hydrophilic properties on lipase immobilized activity and stability

Journal of Bioscience and Bioengineering VOL. 113 No. 2, 166 – 172, 2012 www.elsevier.com/locate/jbiosc

Effect of membranes with various hydrophobic/hydrophilic properties on lipase immobilized activity and stability Guan-Jie Chen, 1 Chia-Hung Kuo, 2 Chih-I Chen, 3 Chung-Cheng Yu, 1 Chwen-Jen Shieh, 2 and Yung-Chuan Liu 1,⁎ Department of Chemical Engineering, National Chung Hsing University, 250 Kuo-Kuang Road, Taichung, 402, Taiwan, 1 Biotechnology Center, National Chung Hsing University, 250 Kuo-kuang Road, Taichung, 402, Taiwan, 2 and Department of Chemical Engineering, Hsiuping Institute of Technology, 11 Gongye Road, Dali District, Taichung, 412, Taiwan 3 Received 22 July 2011; accepted 30 September 2011 Available online 8 November 2011

In this study, three membranes: regenerated cellulose (RC), glass fiber (GF) and polyvinylidene fluoride (PVDF), were grafted with 1,4-diaminobutane (DA) and activated with glutaraldehyde (GA) for lipase covalent immobilization. The efficiencies of lipases immobilized on these membranes with different hydrophobic/hydrophilic properties were compared. The lipase immobilized on hydrophobic PVDF-DA-GA membrane exhibited more than an 11-fold increase in activity compared to its immobilization on a hydrophilic RC-DA-GA membrane. The relationship between surface hydrophobicity and immobilized efficiencies was investigated using hydrophobic/hydrophilic GF membranes which were prepared by grafting a different ratio of n-butylamine/1,4-diaminobutane (BA/DA). The immobilized lipase activity on the GF membrane increased with the increased BA/DA ratio. This means that lipase activity was exhibited more on the hydrophobic surface. Moreover, the modified PVDF-DA membrane was grafted with GA, epichlorohydrin (EPI) and cyanuric chloride (CC), respectively. The lipase immobilized on the PVDF-DA-EPI membrane displayed the highest specific activity compared to other membranes. This immobilized lipase exhibited more significant stability on pH, thermal, reuse, and storage than did the free enzyme. The results exhibited that the EPI modified PVDF is a promising support for lipase immobilization. © 2011, The Society for Biotechnology, Japan. All rights reserved. [Key words: Lipase; Covalent immobilization; Polyvinylidene fluoride; Hydrophobicity; Surface modification]

Lipases (triacylglycerol ester hydrolase, EC 3.1.1.3) are versatile biocatalysts in biological systems capable of catalyzing the hydrolysis of triacylglycerol to glycerol and fatty acids (1) and the reverse synthesis reaction to esters (2). To date, lipases have attracted considerable research interest because of their wide applications, including detergent formulation, oils/fats degradation, pharmaceuticals synthesis, biodiesel, and cosmetics production (3). For industrial applications, the immobilization process for lipase always requires improvement of its lifetime and stabilities, reduction of its operation costs and simplification of the purification process (4). A variety of support materials have been used for lipase covalent immobilization. These include: chitosan beads (5), poly(acrylonitrile-comaleic acid) (6), polyacrylonitrile (7), poly(acrylonitrile-co-2-hydroxyethyl methacrylate) (8), chitosan/polyvinyl alcohol (9), and poly(acylonitrile-co-acrylic acid) (10). These supporting materials change

⁎ Corresponding author. Tel.: + 886 4 22853769; fax: + 886 4 22854734. E-mail address: [email protected] (Y.-C. Liu). Abbreviations: Al/Am, alkyl/amine tail ratio; BA, n-butylamine; CC, cyanuric chloride; DA, 1,4-diaminobutane; EPI, epichlorohydrin; GA, glutaraldehyde; GF, glass fiber; H/H, hydrophobic/hydrophilic; I/F, immobilization/free; PVDF, polyvinylidene fluoride; RC, regenerated cellulose.

