Lipase-catalyzed synthesis of sucrose monoester: Increased productivity by combining enzyme pretreatment and non-aqueous biphasic medium

Accepted Manuscript Title: Lipase-catalyzed synthesis of sucrose monoester: Increased productivity by combining enzyme pretreatment and non-aqueous biphasic medium Authors: Pranee Inprakhon, Natcha Wongthongdee, Taweechai Amornsakchai, Thunyarat Pongtharankul, Panya Sunintaboon, Lars O. Wiemann, Alain Durand, Volker Sieber PII: DOI: Reference:

S0168-1656(17)31546-8 http://dx.doi.org/doi:10.1016/j.jbiotec.2017.07.021 BIOTEC 7964

To appear in:

Journal of Biotechnology

Received date: Revised date: Accepted date:

18-3-2017 6-6-2017 21-7-2017

Please cite this article as: Inprakhon, Pranee, Wongthongdee, Natcha, Amornsakchai, Taweechai, Pongtharankul, Thunyarat, Sunintaboon, Panya, Wiemann, Lars O., Durand, Alain, Sieber, Volker, Lipase-catalyzed synthesis of sucrose monoester: Increased productivity by combining enzyme pretreatment and non-aqueous biphasic medium.Journal of Biotechnology http://dx.doi.org/10.1016/j.jbiotec.2017.07.021 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Lipase-catalyzed synthesis of sucrose monoester: Increased productivity by combining enzyme pretreatment and non-aqueous biphasic medium.

Pranee INPRAKHONa,b*, Natcha WONGTHONGDEEa,d,e, Taweechai AMORNSAKCHAIc, Thunyarat PONGTHARANKULa, Panya SUNINTABOONc, Lars O. WIEMANNb, Alain DURANDd,e, Volker SIEBERb

a

Department of Biotechnology, Faculty of Science, Mahidol University, Rama VI Rd.,

Phayathai, Bangkok 10400, Thailand. b

Bio, Electro and Chemocatalysis BioCat, Straubing branch, Fraunhofer Institute for

Interfacial Engineering and Biotechnology IGB, Schulgasse 11a, 94315 Straubing, Germany. c

Department of Chemistry, Faculty of Science, Mahidol University, Rama VI Rd., Phayathai,

Bangkok 10400, Thailand. d

e

CNRS, LCPM, UMR 7375, Nancy, F-54001, France.

Université de Lorraine, LCPM, UMR 7375, Nancy, F-54001, France.

*

Corresponding author. Tel.: + 66 2 201 5311 ; fax: + 66 2 354 7160.

E-mail address: [email protected] (P. Inprakhon)

1

Highlights

Lipase-catalyzed transesterification of sucrose and vinyl caprate was carried-out Vinyl caprate was dispersed in a concentrated solution of sucrose and lipase in DMSO Lipase pretreatment greatly improved catalytic activity and reaction productivity 2-O-acylated sucrose monoester was obtained as the major isomer in the final product

3

Abstract Sucrose monocaprate was synthesized by carrying out a lipase-catalyzed transesterification in a non-aqueous biphasic medium. Vinyl caprate was mechanically dispersed into a solution of sucrose in DMSO. The use of DMSO allowed increasing sucrose concentration up to 0.7 M (in DMSO). The denaturing effect of DMSO on lipase was avoided by pretreatment of lipase by pH adjustment in the presence of crown ether. This pretreatment maintained a significant catalytic activity which led to 0.2 M sucrose monoester within 1 h at 50 °C, which represented higher productivity than already reported. Detailed structural characterization revealed that only monoester was recovered and the 2-O-acylated sucrose monocaprate was the major isomer in the final product.

Keywords:

Lipase, sucrose, sucrose ester, transesterification.

4

1. Introduction Sucrose esters have appeared as an important class of carbohydrate-based surfactants (Behler et al. 2001; Daudé et al. 2012). They were found very convenient ingredients for food and cosmetic formulations due to very mild dermatological properties, non-toxicity, biodegradability and the possibility to be synthesized from bio-based reactants (Khan and Rathod 2015; Queneau et al. 2008; Shi et al. 2011). Sucrose esters have been synthesized by esterification or transesterification of sucrose with either fatty acids or methyl or vinyl esters with the aim to obtain preferably higher content of mono- and diesters. Indeed, these compounds found more potential applications as compared to highly substituted derivatives which exhibited low solubility in water. In this respect, one important characteristic of the esterification was its selectivity which was quite difficult to achieve considering the high functionality of sucrose (eight hydroxyl groups). Current manufacturing processes rely on base-catalyzed transesterification in polar organic solvents (dimethyl sulfoxide, DMSO, and dimethyl formamide, DMF) or in bulk at about 130 oC and exhibit low selectivity (Zhao et al. 2014). The use of enzymes as alternative catalysts has been actively explored over years at the laboratory scale (Shi et al. 2011). Although, the selectivity of the reaction was greatly improved, the productivity remained quite low. The main difficulties encountered for improving productivity arose from the low solubility of carbohydrates in organic solvents except in the very polar ones. Unfortunately, most of biocatalysts had a low activity in these polar solvents (like DMSO and DMF). Only the proteases of the subtilisin-family were shown to be active in such solvents (Ritthitham et al. 2009). Adding water to the organic solvent or using organic solvent with high polarity increased the solubility of sucrose, but decreased enzyme activity and favored unwanted side reactions like hydrolysis (Chamouleau et al. 2001; Humeau et al. 1998). Similarly, increasing the temperature increased sucrose solubility, but also affected the catalytic activity of the enzyme (Arcos et al. 1998; Coulon et

