Bioresource Technology 158 (2014) 39–47

Contents lists available at ScienceDirect

Bioresource Technology journal homepage: www.elsevier.com/locate/biortech

A novel continuous flow biosynthesis of caffeic acid phenethyl ester from alkyl caffeate and phenethanol in a packed bed microreactor Jun Wang a,b,⇑, Shuang-Shuang Gu a, Hong-Sheng Cui a, Xiang-Yang Wu b,⇑, Fu-An Wu a,c a

School of Biology and Chemical Engineering, Jiangsu University of Science and Technology, Zhenjiang 212018, PR China School of the Environment, Jiangsu University, Zhenjiang 212013, PR China c Sericultural Research Institute, Chinese Academy of Agricultural Sciences, Zhenjiang 212018, PR China b

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

 Ten alkyl caffeates were screened as

suitable substrates via transesterification.  MC was the best substrate for enzymatic synthesis of CAPE at a lower temperature.  A packed bed microreactor was firstly used for the continuous synthesis of CAPE.  The highest CAPE yield of 93.21% was obtained in 2.5 h using RSM optimization.  Novozym 435 was reused without a loss of activity for 20 cycles or 9 days.

a r t i c l e

i n f o

Article history: Received 21 December 2013 Received in revised form 25 January 2014 Accepted 27 January 2014 Available online 10 February 2014 Keywords: Biocatalysis Continuous flow biosynthesis Packed bed microreactor Caffeic acid phenethyl ester Transesterification

a b s t r a c t Caffeic acid phenethyl ester (CAPE) is a rare natural ingredient with several biological activity, but the industrial production of CAPE using lipase-catalyzed esterification of caffeic acid (CA) and 2-phenylethanol (PE) in ionic liquids is hindered by low substrate concentrations and a long reaction time. To establish a high-efficiency bioprocess for obtaining CAPE, a novel continuous flow biosynthesis of CAPE from alkyl caffeate and PE in [Bmim][Tf2N] using a packed bed microreactor was successfully carried out. Among the tested alkyl caffeates and lipases, methyl caffeate and Novozym 435, respectively, were selected as the suitable substrate and biocatalyst. Under the optimum conditions selected using response surface methodology, a 93.21% CAPE yield was achieved in 2.5 h using a packed bed microreactor, compared to 24 h using a batch reactor. The reuse of Novozym 435 for 20 cycles and continuous reaction for 9 days did not result in any decrease in activity. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Caffeic acid phenethyl ester (CAPE), a natural flavonoid-like compound, is one of the main active components of propolis ⇑ Corresponding authors. Address: School of Biology and Chemical Engineering, Jiangsu University of Science and Technology, Zhenjiang 212018, PR China. Tel.: +86 511 85616571; fax: +86 511 85635850 (J. Wang). Tel.: +86 511 85038750; fax: +86 511 85038451 (X.-Y. Wu). E-mail addresses: [email protected], [email protected] (J. Wang), [email protected] (X.-Y. Wu). http://dx.doi.org/10.1016/j.biortech.2014.01.145 0960-8524/Ó 2014 Elsevier Ltd. All rights reserved.

(Suzuki et al., 2006). This compound has strong anticancer (Ozturk et al., 2012), antiviral (Noelker et al., 2005), antioxidant (Gocer and Gulcin, 2011), and immunomodulatory properties in a diverse set of systems (Lee et al., 2009). However, the isolation of this highly valuable CAPE from natural product extracts is inefficient, time-consuming and uneconomical. Therefore, to produce CAPE at a reasonable price, biosynthesis is becoming increasingly attractive due to its economic benefits when compared to extraction from natural sources and greater ecological acceptability compared to chemical synthesis (Kim et al., 2011).

40

J. Wang et al. / Bioresource Technology 158 (2014) 39–47

In recent years, the lipase-catalyzed esterification of CAPE from caffeic acid (CA) and 2-phenylethanol (PE) in organic solvents and ionic liquids (ILs) has been successively established. In most cases, CAPE could be obtained from CA and PE catalyzed by Novozym 435 in isooctane (Widjaja et al., 2008) or [Emim][Tf2N] (Wang et al., 2013a) and the conversion of CA was nearly 100% in 48 h. Although the reaction time was shortened to 9.6 h when ultrasound-accelerated enzymatic technology was used to synthesize CAPE (Chen et al., 2011a,b), this procedure is not suitable for industrial production due to the required special equipment and energy consumption. However, above all, the CAPE yield is still not high enough (only approximately 64.55%) (Wang et al., 2013a). A moderate yield in the esterification of CA with PE to afford CAPE is observed because CA possesses two hydroxyl groups on its aromatic ring and a double bond on the side chain that cause it to inhibit lipase (Tan and Shahidi, 2012). In addition, water is formed as one of the products of the esterification of CA and PE. The water concentration increases during the reaction and has a negative influence not only on substrate conversion but also on lipase activity. In our previous work, lipase-catalyzed esterification of propyl caffeate was performed using methyl caffeate (MC) and 1-propanol in [Bmim][CF3SO3] with a maximum yield of 98.5% in 24 h, which indicated that the transesterification process of alkyl caffeate and the corresponding alcohol provided an efficient pathway for the synthesis of some caffeate esters (Pang et al., 2013). [Bmim][Tf2N] was chosen as the suitable reaction medium to enzymatic transesterification synthesis of CAPE analogues (Kurata et al., 2010). It shows environmentally friendly and can enhance enzyme activity, selectivity, and stability. In addition, the synthesis of CAPE by the lipase-catalyzed transesterification of alkyl caffeate and the corresponding alcohol in [Bmim][Tf2N] has been proven a feasible procedure because alkyl caffeates are much more easily prepared and have cost less. However, there are still some drawbacks for industrial-scale production in a traditional batch reactor, such as long reaction times. Thus, there is a need to explore more efficient processes for the lipase-catalyzed synthesis of CAPE from alkyl caffeate and PE via transesterification. Nowadays, microreactor technology has shown promise as a novel method in biocatalysis, in which the reactions generally produce the desired product with a higher yield, in a shorter time, and more efficiently, safely, and selectively than traditional batch-scale reactions (Chen et al., 2013; Woodcock et al., 2008). These microreactors take advantage of rapid heat and mass transfer rates that cannot be achieved by conventional batch systems (Lévesque and Seeberger, 2012; Seo et al., 2012). In addition, microreactors contribute to the rationalization of process development with significant reductions in manpower, the quantity of reagents required, the amount of waste solvent generated and cost (Wen et al., 2009; Marques et al., 2012). For industrial-scale production, the use of packed bed reactors with immobilized enzymes is more cost effective than the use of reactors operated in batch mode (Chen et al., 2011a,b). The advantages of packed bed reactors include continuous operation, effective reuse of enzyme without prior separation, reduction of labor costs, and protection of enzymes from mechanical shear stress (Kundu et al., 2011; Martin-Rapun et al., 2013). Recent publications in the field of miniaturization of packed bed reactors have shown promising results (Wang et al., 2013b), including improved mass transport rates leading to higher yields and shorter reaction times; for example, a relatively cheap and reusable miniaturized packed bed reactor was used for the lipase-catalyzed synthesis of anhydride and isoamyl alcohol (Cvjetko et al., 2012). Thus, lipase-catalyzed synthesis of CAPE in a miniaturized packed bed reactor is an effective strategy to obtain a greater yield in a shorter amount of time. However, to the best of our knowledge, no report has been published concerning the