the microenvironments and enhance the enzyme activity (11-13). For example, Shakeri et al. (14) reported that the specific activity of transesterification reaction by Rhizopus oryzae lipase immobilized on C18- silica-based mesocellular foam was 73 times higher than that of the free enzyme. Palla et al. (15) observed a hyperactivation of lipase from Rhizomucor miehei (about 3-fold) after immobilization on the aliphatic chain of modified chitosan. The membrane-type operation has many advantages, such as no intraparticle diffusion, low-pressure drop, easy operation and maintenance, simpler scale-up, etc., over the traditional packed bed reactors (1618). Generally, PVDF, RC and GF membranes are commonly employed in microfiltration and ultrafiltration due to their excellent chemical resistance, well-controlled porosity and good thermal property. These membranes have been used as support for various enzymes’ immobilization (19-21). However, few reports have been concerned with lipase immobilization on these membrane matrices (12,22,23). Therefore, it was necessary to systematically investigate lipase immobilization on membranes with various hydrophobic/hydrophilic (H/H) properties. In this study, three different membranes: RC, GF and PVDF, with different H/H properties were employed for lipase covalent immobilization. The effect of H/H surfaces of membranes on the immobilized lipase activity was tested. The immobilized efficiency on GF membranes

1389-1723/$ - see front matter © 2011, The Society for Biotechnology, Japan. All rights reserved. doi:10.1016/j.jbiosc.2011.09.023

VOL. 113, 2012

MEMBRANES EFFECT ON LIPASE COVALENT IMMOBILIZATION

with a different ratio of hydrophobic (alkyl) to hydrophilic (amine) tail were compared. The effect of lipase immobilization by using different coupling reagents for PVDF membrane modification was also investigated. Moreover, the membrane with the highest immobilization efficiency was selected to study its pH, thermal, reaction, and storage stabilities. MATERIALS AND METHODS Materials The regenerated cellulose (RC) membrane (d: 47 mm, pore size: 0.45 μm, thickness: 160 μm) was purchased from Sartorius (Germany). The glass fiber (GF) membrane (d: 47 mm, pore size: 0.45 μm, thickness: 220 μm) was purchased from

Whatman (Japan). The polyvinylidene fluoride (PVDF) membrane (d: 47 mm, pore size: 0.45 μm, thickness: 140 μm) was purchased from Pall (Mexico). Lipase from C. rugosa (Amano AY-30, protein content 8.3% and 30,000 U/g) was purchased from Amano International Enzyme Co. (Nagoya, Japan). p-Nitrophenyl palmitate (p-NPP) was purchased from Sigma-Aldrich Chemical Corporation (MO, USA). The protein assay dye was purchased from Bio-Rad Laboratories (CA, USA). n-Butylamine (BA) was obtained from TEDIA (OH, USA). 1,4-Diaminobutane (DA), glutaraldehyde (GA), epichlorohydrin (EPI) and cyanuric chloride (CC) were obtained from Acros Organics, USA (NJ, USA). All other chemicals were of analytical reagent grade. Surface activation of regenerated cellulose membrane The reaction scheme for coupling EPI, DA and GA onto RC membrane is shown in Fig. 1A. In brief, one piece of RC membrane was treated with EPI solution (5 ml of EPI and 10 ml of 1.5 M NaOH) at 25°C for 12 h (24). The membrane was subsequently rinsed twice with DI water and then placed in 10 ml of 1 M DA solution (with 1 M carbonate buffer, pH 11.0) at 25°C for

B

A RC Membrane

GF Membrane O

O

OH

NH2 EPI NaOH

OH

1,4-DA

EPI

H2O2

Na2CO3

O

OH

NaOH

H2SO4

GA

NH2

CH=NH Lipase

Lipase CH=O

O

OH

NH2

CH=O

CH=O (from 1,4-DA)

*

NH2

1,4-DA : n-BA

CH=NH

GA

CH=O

Na2CO3 CH3

C

CH3 (from n-BA)

PVDF Membrane

CH3

CH3

NH2 1,4-DA Na2CO3

CH=NH NH2

Lipase

* CH=NH Lipase

CH=O GA

CH=NH Lipase

Lipase

CH3

*

CH=O

CH=NH

CH3

Lipase

Lipase

CC

*

O CH2-NH EPI

Lipase

167

Lipase

*

O CH2-NH

FIG. 1. Schematic graphs of lipase immobilized on (A) RC; (B) GF; and (C) PVDF membranes.

