Effects of immobilization, pH and reaction time in the modulation of α-, β- or γ-cyclodextrins production by cyclodextrin glycosyltransferase: Batch and continuous process

This study reports the immobilization of a β-CGTase on glutaraldehyde pre-activated silica and its use to production of cyclodextrins in batch and continuous reactions. We were able tomodulate the cyclodextrin production (α-, β- and γ-CD) by immobilization and changing the reaction conditions.

Carbohydrate Polymers 169 (2017) 41–49 Contents lists available at ScienceDirect Carbohydrate Polymers journal homepage: www.elsevier.com/locate/carbpol Effects of immobilization, pH and reaction time in the modulation of ˛-, ˇ- or -cyclodextrins production by cyclodextrin glycosyltransferase: Batch and continuous process Jéssie da Natividade Schöffer a , Carla Roberta Matte a , Douglas Santana Charqueiro b , Eliana Weber de Menezes b , Tania Maria Haas Costa b , Edilson Valmir Benvenutti b , Rafael C Rodrigues a , Plinho Francisco Hertz a,∗,1 a Grupo de Biotecnologia, Bioprocessos e Biocatálise, Instituto de Ciência e Tecnologia de Alimentos, Universidade Federal Rio Grande Sul (UFRGS), Porto Alegre, RS, Brazil b Laboratório de Sólidos e Superfície, Instituto de Química, Universidade Federal Rio Grande Sul (UFRGS), Porto Alegre, RS, Brazil a r t i c l e i n f o Article history: Received February 2017 Received in revised form 28 March 2017 Accepted April 2017 Available online April 2017 Keywords: CGTase immobilization CDs modulated production Packed-bed reactor CDs continuous production a b s t r a c t This study reports the immobilization of a ˇ-CGTase on glutaraldehyde pre-activated silica and its use to production of cyclodextrins in batch and continuous reactions We were able to modulate the cyclodextrin production (˛-, ˇ- and -CD) by immobilization and changing the reaction conditions In batch reactions, the immobilized enzyme reached to maximum productions of 4.9 mg mL−1 of ␣-CD, 3.6 mg mL−1 of ˇ-CD and 3.5 mg mL−1 of -CD at different conditions of temperature, pH and reaction time In continuous reactor, varying the residence time and pH it was possible to produce at pH 4.0 and 141 of residence time preferentially -CD (0.75 and 3.36 mg mL−1 of ␣- and -CD, respectively), or at pH 8.0 and 4.81 ␣- and ˇ-CDs (3.44 and 3.51 mg mL−1 ) © 2017 Elsevier Ltd All rights reserved Introduction Cyclodextrins (CDs) are cyclic oligosaccharides containing mainly six (␣-CD), seven (ˇ-CD) or eight ( -CD) glucose residues, linked by ␣(1-4) glycosidic bonds Due to the conformation of their glucose residues and the links established among them, CDs have an unique spatial configuration, showing a cylindrical hollow truncated cone shape with a hydrophobic nanoscale cavity and a hydrophilic outer surface (Loftsson & Duchene, 2007; Szejtli, 1982, 1998) The most notable feature of CDs is their ability to form inclusion complexes with solid, liquid and gaseous molecules that fit into their hydrophobic cavity (Del Valle, 2004; Singh et al., 2002) The CDs architecture provides a wide range of applications in phar- ∗ Corresponding author at: Instituto de Ciência e Tecnologia de Alimentos (ICTA), Universidade Federal Rio Grande Sul (UFRGS), Av Bento Gonc¸alves, 9500, P.O Box: 15095, ZC 91501-970, Porto Alegre, RS, Brazil E-mail addresses: rafaelcrodrigues@yahoo.com.br (R.C Rodrigues), plinho@ufrgs.br (P.F Hertz) Web: http://www.ufrgs.br/bbb http://dx.doi.org/10.1016/j.carbpol.2017.04.005 0144-8617/© 2017 Elsevier Ltd All rights reserved maceuticals (Lima et al., 2016; Moussa, Hmadeh, Abiad, Dib, & Patra, 2016; Pereva, Sarafska, Bogdanova, & Spassov, 2016), food (Astray, Gonzalez-Barreiro, Mejuto, Rial-Otero, & Simal-Gandara, 2009; Astray, Mejuto, Morales, Rial-Otero, & Simal-Gandara, 2010; Li, Chen, & Li, 2017; Yuan, Du, Zhang, Jin, & Liu, 2016; Zhao & Tang, 2016), cosmetic and textile processing industries (Mihailiasa et al., 2016) As described by Singh et al (2002), CDs are multipurpose technological tools, they can stabilize and protect the encapsulated molecules from volatility and oxidation, enhance their apparent solubility, hydrophilicity and bioavailability, reduce adverse effect of pharmaceuticals and protect the substances against any undesirable reactions (Astray et al., 2009; Del Valle, 2004; Singh et al., 2002) CDs are produced from starch as a result of intramolecular transglycosylation, one of the four reactions catalyzed by the enzyme cyclodextrin glycosyltransferase (CGTase) The specificity and the yield of each CD depends on the enzyme, resulting in a mixture of linear, branched and cyclic dextrins (␣-, ˇ-, and -CD) at different concentrations Furthermore, for most of the enzymes, the main products are ␣- and ˇ-CDs, and few CGTases produce -CD as the main product (Kamaruddin, Illias, Aziz, Said, & Hassan, 2005; Li et al., 2007) Because the specificity of the molecular inclusion 42 J.d.N Schöffer et al / Carbohydrate Polymers 169 (2017) 41–49 Table Experimental designs and results of the CCD Experiments 10 11 12 13 14 15 16 17 18 Variables Results X1 (temperature, ◦ C) X2 (pH) X3 (time, h) ˛-CD (mg mL−1 ) ˇ-CD (mg mL−1 ) −1 (58.1) −1 (58.1) −1 (58.1) −1 (58.1) +1 (81.9) +1 (81.9) +1 (81.9) +1 (81.9) −1.68 (50) +1.68 (90) (70) (70) (70) (70) (70) (70) (70) (70) −1 (4.8) −1 (4.8) +1 (7.2) +1 (7.2) −1 (4.8) −1 (4.8) +1 (7.2) +1 (7.2) (6) (6) −1.68 (4) +1.68 (8) (6) (6) (6) (6) (6) (6) −1 (9.9) +1 (38.3) −1 (9.9) +1 (38.3) −1 (9.9) +1 (38.3) −1 (9.9) +1 (38.3) (24.1) (24.1) (24.1) (24.1) −1.68 (0.25) +1.68 (48) (24.1) (24.1) (24.1) (24.1) 2.29 1.07 3.28 1.42 2.70 1.04 3.16 1.25 1.87 4.93 1.36 2.81 2.45 0.72 1.27 1.37 1.20 1.29 2.24 0.39 3.63 1.18 2.11 0.32 2.88 0.67 1.30 3.09 0.42 2.92 1.34 0.14 0.87 0.92 0.78 0.91 Fig Thermogravimetric analyses of silica and its derivatives with different amounts of amino groups (solid black line) Si, (dashed line) Si-NH-0.29, (dotted line) Si-NH-0.45, (dash-dot line) Si-NH-0.65 process, as well as the different properties and possibilities of application of each CD (Lima et al., 2016; Szente et al., 2016), the target product will define the choice of the appropriate enzyme by the industry On contrary, exhaustive crystallization steps or the use of solvents will be necessary to obtain and purify a particular CD (Blackwood & Bucke, 2000; Ferrarotti et al., 2006; Li, Chen, Gu, Chen, & Wu, 2014; Li et al., 2007; Szejtli, 1998) In this sense, changes in yield of the three main CDs by varying the sources of the enzyme and substrate have been studied by several authors Currently, the focus is the use of genetic mutations in order to direct the production of specific CDs (van der Veen, Uitdehaag, Dijkstra, & Dijkhuizen, 2000; van der Veen, Uitdehaag, Penninga et al., 2000; Xie, Song, Yue, Chao, & Qian, 2014; Yamamoto et al., 2000) Since CDs are produced exclusively by enzymatic means, techniques such as enzyme immobilization are widely assessed in order to improve the enzyme stability and achieve greater control of the reaction (Graebin et al., 2016) Supports such as silica, chitosan, polyethylene films and agarose are among the most used for immobilization of this enzyme using different approaches such as adsorption, entrapment and covalent binding (Matte et al., 2012; Schöffer, Klein, Rodrigues, & Hertz, 2013; Sobral et al., 2003; Sobral, Rodrigues, de Oliveira, de