biosynthesis of CAPE from alkyl caffeate and PE using lipasecatalyzed transesterification in a packed bed microreactor. The aim of this work was to develop a simple and efficient approach to the enzymatic transesterification synthesis of CAPE from alkyl caffeate and PE in [Bmim][Tf2N] using a novel miniaturized packed bed reactor. For this purpose, a series of alkyl caffeates were synthesized, and the optimal substrate for lipase-catalyzed transesterification of CAPE was chosen by screening. Furthermore, the effect of substrate concentration, substrate molar ratio, reaction temperature, and flow rates on the CAPE yield as well as the stability of the system were investigated in a continuous flow packed bed microreactor. Additionally, kinetic models for the biosynthesis of CAPE in batch reactors and microreactors were proposed and compared. 2. Methods 2.1. Materials [Bmim][Tf2N] was obtained from Shanghai Cheng-Jie Chemical Co., Ltd. (Shanghai, China). Lipases including Novozym 435, Lipozyme TL IM and Lipozyme RM IM were purchased from Novozymes (Bagsvaerd, Denmark). Caffeic acid was purchased from Nanjing Zelang Pharmaceutical Sci. & Tech. Co., Ltd. (Nanjing, China). Methanol and acetonitrile were HPLC grade (Tedia Co., Fairfield, OH, USA), and all other reagents and solvents were analytical grade (Sinopharm Chemical Reagent Co., Ltd., Shanghai, China). 2.2. Batch reaction Initial screening of the suitable alkyl caffeate and lipase for enzymatic transesterification from alkyl caffeate and PE to obtain CAPE was performed in 5 mL glass vials with screw caps (Pang et al., 2013). Table 1 shows the abbreviation of the alkyl caffeates used as substrates in the present study. Alkyl caffeate and PE were added to 1 mL [Bmim][Tf2N]. The reaction was initiated by the addition of lipase. All reactions were performed at temperatures of 55 °C, 70 °C, and 85 °C with a constant stirring speed of 120 rpm. After 3 days, 20 lL aliquots were taken from the reaction mixture and diluted using 980 lL of methanol for HPLC analysis. All experiments were performed in triplicate. 2.3. Continuous flow microreactor reaction Microchannels with dimensions of 1 cm, 500 lm, and 75 mm in width (W), height (H), and length (L), respectively, were milled on a PDMS plate. The designed packed-bed microreactor was easily assembled, operated and cleaned. 90 mg of the Novozym 435 beads was uniformly put in the 75 mm long channel. Then the channel was covered with another PDMS plate. Due to the PDMS material is elastic and two PDMS plates can be sealed by using

Table 1 The abbreviation of the alkyl caffeates used as substrates in the present study. Antioxidant

Abbreviation

—R

Methyl caffeate Ethyl caffeate Propyl caffeate Isopropyl caffeate Butyl caffeate Amyl caffeate Isoamyl caffeate Hexyl caffeate Heptyl caffeate Octyl caffeate