168

CHEN ET AL.

another 12 h. After being rinsed with DI water, the membrane was activated with 10 ml of 0.1% (v/v) GA solution at 25°C for 2 h. Surface activation of glass fiber membrane GF membranes with different hydrophobicity were prepared by being grafted with a different ratio of BA/DA. Fig. 1B shows the reaction scheme for coupling EPI, spacer arm with a different ratio of BA/DA, and GA onto the GF membrane. In brief, one piece of GF membrane was treated with 10 ml of H2O2: H2SO4 =3:7 (v/v) solution at 95°C for 1 h to generate additional silanol groups. The activated GF membrane was then treated with EPI solution (5 ml of EPI and 10 ml of 1.5 M NaOH) at 25°C for 12 h (25). The modified membrane was washed with DI water and then immersed in 10 ml of 1 M different BA/DA ratios (2/1, 1/1, 1/2, 1/0) solution (with 1 M carbonate buffer, pH 11.0) for 12 h. In all cases, the total amounts of BA and DA summed up to 10 mmol. After being rinsed with DI water, the membrane was activated with 10 ml of 0.1% (v/v) GA solution at 25°C for 2 h. Surface activation of polyvinylidene fluoride membrane First, one piece of PVDF membrane was rinsed with 95% ethyl alcohol and subsequently washed twice with DI water. Then the membrane was immersed in 10 ml of 1 M DA solution (with 1 M carbonate buffer, pH 11.0) at 25°C for 12 h to graft the spacer arms onto the membrane (16). Afterwards, the membranes were activated by the interaction with GA (10 ml of 0.1% (v/v) GA solution for 2 h at 25°C), EPI (5 ml of EPI solution and 10 ml of 1.5 M sodium hydroxide for 12 h at 25°C), and CC (10 ml of cyanuric chloride solution for 5 h at 65°C), respectively to couple the functional groups as shown in Fig. 1C. Lipase immobilized onto the pre-activated membranes One piece of the preactivated membrane described in the above-mentioned sections was washed with distilled water and immersed in 10 ml of lipase solution (5 mg/ml in 50 mM phosphate buffer pH 7.0) at 25°C for 24 h. Then the membranes were washed with DI water to remove the unbound enzymes and stored at 4°C until used. Assays Lipase activity was measured according to the method described by Lotrakul and Dharmsthiti (26). 0.1 ml of free lipase or one piece of lipase immobilized membrane were added to a mixture of 2 ml of 0.5% (w/v) p-NPP solution and 2 ml of 0.05 M PBS (pH 8.0), followed by incubating for 5 min at 37°C. The reaction was terminated by adding 4 ml of 0.5 N Na2CO3 followed by centrifuging for 10 min (10,000 rpm). The supernatant (0.5 ml) was diluted 10-folds with distilled water, and the absorbance at 410 nm was measured with a UV/VIS spectrophotometer (Beckman DU-530, Fullerton, CA, USA). A molar extinction coefficient for p-nitrophenol of 15,000 M-1 cm-1 was used. One unit (U) of enzyme activity was defined as the amount of enzyme required to produce 1 mmol p-nitrophenol per min at 37°C, pH 8.0. Protein concentration was estimated by the Bradford protein assay method using bovine serum albumin as a standard (27). The initial and final protein concentrations in the lipase solutions were measured to calculate the amount of protein immobilized onto the membrane. Contact angles were measured using the sessile drop method at the ambient temperature by using a digital optical contact angle meter (G10, Oldinburgh, Taiwan). Three measurements were done on each membrane. Native and modified membranes were dried with nitrogen at room temperature before measurements (28). Determination of kinetics parameters Km and Vmax The maximum reaction rate of the enzymatic reaction (Vmax) and the Michaelis–Menten constant (Km) were determined according to the calculation from Lineweaver–Burk plot, a plot of 1/V against 1/S for systems obeying the Michaelis–Menten equation, where V is the reaction rate and S is the substrate concentration (29). pH and thermal stability tests The pH stability tests for the free and immobilized lipases were carried out by immersing 1 ml of free lipase or one piece of the immobilized membrane in phosphate buffer (50 mM) in the pH range of 5–9 for 2 h at 25°C and then the activities were measured. The relative activity was calculated as the ratio of the lipase activity to that of the maximal lipase activity. The thermal stability tests of the free and immobilized lipases were conducted by immersing 1 ml of free lipase or one piece of the immobilized membrane in phosphate buffer (50 mM, pH 7) at different temperatures ranging from 20°C to 70°C for 2 h and then the activities were determined. The relative activity was calculated as the ratio of the lipase activity to that of the maximal lipase activity. Reusability and storage stability tests To test the reusability of the immobilized lipase, one piece of the prepared lipase membrane was immersed in a mixture of 2 ml of 0.5% (w/v) p-NPP solution and 2 ml of 0.05 M PBS (pH 8.0) in a flask. The reaction condition was the same as that for the activity analysis. After that, the membrane was washed three times with DI water. The relative activity was calculated as the ratio of the residual lipase activity after the test to that of the original lipase activity. The storage stability was studied by preserving free lipase (5 mg/ml, 1 ml per vial) or lipase immobilized membranes in phosphate buffer (50 mM, pH 7.0) at 4°C. In different time intervals, 1 vial of free lipase and one piece of the immobilized membrane were taken and their lipase activities were determined. The relative activity was calculated as the ratio of the residual activity of enzyme after storage to that of the original lipase activity.