Moraes, & Zanin, 2002) Immobilized CGTases generally exhibit greater resistance to changes in conformations caused by denaturing conditions, -CD (mg mL−1 ) 0.89 2.41 0.81 1.63 1.26 2.71 1.02 2.53 1.56 1.44 3.52 2.01 0.38 2.75 1.93 2.17 2.01 2.01 making them more resistant to variations of pH and temperature (Abdel-Naby, 1999; Martín, Plou, Alcalde, & Ballesteros, 2003; Tardioli, Zanin, & de Moraes, 2006) Moreover, the immobilization allows the use of the biocatalyst in continuous reactors, achieving higher productivity and lower production cost (Garcia-Galan, Berenguer-Murcia, Fernandez-Lafuente, & Rodrigues, 2011; Mateo, Palomo, Fernandez-Lorente, Guisan, & Fernandez-Lafuente, 2007; Rodrigues, Ortiz, Berenguer-Murcia, Torres, & Fernandez-Lafuente, 2013) Thus, the objective of the present study was to evaluate the effects of reaction conditions in the specific production of ␣-, ˇ- and -CD by an immobilized ˇ-CGTase The enzyme from Thermoanaer® obacter sp (Toruzyme 3.0 L ) was immobilized in mesoporous silica functionalized with 3-aminopropyltrimethoxysilane and activated with glutaraldehyde, that provide a covalent reaction by the most reactive amino group on the enzyme surface (at neutral pH value, the reaction occurs with the terminal amino group) (Barbosa et al., 2014) The immobilized preparation was applied in batch and continuous reactions varying the temperature, pH and time using a central composite design (CCD) and the response surface methodology (RSM), so it was possible to modulate the production of specific CDs according to reaction conditions Experimental section 2.1 Materials ® Thermoanaerobacter sp CGTase (Toruzyme 3.0 L) was kindly provided by Novozymes A/S (Bagsvaerd, Denmark) Tetraethylorthosilicate 98% (TEOS), 3-aminopropyltrimethoxysilane 97% (APTMS), ␣- ˇ- and -CD were obtained from Sigma-Aldrich (St Louis, USA) All other reagents used were of analytical grade 2.2 Synthesis and functionalization of mesoporous silica The mesoporous silica support for enzyme immobilization was synthesized by sol-gel method A mixture containing mL of TEOS, mL of ethanol and catalyst were stirred The catalyst consists of 36 drops of HF/HCl (6/6 mol L−1 ) mixture in mL of water solution This mixture was stored during weeks at ambient temperature for gelation Then, the formed xerogel was comminuted, washed with water and ethanol, and dried for h at 90 ◦ C, under vacuum (Caldas et al., 2017) The silica support was functionalized with APTMS providing amino groups on its surface The organofunction- J.d.N Schöffer et al / Carbohydrate Polymers 169 (2017) 41–49 Fig Cyclodextrins production by (a) free CGTase, 50 ◦ C; (b) free CGTase, 70 ◦ C; (c) immobilized CGTase, 50 ◦ C; (d) immobilized CGTase, 70 ◦ C (᭿) ␣-CD, ( ␥-CD alization was made using 0.25, 0.5 and mmol of APTMS precursor per gram of silica The reaction was performed in toluene at 80 ◦ C, in argon atmosphere, under mechanical stirring, for 24 h Afterwards, the supernatant was removed and the silica supports were washed with toluene, ethanol, water, and dried in vacuum at 80 ◦ C, for h Further, the samples of silica functionalized with amino groups were activated with glutaraldehyde to make them able to bind the enzyme via amino-terminal Then, 0.5 mL of glutaraldehyde solution (5% v/v at pH 7.0) was mixed to 10 mg of support during h and then, washed several times to remove the excess of activation agent Before and after functionalization, activation and immobilization, the materials were lyophilized and submitted to N2 adsorption-desorption and thermogravimetric analysis (TGA) for textural characterization and to ensure the binding of the organic groups The thermogravimetric