MC EC PC IpC BuC AC IaC HexC HepC OC

CH3 CH2CH3 CH2CH2CH3 CH2CH2CH3 (CH2)3CH3 (CH2)4CH3 (CH2)4CH3 (CH2)5CH3 (CH2)6CH3 (CH2)7CH3

J. Wang et al. / Bioresource Technology 158 (2014) 39–47

some long tail clips. When the reaction completed, clips were took off and the exhausted biocatalysts were extracted. Fresh enzyme was loaded into microreactor for next reaction. The microreactor was placed on a temperature-controlled box (Wang et al., 2013b). [Bmim][Tf2N] containing a specific molar ratio of MC and PE was introduced into the microreactor by a syringe pump at different rates ranging from 2 lL/min to 20 lL/min. At regular time intervals, 20 lL aliquots were collected from the reaction mixture and were diluted using 980 lL of methanol for HPLC analysis. The optimum condition was selected by response surface methodology, and the corresponding information was included in an Electronic Supporting Information. All the experiments were performed in triplicate to determine the experimental variability. 2.4. Kinetic model of a continuous flow packed bed microreactor and a batch reactor For the determination of kinetics in a continuous flow packed bed microreactor, the flow rate dependence of Km for MC was examined at 60 °C. Substrate solutions containing different concentrations of MC ranging from 0.5 mg/mL to 9 mg/mL were fed at flow rates of 2–10 lL/min. The synthesis of CAPE was measured at steady state, and the conversion of MC was calculated from the difference in the initial MC and the product concentration. The kinetics of transesterification in the batch reactor were investigated by studying the effects of the concentrations of both MC and PE on the initial rate of reaction (Tran et al., 2013). The MC concentration was varied at different fixed concentrations of PE and vice versa. The concentrations of MC and PE were varied within the ranges of 2.5–46 mM and 0.16–0.64 M, respectively. Initial reaction rates, expressed as mM CAPE per minute and per gram of enzyme, were determined from the time course of the CAPE concentration using a second-order polynomial regression analysis of the product concentration and by determining the initial slope of the line tangent to the curve. 2.5. Reusability and stability of enzyme in the continuous flow microreactors Reusability of Novozym 435 was tested by reusing the enzyme in the synthesis of CAPE in the microreactor. In each cycle of the assay, the flow rate of the reaction system was set at 2 lL/min, 3 mg/mL MC, and the reaction time was 3 h at 60 °C. After this time, the enzyme was washed with fresh [Bmim][Tf2N] and fresh reaction substrates were then injected for the next cycle. For each cycle, the CAPE yield after 3 h was measured. The stability of Novozym 435 was investigated continuously for 15 days in the microreactor at 60 °C. The flow rate of the reaction system was set at 2 lL/min and contained MC (concentration: 3 mg/mL). At 12 h time intervals, 20 lL samples from the outflows were collected to determine the stability of the enzyme yield. 2.6. Confirmation and analysis of products by LC–MS and NMR To confirm the product CAPE, reaction mixtures were injected into an LC-MS. The experimental conditions were identical to those employed in previous studies (Wang et al., 2013a). The data were processed using Xcalibur 1.2 software (Pang et al., 2013). The intense peaks at m/z 193.1 and m/z 283.1 in the ESI-MS spectra under negative ion mode corresponded to the deprotonated [M–H] ions of MC and CAPE, respectively (see Electronic Supporting Information). 1 H NMR data of CAPE and MC were acquired (at room temperature) on a Bruker AVANCE spectrometer 400 (BrukerBiospin Co., Billerica, MA, USA). Dimethylsulfoxide (DMSO)-d6 was used as a solvent, and 5 mg of sample was dissolved in 0.5 mL of DMSO-d6.

41

Chemical shifts are expressed in d (ppm) values relative to tetramethylsilane (TMS) as an internal standard. The NMR data for CAPE and alkyl caffeate were in accordance with the values in the literature (Chalas et al., 2001).

2.7. HPLC analysis The product CAPE concentration was determined using an HPLC system equipped with a constant flow pump (2PB0540, Beijing Satellite Factory., Beijing, China) and a UV–VIS detector (L-7420, Techcomp Co., Ltd., Shanghai, China). Separation was performed on a Kromasil C18 column (250 mm  4.6 mm, i.d.: 5 lm, Akzo Nobel Co., Amsterdam, Netherlands) at 30 °C. The determinations of CAPE and MC were performed using acetonitrile/0.05% acetic acid solution (50:50, v/v)/water (50:50, v/v) and methanol/water (65:35, v/v), respectively. The flow rate was 1 mg/mL, and samples were detected under UV light at 325 nm (Wang et al., 2011). The CAPE yield was calculated according to Eq. (1).

CAPE yield ð%Þ ¼

CAPE generated ðmolÞ  100 Initial MC ðmolÞ

ð1Þ

The rate of CAPE production (r) under continuous-flow condition was calculated according to Eq. (2).

r ¼ ½P  f ½lmol=min

ð2Þ

where [P] is the concentration of the product (lmol/mL), f is the flow rate (mL/min).

3. Results and discussion 3.1. Alkyl caffeate selection Fig. 1A shows the effect of alkyl caffeates with different chain lengths on the lipase-catalyzed synthesis of CAPE via transesterification of alkyl caffeate and PE at temperatures of 55 °C, 70 °C, and 85 °C. In general, alkyl caffeates had relatively higher CAPE yields when the enzymatic reaction was conducted at 70 °C. At 70 °C, the CAPE yields were highest when MC, PC, and EC were used as substrates, and the yields were 71.4%, 70.0%, and 69.6%, respectively. Further increases in the number of carbon atoms of the alkyl caffeate side chain decreased the CAPE yield, and the lowest yield occurred with OC. A possible reason for this result is that the longer the alkyl group, the higher the steric hindrance, which makes substrate contact with the enzyme more difficult (Vafiadi et al., 2005). Thus, additional experiments were performed using MC, EC and PC as substrates. Further, the effect of temperature (25 °C, 40 °C, 55 °C, 70 °C, and 85 °C) was tested in the substrate screening (Fig. 1B). MC, EC, and PC were used as substrates. The three substrates showed a similar trend. With the increase of temperature, the CAPE yield increased, and the highest yield was obtained at 70 °C. The MC substrate had the highest gain in CAPE yield (71.4%) among the three substrates. Compared with ethanol and propyl alcohol, the by-products in the transesterification of EC and PC, methanol is more volatile and is thus more conducive to higher reaction yields. The CAPE yield decreased in all of the reactions after further increasing the temperature. Some references have reported that the enzymatic activity of Novozym 435 is highest between 70 °C and 80 °C, and a higher temperature would therefore lead to a decrease in enzymatic activity (Yadav and Devendran, 2012). Thus, MC was the most suitable substrate, and the optimal temperature was 70 °C.