RESULTS AND DISCUSSION Effect of membrane hydrophobic/hydrophilic on immobilized lipase activity Three different membranes: RC, GF and PVDF, with various H/H properties were activated by GA and then employed for lipase immobilization, respectively. The results of the contact angle,

J. BIOSCI. BIOENG., TABLE 1. Effect of lipase immobilized on the membranes with various hydrophobic levels. Matrix

RC-DA-GAb GF-BA/DA-GAc PVDF-DA-GAd

Contact anglea (o) 73.51 114.97

Protein loaded (mg/g membrane)

Activity (U/g membrane)

Specific activity (U/mg protein)

1.35 1.59 1.06

2.97 16.38 34.48

2.2 10.3 32.5

a

The contact angles were measured after surface modification. The RC membrane was pre-activated by GA using 1,4-DA as the spacer. c The GF membrane was pre-activated by GA using mixture BA/DA (1/2) as the spacer. d The PVDF membrane was pre-activated by GA using 1,4-DA as the spacer. b

protein loading amount (mg/g membrane), lipase activity (U/g membrane) and specific activity (U/mg protein) for the three membranes are shown in Table 1. The pre-activated PVDF membrane with a contact angle value of 115.0° exhibited a hydrophobic surface; the pre-activated GF membrane (BA/DA = 1/2) with a contact angle value of 73.5° displayed a less hydrophobic property (b90°), and the pre-activated RC membrane showed a hydrophilic property with an undetectable contact angle. The protein bound onto the three activated RC, GF and PVDF membranes were 1.35, 1.59 and 1.06 mg/g membrane, respectively. Among them, the lipase exhibited the highest activity (34.48 U/g membrane) and specific activity (32.53 U/mg protein) on the activated PVDF membrane, whereas the lowest activity (2.97 U/g membrane) and specific activity (2.2 U/mg protein) appeared on the RC membrane. It was noted that the lipase coupled onto the hydrophilic RC displayed the lowest lipase activity. When lipase was coupled onto the GA membrane, a medium lipase activity was expressed though the largest amount of protein was bound. In contrast, when lipase was immobilized onto the hydrophobic PVDF membrane, it exhibited the highest lipase activity. This result indicates that lipase exhibits a higher activity when the membrane has a higher hydrophobicity. Effect of alkyl/amine of the glass fiber membrane on immobilized lipase activity To further test the hydrophobicity of membrane for lipase immobilization, different BA and DA ratios were grafted onto the GF membrane to give different alkyl/amine (Al/Am) tails or H/H properties for the prepared GF membrane, as shown in Fig. 1B. The resultant GF-BA/DA membranes were then coupled with GA and used for lipase immobilization. The comparisons of the contact angle, protein-binding amount (mg/g membrane), lipase activity (U/g membrane) and specific activity (U/mg protein) for the immobilized lipase membranes are shown in Table. 2. It can be seen that the contact angle, specific activity and reusability were enhanced with the increased amount of hydrophobic (alkyl) tail. The native GF membrane had a contact angle of 85.63°. The most hydrophobic GF membrane, i.e., BA/DA (2/1) with a contact angle of 96.9°, showed the highest specific activity of 19.3 U/mg and the highest residual activity of 62% after reuse for ten batches. In contrast, the most hydrophilic GF membrane, i.e., BA/DA (0/1) with an undetectable contact angle, had the lowest specific activity of 6.9 U/mg and 29% activity retained after 10-time reuse. The results demonstrated the immobilized lipase activity and stability increased as the surface hydrophobicity TABLE 2. Effect of the Al/Am ratio on lipase immobilized on the GF membrane. (Al/Am) a