5

al. 1996). Thus there was a great challenge to find treatments for enzymes that could enhance their stability and their activity in polar solvents like DMSO or DMF. This would allow increasing the concentration of sucrose in the reaction medium and consequently increasing the productivity while maintaining selectivity of esterification. According to the state of the art, the highest reported sucrose loaded concentration was 0.15 mol/L in reaction media containing either DMF or DMSO (Ashrafuzzaman et al. 2014; Pedersen et al. 2003; Shi et al. 2011; Wang et al. 2012). Nevertheless, the maximal solubility of sucrose in DMSO has been reported to exceed 0.8 mol/L (Gaylord Chemical Company 2014). Thus the limiting parameter has been enzyme catalytic activity which prevented the use of good solvents for sucrose substrate. In addition, the times needed for reaction ranged between 6 hours and several days, which could be serious drawback for industrial developments. This

work

examined

the

synthesis

of

sucrose

monoesters

by

lipase-catalyzed

transesterification in a non-aqueous biphasic reaction medium in which vinyl caprate was dispersed into a concentrated solution of sucrose in DMSO. We selected vinyl ester as the substrate in order to suppress possible limitation by thermodynamic equilibrium. In addition, the choice of vinyl caprate was made by reference to substrate preference of Candida rugosa regarding the number of carbon atoms in the alkyl chain of the acyl donor (Janssen et al. 1996; Kaewprapan et al. 2011). The main goal of this work was to increase sucrose concentration as compared to previously reported conditions. The effect of several operational parameters (pre-treatment of lipase by pH-adjustment, sucrose concentration in reaction medium) on monoester production was examined. The chemical structure of sucrose ester was determined using different techniques (LC-MS, 1H NMR, 13C NMR). The surface activity of synthesized sucrose ester was characterized and compared to that of a commercial product.

6

2. Materials and Methods 2.1. Materials The lipase from Candida rugosa (Lipase AY) with 32,800 U.g-1 was purchased from Amano Enzyme Co. (Nagoya, Japan). Vinyl caprate, sucrose and dimethyl sulfoxide (DMSO, CROMASOV®) were purchased from Sigma-Aldrich (Buchs, Switzerland) and used without further purification. All other chemicals were of highest commercial purity and used without further purification.

2.2. Methods 2.2.1. Preparation of pH-adjusted lipase AY co-lyophilized with 18-Crown-6 ether Firstly, the mixed solution of 18-Crown-6 ether (43 mM) in phosphate buffer solution was prepared according to literature (Kaewprapan et al. 2011). Twenty milliliters of 20 mM phosphate buffer pH 7.5 were mixed thoroughly with 0.23 g (0.86 mmol) of 18-Crown-6 ether. Subsequently, 1 g of Candida rugosa lipase AY was added to the solution of 18Crown-6 ether in 20 mM phosphate buffer at pH 7.5 and the mixture was stirred at room temperature, with 250 rpm of agitation speed for 1 h. Lastly, the mixed solution was flashfrozen in liquid nitrogen and lyophilized with a freeze drier (Labcongo Corp., USA). The pretreated Candida rugosa lipase AY was recovered as a white powder and yielded 1.15 g (approximately 94%). When required, the protein content of native and pretreated lipase AY was increased from 2.3 wt% to 22 wt% by ultrafiltration The crude enzyme was dissolved in 50 mM KPi (pH 6) and concentrated using an Amicon ultrafiltration unit (MWCO 10 KDa, Millipore, Massachusetts, USA) according to the manufacturer’s recommendation.

2.2.2. Determination of lipase activity by pH-stat

7

The lipase activity in catalyzed-hydrolysis of standard tributyrin (substrate) was measured by automatically titrating the amount of released acid with a 0.01 M NaOH solution (Kaewprapan et al. 2007). The equivalence of mole of NaOH and released acid by pH-stat titration (C30 Compact Karl Fischer Coulometer, Mettler Toledo, Germany) was used to calculate the catalytic activity with the reaction mixture sample. The initial rate was calculated from the slope of the linear variation of the amount of released acid as a function of time (0-10min). The experiments were repeated in triplicate at fixed concentration of Candida rugosa lipase AY (1 mL, concentration of 1 mg.mL-1), stabilizer (4 mL, 10% w/w gum arabic solution in Tris-HCl buffer) and standard tributyrin (146.5 μL, concentration 25 mM) in the final reaction mixture of Tris-HCl buffer (20 mL, 0.28 mM Tris-HCl, 150 mM NaCl, 1.4 mM CaCl2). The results were reported as mean values ± standard deviation (SD).

2.2.3. Deactivation of pretreated lipase AY co-lyophilized with 18-Crown-6 ether Thermal deactivation of pretreated lipase AY was carried out by suspending 0.125 g of pretreated lipase AY in 5 mL of distilled water. The solution of pretreated enzyme was refluxed at 100°C for 5 h and then allowed to cool down at room temperature. Afterward, the samples were flash-frozen in liquid nitrogen and lyophilized with a freeze drier (Labcongo Corp., USA) for 48 h (Kaewprapan et al. 2007). Total deactivation of enzyme activity could be verified by using the methodology for determination of lipase activity described in section 2.2.2.

2.2.4. Lipase-catalyzed transesterification in a non-aqueous biphasic system The experiment was designed to study catalytic ability of different forms of Candida rugosa lipase including native lipase AY (non-pretreated lipase), pretreated lipase AY (pH-adjusted and co-lyophilized with 18-Crown-6 ether), deactivated pretreated lipase AY, and without

8

adding any lipase (using salts as control-catalyst instead of enzyme) for catalyzedtransesterification of sucrose with vinyl caprate in an organic biphasic system. Typically, a 0.3 M sucrose solution in DMSO was first prepared by dissolving 0.103 g of sucrose in 1 mL of DMSO under stirring with magnetic barrel at 50°C. A 0.84 M of vinyl caprate (188 μL) was mixed thoroughly with the sucrose solution and kept under stirring at 50°C for 5 min before adding 8.1 mg of enzyme (6.8 mg.mL-1). The enzyme-catalyzed synthesis reaction was allowed to proceed for 1 h. The progress of the reaction was detected by TLC. For comparison, salts (0.07 mg of potassium phosphate monobasic (KH2PO4) and 0.49 mg of potassium phosphate dibasic (K2HPO4) were added separately in the reaction medium of DMSO (1 mL) instead of Candida rugosa lipase AY as a control experiment. Afterwards, 0.10 g of sucrose was dissolved in the salt solution and kept under stirring with magnetic barrel at 50°C until complete dissolution. Subsequently, 0.84 M of vinyl caprate (188 μL) was mixed thoroughly with the sucrose solution and kept under stirring with magnetic barrel at 50°C, 500 rpm for 1 h. The progress of reaction was monitored by TLC. Additionally, vinyl caprate droplet size and density were visualized by inverted microscope BX51 & DP70 Digital Camera System (Olympus Corporation, Tokyo, Japan) at high power objective (x40 Phase contrast Microscopy) coupled with a microscope imaging software (Olympus DP Controller software) for imaging and size measurement.

