Ultrasonication-assisted lipase-catalyzed esterification of chlorogenic acid: A comparative study using fatty alcohol and acids in solvent and solvent-free conditions

Abstract

Chlorogenic acid, a well-known antioxidant, has potential applications in health care, food, and cosmetic sectors. However, its low solubility hinders its application at the industrial scale. The primary goal of the present study was to increase the lipophilic property of chlorogenic acid through esterification using an ultrasonication approach and Novozym® 435 as the catalyst. The esterification was executed in two ways. In the first method, chlorogenic acid was converted to chlorogenic acid ester using octanol in a solvent-free reaction. Catalytic factors such as reaction time (12 h ∼ 36 h), enzyme dosage (10 ∼ 50 mg), and ultrasonication power (90 ∼ 150 W) were optimized using Box-Behnken design (BBD) while temperature (60 ℃) and molar ration (chlorogenic acid/octanol, 1:500) were kept constant. A maximum conversion rate of 95.3 % was achieved when the esterification was performed for 12 h at 120 W ultrasonication power and 50 mg enzyme dosage. Contrary to the first method, when esterification was done using caprylic acid in the presence of 2-methyl-2-butanol as a solvent, the conversion rate was relatively low. Despite optimization of factors including molar ratio, enzyme dosage, and reaction time, the highest conversion rate achieved was of only 36.8 %. Moreover, molecular docking results revealed that the lowest binding energy was between lipase and octanol. The finding of the study clearly stated that the esterification of chlorogenic acid was more effective in a solvent-free condition as compared to in the presence of solvent.

1. Introduction

The growing awareness of health and healthy lifestyle in society has increased the importance of antioxidants, especially in pharmaceuticals, cosmetics, and food. In 2020 the worldwide market for antioxidants exceeded 3 billion dollars [1]. In recent years, natural antioxidants such as polyphenols have garnered immense interest compared to their synthetic counterparts owing to the harmful effects of the latter, for example, carcinogenicity. One of the naturally occurring representatives of polyphenols with known antioxidant properties is chlorogenic acid, which is mostly found in coffee beans, peaches, apples, sweet potatoes, grapes, and tea [2]. It is formed by the condensation of caffeic acid with quinic acid and exhibits numerous biological activities including antioxidant, anti-inflammatory, anti-obesity, anti-diabetes, anti-bacterial, anti-hypertensive, hepatocellular protective, and myocardial cell protective [3], [4]. It is involved in synthesizing an essential structural polymer, lignin, and is produced during aerobic respiration in plants through the shikimic acid pathway, a key route in the biosynthesis of aromatic compounds. Additionally, chlorogenic acid plays a role in the plant defense system [5].

The antioxidant ability of chlorogenic acid is greatly influenced by its molecular structure, which consists of one carboxyl group and five active hydroxyl groups [6]. The phenolic hydroxyl group structure has a strong antioxidant impact as it swiftly reacts with free radicals to form hydrogen radicals, which have an antioxidant effect and suppress the activity of hydroxyl radicals and superoxide anions [7]. Additionally, chlorogenic acid forms protein–polyphenol complexes through hydrogen bonding, which thickens the interface layer, restricts free radical diffusion, and enhances the antioxidant capacity of the protein [6]. While chlorogenic acid has multiple bioactive properties, its use in human health and lipid-based pharmaceuticals, nutraceuticals, and cosmetics products is restricted by its highly hydrophilic structure. The presence of one carboxylic and five hydroxyl groups, besides contributing to its strong antioxidant properties, also increases its hydrophilicity, resulting in poor miscibility with lipid-based formulations and limited skin permeability [1]. Its susceptibility to oxidation also lessens its ability to shield oil from oxidation and hence loses its nutritious qualities.

The lipophilicity of chlorogenic acid can be improved by esterification reactions which aid in modifying the solubility of chlorogenic acid in oil-based formulations. The enzymatic esterification of polyphenols offers advantages over its chemical counterparts, including less unwanted reactions and byproducts, high selectivity, no use of toxic solvents, mild conditions, and greater environmental friendliness [8]. Despite being non-hazardous to the environment and working in pleasant conditions, the application of enzymes in any process is constrained owing to their high cost, loss of activity, and instability [2]. This shortcoming can be overcome by using immobilized enzymes for biocatalytic reactions. Immobilization allows the application of enzymes at a wider scale and promotes enzyme recovery and reusability [9]. Mostly for esterification reactions, lipases are utilized and commercially available immobilized lipase Novozym® 435 has been successfully reported for the production of different esters of formic acid, esculin, naringin, phloridzin, anthocyanins [10], [11], [12].

The esterification of chlorogenic acid can be carried out mainly by two methods i.e. first, the reaction of acid moieties with short or long alkyl chain alcohols for example octanol, and second, the reaction of hydroxyl functions with short or long alkyl chain acids such as palmitic acid and oleic acid [1]. Octanol, a medium-chain alcohol with moderate polarity, serves as both an alcohol substrate for esterification and a solvent that dissolves chlorogenic acid while providing a hydrophobic medium to enhance lipase interfacial activity. Moreover, the reaction can be carried out in a solvent-free environment or with an organic solvent with lower toxicity concerning the environmental impact. Guyot et al. examined the differences in conversion yield when various alcohols reacted with chlorogenic acid. The experiment evaluated four alcohols (octanol, dodecanol, hexadecanol, and 9-octadecen-ol) under both solvent-free conditions and with 2-methyl-2-butanol (2M2B) as a solvent. Under solvent-free conditions, octanol achieved the highest conversion yield of approximately 60 %, while the other alcohols reached around 40 %. The use of 2M2B as a solvent in the case of octanol significantly enhanced the conversion yield to 75 %, while the conversion yields for the other alcohols ranged between 55 % and 75 %. However, the reaction required 30 days to reach completion [13].