Contact angleb (o)

2/1 1/1 1/2 0/1

96.91 90.01 73.51

a

Specific activity (U/mg protein)

Residual activity after 10 batches (%)

19.3 ± 2.89 15.2 ± 2.62 10.3 ± 2.11 6.91 ± 1.32

62 49 41 29

Al/Am is the molar ratio of alkyl tail to amine tail, where alkyl tail is the CH3 group of nBA, and amine tail is the NH2 group of 1,4-DA. b The native contact angle for GF membrane is 85.63o.

VOL. 113, 2012

MEMBRANES EFFECT ON LIPASE COVALENT IMMOBILIZATION

169

TABLE 3. Effect of activation agents on lipase immobilized on the PVDF membrane.

Free lipase Native PVDF PVDF-DA-GA PVDF-DA-EPI PVDF-DA-CC a

Contact angle (o)

Protein attached (mg/g membrane)

Activity (U/g membrane)

119.38 114.97 116.18 122.97

0.91 1.06 1.01 1.22

15.02 34.48 37.87 32.73

Specific activity (U/mg protein)

Km (mM)

Vmax (U/mg)

46.46 16.51 32.53 37.49 29.49

1.98 3.72 3.64 3.59 3.76

59.4 33.4 49.8 57.8 45.6

Activity retention (%)a 100 56.2 83.8 97.1 76.8

Activity retention = the maximum reaction rates of the immobilized enzyme over that of the free counterpart, η = νimmobilized/νfree.

increased. It was assumed that the increased activity of lipase immobilized on the GF membrane might be attributed to the alkyl tail coupled to the modified GF membrane. The hydrophobic alkyl tail might orient towards lipase surface and reorganize as the packed hydrophobic domains around lipase's active site, which would further make lipase to form a more stable and reactive conformation. Effect of coupling reagents for lipase immobilization The modified PVDF-DA membranes were further activated with three coupling reagents, i.e., GA, EPI and CC respectively, and the resultant membranes were used for lipase immobilization, as seen in Fig. 1C. The results of the contact angle, protein-binding amount (mg/g membrane), lipase activity (U/g membrane) and specific activity (U/mg protein) are shown in Table 3. The contact angles of PVDF-DA membranes activated under the three reagents were in the range of 114° to 122°, displaying a hydrophobic property after modification. Lipase directly immobilized on the native PVDF membrane showed a much lower activity (15.02 U/g membrane) and specific activity (16.51 U/mg protein) as compared to that of the activated PVDF membranes (all higher than 32 U/g membrane and 29 U/mg protein). It is likely that when using the coupling reagents for lipase immobilization, they would bind with the amine, hydroxyl or sulfhydryl groups of the amino acids on lipase and form a flexible and longer spacer arms between the enzyme and the support. This might further result in a reduction in enzyme denaturation and enhanced substrate accessibility to the lipase's active site. Among the modifications, PVDF-DA-EPI membrane, showing the highest activity of 37.87 U/g membrane and specific activity of 37.49 U/mg protein, was chosen for the following experiments. FTIR characterization of lipase immobilized polyvinylidene fluoride membrane To further prove the lipase was accurately immobilized onto the PVDF membrane, the infrared spectra of native, aminated, and lipase bounded PVDF membranes were analyzed using