2.2.5. Scale up of sucrose monoester synthesis and purification procedure Sucrose (3.08 g, 9 mmol) was dissolved in DMSO (30 mL) at 50°C with magnetic stirrer agitation speed of 500 rpm. Then vinyl caprate (5.64 mL, 25.31 mmol) was added continuously and kept under magnetic stirring for 30 min until vinyl caprate was well dispersed in the reaction mixture (a cloudy white emulsion reaction mixture formed). The enzymatic reaction was started by adding 0.24 g pretreated lipase AY as the biocatalyst and

9

the reaction was carried out during 3 h of incubation time. At the end of reaction time, the turbid reaction mixture turned into a clear solution. The reaction was stopped and the presence of sucrose ester was checked by 1H NMR analysis. Afterwards, the separation of products was carried out by liquid-liquid extraction to eliminate residual vinyl caprate. Hundred milliliters of n-hexane were added to the reaction mixture (50 mL) and mixed vigorously at room temperature. The n-hexane phase (clear upper phase) was separated to eliminate the residual vinyl deaconate. The remaining turbid liquid phase that contained sucrose, sucrose ester and DMSO was mixed with 50 mL of water and extracted with 200 mL of cyclohexane: 1-butanol (95:5, v/v) three times. Subsequently, the organic phases (upper phase) of each extraction were pooled together and the organic solvent was evaporated under reduced pressure at 40 °C until obtaining a crude oily product. The crude oily product (30.95 g) was added on the top of silica gels. A glass chromatography column (4 cm diameter x 25 cm height of gels) was packed with silica gels (particle size of 40-63 μm) and equilibrated with dichloromethane. Firstly, the column was washed with dichloromethane to remove free vinyl caprate and DMSO until there was no DMSO spot (Rf value of 0.818) detected on TLC plate. Then, sucrose monocaprate was eluted by a mixture of dichloromethane and methanol (90:10, v/v) with the Rf value of 0.1. The separated products were dried in a vacuum incubator (BIOBLOCK Scientific co., Ltd., France) at 65°C, 30 mbar for 24 h. Then the purity of sucrose monocaprate was verified by HPLC analysis.

2.3. Analytical procedures Analytical TLC was performed on TLC plate (TLC siliga gel 60 F425, Merck, Germany) with chloroform:methanol (8.5:1.5, v/v) as mobile phase in a developing chamber (50 mL glass bottle). Samples of reaction mixtures were taken by a capillary tube and spotted directly on the base line of TLC plate (3 cm width x 7 cm length). Later, each spot was visualized by

10

immersing TLC plate into the dye solution of potassium permanganate followed by drying with an air dryer. To obtain qualitative data, the spots corresponding to sucrose, sucrose monocaprate, DMSO and vinyl caprate were identified by comparing their retention factor (Rf) with those of standard compounds. The dye solution was prepared and stored in a dark bottle. The dye mixture consisted of 1.5 g of potassium permanganate, 10 g of potassium carbonate, 1.25 mL of 10% NaOH, w/w) and the volume was adjusted to 200 mL with distilled water. The HPLC analysis of sucrose esters was performed by modifying the method described in a previous study (Ritthirham et al. 2009). Sucrose monoester concentrations of the reaction mixture were determined by HPLC (Waters 2690 Alliance HPLC Separations Module) equipped with Agilent Zorbax® ODS C18 HPLC Column, Particle size: 5 μm, Dimensions: 250 x 3 mm (Agilent technology, Inc., USA) and RI detector (Waters 410 Differential Refractometer detector). The column temperature and detector were operated at 45°C. The samples were diluted with methanol (1:1, 1:2 or 1:3) and filtered through a PTFE syringe filter membrane, 0.45 μm pore size (Merck, Germany). Sucrose monoester was quantified based on the calibration of standard sucrose monocaprate obtained from the column chromatography purification. The injection volume was 10 µL and the mobile phase was 30% acetonitrile: 70 % water. The flow rate was 1 mL.min-1. For LC-MS analysis, Sucrose monocaprate (0.02 g), chemical formula C22H40O12, with a molar mass of 496.56 g.mol-1, was dissolved in methanol (2 mL) and then filtered through a PTFE syringe filter membrane, 0.45 μm pore size (Merck, Germany), 20 μL of sample solution (20 mM) was directly introduced into LC-MS analysis system coupled with ESI. The LC-MS analysis system (LC-MS-2020, SHIMADZU USA MANUFACTURING Inc, USA) was coupled with ESI in the electrospray positive mode. The mass ions (m/z) were recorded in a full scan mode with a mass range of 200-700. In positive mode, the ion source condition

11

was ion-spray voltage +4.5 kV. The chromatographic separation was performed on Nucleodur® RP C18 Column, Particle size: 10 μm, Dimensions: 250 x 4.6 mm (MACHEREY-NAGEL GmbH & Co, KG, Germany) and used gradient mobile phase program from A: 10% Acetonitrile/90% water/0.1% Formic acid and B: 100% Acetonitrile/0.1% Formic acid at a flow rate of 1 mL.min-1. Gradient condition was conducted as follows: 20%-100% B from 0-30 min and 100% B from 30-35 min. The eluent was nebulized in the system by a stream of dried air at a flow rate of 1.5 mL.min -1 at room temperature. For NMR analyses, the sucrose monocaprate (0.01 g) and/or the reaction mixture (0.1 mL) samples were dissolved in DMSO-d6 (0.7 mL) and transferred to NMR tubes. 1H NMR spectra were recorded on a Bruker Avance 300 MHz spectrometer. All chemical shifts were denoted in δ-units (ppm) relative to 1H residual signal of DMSO-d6 (2.5 ppm).

2.4. Surface tension measurements A series of aqueous solutions (20 mL) at various concentrations (1x10-8 to 6x10-2 M) of synthesized sucrose monocaprate, and n-decanoylsucrose, Calbiochem®, were prepared for the experiments and aged at room temperature for 24 h before the first measurement of surface tension in order to reach adsorption equilibrium at the surface. The surface tension with Milli-Q® ultrapure water was determined at room temperature (25°C) with FORCE Tensiometer (KRÜSS Gmbh, Germany) by du Nouy ring method. Measurements were repeated six times, and the results given as mean values ± SD.