The use of ultrasonication as a green process method has increased the effectiveness of lipase catalyzed reactions [14], [15], [16]. Ultrasonication induces an immense amount of heat and pressure and causes several cavities to collapse rigorously. This leads to phase disruptions between compounds, which facilitates mass transfer, reduces the activation energy, and ultimately quickens the rate of the reaction [14], [15]. Ultrasonication in any process has gained immense recognition for promoting heat and mass transfer and enhancing the efficiency of the process. It also increases enzyme activity and hence facilitates enzymatic catalysis. However, it should be kept in mind that excessive heat might inactivate enzymes. The optimal temperature for most lipase-catalyzed sonochemical synthesis processes typically ranges between 45–65 °C [17]. The ultrasound-assisted Novozym® 435 catalyzed synthesis of xylityl acyl esters (95 % yield) using fatty acids such as caprylic, capric, lauric, and myristic in a solvent-free condition has been successfully reported [18]. Ultrasonication is extensively used for the modifications of various value-added products that contribute toward Sustainable Development Goals (SDGs) (Table 1). Further, the efficacy of the enzymatic process can also be enhanced by process optimization using response surface methodology which helps in studying the impact of factors involved in the process such as reaction time, temperature, molar ratio, etc. on the product yield [15], [19].

Table 1.

Application of ultrasonication for the modification of various value-added products contributing towards SDGs.

The Focus of the Study	Key Points	SDGs and Targets	Countries of affiliation (authors)	References
Ultrasound-accelerated lipase-mediated acetylation	•Combined ultrasonic and lipase (Candida antarctica; Novozym® 435) catalyzed synthesis of 4′-acetoxyresveratrol from resveratrol and vinyl acetate •Enhanced yield and decreased acetylation time	SDG-3: Ensure healthy lives and promote well-being for all at all ages. Target 3.4	Taiwan	[16]
Synthesis of xylitol fatty acid esters using ultrasound-assisted enzymatic approach	•Higher yield (more than 96 %) of two main products (xylityl monoacyl ester) and (xylitol diacyl ester)	SDG-3: Ensure healthy lives and promote well-being for all at all ages. Target 3.3 and 3.4	Spain	[18]
Enzymatic transesterification of ethyl phenates (EPs) with glycerol by ultrasonic assistance for the synthesis of phenolic acid glycerols	•Ultrasound assistance increased catalytic efficiency as compared to stirring methods •Exceptional antimicrobial activity against E. coli	SDG-2: End hunger, achieve food security and improved nutrition and promote sustainable agriculture. Target 2.1 SDG-3: Ensure healthy lives and promote well-being for all at all ages. Target 3.3 and 3.4	China	[20]
Ultrasound assisted Novozym® 435 catalyzed transesterification of cinnamyl alcohol and vinyl acetate for synthesis of cinnamyl acetate	•Higher yield of 99.99 % •Reduced time for maximum conversion •Reusability of enzyme up to 7 times without loss in activity	SDG-3: Ensure healthy lives and promote well-being for all at all ages. Target 3.9 SDG-12: Ensure sustainable consumption and production patterns. Target 12.4	India	[21]
Synthesis of D-isoascorbyl palmitate from D-isoascorbic acid using immobilized lipase Novozym 435 under ultrasonic conditions	•Decreased in the reaction time by 50 % due to ultrasonication •Enhanced D-isoascorbyl palmitate yield and productivity •Reusability of Novozym 435 up to 7 cycles.	SDG-3: Ensure healthy lives and promote well-being for all at all ages. Target 3.4	China	[22]
Ultrasound-accelerated Novozym® 435 mediated synthesis of caffeic acid phenethyl ester from caffeic acid and phenethyl alcohol	•Shorter reaction time •Maximum molar conversion (96.03 ± 5.18 %)	SDG-3: Ensure healthy lives and promote well-being for all at all ages. Target 3.4	Taiwan	[23]
Ultrasound assisted enzymatic esterification of rutin and naringin	•Increased conversion of flavonoid esters in a short reaction time	SDG-3: Ensure healthy lives and promote well-being for all at all ages. Target 3.3 and 3.4	China	[24]
Ultrasound-assisted enzymatic synthesis of caffeic acid phenethyl ester (CAPE) from caffeic acid and phenyl ethanol	•Enhanced CAPE molar conversion	SDG-3: Ensure healthy lives and promote well-being for all at all ages. Target 3.4	Taiwan	[25]

In this study, the esterification of chlorogenic acid was attempted using a green ultrasonication technique with commercially available lipase Novozym® 435 as the catalyst. Two methods of esterification were employed, one solvent-free, and the other utilizing 2M2B solvent. In the solvent-free method, octanol was used as substrate and solvent, while in the solvent-based method, esterification was carried out with different fatty acids including palmitic acid, lauric acid, and caprylic acid in the presence of 2M2B. This study focuses on the comparative evaluation of the chlorogenic acid esterification at the carboxylic or the phenolic OH groups of the quinic moiety. Two possible pathways to increase lipophilicity were studied, involving different reaction media. The catalytic factors, such as ultrasonication power, enzyme dosage, reaction time, and molar ratio were optimized for the enhanced conversion using the Box–Behnken design of response surface methodology. In order to explore the difference in the conversion yield of lipase-catalyzed esterification of chlorogenic acid, the binding mechanism between lipase Novozym® 435 and octanol as well as caprylic acid was investigated using the molecular docking technique.