a FTIR spectrophotometer (Paragon 500, Perkin Elmer) and the results are shown in Fig. 2. The bending of C–H and stretching of C–C bonds at 1454 and 1371 cm-1 (30) and vibration of CF2 = CH2 bonds near 1735 cm-1 (31) are characteristic vibrations of PVDF membrane (Fig. 2A). After the native PVDF membrane being aminated, the peak near 1735 cm-1 became less sharp than that native PVDF membrane and the adsorption band at 1602 cm-1 representing N–H bending is evident to 1,4-diaminobutane (1,4-DA) bonded to the polymer backbone (28) (Fig. 2B). After lipase immobilization, a peak at 1647 cm− 1 in the spectra presumably represented the imine (N = C) Schiff-base produced by the amine-glutaraldehyde reaction (32) (Fig. 2C). All these results demonstrate the surface modification and lipase coupled onto the PVDF membrane. Kinetics study for free and immobilized lipases To analyze the effect of lipase immobilized on PVDF membrane, the kinetics constants (Km and Vmax) of free and immobilized lipases (on the modified PVDF) were determined according to the procedure in Section “Determination of kinetics parameters Km and Vmax”. The results are shown in Table 3 (with all the correlation coefficients above 0.99). The Km value was found to be 1.98 mM for free lipase. For lipase immobilized on various modified PVDF membranes, the Km values were around 3.7 mM, which was approximately 1.86 times higher than the value of the free enzyme. This meant that the affinity between immobilized lipase and the substrate is weaker than that of free lipase. In addition, the Vmax value for the free lipase was 59.4 U/mg. The Vmax value for immobilized lipase on native PVDF membrane was 33.4 U/mg. In contrast, all the modified PVDF membranes with the space arms yielded an average Vmax value of 51.07 ± 5.06 U/mg, which is much higher than that on the native PVDF membrane. Several reasons can be explained for the variations of the Vmax values of the lipase enzyme after immobilization (8,33,34). In general,

Residual activity (%)

100

80

60

40

20 Immobilized lipase Free lipase

0 5

6

7

8

9

pH

FIG. 2. FTIR spectra of (A) native, (B) aminated, and (C) lipase immobilized PVDF membranes.

FIG. 3. pH stability of lipases, where (rectangles) represents free lipase and (circles) represents lipase immobilized on the PVDF-DA-EPI membrane. Ranges denote the standard deviations of three tests.

170

CHEN ET AL.

J. BIOSCI. BIOENG.,

Residual activity (%)

100

80

60

40

20 Immobilized lipase Free lipase 0 20

30

40

50

60

70

Temperature (oC) FIG. 4. Thermal stability of lipases, where (rectangles) represents free lipase and (circles) represents lipase immobilized on the PVDF-DA-EPI membrane. Ranges denote the standard deviations of three tests.

120

Relative activity (%)

100

80

60

40

20

PVDF-DA-EPI Native PVDF

0 0

1

2

3

4

5

6

7

8

9

10

11

Reuse time (Times) FIG. 5. Reuse tests of lipase immobilized on the native PVDF (circles) and PVDF-DA-EPI (rectangles) membranes respectively. Ranges denote the standard deviations of three tests.

the immobilization process does not control the orientation of the immobilized enzyme bound on the support. The improper fixation might change the lipase conformation and/or hinder the active site of the immobilized lipase molecules. From the study involving the GF membrane, it was found that the greater the hydrophobicity of the membrane, the more lipase activity was retained on the immobilized membrane. Lipases are known to be a more hydrophobic enzyme than other proteins (35). Talukder et al. (36) have been reported that the activation of R. oryzae lipase by pretreatment with isopropanol (polar solvent) was accomplished by a conformational change leading to more hydrophobic open form, increasing the access of the substrate to the lipase active site. Therefore, it is reasonable to preserve more lipase activity on the hydrophobic membrane such as PVDF membrane than on the other membranes. In addition, when practicing the direct lipase immobilization on the native PVDF membrane, the enzyme might be attached more closely to the membrane surface and result in the hindrance for substrate to approach the enzyme. However, as seen in Table 3, it is noted that all the lipase immobilized membranes exhibit almost the same Km values, which indicates the substrate accessibility is not the main issue for the increase of lipase activity on the PVDF membrane with spacer arms. It is found that space arms would bring about a dramatic increase on Vmax as compared to that of the native PVDF membrane. Therefore, it is inferred that spacer arms on the PVDF membrane might contribute to align the enzyme conformation and make it more effective in the lipase hydrolysis reaction. pH and thermal stabilities To compare the pH stabilities of free and immobilized lipases, the individual enzyme was immersed in 50 mM phosphate buffer (pH ranging from 5 to 9) for 2 h at 25°C. The lipase activity was determined, and the results are shown in Fig. 3. The immobilized lipase displays a stable pH range between 4 and 7, whereas the free lipase is only stable around pH 6. This indicates that immobilization significantly improves the stability of the lipase for a larger pH range operation. Thermal stability was investigated by incubating the free lipase and immobilized lipase at different temperatures ranging from 20°C to 70°C for 2 h. The results are shown in Fig. 4. For the free lipase, the temperature stable range is up to 40°C, whereas, for the immobilized lipase, no activity loss was observed for temperatures up to 50°C. At 60°C, the immobilized lipase retained 88.7% of the activity, whereas the residual activity of the free enzyme was rapidly decreased to 33%. These results indicate that the thermal stability of immobilized lipase is much better than that of the free one. Reusability and storage stability The reusability of the immobilized enzyme is an important index for industrial applications. The operational stability of the PVDF-DA-EPI and of the native PVDF