2.5. Dynamic light scattering (DLS) Dynamic light scattering (DLS) measurements were carried out using Zetasizer Nano ZS, Malvern Instruments, Malvern, UK. For these measurements, each sample was diluted in

12

distilled water to obtain count rates within the range of 200-300 kcps. Within that range of count rates we checked that varying the dilution did not modify significantly the obtained values. Each experiment was the average of 14 successive measurements. Average diameter (Dz) was calculated from the analysis of the autocorrelation function by the method of cumulants. The “polydispersity index” (PDI) provided by HPPS was a reasonable way to appreciate whether the sample size distribution could be considered as monomodal. The PDI was calculated by the software and its value relied on the comparison between the fitted and experimental autocorrelation functions. PDI values higher than 0.30 indicated a broad and potentially multimodal size distribution.

2.6. Transmission electron microscopy (TEM) The size and morphology of protein samples were observed with transmission electron microscope (TEM, Phillip TECNAI 20). An aliquot of 50 L was deposited on a copper grid and stained with 1.5 wt% phosphotungstic acid (PTA). The particle diameter was denoted with arithmetic average diameter.

3. Results and Discussion 3.1. Design and characterization of the reaction medium In order to operate at low temperature in a liquid reaction medium, sucrose was solubilized together with lipase in DMSO. DMSO has been reported as a good solvent for both sucrose and sucrose esters (Voutsas et al. 2002). Vinyl caprate was not fully soluble in DMSO and thus was dispersed under vigorous agitation using magnetic stirring forming a non-aqueous disperse system. Macroscopically a turbid mixture was observed (Fig. 1). At this stage two questions needed to be solved. First it was necessary to produce a sufficient interfacial area for reaching significant sucrose consumption. Second since enzyme deactivation had been

13

reported in polar organic solvents, we had to ensure that lipase catalysis could occur in our conditions. Regarding enzyme catalytic activity, we have previously demonstrated that pHadjustment of lipase AY significantly improved its stability in DMSO (Kaewprapan et al. 2011). Similar results had been found by other authors in ionic liquids (Dang et al. 2007). Preliminary experiments showed that with pretreated lipase AY, the initially turbid reaction medium became completely transparent after 30 min stirring. After sampling the reaction medium, TLC analysis confirmed the presence of sucrose ester. These preliminary observations confirmed that our conditions could be used for the synthesis of sucrose ester in the presence of pretreated lipase AY. The characterization of the biphasic system containing sucrose and vinyl fatty acid ester was performed by microscope observations. The reaction medium containing pretreated lipase AY was sampled and observed immediately after the addition of the enzyme (Fig. 2). We observed droplets of vinyl caprate dispersed into a continuous phase containing DMSO, sucrose and enzyme. Droplet diameters were in the range between 1 and 20 µm.

3.2. Effect of lipase pretreatment and sucrose concentration on monoester production Two important operational parameters were examined in more details: the role of lipase AY pretreatment and that of sucrose concentration in DMSO on the final amount of produced monoester.

3.2.1. Lipase pretreatment We first focused on lipase AY pretreatment. When native lipase AY was used as catalyst in the reaction medium, no sucrose ester could be detected by TLC analysis. As a conclusion, it was proved that the effect of the pretreatment of lipase AY was to improve the catalytic activity of enzyme in DMSO. The influence of the salt concentration during lipase AY

14

pretreatment was further examined. Indeed, it has been shown that phosphate buffer acted as a catalyst in the case of acylation of sucrose with vinyl laurate in DMSO (Plou et al. 1999). A control experiment was carried out in which only salt was added to reaction medium, with an equivalent quantity to that added with enzyme during a pretreatment at 20 mM phosphate buffer concentration. After 30 min the reaction medium was still turbid and TLC analysis could not evidence the presence of sucrose ester. Similar result was obtained when thermally deactivated pretreated lipase AY was used. These results demonstrated that significant sucrose consumption was detected during the first 30 min only when pretreated and catalytically active lipase AY was added into the reaction medium. Thus we could exclude any contribution of salts as chemical catalysts of esterification under the proposed conditions. A systematic study of salt concentration used for pretreatment was carried out by varying salt concentration between 2 and 100 mM (Table 1). When reducing potassium phosphate buffer concentration from 20 to 2 mM no formation of sucrose ester could be detected by TLC analysis after 1 h stirring (Fig. 3). Thus preservation of lipase AY catalytic activity in DMSO required a minimal salt concentration during enzyme pretreatment. When increasing salt concentration from 20 mM up to 40 and 60 mM, significantly higher initial rates of monoester production were obtained. Increasing further salt concentration led to the decrease of the rate. Thus the optimal salt concentration for pretreatment step was between 40 and 60 mM. Nevertheless, in order to avoid any possibility of auto-catalysis of transesterification, the concentration of salt was kept at 20 mM during pretreatment for all other experiments. Regarding the role of 18-Crown-6 ether concentration used for pretreatment, it has already been reported that the optimal molar ratio was 50 as compared to Lipase AY (Kaewprapan et al. 2011). As a consequence, we kept this optimal ratio for all other experiments. In order to get further insight about the role of the pretreatment on the state of lipase AY in DMSO, the reaction medium was observed by TEM. To that goal, the protein content of

15

native and pretreated lipase AY was increased from 2.3 wt% to 22 wt% by ultrafiltration (see Materials and Methods). When pH-adjusted lipase AY was dissolved in water (Fig. 4, left) and DMSO (Fig. 4, right), colloidal aggregates were observed with similar sizes within the range 50-100 nm. Protein samples were incubated in DMSO at 50 oC to simulate the real reaction conditions. It turned out from TEM observations that only the pretreated lipase AY maintained its aggregated form of approximately 50 nm in diameter while the native one lost its tridimensional structure under these conditions (Fig. 5). Thus, we concluded that salts and crown ethers might be involved in lipase AY aggregation which preserved catalytic activity by limiting unfolding during full dissolution in a polar solvent like DMSO. Several complexation processes resulting from crown ethers may contribute to the enzyme activation such as buffer cations present in the enzyme preparation, charged residues on the enzyme surface, and water at the surface and/or in the active site of the enzyme (Engbersen et al. 1996; Unen et al. 2001). Regarding the phosphate salts, they might be involved in the aggregation of several molecules of lipase AY by the formation of salt bridges between cations found on the surface of the colloidal aggregates. Our results demonstrated the great interest of lipase pretreatment for maintaining a significant catalytic activity in a polar solvent like DMSO. The possibility to use this solvent was also very convenient for increasing the loaded concentration of sucrose in the reaction medium.