42

J. Wang et al. / Bioresource Technology 158 (2014) 39–47

and Lipozyme RM IM. When Novozym 435 was used as the catalyst and MC as the substrate, the CAPE yield was 71.4%. Very low yields were measured with Lipozyme TL IM and Lipozyme RM IM, indicating that the two enzymes are clearly not adapted to catalyze the synthesis of CAPE. This result was in agreement with the literature (Ha et al., 2012). Thus, Novozym 435 is an ideal candidate to catalyze the transesterification of MC with PE to obtain CAPE. 3.3. Continuous flow transesterification in a packed bed microreactor

Fig. 1. The effects of different alkyl caffeates as the substrate on the yield of CAPE in [Bmim][Tf2N] at different temperatures. (A) The effects of different alkyl caffeates as the substrate on the yield of CAPE in [Bmim][Tf2N] at 55 °C, 70 °C and 85 °C. (B) The effects of MC, EC and PC as the substrate on the yield of CAPE in [Bmim][Tf2N] at different temperatures. Conditions: molar ratio of alkyl caffeate to 2-phenylethanol = 1:14, mass ratio of alkyl caffeate to lipase = 1:16, [Bmim][Tf2N] 1 mL, 120 rpm for 72 h.

3.2. Lipase selection Three commercial immobilized lipases (Novozym 435, Lipozyme TL IM, and Lipozyme RM IM) were tested at different temperatures to compare their effects on the transesterification reactions of MC and EC as well as PC with PE (Table 2). For the three substrates, Lipozyme TL IM and Lipozyme RM IM preferentially catalyzed the transformation of PC, whereas Novozym 435 always acted in favor of the transformation of MC. Novozym 435 had the highest catalytic activity compared to Lipozyme TL IM

Further experiments using [Bmim][Tf2N] as the media and Novozym 435 as the catalyst for CAPE synthesis were conducted in a packed bed microreactor, and the effects of reaction parameters on the CAPE yield at different flow rates were systematically studied. Temperature had a significant effect on the yield of CAPE and the rate of CAPE production using a continuous flow packed bed microreactor in the evaluated range (40–70 °C), respectively (see Fig. 2A and B). Although the rate of CAPE production reached the maximum 0.022 lmol/min at 70 °C and 10 lL/min, the CAPE yield was low (41.8%) due to the length limitation of the microreactor. An improvement in performance would be expected by increasing the length of microreactor to obtain enough residence time. Hence, the CAPE yield was chosen as the basis of comparison to study the effect of temperature. The CAPE yield was the lowest at 40 °C and achieved its highest value at 60 °C at low flow rates (2 lL/min and 5 lL/min). The CAPE yield did not change significantly with further increases in temperature. However, the CAPE yield still increased with further increases in temperature at high flow rates (10 lL/ min and 20 lL/min). When the reaction was conducted at 40 °C and 20 lL/min, only a 3.5% CAPE yield was obtained, while at 60 °C and 2 lL/min, a maximum 84.5% CAPE yield was achieved. The higher temperature resulted in an increase in the catalytic activity. Compared to the batch reactor (80 °C), the optimal temperature was much lower in the microreactor (60 °C). This is due to the higher heat and mass transfer that exists in the microreactor (Pohar et al., 2012). Therefore, lower temperature provides industrial production with beneficial energy savings. Under identical conditions of temperature and amount of lipase, the effects of the concentration of MC on the CAPE yield and the rate of CAPE production at different flow rates were observed, respectively (see Fig. 2C and D). In view of the solubilities of substrate MC and product CAPE in [Bmim][Tf2N] were 9.2 mg/mL and 19.2 mg/mL at 60 °C, respectively. Thus the concentration of MC in the range of 0.5–9 mg/mL was investigated. The conversion between MC and CAPE was enhanced with increasing MC concentrations up to 3 mg/mL. As expected, the CAPE yield at steady-state conditions increased by lowering the flow rate. A yield of approximately 87.0% was achieved at an inlet MC concentration of 3 mg/ mL and a flow rate of 2 lL/min. With increasing concentration of MC, the reaction rate was observed to increase. This is due to that the concentration of the substrate is increased and more substrate molecules bind to the lipase which makes reaction rate quick. The CAPE yield and the rate of CAPE production increased as the molar ratio of MC to PE increased and did not decrease at a ratio above 1:25 (see Fig. 2E and F). The highest CAPE yields of 92.6%, 82.3% and 42.6% were obtained at flow rates of 2 lL/min, 5 lL/ min and 10 lL/min, respectively. At a flow rate of 20 lL/min and a molar ratio of MC to PE of 1:40, the highest CAPE yield had only 25.8%; however, the maximal rate of CAPE production (0.027 lmol/min) was achieved. One reason for adding this large amount of PE is that transesterification is a reversible reaction, and the excess PE would shift the equilibrium toward CAPE production. Additionally, PE can enhance mass transfer by reducing the viscosity of IL (Yu et al., 2010). Excess PE had an inhibitory effect on lipase in the continuous flow microreactor, especially at

43

J. Wang et al. / Bioresource Technology 158 (2014) 39–47 Table 2 The effects of different lipases as biocatalysts on the yield of CAPE in [Bmim][Tf2N] at different temperatures. Temp. (°C)