2. Materials and methods

2.1. Chemicals

Chlorogenic acid (99 %) and enzyme Novozym® 435 (10000 PLU g⁻¹; propyl laurate units) were purchased from Sigma Aldrich, St. Louis, MO, USA. Chemicals such as Octanol (1-octanol 99 %), Methanol (99.9 %) and 2-Methyl-2-butanol (98 %, 2M2B) were obtained from UNI-ONWARD Corp., Taipei, Taiwan. Palmitic acid, lauric acid and caprylic acid were bought from Echo Chemical Co., Ltd., Miaoli County, Taiwan. All the chemicals used in the study were of analytical grade.

2.2. Synthesis of chlorogenic acid esters

The synthesis of chlorogenic acid esters was carried out using two different methods. In the first method, chlorogenic acid and octanol were used as substrates and mixed in a ratio of 1:500 followed by the addition of Novozym® 435 and ultrasonication (Delta DC150H, New Taipei, Taiwan) for the synthesis of chlorogenic acid esters (Fig. 1a). In the second method, chlorogenic acid was separately reacted with either palmitic acid, lauric acid or caprylic acid (Fig. 1b) to catalyze the synthesis of chlorogenic acid ester using the immobilized enzyme Novozym® 435 and ultrasonication.

Fig. 1.

Ultrasound assisted esterification of chlorogenic acid with octanol (a) and caprylic acid (b).

2.3. Ultrasound-Assisted synthesis of chlorogenic acid esters

Before synthesis, all reagents were dehydrated for 24 h using a 4 Å molecular sieve. For the first method, chlorogenic acid and octanol (1:500) were added to 2 mL Eppendorf tubes, and in it lipolytic enzyme Novozym® 435 was added (10–50 mg; 1.0 – 5.4 %, w/w, calculated from the total weight of both substrates). After mixing thoroughly, the reaction was executed in an ultrasonic system. The effect of reaction time (12–72 h) was also studied using one factor at a time approach in which one factor was varied. The reaction temperature, enzyme dosage, and ultrasonication power were kept constant at 60 ℃, 25 mg, and 150 W, respectively. For the second method, chlorogenic acid was mixed with either palmitic acid, lauric acid or caprylic acid at different molar ratios (1:70–1:400) for different time intervals (1–5 days) using different enzyme dosages (20–140 mg) in solvent 2-methyl-2-butanol. The reaction temperature and ultrasonication power were kept constant at 60 ℃ and 150 W, respectively. Upon completion of the reaction, the mixture was centrifuged at 13,000 rpm for 10 min. An appropriate amount of supernatant was extracted and stored in a refrigerator at 4 °C. For analysis, 10 µl of the reaction solution was extracted, diluted 100 times, and analyzed using high-performance liquid chromatography (HPLC). All reactions were performed in triplicate and results were expressed as the mean ± standard deviation.

2.4. Statistical optimization for chlorogenic acid ester production

Statistical optimization was executed using Box-Behnken design (BBD) of response surface methodology (RSM) to increase chlorogenic acid ester production. For the first method, three factors including reaction time (12–36 h), enzyme dosage (10–50 mg), and power (90–150 W) were optimized. The molar ratio (1:100–1:300), enzyme dosage (20–140 mg), and reaction time (1–3 days) were the three factors chosen for the optimization in the second method of chlorogenic acid ester synthesis. The program used was Design-Expert® (Version 8.0.6.1, Stat-Ease Inc., Minneapolis, MN, USA), which employs multiple regression techniques to fit a second-order model to the data. This model describes the changes in the response surface, allowing the determination of the optimal reaction values through the analysis of the second-order response surface. The corresponding quadratic polynomial equation generated is as follows:

$$\mathrm{Y}=b\text{0}+\sum _{i=1}^{3}b\text{i}x\text{i}+\sum _{i=1}^{3}b\text{ii}x\text{i}\text{2}+\sum _{i=1}^2\sum _{j=i+1}^{3}b\text{ij}x\text{i}x\text{j}$$ (1)

Where Y is response (yield %); Xi and Xj are the unstandardized independent variables; b0 is the constant, bi is linear constant, bii is the secondary effect coefficient and bij is the variable interaction coefficient. The appropriateness of this mathematical model in describing experimental data is verified by statistical testing. The main effects of each factor on the response and the interaction effects between factors were analyzed using this mathematical model.

In total 15 experiments were performed in random order to avoid bias for each method. The determination coefficient (R2) was used to determine the adequacy of the model. It is defined as the ratio of the predicted model to the actual experimental data. Therefore, the closer the value is to 1, the better the accuracy of the experimental design. The F value was used to determine whether the influencing variable has a significant impact on the experimental target serving as a basis for determining the effect of a factor on the esterification conversion rate. The p value was used to determine the significance of the model as well as factors. The smaller the p value, the more obvious the effect of the variable on the target value. Conversely, when the p value is greater than 0.05, the significance of this variable is lower than the 95 % confidence level.