3.2.2. Sucrose loaded concentration in DMSO HPLC analysis of final reaction medium allowed isolation of unreacted sucrose, sucrose monoesters (three isomers were evidenced among which one was largely predominant) and sucrose polyesters (not identified). After extraction and purification of the major monoester isomer, it was used as a standard for calibrating the signal of the refractometer. As a result,

16

we expressed monoester production (in mol %) as the ratio of the final amount of major monoester isomer to the initial sucrose load. For an initial load of 0.3 M sucrose in DMSO, we compared the relative amount of sucrose monoester to the overall consumption of sucrose ester (as determined by HPLC) as a function of reaction time (Fig. 6). HPLC analysis revealed that after 10 min reaction sucrose di- and triester started to form and thus slowed down the production of sucrose monoester. As a result, sucrose monoester production reached a maximum and decreased further because of the progressive conversion of monoester into di- and triester. Since monoester was our targeted product, we particularly investigated the role of reaction conditions on the time corresponding to maximal monoester production and the corresponding amount of sucrose monoester. A series of experiments was carried out in which the concentration of sucrose dissolved in DMSO was increased while keeping unchanged the sucrose:vinyl caprate molar ratio at 2.8 as well as all other operational parameters (Table 2). The initial reaction rate (amount of produced monoester per liter and min) as well as the maximal sucrose monoester production (as determined by HPLC analysis) decreased regularly when increasing sucrose concentration in DMSO. This result was attributed to the increase of vinyl caprate disperse phase which led to an increase of average droplet size and thus a decrease of interfacial area. In order to support that assumption, the average droplet size was determined by DLS for reaction media containing initial sucrose concentration higher than 0.3 M in DMSO. It was found that at 0.3 M the average droplet size after enzyme addition was of the order of 0.5 µm while for higher sucrose concentrations, the average droplet size exceeded 1 µm. Initially, after enzyme addition, the reaction medium appeared turbid because of the dispersion of vinyl caprate droplets into DMSO continuous phase. After 3 h the reaction mixture appeared as a clear and transparent liquid medium, which was consistent with the consumption of vinyl caprate by the transesterification reaction producing sucrose ester which was soluble in DMSO. Thus, we decided to scale up the

17

production of sucrose ester from 1 mL DMSO to 30 mL with a loaded sucrose concentration equal to 0.3 M. Our results demonstrated that it was possible to operate with loaded sucrose concentrations much higher (up to 0.7 M) than the current values reported in literature (0.15 M at the maximum) (Shi et al. 2011). No further optimization of agitation of reaction medium was attempted in that work. Nevertheless, this would be a relevant extension of the present results which could further increase the productivity of the process.

3.3. Scale up of transesterification reaction and purification of sucrose ester After optimization of reaction conditions for maximizing sucrose ester productivity in small volumes (1 mL DMSO in reaction media), selected reaction conditions (0.3 mol/L sucrose load in DMSO) were scaled up to reaction medium containing 30 mL DMSO. The initially turbid reaction mixture was converted into a clear and transparent liquid after 3 h reaction, which was consistent with the observations of experiments carried out with 1 mL DMSO (see above). After 3 h, the sucrose ester was isolated using several liquid-liquid extraction steps (see Materials and Methods) and its purity was checked using HPLC analysis (Fig. 7). HPLC chromatogram evidenced the presence of three different sucrose ester isomers with a largely predominant one (85 % of total peak surface area) and two other minor constituents (10 % and 5 % of the total peak surface area). The chemical structure of reaction product as well as the position of substitution was further investigated using spectroscopic techniques (1H NMR and 13C NMR) as well as mass spectrometry.

3.4. Structural characterization of sucrose ester and position of substitution Two samples were analyzed by LC-MS, one was a sample from the crude reaction medium after 3 h and the other was the purified product obtained after liquid-liquid extractions. Both

18

samples led to identical results which demonstrated that the purification steps did not modify the product distribution from enzyme catalysis. Consequently, only the LC-MS results obtained with purified products will be further commented. LC-MS coupled with ESI analysis revealed two peaks in LC chromatograms (with one largely predominant) which both led to a major signal at m/z equal to 519 in MS spectra and minor signals at m/z equal to 535 and 560 (Fig. 8). Since the theoretical molar mass of sucrose monoester was 496.55 g/mol, the signals in MS spectra could be attributed to [M+Na]+ adduct ion for the major peak at m/z = 519 and to [M+K]+ and [M+ACN+Na]+ adduct ions for the two other minor peaks at m/z = 535 and 560, respectively. These assignments were consistent with previous characterizations of sucrose monoesters by mass spectrometry (Zhu et al. 2009). Thus LC-MS experiments confirmed the formation of sucrose monoester with a major isomer and minor by-products differing by the position of substitution. No sucrose diesters could be detected. This last result indicated that in our conditions, sucrose monoesters were selectively obtained with one major product and limited amounts of other isomers. 1

H NMR analysis of reaction product was fully consistent with the formation of sucrose

monoester regarding both peak positions and areas (Table 3) (Jamroz et al. 2014). To further identify the major position of substitution,

13

C NMR was used. It had been showed that O-

acylated carbons underwent downfield shift while neighboring carbon underwent upfield shift as compared to native sugar (Kitagawa et al. 1999; Rodrigues Borges and Balaban 2007; Yoshimoto et al. 1980). In addition, the observed effects on chemical shifts were rather independent from the solvent and the nature of alkyl substituent. Analysis of

13

C NMR

spectrum of the purified product in DMSO-d6 demonstrated that the major isomer was 2-Oacylated sucrose monocaprate and that the variations of chemical shifts as compared to sucrose were in good agreement with other literature data (Table 4). This selectivity of enzyme-catalyzed transesterification was consistent with previous works using polar solvents

19

like DMSO or DMF (Shi et al. 2011). In addition, calculations of molecular electrostatic potential have shown that C2-OH was the most electropositive, which was attributed to the persistence of an intramolecular hydrogen bond (Lichtenthaler et al. 1995).