Yielda (%) Novozym 435

25 40 55 70 85 a

Lipozyme TL IM

Lipozyme RM IM

MC

EC

PC

MC

EC

PC

MC

EC

PC

14.4 ± 0.84 60.4 ± 1.40 64.7 ± 2.35 71.4 ± 5.36 54.5 ± 0.17

16.5 ± 0.89 59.4 ± 1.33 60.7 ± 2.05 66.6 ± 1.19 49.9 ± 1.20

30.±1.15 58.±1.11 58 ± 0.04 60.±0.39 44.±0.59

0.01 ± 0.04 2.6 ± 0.15 5.0 ± 0.07 5.3 ± 0.42 2.8 ± 0.12

0.6 ± 0.09 9.8 ± 2.61 4.2 ± 0.27 5.7 ± 0.22 2.2 ± 0.31

2.5 ± 0.52 6.6 ± 0.05 12.±0.22 10.0 ± 0.94 3.7 ± 0.05

2.3 ± 0.54 3.2 ± 0.08 8.2 ± 0.14 12.0 ± 0.32 6.9 ± 0.06

3.0 ± 0.27 5.4 ± 0.21 13 ± 0.52 17.0 ± 0.88 9.1 ± 0.28

6.1 ± 0.06 10.2 ± 0.48 18.3 ± 1.09 22.5 ± 1.12 13.8 ± 0.57

Molar ratio of alkyl caffeate to 2-phenylethanol = 1:14, mass ratio of alkyl caffeate to lipase = 1:16, [Bmim][Tf2N] 1 mL, 120 rpm for 72 h.

Fig. 2. The effects of temperature (A and B), MC concentration (C and D), and substrate molar ratio (E and F) on the yield of CAPE and the rate of CAPE production using a continuous flow packed bed microreactor at different flow rates, respectively. Conditions: (A and B) 90 mg of Novozym 435, molar ratio of MC to PE = 1:20, MC concentration = 1 mg/mL; (C and D) 90 mg of Novozym 435, molar ratio of MC to PE = 1:20, 60 °C; (E and F) 90 mg of Novozym 435, MC concentration = 3 mg/mL, 60 °C.

44

J. Wang et al. / Bioresource Technology 158 (2014) 39–47

low flow rates. Dencˇic´ et al. (2013) also reported that the low molar ratio had benefit to lipase transesterification synthesis of ethyl butyrate, especially at longer residence times (at low flow rates). At a high flow rate (20 lL/min), the contact time between PE and the enzyme decreased. Therefore, lipase is more tolerant to higher PE concentrations at a high flow rate. Fig. 3 shows that the CAPE yields increased steadily with the residence time, which ranged from 2.5 h to 15 min, in a packed bed microreactor. The residence time was obtained by actually measuring the time of reaction mixture from inlet to outlet in the microreactor at a flow rate ranging from 2 lL/min to 20 lL/

min. The results indicated that the CAPE yield was low for a short residence time, i.e., only 7.2% of the CAPE yield was obtained for a residence time of 15 min at an MC concentration of 9 mg/mL at 60 °C. The CAPE yield increased with increasing residence time because the interactions between the substrate and enzyme increased (Cvjetko et al., 2012). The residence time (1–2.5 h) did not have a significant effect on the CAPE yield at a low temperature (40–50 °C) (Fig. 3A). The CAPE yield increased linearly with the residence time (0.25–1 h) at different MC concentrations (Fig. 3B) and substrate molar ratios (Fig. 3C). As the residence time increased from 1 to 2.5 h, the increase in the CAPE yield was relatively stable. The substrate may have had contact with most of the enzymes at a flow rate of 5 lL/min (residence time 1 h). The highest yield (92.6%) of CAPE was obtained at a flow rate of 2 lL/min (residence time 2.5 h), which would give the immobilized enzyme a productivity of 19 mmol/g/h, calculated per g of Novozym 435 preparation. Compared to the batch reactor (7.4 mmol/g/h), the microreactor had a higher efficiency. This improvement can be attributed to two factors: (1) the catalyst surface area to volume of reactor in the microreactors is much higher than that in batch reactors, which increases the enzyme active sites available to reactants at any time; (2) the volume of the microreactor is so limited that reactants are forced to be in contact with enzyme active sites because the diffusion paths in microreactors are much smaller (Kundu et al., 2011). 3.4. Kinetic parameters in a packed bed microreactor and a batch reactor For continuous flow kinetics, this investigation aims to determine the variability of apparent kinetic parameters (Km(app)) with flow rates. The influence of flow rate on Km(app) for the packed bed enzyme reactor systems can be easily evaluated according to the data obtained from the flow rate and MC concentration studies. The experimental data were fitted by the Lilly–Hornby model (Lilly et al., 1966) and described by the following equation:

f  ½A0  ¼

Fig. 3. The effect of residence time on the yield of CAPE using a miniaturized continuous flow packed bed microreactor with different temperatures (A), MC concentrations (B) and substrate molar ratios (C). Conditions: (A) 90 mg of Novozym 435, molar ratio of MC to PE = 1:20, MC concentration = 1 mg/mL. (B) 90 mg of Novozym 435, molar ratio of MC to PE = 1:20, 60 °C; (C) 90 mg of Novozym 435, MC concentration = 3 mg/mL, 60 °C.