2.5. Analysis

The product was analyzed by high-performance liquid chromatography (HPLC, Hitachi L-7400, Tokyo, Japan). The column used was Thermo C18 (250 mm × 4.6 mm, Agilent, Waltham, MA, USA). The separation was conducted via gradient elution using 0.1 % acetic acid in water (A) and 100 % methanol (B). The gradient elution was as follows: 50 % A and 50 % B at 0 min, 100 % B from 0 to 5.5 min and 100 % B from 5.5 to 10 min with a flow rate of 1.0 min/mL. The total analysis time was 10 min, and the absorbance was detected at UV 325 nm using UV detector (detector L-7100).

2.6. Calculation of chlorogenic acid ester yield

The chlorogenic acid ester yield was calculated using the peak area of ​​the substrate and product in HPLC after the reaction. The HPLC profile of chlorogenic acid and chlorogenic acid ester is shown in Fig. S1. The peak area of ​​chlorogenic acid and its derivative, chlorogenic acid ester, were used for the estimation, and the calculation formula is as follows:

$$\mathrm{Y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d}(\%)=\frac{B}{(\mathrm{A}+B)}\times 100\%$$ (2)

Where A: Peak area of ​​chlorogenic acid.

B: Peak area of ​​chlorogenic acid esters.

2.7. Molecular docking

The interactions between Novozym® 435 and octanol, caprylic acid, and chlorogenic acid were studied using a commercial molecular docking tool, AutoDock Vina 1.1.2. Lipase from Candida antarctica (lipase B) was used as a receptor and octanol and caprylic acid were used as ligands. The structure of Candida antarctica lipase B was obtained from Protein Data Bank (PDB: 1TCB). The three-dimensional structures of octanol, caprylic acid, and chlorogenic acid were obtained from the PubChem database, NCBI. Initially, the three-dimensional structure of lipase B was prepared by removing water molecules and natural ligands bound in the crystallographic structure. The PDB file of lipase B, octanol, caprylic acid, and chlorogenic acid were converted to PDBQT format using AutoDock Tool 1.5.7. The amino acid Ser105 representing the catalytic site of the enzyme, was located in the center docking box with the coordinates 15.284375, 9.988542, and 6.470042. Autodock Vina was used to determine the free binding energy (Kcal/mol), and the radius and exhaustiveness were adjusted to 12 and 8, respectively. The Discovery Studio Visualizer (21.1, BIOVIA Dassault Systemes, Vélizy-Villacoublay, Paris, France) was used to visualize the molecular interactions.

3. Results and Discussion

3.1. Preliminary study of effect of reaction time on chlorogenic acid esterification with octanol

The optimal reaction time of any process is very crucial for the effective synthesis of the product. Therefore, the optimal reaction time for the esterification of chlorogenic acid was evaluated by carrying out the reaction for various time intervals at a fixed temperature of 60 ℃, chlorogenic acid and octanol molar ratio of 1:500, 25 mg enzyme dosage and ultrasonication power of 150 W. During the enzyme-catalyzed reaction, as the reaction time was increased, the conversion rate of chlorogenic acid to chlorogenic acid ester also increased. The reaction equilibrium was obtained after 48 h of reaction with a conversion rate of 88.49 % (Fig. 2). No increase in the conversion rate was observed after 48 h. Esterification benefits from a high likelihood of intermolecular collision and adequate substrate availability in the early stages of the reaction, which increases the conversion rate of chlorogenic acid. However, as the reaction duration increases, the substrates are consumed, and the reaction reaches equilibrium, indicating no further increase in the conversion rate [2]. The reaction time to reach equilibrium may vary depending on the type of the reaction for example, the green synthesis of 2-ethylhexyl salicylate was achieved using Novozym® 435 in which the molar conversion of 80 % was achieved after 12.5 h of reaction, and equilibrium was achieved at 16 h [26].

Fig. 2.

Effect of reaction time on the conversion rate of chlorogenic acid with octanol. Different lowercase letters indicate a significant difference (p < 0.05) in conversion rate.

3.2. Response surface design for chlorogenic acid esterification with octanol

This study used the lipolytic enzyme Novozym® 435 to esterify chlorogenic acid in the presence of octanol to synthesize the target product-chlorogenic acid ester. The reaction parameters, i.e., reaction time, enzyme dosage, and ultrasonication power were optimized. A total of 15 experiment sets were conducted using the Box-Behnken design (BBD) design and the results are given in Table 2. The temperature (60 ℃) and molar ratio (chlorogenic acid and octanol, 1:500) were kept constant and the reactions listed in Table 2 were carried out in an ultrasonication bath (Delta DC150H, New Taipei, Taiwan). The chlorogenic acid ester yield obtained from the experiment was used to explore the important influencing variables and the interactions between the variables required in the reaction process to understand the impact on the formation of chlorogenic acid esters.

Table 2.

Box-Behnken design and responses of each run for the synthesis of chlorogenic acid esters using octanol.