3.5. Surface activity of sucrose monoester at air-water interface The surface tension of aqueous solutions of sucrose monoester was measured as a function of surfactant concentration (Fig. 9). These measurements allowed determining the more important properties characterizing surfactant efficiency: the critical micelle concentration, CMC in mol.L-1, the surface excess  in mol.m-2 and the minimal surface tension min in mN.m-1 (Table 5). The found values were consistent with those reported in literature (Garofalakis et al. 2000; Makino et al. 1983; Molinier et al. 2005; Zhang et al. 2014). In addition, the sucrose monoester synthesized in our conditions exhibited performances similar to those of a commercial sugar surfactant.

4. Conclusions The synthesis of well-defined sucrose monocaprate using lipase catalysis was carried out in a non aqueous biphasic medium. The alkylating reactant (vinyl caprate) was dispersed mechanically into a solution of sucrose in DMSO. Thanks to a pretreatment of lipase by pH adjustment in the presence of crown ether, its catalytic activity was maintained despite the use of a polar solvent like DMSO. In addition, DMSO allowed increasing the loaded concentration of sucrose much above the previously reported syntheses. Optimization of agitation of reaction medium in order to increase interfacial area appeared promising for increasing further the productivity of the process, i.e. the final concentration of sucrose monoester. Starting from experiments in very small volumes (1 mL), the enzymatic process was scaled up to 30 mL reaction volume and a purification procedure was designed. Detailed

20

structural analysis demonstrated that, in our conditions, only monoester was recovered and that the 2-O-acylated sucrose monocaprate was the major isomer in the final product. Surfactant properties were measured at air/water interface and found close to those of commercial sucrose monoester surfactant.

Acknowledgements This research was supported by a European Marie Curie IIF scholarship (No. 301723) and by the Thailand Research Fund through the Royal Golden Jubilee (RGJ) Ph.D. Program (Grant No. PHD /1010/ 2545). Both institutions are gratefully acknowledged.

21

References Arcos, J.A., Bernabé, M., Otero, C., 1998. Quantitative Enzymatic Production of 6-OAcylglucose Esters Biotechnol. Bioeng. 57, 505-509. Ashrafuzzaman, M., Pyo, J.I., Cheong, C.S., 2014. Sucrose Derivatives Preparation using Thermomyces lanuginosus Lipase and Their Application. Bull. Korean Chem. Soc. 35, 477482. Behler, A., Biermann, M., Hill, K., Raths, H.-C., Saint-Victor, M.-E., Uphues, G., 2001. Industrial Surfactant Syntheses. Surfactant Science Series 100, 1-44. Chamouleau, F., Coulon, D., Girardin, M., Ghoul, M., 2001. Influence of water activity and water content on sugar esters lipase-catalyzed synthesis in organic media. J. Mol. Catal. B Enzym. 11, 949-954. Coulon, D., Ismail, A., Girardin, M., Rovel, B., Ghoul, M., 1996. Effect of different biochemical parameters on the enzymatic synthesis of fructose oleate. J. Biotechnol. 51, 115121. Dang, D.T., Ha, S.H., Lee, S.-M., Chang, W.-J., Koo, Y.-M., 2007. Enhanced activity and stability of ionic liquid-pretreated lipase. J. Mol. Catal. B Enzym. 45, 118-121. Daudé, D., Remaud-Siméon, M., André, I., 2012. Sucrose analogs: an attractive (bio)source for glycodiversification. Nat. Prod. Rep. 29, 945-960. Engbersen, J.F.J., Broos, J., Verboom, W., Reinhoudt, D.N., 1996. Effects of crown ethers and small amounts of co-solvent on the activity and enantioselectivity of α-chymotrypsin in organic solvents. Pure Appl. Chem. 68, 2171-2178. Garofalakis, G., Murray, B.S., Sarney, D.B., 2000. Surface Activity and Critical Aggregation Concentration of Pure Sugar Esters with Different Sugar Headgroups. J. Colloid Interface Sci. 229, 391-398. Gaylord Chemical Company, L.L.C. 2014. Dimethyl Sulfoxide (DMSO) Solubility Data. Humeau, C., Girardin, M., Rovel, B., Miclo, A., 1998. Effect of the thermodynamic water activity and the reaction medium hydrophobicity on the enzymatic synthesis of ascorbyl palmitate. J. Biotechnol. 63, 1-8. Jamroz, M.K., Paradowska, K., Zawada, K., Makarova, K., Kazmierski, S., Wawera, I., 2014. 1 H and 13C NMR-based sugar profiling with chemometric analysis and antioxidant activity of herbhoneys and honeys. J. Sci. Food Agric. 94, 246-255. Janssen, A.E.M., Vaidya, A.M., Halling, P.J., 1996. Substrate specificity and kinetics of Candida rugosa lipase in organic media. Enzyme Microb. Technol. 18, 340-346. Kaewprapan, K., Tuchinda, P., Marie, E., Durand, A., Inprakhon, P., 2007. pH-imprinted lipase catalyzed synthesis of dextran fatty acid ester. J. Mol. Catal. B Enzym. 47, 135-142. Kaewprapan, K., Wongkongkatep, J., Panbangred, W., Phinyocheep, P., Marie, E., Durand, A., Inprakhon, P., 2011. Lipase-catalyzed synthesis of hydrophobically dextran: Activity and regioselectivity of lipase from Candida rugosa. J. Biosci. Bioeng. 112, 124-129. Khan, N.R., Rathod, V.K., 2015. Enzyme catalyzed synthesis of cosmetic esters and its intensification: A review. Process Biochem. 50, 1793-1806. Kitagawa, M., Fan, H., Raku, T., Shibatani, S., Maekawa, Y., Hiraguri, Y., Kurane, R., Tokiwa, Y., 1999. Selective enzymatic preparation of vinyl sugar esters using DMSO as a denaturing co-solvent. Biotechnol. Lett. 21, 355-359. Lichtenthaler, F.W., Immel, S., Pokinskyj, P., 1995. Selective 2-0-Benzylation of Sucrose: A Facile Entry to Its 2-Deoxy- and 2-Keto-Derivatives and to Sucrosamine. Liebigs Ann.19391947. Makino, S., Ogimoto, S., Koga, S., 1983. Sucrose monoesters of fatty acids: Their properties and interaction with proteins. Agric. Biol. Chem. 47, 319-326.

