressed as MIC (mM)

Compound	Minimum inhibitory concentration (mM)
	E. coli	B. subtilis	S. aureus	A. niger	C. albicans
SAM30	15	7.5	3.75	15	15
SE78	15	7.5	15	15	7.5
Sinapic acid	15	15	3.75	1.875	7.5
11F	7.5	15	3.75	30	30
FAM34	ND	ND	ND	7.5	7.5
Ferulic acid	7.5	7.5	15	7.5	3.75
CE48	7.5	15	7.5	3.75	30
Caffeic acid	3.75	7.5	7.5	15	15
Amoxicillin	< 1	< 1	< 1	NA	NA
Fluconazole	NA	NA	NA	< 1	< 1

ND not determined, NA not applicable

Phenylpropanoids and their derivatives are well known for their antibacterial and antifungal activities, and same can be corroborated with published literature (Andrade et al. 2015; Barber et al. 2000; Khatkar et al. 2015). Even though the antimicrobial potential of studied compounds was lesser than the standard antibiotics (amoxicillin and fluconazole); some of the derivatives exhibited higher or equivalent antibacterial activity to the free PPs, which is in accordance with the observations made by other researchers (Chochkova et al. 2012; Georgiev et al. 2012). Thus, based on our findings, we suggest that a blend of these tested derivatives has potential to combat microbial infection caused due to E. coli, B. subtilis, S. aureus, A. niger, and C. albicans.

Computational studies

In silico molecular docking analysis

To delve into possible binding efficacy of synthesized PPs derivatives (SAM30, SE78, 11F, FAM34, and CE48), the X-ray crystallographic structure of tyrosinase (2Y9W) from Agaricus bisporus was docked with derivatives and a well-known inhibitor, Kojic acid. The docking score, binding energy, and molecular interactions were calculated for derivatives to obtain a potent inhibitor for tyrosinase inhibition (Table 3). The molecular docking analysis revealed that all the synthesized PPs derivatives had exhibited the binding interaction with one or the other amino acids in the active site of tyrosinase (Fig. 2 and Figs. S2–S6 are provided in the supplementary data). The compound SAM30 interacts with tyrosinase via multiple sites and exhibits stronger interactions with it, as evidenced by lowest Glide score (− 8.117) compared to kojic acid, a standard inhibitor (− 6.42) and other derivatives. The compound SAM30 interacted with V283 residue to establish a hydrogen bond (H-bond) within the backbone of tyrosinase. Additionally, SAM30 formed five H-bonds by interacting with side-chain residues H85, N81, N260, R268, and S282 of the target enzyme. SAM30 contains two SA moieties in its structure that formed two Pi–Pi stacking interactions, for instance phenolic ring of one of the SA moieties established Pi–Pi stacking with aromatic ring of H85, while phenolic ring of another SA moiety exhibited Pi–Pi stacking with aromatic case of F264 residue present in the hydrophobic region of substrate-binding domain of tyrosinase. Moreover, hydrophobic amino acid residues (C83, A246, V247, V248, V283, and P284) and polar amino acid residues (N81, T84, H85, H244, H259, N260, H263, S282, and T324) present in the substrate-binding domain may contribute to stabilize the formation of complexes between SAM30 and tyrosinase.

Table 3.

Molecular docking results of phenylpropanoid derivatives

Compound	Glide score	Glide energy (Kcal/mol)	Interactions			Interacting amino acids
			H-bond	Pi–Pi	Pi–cation
SAM30	− 8.117	− 60.362	6	2	–	N81, H85, N260, F264, R268, S282, V283
SE78	− 6.151	− 72.173	3	1	–	Y65, N260, F264, A323
11F	− 6.009	− 65.563	2	1	1	N81, H85, R268
FAM34	− 6.529	− 52.14	4	2	–	N81, C83, H85, A246, R321
CE48	− 6.393	− 57.893	2	1	–	N81, H85, F264
Kojic acid	− 6.42	− 32.434	4	–	–	N81, T84, H85, N320

Fig. 2.