Run		Factors		Conversion (%)
	Reaction time (h)	Enzyme dosage (mg)	Ultrasonication power (W)
1	24 (0)	10 (−1)	90 (−1)	28.2 ± 1.72
2	24 (0)	50 (1)	150 (1)	93.5 ± 1.27
3	12 (−1)	50 (1)	120 (0)	95.3 ± 3.88
4	24 (0)	30 (0)	120 (0)	88.1 ± 2.60
5	36 (1)	50 (1)	120 (0)	93.7 ± 2.63
6	36 (1)	30 (0)	90 (−1)	78.1 ± 2.37
7	24 (0)	50 (1)	90 (−1)	91.9 ± 1.36
8	36 (1)	10 (−1)	120 (0)	34.8 ± 0.14
9	24 (0)	30 (0)	120 (0)	88.5 ± 2.37
10	24 (0)	30 (0)	120 (0)	88.4 ± 1.04
11	36 (1)	30 (0)	150 (1)	82.5 ± 0.73
12	12 (−1)	30 (0)	150 (1)	70.2 ± 3.95
13	12 (−1)	10 (−1)	120 (0)	16.8 ± 3.29
14	24 (0)	10 (−1)	150 (1)	30.9 ± 1.56
15	12 (−1)	30 (0)	90 (−1)	52.3 ± 5.35

The polynomial regression equation for the quadratic model obtained from the design software is given below as Eq. (3)

$$\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}\left(\%\right)=-243.68125+5.39410\times X_1+5.03603X_2+2.56069X_3-0.020417X_1{X}_2-0.009375X_1{X}_3-0.000458X_2{X}_3-0.064352{X_1}^2-0.047292{X_2}^2-0.009213{X_3}^2$$ (3)

Where, X₁ represents reaction time, X₂ represents enzyme dosage, and X₃ represents ultrasonication power.

For any model to correctly anticipate or yield positive outcomes, data analysis needs to be validated by an adequacy check [27]. Fig. S2 displays a plot of the predicted values produced by the software against the actual values that were obtained through experimentation. It was evident from the generated data that the model was appropriate for the experimental data and could be used to predict the rate of conversion of chlorogenic acid as the data were normally distributed and generally fell on a straight line.

The results of the analysis of variance (ANOVA) indicated the fitness of the model, and the significance of the model was specified by the model F-value of 53.16 and p value 0.0002. The model terms X₁, X₂, X₁ ², X₂ ², and X₃² were significant as their p-values were less than 0.05 whereas factors with p-values more than 0.05 were regarded as insignificant. This indicated that reaction time (X₁) and enzyme dosage (X₂) were significantly important factors for the synthesis of chlorogenic acid esters. The strength of the model was indicated by the R² of 0.9897 which was ≥ 0.8 and a difference of less than 0.2 between the predicted R² (0.8346) and the adjusted R² (0.9710). Additionally, the adequate precision of 20.868, which measures the signal-to-noise ratio indicated an adequate signal.

The three-dimensional response surface graphs were used to show the impact of two independent variables on the conversion rate. In these graphs, the remaining two variables (remnants) are maintained constant at the center points. The effect of reaction time and enzyme dosage on the conversion rate of chlorogenic acid ester is depicted in Fig. 3a. Both reaction time and enzyme dosage had a significant effect on the conversion rate. It was seen that increasing the reaction time and enzyme dosage significantly increased the conversion rate, and the two variables were positively correlated with the conversion rate. It has been reported that the enzyme dosage is the most crucial factor in the Novozym® 435-catalyzed process [28], [29]. Both reaction time and enzyme dosage have been reported to increase the esterification process catalyzed by Novozyme 435 [30]. Fig. 3b shows the effects of enzyme dosage and ultrasonication power on the conversion rate of chlorogenic acid esters. The amount of enzyme significantly increased the conversion rate, however in case of ultrasonication power the conversion rate was highest at about 120 W, but the difference was not substantial after which it decreased. Therefore, it was concluded that the best conversion rate can be obtained by increasing the enzyme dosage and maintaining the power at about 120 W. A significant positive impact on the conversion rate of chlorogenic acid ester was observed as ultrasonication power and reaction time were increased (Fig. 3c).

Fig. 3.

Three-dimensional response surface plots showing the interaction between (a) reaction time and enzyme dosage, (b) enzyme dosage and ultrasonication power and (c), reaction time and ultrasonication power for the esterification of chlorogenic acid with octanol.

3.2.1. Validation of the model

For the validation of the model, three new experimental runs were generated by the software, and the experiments were performed under those conditions (Table 3). The experimental conversion rate was in close proximity to the theoretical value predicted by the software for all three sets, which validated the design model and indicated its appropriateness. However, the maximum conversion rate of 95.3 % was achieved in set 2 when the reaction was carried out for 12 h, with 50 mg enzyme dosage (5.4 %, w/w, calculated from the total weight of both substrates) and 120 W ultrasonication power. The reaction time was reduced from 48 h to 12 h after BBD optimization, which could be due to the effect of ultrasonication [15]. The efficiency of the lipase-catalyzed reaction has been reported to be considerably increased with the help of ultrasonication [31]. The results were consistent with the reported study where in the central composite design-based optimization ultrasound-assisted synthesis of D-isoascorbyl palmitate using Novozym 435 resulted in a conversion rate of 94.32 % which is almost near to the conversion rate obtained in this study [22].