22

Molinier, V., Fenet, B., Fitremann, J., Bouchu, A., Queneau, Y., 2005. PFGSE–NMR study of the self-diffusion of sucrose fatty acid monoesters in water. J. Colloid Interface Sci. 286, 360-368. Patil, D.R., Retwisch, D.G., Dordick, J.S., 1991. Enzymatic Synthesis of a SucroseContaining Linear Polyester in Nearly Anhydrous Organic Media. Biotechnol. Bioeng. 37, 639-646. Pedersen, N.R., Halling, P.J., Haarstrup Pedersen, L., Wimmer, R., Matthiesen, R., Veltman, O.R., 2002a. Efficient transesterification of sucrose catalysed by the metalloprotease thermolysin in dimethylsulfoxide. FEBS Letters 519, 181-184. Pedersen, N.R., Wimmer, R., Emmersen, J., Degn, P., Pedersen, L.H., 2002b. Effect of fatty acid chain length on initial reaction rates and regioselectivity of lipase-catalysed esterification of disaccharides. Carbohydr. Res. 337, 1179-1184. Pedersen, N.R., Wimmer, R., Matthiesen, R., Pedersen, L.H., Gessesse, A., 2003. Synthesis of sucrose laurate using a new alkaline protease. Tetrahedron: Asymmetry 14, 667-673. Plou, F.J., Cruces, M.A., Pastor, E., Ferrer, M., Bernabé, M., Ballesteros, A., 1999. Acylation of sucrose with vinyl esters using immobilized hydrolases: demonstration that chemical catalysis may interfere with enzymatic catalysis. Biotechnol. Lett. 21, 635-639. Queneau, Y., Chambert, S., Besset, C., Cheaib, R., 2008. Recent progress in the synthesis of carbohydrate-based amphiphilic materials: the examples of sucrose and isomaltulose. Carbohydr. Res. 343, 1999-2009. Ritthirham, S., Wimmer, R., Stensballe, A., Pederson, L.H., 2009. Analysis and Purification of O-decanoyl sucrose region-isomers by reverse phase high pressure chromatography with evaporative light scattering detection. J. Chromatogr. A 1216, 4963-4967. Ritthitham, S., Wimmer, R., Stensballe, A., Pedersen, L.H., 2009. Selectivity and stability of alkaline protease AL-89 in hydrophilic solvents. J. Mol. Catal. B Enzym. 56, 266-273. Rodrigues Borges, M., Balaban, R., 2007. Sucrose-Branched Polymer Synthesized by Protease from Bacillus Subtilis. Macromol. Symp. 258, 25-29. Shi, Y.-g., Li, J.-r., Chu, Y.-H., 2011. Enzyme-catalyzed regioselective synthesis of sucrosebased esters. J. Chem. Technol. Biotechnol. 86, 1457-1468. Unen, D.-J., Engbersen, F.J., Reinhoudt, D.N., 2001. Studies on the mechanism of crownether-induced activation of enzymes in non-aqueous media. J. Mol. Catal. B Enzym. 11, 877882. Voutsas, E.C., Tsavas, P., Magoulas, K., Tassios, D., 2002. Solubility Measurements of Fatty Acid Glucose and Sucrose Esters in 2-Methyl-2-butanol and Mixtures of 2-Methyl-2-butanol with Dimethyl Sulfoxide. J. Chem. Eng. Data 47, 1517-1520. Wang, X., Miao, S., Wang, P., Zhang, S., 2012. Highly efficient synthesis of sucrose monolaurate by alkaline protease Protex 6L. Bioresour. Technol. 109, 7-12. Yoshimoto, K., Itatani, Y., Tsuda, Y., 1980. 13C-Nuclear magnetic resonance (NMR) spectra of O-acylglucoses. Additivity of shift parameters and its application to structure elucidations. Chem. Pharm. Bull. 28, 2065-2076. Zhang, X., Song, F., Taxipalati, M., Wei, W., Feng, F., 2014. Comparative Study of SurfaceActive Properties and Antimicrobial Activities of Disaccharide Monoesters. PloS ONE 9, e114845. Zhao, R., Chang, Z., Jin, Q., Li, W., Dong, B., Miao, X., 2014. Heterogeneous base catalytic transesterification synthesis of sucrose ester and parallel reaction control. Int. J. Food Sci. Technol. 49, 854-860. Zhu, J., Tang, Y., Li, J., Zhang, S., 2009. Analysis of Sucrose Esters with Long Acyl Chain by Coupling of HPLC-ELSD with ESI-MS System. Chinese J. Chem. Eng. 17, 1032-1037.

23

Table 1: Effect of salt concentration used during lipase AY pretreatment on the initial rate of sucrose monoester production. The sucrose concentration in the load was 0.3 M. For other conditions see Materials and Methods.

a

Salt concentration used for

Initial rate of sucrose monoester

pretreatment of Lipase AY (mM)

production relative to that at 20 mM

0

0a

2

0a

20

1.00

40

1.84

60

1.76

80

0.89

100

0.80

No sucrose monoester could be detected by HPLC after 30 min.

24

Table 2: Variation of initial rate of consumption and maximal production of sucrose monoester as a function of loaded sucrose concentration in DMSO (for other conditions see Materials and Methods).

Sucrose loaded

Initial rate

Maximal

Time of

Final sucrose

concentration in

(mM of

sucrose

maximal

monoester

DMSO

monoester

monoester

sucrose

concentration

(M)

produced/

production

monoester

(mM)

min)

(%)a

production (min)

a

0.1

1.77

50±4

60

50

0.2

5.31

68±4

50

136

0.3

4.56

45±4

40

135

0.4

3.80

41±3

40

162

0.5

4.45

39±3

30

190

0.6

3.86

34±2

50

204

0.7

2.94

30 ±2

60

211

Experimental values  standard deviation.

25

Table 3: Chemical shifts and relative surface areas of peaks in 1H NMR spectrum of sucrose monoester. The solvent was DMSO-d6. Experimental values of peak areas are compared to calculated values on the basis of sucrose monocaprate formula.