C þ K mðappÞ  lnð1  f Þ Q

ð3Þ

where f is the fraction of substrate converted during the reaction, Q is the flow rate, [A0] is the initial concentration of substrate, C is the reaction capacity of the continuous flow packed bed microreactor, and Km(app) is the apparent Michaelis constant. Fig. 4A illustrates the linear plots of f[A0] versus ln(1  f), and Km(app) values were derived from the slopes of these plots. Over the tested flow rate range, the Km(app) values for the microreactors were in the range of 14.04–39.56 mM (Table 3). As expected, Km(app) values increased with increasing flow rates. Typically, the apparent Michaelis constant depends on two factors, the enzyme and the mass transfer rate. The different Km(app) values suggest that the apparent kinetics of the packed bed microreactor is significantly affected by mass transfer. At 2 lL/min, the Km(app) was lowest, which indicated a more efficient formation between enzyme and substrate. Therefore, the low flow rate is more suitable for the reaction. The kinetics of the enzymatic transesterification synthesis of CAPE was studied in a batch reactor to better understand the mechanism of the reaction. Most kinetics studies on lipasecatalyzed transesterification are described using a ping-pong bi–bi kinetic model (Zhang et al., 2004). During the course of the experiment, substrate inhibition was tested. The enzyme was inhibited at higher concentrations of PE. Therefore, the kinetics of the lipase-catalyzed transesterification synthesis of CAPE suggests that the model is based on the ping-pong bi–bi mechanism, with inhibition by PE. Thus, the rate equation is the following:

45

J. Wang et al. / Bioresource Technology 158 (2014) 39–47 Table 4 Kinetic parameters for the lipase-catalyzed synthesis of CAPE in a batch reactor at different substrate concentrations. Parametera

Value

Vmax (mM min1 g1) KmA (mM L1) KmB (mM L1) KiB (mM L1)

16.95 432.10 1279.00 482.40

a The kinetic assay was performed at 60 °C and 120 rpm with the concentrations of CA and PE varied from 15–40 mM to 0.16– 0.64 M, respectively.

Fig. 4. Kinetic plots of the lipase-catalyzed synthesis of CAPE from MC and PE using different reactors. (A) Lilly–Hornby plots of the continuous flow enzyme microreactor at flow rates of 2 lL/min, 5 lL/min and 10 lL/min under different MC concentrations (2.5–46 mM); (B) Lineweaver–Burk plot of reciprocal initial rates vs. MC concentrations at fixed PE concentrations. Reaction conditions: 90 g enzyme L1, 60 °C and 120 rpm.

Table 3 Kinetic constants for the lipase-catalyzed synthesis of CAPE in [Bmim][Tf2N] using a continuous flow packed bed microreactor at 60 °C. Flow rate (lL/min)

Km(app) (mM)

R

2a 5b 10c

14.04 22.07 39.56

0.872 0.969 0.905

a

Molar ratio of MC to PE = 1:20, mass ratio of MC to Novozym 435 = 1:90, flow rate 2 lL/min. b Molar ratio of MC to PE = 1:20, mass ratio of MC to Novozym 435 = 1:90, flow rate 5 lL/min. c Molar ratio of MC to PE = 1:20, mass ratio of MC to Novozym 435 = 1:90, flow rate 10 lL/min.



V max ½A½B   K mB ½A þ K mA ½B 1 þ K½BiB þ ½A½B

ð4Þ

where v is the initial reaction rate, Vmax the maximum reaction rate, KmA and KmB are the Michaelis constants for substrates CA (A) and PE (B), and KiB is the inhibition constant for PE. To determine the kinetic parameters of this reaction, the MC concentration was varied at different fixed concentrations of PE and vice versa. Reciprocal initial reaction rates (v1) were plotted versus the inverse MC concentration (C1) (Lineweaver–Burk plot)

Fig. 5. Reusability and service life of Novozym 435 in continuous flow transesterification synthesis of CAPE from MC and PE. Reaction conditions: flow rate 2 lL/ min, 90 mg enzyme, molar ratio of MC to PE 1:20, and temperature 60 °C.

for different PE concentrations (Fig. 4B). PE inhibition was confirmed because at high PE concentrations, the slopes of the lines appear to increase. The reaction sequence may be given as follows: MC binds first to the enzyme and forms an acyl–enzyme complex with the acyl donor after the release of the first product, methanol. The second substrate, PE, binds to the acyl–enzyme complex and forms another complex, which again undergoes isomerization to an ester–enzyme complex. Finally, the complex breaks into the second product, CAPE, and the enzyme (Xiong et al., 2008). Initial reaction rates, expressed as mmol produced CAPE per minute and per gram of enzyme, were determined from the time course of CAPE concentration. Graphs were made of CAPE concentrations versus the time, the initial reaction rates were based on the linear part of the slopes. The kinetic parameters were fitted to these