Table 3.

RSM model validation for the yield of chlorogenic acid ester using octanol.

Experiments*	Theoretical value (%)	Experimental value (%)
0/0/-1	77.9	76.7 ± 3.09
−1/+1/0	95.7	95.3 ± 1.62
+1/-1/+1	31.5	30.3 ± 3.39

*-1-low value, 0- mid value and + 1- higher value of the independent factors.

3.3. Preliminary study of the synthesis of chlorogenic acid esters with fatty acids

3.3.1. Effect of molar ratio

The esterification of chlorogenic acid can also occur using saturated/unsaturated fatty acid in the presence of an alcohol solvent such as 2-methyl-2-butanol (2M2B). 2M2B is an ideal solvent for lipase-catalyzed esterification as lipases do not recognize it as substrate [32]. Moreover, chlorogenic acid can be dissolved in 2M2B to provide a favorable reaction environment. Polyphenol compounds, as secondary alcohols with significant steric hindrance, require higher substrate molar ratios for effective esterification reactions. For example, in lipase-catalyzed transesterification of resveratrol with vinyl acetate or vinyl laurate, a molar ratio of 40–55 is typically employed [16], [33], [34]. However, the direct esterification of chlorogenic acid with carboxylic acids is more challenging as compared to transesterification. In the case of phenolic acids, Lorentz et al. reported that conversion was influenced by the molar ratio of palmitic acid/chlorogenic acid, which ranged from 10 to 80 [35]. To enhance the conversion, the molar ratio was increased to 400. However, due to solubility limitations, only caprylic acid can achieve such high molar ratios. In this study, palmitic acid (C16), lauric acid (C12), caprylic acid (C8, and octanoic acid, OA) were used for esterification reaction in the solvent 2M2B at reaction temperature 60 ℃, reaction period 3 days, and ultrasonication power 150 W. The molar ratio (chlorogenic acid: fatty acid) of all fatty acids was varied (1:70–1:400) and the immobilized enzyme Novozym® 435 was used to catalyze the synthesis of chlorogenic acid ester. The results indicated that the conversion rate increased as the ratio increased and the highest conversion rate of about 30 % was obtained when the ratio of chlorogenic acid to caprylic acid was 1:400 (Fig. 4). The minimum conversion rate of approximately 15 % was achieved for palmitic acid at the molar ratio 1:150. The increase in viscosity and higher melting points of lauric acid and palmitic acid, due to their longer alkyl chain lengths compared to caprylic acid, may have restricted mass transfer [18]. For a reaction to be 100 % converted, a molar ratio of 1:1 should ideally be adequate. However, selection of optimal molar ratio is very crucial for obtaining a high yield. In a lipase-catalyzed process, increasing the molar ratio of reactants has the ability to shift the reaction equilibrium in favor of esterification; however, an excessive increase in the molar ratio may have a negative effect on the conversion rate by increasing solvent viscosity, impeding mass transfer, or decreasing enzyme activity [36], [37]. The results were aligned with the study reported by Shin et al. [38] for the esterification of phenethyl alcohol using formic acid catalyzed by Novozym 435. At 1:5 formic acid:phenethyl alcohol molar ratio, a maximum (71.40 %) conversion to phenethyl formate was achieved.

Fig. 4.

Effect of varying molar ratios on the synthesis of chlorogenic acid ester with palmitic acid, lauric acid and caprylic acid. Different lowercase letters with same color indicate a significant difference (p < 0.05) in conversion rate.

3.3.2. Effect of enzyme dosage and reaction time

The impact of other two important factors, i.e., enzyme dosage and reaction time on the esterification of chlorogenic acid using caprylic acid in the presence of 2M2B was also studied. The reaction was carried out at 60 ℃, 150 W ultrasonication power, and molar ratio (1:400). Either enzyme dosage (20–140 mg; 2.4 – 17.1 %, w/w, calculated from the total weight of both substrates) or reaction time (1–5 days) were varied to study their impact. It was revealed that after 3 days of reaction, the conversion rate drastically increased when the enzyme dosage increased from 50 mg (24.10 %) to 80 mg (31.60 %), and a maximum of 35.84 % was achieved at the enzyme dosage of 140 mg (Fig. 5a). Increasing the enzyme dosage may enhance the availability of active sites for substrate interaction, thereby increasing the probability of reactions and ultimately boosting the conversion rate [38]. Lorentz et al. synthesized chlorogenic acid ester using palmitic acid at an enzyme dosage of 40 mg (12.13 %, w/w), the conversion curve leveled off within 20 days, with a conversion yield of 35 % [35]. In comparison, the esterification of chlorogenic acid with octanol required relatively less enzyme dosage (1.0 – 5.4 %, w/w; this study) and achieved high conversion, indicating that esterifying the carboxyl group of chlorogenic acid is much easier than its hydroxy group. In case of varying reaction time, when 140 mg of enzyme dosage was used, the best conversion rate (38.65 %) was attained after 2 days of reaction after which it started to decrease and the minimum conversion rate of 28.84 % was achieved on day 5th (Fig. 5b). This might be the result of the enzyme inactivation or that the rate of the reverse reaction increased as the number of days increased [39].

Fig. 5.

Effect of varying enzyme dosage (a) and reaction time (b) on the synthesis of chlorogenic acid ester with caprylic acid. Different lowercase letters indicate a significant difference (p < 0.05) in conversion rate.