Binding interactions of the newly synthesized phenylpropanoid derivatives with mushroom tyrosinase. a, b represents the interaction of SA derivatives SAM30 and SE78; C-D represents the interaction of FA derivatives 11F and FAM34; and E depicts CA derivative CE48 interaction with tyrosinase (PDB code: 2Y9W). Ligands are represented in brown ball-and-stick model. Two red spheres represent the copper atoms. Hydrogen bond interactions are presented in pink dotted lines, pi–cation interaction in red dotted line and green dotted line represent the pi–pi stacking interaction

It has been observed that N81, C83, H85, H244, V248, N260, H263, F264, S282, V283, and P284 spotted in the binding pocket of tyrosinase are the key residues that play a vital role in ligand binding and their stabilization. These findings are well supported by literature (Cui et al. 2015; Liu et al. 2017; Ullah et al. 2019). The proposed PPs ester derivatives (SE78, 11F, and CE48) interacted with active site residues A323, R268, F264, N260, H85, N81, and Y65. Molecular docking of cinnamic acid ester derivatives has been reported to occur at the active site residues of F264 and R268 (Sheng et al. 2018). It is interesting to notice that all the synthesized derivatives exhibited H-bonding and Pi–Pi stacking interactions, among which amide derivatives established the maximum number of interactions, that is, SAM30 formed six H-bonds and two Pi–Pi stacking interactions, while FAM34 build four H-bonds and two Pi–Pi stacking interactions. The computational studies (glide docking score) and tyrosinase inhibition assay studies revealed that except for SAM30 and SE78, all other compounds followed the same order (FAM34 > CE48 > 11F) as observed in our experimental studies. Thus, the overall findings suggest that our experimental results correlate with our molecular docking results.

In silico ADME and toxicity profiling

This study has proposed derivatives for inhibition of tyrosinase activity and to reduce the oxidation and microbial contamination of food products, with the least side effects in humans. To this effect, we have conducted computational studies to assess the ADME profile of derivatives, and their results are given in Table 4. All the derivatives fall in an acceptable range of Lipinski violations rule and were found to be inactive toward the central nervous system (CNS). While Cancer coli-2 (Caco-2) cells represent the gut–blood barrier and Madin-Darby Canine Kidney (MDCK) cells represent the blood–brain barrier, none of the compounds showed moderate or reasonable permeability to Caco-2 cells (36.67–209.88 nm/s) and MDCK cells (19.45–91.51 nm/s) (Table 4). The proposition that these compounds be used in food products and not as a drug for humans requires that they possess low permeability toward the Caco-2 and MDCK cells. Except for 11F and CE48, all other compounds do not cross the blood–brain barrier as well as have low HOA. All compounds have good aqueous solubility ranging from − 3.34 to − 6.5 mol dm^–3, except for SAM30 (Hosen et al. 2017).

Based on in silico toxicity prediction, none of the compounds was found to be possessing mutagenic and carcinogenic properties. Therefore, the overall results of ADME and toxicity studies revealed that the proposed derivatives of PPs are safe and non-toxic. These compounds possess good aqueous solubility, low human oral absorption, gut–blood barrier, blood–brain barrier penetration, and are non-mutagenic and non-carcinogenic.

Conclusion

To obtain innovative and safe tyrosinase inhibitors, phenylpropanoid esters (SE78, 11F, and CE48) and amides (SAM30 and FAM34) were synthesized using phenylpropanoid (C₆–C₃)—a natural scaffold, as a starting compound. Most of the synthesized compounds displayed better tyrosinase inhibition activities than kojic acid, except for SAM30 and SE78, and their order was as follows: FAM34 > CE48 > 11F > Kojic acid > SAM30 > SE78. Compound CE48 was found to be more efficient antioxidant than ascorbic acid. Although the antimicrobial activity was lower than that of the standard antibiotics, some of the compounds SAM30, SE78, 11F, and CE48 presented superior anti-B. subtilis, anti-S. aureus, and anti-A. niger activity, respectively, compared to their parents. Furthermore, computational tools were used to study the binding interactions of inhibitors with tyrosinase, and to analyze their pharmacokinetics and safety profile. Amongst all the newly synthesized compounds, CE48 was found to be an excellent anti-tyrosinase, antioxidant, and antifungal compound. These findings suggest that the multifunctional PPs hybrids may contribute in designing and development of novel tyrosinase inhibitors along with the potential to scavenge free radicals and inhibit microbial growth.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Neelam conducted the research, analyzed data, and prepared the manuscript. SA conducted antimicrobial study. AS performed tyrosinase assay and AL provided technical support in the compound synthesis. AK and KKS planned the experiment, edited the manuscript, supervised the research, and provided financial assistance. All authors declare that they have no conflict of interest.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest in the publication.