46

J. Wang et al. / Bioresource Technology 158 (2014) 39–47

initial reaction rates and estimated using the non-linear regression of Eq. (4) implemented in SPSS statistic software. Based on the results of the calculated kinetics parameters indicating that KmA was lower than KmB (Table 4), the enzyme has higher affinity toward MC than PE. 3.5. Reusability and stability of the enzyme in a continuous flow microreactor Fig. 5A shows the reusability of the enzyme over 20 runs. At optimal conditions, the CAPE yield was maintained at approximately 90%; this result indicated that the immobilized lipase could be reused at least 20 times without losing catalytic activity. However, Novozym 435 could only be reused three times and maintain more than 90% of its activity in the batch reactors (Widjaja et al., 2008). Meanwhile, system stability was tested by following CAPE productivity over 15 consecutive days of the enzymatic transesterification process run at a 2 lL/min flow rate and a 3 mg/mL inlet MC concentration (Fig. 5B). The conversion of MC to CAPE at a 2 lL/min flow rate was constant for 9 days. After day 9, the CAPE yield started to decrease with a corresponding increase in the remaining unconverted substrate and methanol residual. After 11 days, the microreactor had only 50% transformation efficiency. Hence, a reasonable assumption is that removing the methanol can lead to continuous synthesis over a prolonged period. Moreover, the separation of the product CAPE from [Bmim][Tf2N] is very easy. In our follow-up study, the partition coefficient of CAPE between a new complexation extraction system (consisted of trioctylphosphine oxide and cyclohexane) and [Bmim][Tf2N] is 9.6, so the developed complexation extraction process could be used to extract CAPE, and the solvents could be recycled and reused. In addition, [Bmim][Tf2N] can be reused at least 20 times without decreasing in yield. Finally, the residue is easy to obtain white product with high purity by simple recrystallization. 4. Conclusions The Novozym 435-catalyzed synthesis of CAPE by the transesterification of MC with PE in [Bmim][Tf2N] was first carried out, and a packed bed microreactor was used to strengthen this enzymatic reaction in a continuous flow mode. The microreactor had a more efficient immobilized enzyme productivity compared with the batch reactor. RSM was successfully used to determine the optimal conditions for CAPE synthesis. A maximum yield of 93.21% was achieved in 2.5 h under optimized conditions. The microreactor was reusable multiple times without loss of enzyme activity. A ping-pong bi–bi model with inhibition by PE was proposed for the transesterification process. Acknowledgements This work was supported by the Natural Science Foundation of China (21206061), the China Postdoctoral Science Foundation Funded Projects (2012M510124, 2013T60505), the Postdoctoral Science Foundation Funded Project of Jiangsu University (1143002085), the Qing Lan Project of Jiangsu Province, the Graduate Innovation Project of Jiangsu Province (CXZZ13_0713), and the Start Project for Introduced Talent at Jiangsu University of Science and Technology (35211002). Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biortech.2014.01. 145.

References Chalas, J., Claise, C., Edeas, M., Messaoudi, C., Vergnes, L., Abella, A., Lindenbaum, A., 2001. Effect of ethyl esterification of phenolic acids on low-density lipoprotein oxidation. Biomed. Pharmacother. 55, 54–60. Chen, H.C., Chen, J.H., Chang, C., Shieh, C.J., 2011a. Optimization of ultrasoundaccelerated synthesis of enzymatic caffeic acid phenethyl ester by response surface methodology. Ultrason. Sonochem. 18, 455–459. Chen, H.C., Ju, H.Y., Wu, T.T., Liu, Y.C., Lee, C.C., Chang, C., Chung, Y.L., Shieh, C.J., 2011b. Continuous production of lipase-catalyzed biodiesel in a packed-bed reactor: optimization and enzyme reuse study. J. Biomed. Biotechnol. 2011. http://dx.doi.org/10.1155/2011/950725. Chen, Y.Z., Zhao, Y.C., Han, M., Ye, C.B., Dang, M.H., Chen, G.W., 2013. Safe, efficient and selective synthesis of dinitro herbicides via a multifunctional continuousflow microreactor: one-step dinitration with nitric acid as agent. Green. Chem. 15, 91–94. Cvjetko, M., Vorkapic´-Furacˇ, J., Zˇnidaršicˇ-Plazl, P., 2012. Isoamyl acetate synthesis in imidazolium-based ionic liquids using packed bed enzyme microreactor. Process. Biochem. 47, 1344–1350. Dencˇic´, I., de Vaan, S., Noël, T., Meuldijk, J., de Croon, M., Hessel, V., 2013. Lipasebased biocatalytic flow process in a packed-bed microreactor. Ind. Eng. Chem. Res. 52, 10951–10960. Gocer, H., Gulcin, I., 2011. Caffeic acid phenethyl ester (CAPE): correlation of structure and antioxidant properties. Int. J. Food. Sci. Nutr. 62, 821–825. Ha, S.H., Anh, T.V., Lee, S.H., Koo, Y.M., 2012. Effect of ionic liquids on enzymatic synthesis of caffeic acid phenethyl ester. Bioprocess. Biosyst. Eng. 35, 235– 240. Kim, H., Choi, Y.K., Lee, J., Lee, E., Park, J., Kim, M.J., 2011. Ionic-surfactant-coated Burkholderia cepacia lipase as a highly active and enantioselective catalyst for the dynamic kinetic resolution of secondary alcohols. Angew. Chem. Int. Ed. 50, 10944–10948. Kundu, S., Bhangale, A.S., Wallace, W.E., Flynn, K.M., Guttman, C.M., Gross, R.A., Beers, K.L., 2011. Continuous flow enzyme-catalyzed polymerization in a microreactor. J. Am. Chem. Soc. 133, 6006–6011. Kurata, A., Kitamura, Y., Irie, S., Takemoto, S., Akai, Y., Hirota, Y., Fujita, T., Iwai, K., Furusawa, M., Kishimoto, N., 2010. Enzymatic synthesis of caffeic acid phenethyl ester analogues in ionic liquid. J. Biotechnol. 148, 133–138. Lee, E.S., Lee, J.O., Lee, S.K., Kim, J.H., Jung, J.H., Keum, B., Park, S.H., Kim, H.S., 2009. Caffeic acid phenethyl ester accumulates b-catenin through GSK-3b and participates in proliferation through mTOR in C2C12 cells. Life Sci. 84, 755–759. Lévesque, F., Seeberger, P.H., 2012. Continuous-flow synthesis of the anti-malaria drug artemisinin. Angew. Chem. Int. Ed. 51, 1706–1709. Lilly, M.D., Hornby, W.E., Crook, E.M., 1966. The kinetics of carboxymethylcellulose– ficin in packed beds. Biochem. J. 100, 718–723. Marques, M.P.C., Fernandes, P., Cabral, J.M.S., Zˇnidaršicˇ-Plazl, P., Plazl, I., 2012. Continuous steroid biotransformations in microchannel reactors. New Biotechnol. 29, 227–234. Martin-Rapun, R., Sayalero, S., Pericas, M.A., 2013. Asymmetric anti-mannich reactions in continuous flow. Green. Chem. 15, 3295–3301. Noelker, C., Bacher, M., Gocke, P., Wei, X., Klockgether, T., Du, Y.S., Dodel, R., 2005. The flavanoide caffeic acid phenethyl ester blocks 6-hydroxydopamine-induced neurotoxicity. Neurosci. Lett. 383, 39–43. Ozturk, G., Ginis, Z., Akyol, S., Erden, G., Gurel, A., Akyol, O., 2012. The anticancer mechanism of caffeic acid phenethyl ester (CAPE): review of melanomas, lung and prostate cancers. Eur. Rev. Med. Pharmacol. Sci. 16, 2064–2068. Pang, N., Gu, S.S., Wang, J., Cui, H.S., Wang, F.Q., Liu, X., Zhao, X.Y., Wu, F.A., 2013. A novel chemoenzymatic synthesis of propyl caffeate using lipase-catalyzed transesterification in ionic liquid. Bioresour. Technol. 139, 337–342. Pohar, A., Zˇnidaršicˇ-Plazl, P., Plazl, I., 2012. Integrated system of a microbioreactor and a miniaturized continuous separator for enzyme catalyzed reactions. Chem. Eng. J. 189–190, 376–382. Seo, T.Y., Eum, K.W., Han, S.O., Kim, S.W., Kim, J.H., Song, K.H., Choe, J., 2012. Immobilized cell microchannel bioreactor for evaluating fermentation characteristics of mixed substrate consumption and product formation. Process. Biochem. 47, 1011–1015. Suzuki, K., Tanaka, I., Nakanishi, I., Kurematsu, A., Yakumaru, H., Ikota, N., Ishihara, H., 2006. Drastic effect of several caffeic acid derivatives on the induction of heme oxygenase-1 expression revealed by quantitative real-time RT-PCR. Biofactors 28, 151–158. Tan, Z.L., Shahidi, F., 2012. A novel chemoenzymatic synthesis of phytosteryl caffeates and assessment of their antioxidant activity. Food Chem. 133, 1427– 1434. Tran, D.T., Lin, Y.J., Chen, C.L., Chang, J.S., 2013. Kinetics of transesterification of olive oil with methanol catalyzed by immobilized lipase derived from an isolated Burkholderia sp. strain. Bioresour. Technol. 145, 193–203. Vafiadi, C., Topakas, E., Wong, K.K.Y., Suckling, I.D., Christakopoulos, P., 2005. Mapping the hydrolytic and synthetic selectivity of a type C feruloyl esterase (StFaeC) from Sporotrichum thermophile using alkyl ferulates. Tetrahedron: Asymmetry 16, 373–379. Wang, J., Gu, S.S., Zhang, L.X., Wu, F.A., Guo, X.J., 2011. Pentyl E-3-(3,4droxyphenylacrylate. Acta. Crystallogr. Sect. E 67, o2871. Wang, J., Li, J., Zhang, L.X., Wu, F.A., 2013a. Lipase-catalyzed synthesis of caffeic acid phenethyl ester in ionic liquids: effect of specific ions and reaction parameters. Chin. J. Chem. Eng. 21, 1376–1385.