analysis was performed on a Shimadzu Instrument model TGA-50H, under argon flow at 50 mL min−1 , with a heating rate of 10 ◦ C min−1 , from room temperature up to 650 ◦ C N2 isotherms were obtained at liquid nitrogen boiling point using a Tristar 3020 Kr Micromeritics equipment Samples were previously degassed at 120 ◦ C, under vacuum, for 12 h The specific 43 ) ␤-CD, ( ) surface areas were estimated by the Brunauer, Emmett and Teller (BET) multipoint method and the pore size distributions were calculated by using Barret, Joyner and Halenda (BJH) model applied to the desorption branch of isotherms (Gregg & Sing, 1982) 2.3 Enzyme immobilization For enzyme immobilization, 0.5 mL of CGTase solution (in sodium phosphate buffer 10 mmol L−1 , pH 6.0) and 10 mg of activated support (protein load of 10 mg g−1 ) were mixed overnight under gentle stirring After that, the support was separated by decantation from the solution and washed with sodium phosphate buffer (10 mmol L−1 , pH 6.0), ethylene glycol (30%) and NaCl (1 mol L−1 ) to remove non-covalently bonded proteins Enzyme concentration of immobilization solution and washing fractions were determined by enzymatic activity assay These values were used to calculate the immobilization yield and efficiency, using the equations below, according to Sheldon and van Pelt (2013) The protein concentrations were determined according to Lowry method (Lowry, Rosebrough, Farr, & Randall, 1951) Yield (%) = 100 × (immobilized activity/starting activity) 44 J.d.N Schöffer et al / Carbohydrate Polymers 169 (2017) 41–49 Table Textural and thermal analyses data Sample Specific surface area (±7 m2 g−1 ) Pore volume ±0.01 cm3 g−1 ) Pore diametera (nm) Amino groupb (mmol g−1 ) Si Si-NH-0.29 Si-NH-0.45 Si-NH-0.65 133 121 114 109 0.79 0.72 0.65 0.61 24 23 23 21 – 0.29 0.45 0.65 a b Maximum of BJH pore diameter distribution curve Aminopropyl group amount estimated by TGA Table CGTase immobilization parameters on Si-NH-G 2.7 Central composite design Sample Yield (%) Efficiency (%) U g−1 a Si-NH-0.29-G Si-NH-0.45-G Si-NH-0.65-G 96.54 98.18 98.63 3.41 3.92 5.37 6333 7405 10173 a Measured by the phenolphthalein method Efficiency (%) 100 × = (observed activity/immobilized activity) 2.4 Enzymatic activity The CGTase activity was determined by using the phenolphthalein method developed by Vikmon (1982), with some modifications This method is based on the decrease in the color of a phenolphthalein solution caused by its encapsulation by ˇCD formed during the reaction The substrate solution was soluble starch 4% (w/v) in sodium phosphate buffer 10 mmol L−1 , pH 6.0 For free CGTase activity, 1.05 mL of enzyme solution and 1.95 mL of substrate were mixed and incubated at 60 ◦ C for 15 For immobilized CGTase, 1.95 mL of substrate solution was added to 1.05 mL of sodium phosphate buffer containing 10 mg of immobilized enzyme and incubated at 60 ◦ C for 10 min, under gentle stirring At the end of the reaction time, 0.5 mL of this mixture was added to mL of phenolphthalein solution (0.04 mmol L−1 phenolphthalein dissolved in 125 mmol L−1 Na2 CO3 ) The decrease in color intensity was measured in a spectrophotometer at 550 nm, and the ˇ-CD concentration was determined by applying the absorbance value at a standard curve with the range of 40–400 ␮g mL−1 of a commercial ˇ-CD One unit of CGTase activity (U) was defined as the amount of enzyme that produces ␮g ˇ-CD min−1 under the reaction conditions 2.5 Cyclodextrins quantification Quantifications of ␣-, ˇ- and -CDs were determined using a HPLC system (Shimadzu, Tokyo, Japan) equipped with a refractive index detector (Shimadzu, RID-10A) and Aminex HPX-42A column (Bio-Rad) Distilled