Calculated area

Range of 

Experimental Assignment

for sucrose area

(ppm)

monocaprate 4.3 – 5.3

Sucrose moiety (OH and H on C1)

8.7

9

11.9

12

Sucrose moiety (H on C3 to C6 and 3.10 – 4.10 C1’ to C6’) 2.26 – 2.33

Methylene group close to ester

2.2

2

1.50 – 1.53

Methylene group

2.4

2

1.25

Methylene groups of hydrocarbon tail

12.5

12

0.83 – 0.88

Methyl groups of hydrocarbon tail

3.0

3

26

Table 4:

13

C NMR signals from sucrose monoester and comparison of chemical shifts to

those of sucrose. The solvent was DMSO-d6. Comparison with literature values obtained in other solvents.



Sucrose monocaprate prepared

Literature data for

in this work

sucrose monoesters

Sucrose ester

Carbon

 (ppm)

a

Sucrose



 (ppm)a







(ppm)

(ppm)b

(ppm)c

C1

88.7

92.4

‒3.7

‒2.7

‒3.3

C2

73.7

72.3

+1.4

+1.2

+0.8

C3

70.0

73.5

‒3.5

‒2.7

‒3.4

C4

69.8

70.5

‒0.7

+0.3

‒0.1

C5

72.9

73.5

‒0.6

‒0.2

‒0.5

C6

60.3

61.2

‒0.9

+0.1

‒0.5

C1’

61.1

62.8

‒1.7

‒1.0

‒1.6

C2’

104.2

104.7

‒0.5

---d

---d

C3’

75.2

77.7

‒2.5

‒2.3

‒2.6

C4’

72.5

75.0

‒2.5

‒0.5

‒1.0

C5’

82.6

83.2

‒0.6

‒0.4

‒0.7

C6’

62.2

62.4

‒0.2

+0.1

‒0.3

Spectrum of sucrose in DMSO-d6.(Jamroz et al. 2014; Patil et al. 1991; Rodrigues Borges and Balaban 2007;

Wang et al. 2012). b Recalculated from spectra of 2-O-stearoyl sucrose and sucrose in chloroform:methanol 1:1 v:v.(Ritthitham et al. 2009).

c

Recalculated from spectra of 2-O-lauroyl sucrose and sucrose in pyridine-

d5.(Pedersen et al. 2002a; Pedersen et al. 2002b). d Not given.

27

Table 5: Surfactant properties of sucrose monoester synthesized by lipase-catalyzed transesterification and commercial surfactant in aqueous solution at 25 °C. Comparison with literature values. Critical micelle concentration (CMC), surface excess concentration () and surface tension at CMC (CMC).

CMC

CMC



Sucrose monocaprate

Reference -1

2-O-ester Water 25°C

-2

-1

(mol.L )

(mol.m )

(mN.m )

0.0010

2.1 10-6

29.1

0.0003

2.2 10

-6

3.4 10

-6

This work

Commercial mixture of 6-O- and 36.0

6’-O-esters Water 25°C (Zhang et 6-O-ester Water 25°C

0.0006

33.8 al. 2014) (Makino et

Position not specified Water 25°C

0.0025

---

--al. 1983) (Garofalakis

6-O- + 6’-O-ester Water 32°C

0.0006

3.5 10-6

32.9 et al. 2000) (Molinier et

Mixture of isomers Water 25°C

0.0042

---

--al. 2005)

28

Figure captions

Figure 1: Visual appearance of reaction mixture in glass vials. (a) Clear reaction mixture containing sucrose (0.1027 g) dissolved in DMSO (1 mL) for 30 min at 50°C. (b) Turbid reaction mixture after adding vinyl caprate (188 µL) in previous sucrose solution. (c) Turbid reaction mixture after adding pretreated lipase AY (6.73 mg.mL-1).

Figure 2: Visual microscope image of non-aqueous biphasic reaction medium containing sucrose (0.308 g), vinyl caprate (0.564 mL), pretreated lipase AY (6.95 mg.mL-1), DMSO (3 mL). The image was taken after 2 h at 50 °C under agitation rate of 500 rpm.

Figure 3: Visual appearance of reaction mixture after 30 min at 50 °C. Reaction medium containing pretreated lipase AY by phosphate buffer (2 or 20 mM) coupled with 18-crown-6 ether. (a) Pretreatment with 2 mM phosphate buffer. (b) Pretreatment with 20 mM phosphate buffer. Corresponding TLC chromatograms obtained with sampled reaction medium after 1 h at 50 °C. (c) Pretreatment with 2 mM phosphate buffer. (d) Pretreatment with 20 mM phosphate buffer.

Figure 4: TEM images (29K) of pH-adjusted lipase AY co-lyophilized with crown ether after incubation in water (left) and in DMSO (right).

Figure 5: TEM images (19000x) of carbon grid, native lipase AY and pretreated lipase AY.

29

Figure 6: Sucrose conversion and sucrose monoester production (expressed in mol % of initial sucrose load) as a function of reaction time for an initial sucrose concentration equal to 0.3 M in DMSO. Data were obtained by HPLC analysis (see Materials and Methods).

Figure 7: HPLC chromatogram of purified sucrose monoester. The arrows indicate retention times of the three peaks: 8.88 min, 13.40 min, and 14.25 min, respectively. For details see Materials and Methods.

Figure 8: LC chromatogram (a) and MS spectra (b) of sucrose monoester isomers eluted at retention times 10.05 min and 11.02 min.

Figure 9: Surface tension of aqueous solution of surfactants at 25 °C. Sucrose monocaprate synthesized by lipase-catalyzed transesterification () and commercial n-decanolysucrose Calbochem® (◯).

30

Figure

Figure 1

Figure

Figure 2

Figure

Figure 3

Figure

Figure 4

Figure

Figure 5

Sucrose monoester production or sucrose conversion (%)

Figure

Figure 6

100

80

60

40

20

0 0 10 20 30

Time (min) 40 50 60

Figure

Figure 7

Figure

Figure 8

Figure

Figure 9

80

Surface tension (mN/m)

70 60 50 40 30 20 -9 10

-7

10

10

-5

-3

10

Surfactant concentration (mol/L)

-1

10