J. Wang et al. / Bioresource Technology 158 (2014) 39–47 Wang, J., Gu, S.S., Cui, H.S., Yang, L.Q., Wu, X.Y., 2013b. Rapid synthesis of propyl caffeate in ionic liquid using a packed bed enzyme microreactor under continuous-flow conditions. Bioresour. Technol. 149, 367–374. Wen, Z.Z., Yu, X.H., Tu, S.T., Yan, J.Y., Dahlquist, E., 2009. Intensification of biodiesel synthesis using zigzag micro-channel reactors. Bioresour. Technol. 100, 3054– 3060. Widjaja, A., Yeh, T., Ju, Y., 2008. Enzymatic synthesis of caffeic acid phenethyl ester. J. Chin. Inst. Chem. Eng. 39, 413–418. Woodcock, L.L., Wiles, C., Greenway, G.M., Watts, P., Wells, A., Eyley, S., 2008. Enzymatic synthesis of a series of alkyl esters using novozyme 435 in a packedbed, miniaturized, continuous flow reactor. Biocatal. Biotransfor. 26, 466–472.

47

Xiong, J., Wu, J.P., Xu, G., Yang, L.R., 2008. Kinetic study of lipase catalyzed asymmetric transesterification of mandelonitrile in solvent-free system. Chem. Eng. J. 138, 258–263. Yadav, G.D., Devendran, S., 2012. Lipase catalyzed synthesis of cinnamyl acetate via transesterification in non-aqueous medium. Process. Biochem. 47, 496–502. Yu, D.H., Tian, L.X., Ma, D.X., Wu, H., Wang, Z., Wang, L., Fang, X.X., 2010. Microwave-assisted fatty acid methyl ester production from soybean oil by Novozym 435. Green. Chem. 12, 844–850. Zhang, J., Wu, J.P., Yang, L.R., 2004. The kinetic study on lipase-catalyzed asymmetric alcoholysis of a-cyano-benzyl acetate in organic media. J. Mol. Catal. B: Enzym. 31, 67–72.

A novel continuous flow biosynthesis of caffeic acid phenethyl ester from alkyl caffeate and phenethanol in a packed bed microreactor.

Caffeic acid phenethyl ester (CAPE) is a rare natural ingredient with several biological activity, but the industrial production of CAPE using lipase-...
2MB Sizes 0 Downloads 3 Views