water was used as mobile phase at 70 ◦ C and a flow rate of 0.5 mL min−1 The samples and mobile phase were filtered through Millipore membranes of 0.22 ␮m 2.6 Production of cyclodextrins as a function of time Preliminary tests were carried out for cyclodextrins production under different conditions of pH, temperature and reaction time using free (95.82 U) and immobilized enzyme (101.73 U) The temperatures evaluated were 50 and 70 ◦ C The substrate was soluble starch 4% (w/v) at pH 4.0 and 7.0 (sodium phosphate buffer, 10 mmol L−1 ) Samples were taken at 1, 24 and 48 h, filtered and analyzed by HPLC for CDs quantification A 23 central composite design (CCD) with three variables was performed to evaluate the combined effects of temperature, pH and reaction time in the production of ␣-, ˇ- and -CD by immobilized CGTase (10 mg) The temperature varied from 50 to 90 ◦ C, the pH from 4.0 to 8.0, while the reaction time was from 0.25 to 48 h The factorial design consisted of eight factorial points, six axial points (two axial points on the axis of design variable), and four replications at the central point, leading to 18 experiments as shown in Table In each case, the CDs concentrations were determined by HPLC The experimental design and analyses of results were carried out using Statistica 13.0 (Statsoft, USA) Statistical analysis of the model was performed as analysis of variance (ANOVA) The variance explained by the model was given by the multiple determination coefficients, R2 For each variable, the quadratic models were represented as contour plots (2D) 2.8 Continuous production of CD in packed-bed reactor The production of CDs was tested using a continuous packedbed reactor (Ø = cm; height = 12 cm) with g of silica with immobilized enzyme The reaction temperature was maintained at 70 ◦ C by circulating water inside the jacket around the reactor Substrate was 4% of soluble starch (w/v) and the pH was 4.0 (sodium acetate buffer, 10 mmol L−1 ) or pH 8.0 (sodium phosphate buffer, 10 mmol L−1 ) The substrate was fed through the column using a peristaltic pump with variable flow rate, from 0.01 to 0.90 mL min−1 , representing residence times from 1.96 to 141.12 Samples were collected after reaching the steady state, filtered at 0.22 ␮m Millipore membranes and analyzed by HPLC for ␣-, ˇ- and -CD quantification Results and discussion 3.1 Synthesis and characterization of support Silica (hereafter assigned as Si) with controlled pore size was synthesized using tetraethylorthosilicate (TEOS), which provides highly reactive silanol groups on its surface The material was functionalized by silanization with different concentrations of APTMS (3-aminopropyltrimethoxysilane), incorporating amino groups to the support (Si-NH) These amino groups were activated with glutaraldehyde (Si-NH-G), enabling the covalent bond with the enzyme Thermogravimetric analyses (TGA) were performed in order to check the amount of organics incorporated in each step The weigh losses can be seen in the thermograms presented in Fig The weight losses between 150 and 650 ◦ C represent the decomposition of APTMS (Pavan, Gobbi, Costa, & Benvenutti, 2002), and were used to estimate the organic quantities According to Table 2, the silica was coated with increasing amounts of amino groups (0.29, 0.45 and 0.65 mmol g−1 of support), providing different activation degrees for CGTase immobilization The resulting modified silica J.d.N Schöffer et al / Carbohydrate Polymers 169 (2017) 41–49 45 Table Statistical analysis of the CCD for CDs production Variable ˛-CD ˇ-CD -CD Effect Standard error p-Value Effect Standard error p-Value Effect Standard error p-Value Mean 1.30a 0.03