TABLE 4. Comparison of lipase immobilization on different matrices. Matrices Chitosan beads PANCMA HFM b PAN NFM c PANCHEMA EFM d Chitosan/poly(vinyl alcohol) NFM PANCAA EFM e RC membrane f GF membrane g PVDF membrane h a b c d e f g h

Km ratio (Imm/Free)

Vmax ratio (Imm/Free)

Residual activity a (%)

Storage stability (%/days)

Reference

10.8 (18.0/1.67) 3.1 (1.36/0.45) 1.2 (0.55/0.46)

103.9% (96.1/92.5) 34.7% (16.1/46.4) 78.6% (31.2/39.7)

67/7

2.2 (0.97/0.44) 2.1 (0.98/0.48)

51.8% (23.1/44.6) 65.5% (16.5/25.2)

1.8 (3.59/1.98)

97.1% (57.8/59.4)

74 62 70 30 46 62 16 41 97

5 6 7 8 9 10 This study This study This study

The percentage shows the immobilized lipase activity to the original activity after 10 repeated uses. PANCMA HFM, poly(acrylonitrile-co-maleic acid) hollow fiber membrane. PAN NFM, polyacrylonitrile nanofibrous membrane. PANCHEMA EFM, poly(acrylonitrile-co-2-hydroxyethyl methacrylate) electrospun fibrous membrane. PANCAA EFM, poly(acrylonitrile-co-acrylic acid) electrospun fibrous membrane. The RC membrane was pre-activated by GA using DA as a spacer. The GF membrane was pre-activated by GA with 66% DA and 33% BA. The PVDF membrane was pre-activated by EPI using DA as a spacer.

95/20 80/30 56/30 70/30

95/30

VOL. 113, 2012

MEMBRANES EFFECT ON LIPASE COVALENT IMMOBILIZATION

EPI-modified PVDF membrane exhibited a higher pH and thermal stability. It also displayed the remarkable reuse and storage stabilities. From the economic viewpoint, this approach is considered a potential alternative for lipase immobilization in industrial application.

100 90 80

Relative activity (%)

171

70

ACKNOWLEDGMENTS

60

This work was supported by research funding grants provided by the National Science Council of Taiwan, ROC (Grants No. 96-2622-E164 -005-CC3) and was also supported in part by the Ministry of Education, Taiwan, R.O.C. under the ATU plan.

50 40 30 20

References

Free lipase Immobilization lipase

10 0 0

5

10

15

20

25

30

Storage time (Days) FIG. 6. Storage stability of lipases, where (rectangles) represents free lipase and (circles) represents lipase immobilized on the PVDF-DA-EPI membrane. Ranges denote the standard deviations of three tests.