3.4. Response surface design for chlorogenic acid esterification using caprylic acid

After studying the effect of one factor at a time on the conversion rate of chlorogenic acid, the effect of all three factors i.e., molar ratio of chlorogenic acid to caprylic acid, enzyme dosage, and reaction time were studied together using BBD to understand the impact of the interactions between these factors on conversion rate. Based on these values, a total of 15 experimental runs were generated by the software. The conversion rate of chlorogenic acid obtained in 15 runs is mentioned in Table 4. Each reaction was executed at a constant temperature of 60 ℃ and ultrasonication power of 150 W.

Table 4.

Box-Behnken design and responses of each run for the synthesis of chlorogenic acid esters using caprylic acid.

Run		Factors		Conversion (%)
	Molar ratio	Enzyme dosage (mg)	Reaction time (days)
1	1:200 (0)	80 (0)	2 (0)	20.1 ± 0.91
2	1:200 (0)	140 (1)	3 (1)	24.7 ± 2.46
3	1:200 (0)	20 (−1)	1 (−1)	3.18 ± 0.67
4	1:200 (0)	140 (1)	1 (−1)	26.7 ± 1.20
5	1:300 (1)	80 (0)	1 (−1)	21.5 ± 0.75
6	1:200 (0)	20 (−1)	3 (1)	6.5 ± 0.48
7	1:200 (0)	80 (0)	2 (0)	20.5 ± 2.46
8	1:100 (−1)	80 (0)	1 (−1)	10.3 ± 1.30
9	1:300 (1)	80 (0)	3 (1)	27.4 ± 1.03
10	1:300 (1)	140 (1)	2 (0)	36.5 ± 0.83
11	1:100 (−1)	20 (−1)	2 (0)	2.7 ± 0.22
12	1:100 (−1)	140 (1)	2 (0)	13.5 ± 0.62
13	1:200 (0)	80 (0)	2 (0)	20.6 ± 1.42
14	1:300 (1)	20 (−1)	2 (0)	6.49 ± 1.05
15	1:100 (−1)	80 (0)	3 (1)	9.4 ± 0.79

The polynomial equation obtained from the design is given below as Eq. (4)

$$\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}\left(\%\right)=-12.56361+0.046404X_1+0.22247X_2+4.71833X_3+0.0008X_1{X}_2+0.017X_1{X}_3-0.022167X_2{X}_3-0.000186{X_1}^2-0.001039{X_2}^2-1.38875{X_3}^2$$ (4)

Where, X₁ represents molar ratio, X₂ represents enzyme dosage, and X₃ represents reaction time.

The plot of the predicted values produced against the actual values obtained after experimentation was a straight line indicating the appropriateness of the model (Fig. S3).

According to the ANOVA results, the model was significant with model F-value of 297.08 and p-value < 0.0001. In this case X₁, X₂, X₃, X₁ X₂, X₁ X₃, X₂ X₃, X₁², X₂², X₃² were significant model terms with p-values less than 0.0500. It was evident from the results that all three factors i.e. molar ratio, enzyme dosage and reaction time were very crucial and had a positive impact on the conversion rate of chlorogenic acid as confirmed by polynomial equation (2). The strength and fitness of the model was also indicated by R² value of 0.9981, reasonable agreement between predicted R² of 0.9715 and adjusted R² of 0.9948 and the signal to noise ratio of 58.124.

The effect of molar ratio and enzyme dosage on the conversion rate of chlorogenic acid ester is shown in Fig. 6a. It was found that the two variables were positively correlated with the conversion rate as increasing the molar ratio and enzyme dosage significantly increased the conversion rate. The interaction between molar ratio and reaction time also exhibited a positive impact on the conversion rate (Fig. 6b). A significant increase in conversion was observed when the molar ratio increased, and the conversion rate was the highest when the reaction time was about 2 days. However, as the number of days increased, the conversion rate decreased, therefore, the maximum reaction time was 2 days to obtain the best conversion rate. Similar results were obtained for the interaction between enzyme dosage and reaction time on the conversion rate of chlorogenic acid esters (Fig. 6c). The conversion rate was enhanced as the enzyme dosage increased and in case of reaction time the highest conversion rate was achieved after 2 days beyond which it decreased. For any esterification reaction, molar ratio, enzyme dosage and reaction time are important factors. The esterification of dihydrocaffeic acid with hexanol in ionic liquid was optimized using response surface methodology. A conversion rate of 84.4 % was predicted at 39.4 °C, temperature, 77.5  h reaction time, 41.6 % enzyme dosage, and mole ratio of 2.1 [30].

Fig. 6.

Three-dimensional response surface plots showing the interaction between (a) molar ratio and enzyme dosage, (b) molar ratio and reaction time, and (c), reaction time and enzyme dosage for the esterification of chlorogenic acid with caprylic acid.

Ridge analysis was performed to obtain the optimal conditions for the synthesis of chlorogenic acid ester. The process of ridge analysis makes it possible to determine the optimal values and maximum response [40]. According to the results of ridge analysis, the best conversion rate was obtained by conducting experiments under the analysis condition when the coding radius increased to 2.0 and 2.5. The optimal conditions for the reaction were: molar ratio 1:348.5, enzyme dosage 159.6 mg, and reaction time 2.19 days. Based on this ridge analysis, the conversion rate obtained in the experiment was 36.8 %. Ridge analysis has been applied to various studies to determine the optimal solution within the experimental design space and to evaluate the sensitivity of responses [41], [42], [43]. It has been employed to identify optimal conditions for factors including temperature, ethanol concentration, liquid-to-solid ratio, and ultrasonic power for ultrasound-assisted extraction for chlorogenic acid from Lonicera japonica. The actual experimental yield obtained was 43.13 mg/g [44]. Lipase-catalyzed esterification of chlorogenic acid with palmitic acid has reported that the acylated positions were major at the C-4 of the quinic moiety and minor at the C-3, but the conversions obtained in 7 days ranged from 14 to 60 % [35]. Upon comparing the results of the conversion of chlorogenic acid to its respective ester, it was evident that in the first case where esterification was done in the presence of octanol, the conversion rate was highest compared to the second method when chlorogenic acid and caprylic acid were used. This is because the OH groups of the caffeic acid moiety are secondary alcohols, which are less reactive than primary alcohols (octanol). As the size of the carbon group attached to secondary alcohol increases, lipase B's relative activity can decrease by over 1000-fold [45]. Therefore, the method of esterifying chlorogenic acid in the presence of octanol proved to be more efficient as it included less time, and enzyme dosage and was solvent-free.

3.5. Molecular docking analysis

As discussed above the conversion rate of chlorogenic acid esterification with octanol was higher than the conversion rate obtained with caprylic acid. Molecular docking analysis was performed to investigate the role of interactions between the enzyme and either caprylic acid or octanol to gain further insights into these results. Ser105 is one of the amino acids that form the catalytic triad of lipase B and therefore, was used to dock octanol or caprylic acid with lipase B [46]. The results of molecular docking are shown in Fig. 7. The main interaction force between lipase B and octanol, was hydrogen bonding between the Ser105 of lipase and OH group of octanol (Fig. 7a). In addition, alkyl interaction between His224 which is also a part of the catalytic triad of lipase B and octanol was also observed. Similarly, in case of lipase B and caprylic acid, the main hydrogen bonding was also visible between caprylic acid and Ser105 (Fig. 7b). However, the binding energy of lipase with octanol was −4.2 kcal/mol and with caprylic acid was −4.0 kcal/mol indicating that the affinity of octanol to lipase was higher than that of caprylic acid. Moreover, the hydrogen bonding between octanol and Ser105 was stronger as the distance was shorter (2.55 Å) as compared to between caprylic acid and Ser105 (3.16 Å), which may also aid in stabilizing the secondary structure of the enzyme [47]. Additionally, the interactions between lipase B and chlorogenic acid also indicated hydrogen bonding between the carboxyl group of chlorogenic acid with Ser105 (Fig. 7c), but no interactions between Ser105 and 3-OH or 4-OH groups of chlorogenic acid. In an enzyme reaction, the substrate must enter the enzyme's active site. This site contains specific amino acid residues that provide a unique environment for substrate binding and facilitate enzymatic esterification. The results of molecular docking analysis explained that the esterification of chlorogenic acid with octanol is easier than that of caprylic acid. Similar results were also found in lipase-catalyzed esterification of chlorogenic acid with alcohol [48] and fatty acids [35].

Fig. 7.

The binding process of (a) octanol, (b) caprylic acid and (c) chlorogenic acid to lipase B. Ⅰ represents three-dimensional docking results and Ⅱ represents two-dimensional interactions.

4. Conclusions

In the present study, the immobilized enzyme Novozym® 435 was used to catalyze the esterification of chlorogenic acid, resulting in the synthesis of its derivative chlorogenic acid ester. The esterification was done in two ways and the conditions for the esterification of chlorogenic acid were optimized using Box–Behnken design of response surface methodology. In the first case when no solvent was used and esterification was done in the presence of octanol, the maximum conversion rate of 95.3 % was achieved and the model was significant. However, in the second method when caprylic acid was used, the highest conversion rate of 36.8 % was achieved at a molar ratio of 1:348.5, enzyme dosage of 159.6 mg, and reaction time of 2.19 days. The conversion rate in the second case was quite low and the process was lengthy, requiring approximately 2 days. Based on the results and molecular docking analysis, it was evident that esterifying chlorogenic acid in the presence of octanol and in a solvent-free condition was a more effective method that required less time, less enzyme dosage, and no addition of stronger organic solvent. The first process of chlorogenic acid esterification paved the way for the cost-effective and economically viable synthesis of chlorogenic acid ester. The synthesized chlorogenic acid ester can further be purified, characterized, and assessed for its potential applications in food and pharmaceutical industries. The second process of chlorogenic acid esterification with caprylic acid employed optimized reaction parameters to enhance conversion; however, the results remained suboptimal. Investigating alternative co-solvents could be a promising avenue for future research to improve conversion efficiency.

CRediT authorship contribution statement

Chia-Hung Kuo: Writing – review & editing, Validation. Parushi Nargotra: Writing – original draft, Formal analysis. Tsung-Han Lin: Methodology, Investigation, Formal analysis. Chwen-Jen Shieh: Supervision, Resources, Project administration, Data curation. Yung-Chuan Liu: Writing – review & editing, Visualization, Supervision, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

Data will be made available on request.