immobilized lipases was evaluated in a batch process. The results are shown in Fig. 5. After 10 repeated uses, the native and the modified PVDF membranes retained 70% and 97% of lipase activities, respectively. The reuse ability of lipase immobilized on different membranes is shown in Table 4 for comparison. The lipases immobilized on the modified RC or GF membranes retained only 16% and 41% of the original activity, respectively. In contrast, as compared with the other immobilized approaches, the modified PVDF-DA-EPI membrane showed the highest residual activity after reuse. In order to test the storage stability, the free and immobilized lipases were stored in a phosphate buffer (50 mM, pH 6.0) at 4°C for 30 days. The activities were measured, and the results are shown in Fig. 6. The free enzyme lost 60% of its original activity after 30 days, whereas immobilized lipase retained almost all of its original activity in the same period. The immobilized lipase showed a significantly improved stability in storage over that of the free enzyme. Comparison of different immobilization methods Table 4 lists the kinetics performance of immobilized C. rugosa lipase on various supports when using p-NPP as the substrate in the hydrolysis reaction. The values of Km (Immobilization/Free, I/F) ratio and Vmax (I/F) ratio of the immobilized lipases were collected. In general, the Km values would increase, and the Vmax values would decrease after enzyme immobilization due to the lower accessibility of the substrate to approach the active site of the immobilized enzyme. In Table 4, it can be seen that Km (I/F) ratio is the lowest (1.8) for PVDF among all the matrices except that for PAN NFM. However, the PVDF membrane showed a higher retained lipase activity (97.1%) than did PAN NFM (78.6%) in the repeated use tests. In addition, as compared with other supports, the PVDF membrane gave the highest Vmax (I/F) ratio value of 97.1% among all the matrices except that for chitosan beads. However, the Km ratio for chitosan beads increased up to 10.8 times after immobilization. In addition, it can be seen that the PVDF membrane gave the best reusability (97% residual activity after 10 batches) and storage stability (95% residual activity after 30 days) than those of the other matrices. The EPI-modified PVDF membrane did provide a simple and efficient approach for lipase immobilization in industrial application. The EPI-modified PVDF membrane constructed in this study showed the highest lipase activity of 37.87 U/g with a specific activity of 37.49 U/mg and the highest Vmax value of 57.8 U/mg, which was 97% activity retention to that of free lipase. The immobilized lipase on

1. Mateo, C., Palomo, J. M., Fernandez-Lorente, G., Guisan, J. M., and FernandezLafuente, R.: Improvement of enzyme activity, stability and selectivity via immobilization techniques, Enzyme Microb. Technol., 40, 1451–1463 (2007). 2. Yang, J., Ma, X., Zhang, Z., Chen, B., Li, S., and Wang, G.: Lipase immobilized by modification-coupled and adsorption-cross-linking methods: A comparative study, Biotechnol. Adv., 28, 644–650 (2010). 3. Villeneuve, P., Muderhwa, J. M., Graille, J., and Haas, M. J.: Customizing lipases for biocatalysis: a survey of chemical, physical and molecular biological approaches, J. Mol. Catal. B: Enzym., 9, 113–148 (2000). 4. Pandey, A., Benjamin, S., Soccol, C. R., Nigam, P., Krieger, N., and Soccol, V. T.: The realm of microbial lipases in biotechnology, Biotechnol. Appl. Biochem., 29, 119–131 (1999). 5. Hung, T. C., Giridhar, R., Chiou, S. H., and Wu, W. T.: Binary immobilization of Candida rugosa lipase on chitosan, J. Mol. Catal. B: Enzym., 26, 69–78 (2003). 6. Ye, P., Xu, Z. K., Che, A. F., Wu, J., and Seta, P.: Chitosan-tethered poly (acrylonitrile-co-maleic acid) hollow fiber membrane for lipase immobilization, Biomaterials, 26, 6394–6403 (2005). 7. Li, S. F., Chen, J. P., and Wu, W. T.: Electrospun polyacrylonitrile nanofibrous membranes for lipase immobilization, J. Mol. Catal. B: Enzym., 47, 117–124 (2007). 8. Huang, X. J., Yu, A. G., and Xu, Z. K.: Covalent immobilization of lipase from Candida rugosa onto poly (acrylonitrile-co-2-hydroxyethyl methacrylate) electrospun fibrous membranes for potential bioreactor application, Bioresour. Technol., 99, 5459–5465 (2008). 9. Huang, X. J., Ge, D., and Xu, Z. K.: Preparation and characterization of stable chitosan nanofibrous membrane for lipase immobilization, Eur. Polym. J., 43, 3710–3718 (2007). 10. Huang, X. J., Yu, A. G., Jiang, J., Pan, C., Qian, J. W., and Xu, Z. K.: Surface modification of nanofibrous poly (acrylonitrile-co-acrylic acid) membrane with biomacromolecules for lipase immobilization, J. Mol. Catal. B: Enzym., 57, 250–256 (2009). 11. Balcao, V. M., Paiva, A. L., and Xavier Malcata, F.: Bioreactors with immobilized lipases: state of the art, Enzyme Microb. Technol., 18, 392–416 (1996). 12. Tsai, S. W. and Shaw, S. S.: Selection of hydrophobic membranes in the lipasecatalyzed hydrolysis of olive oil, J. Membr. Sci., 146, 1–8 (1998). 13. Lozano, P., Perez-Marin, A., De Diego, T., Gomez, D., Paolucci-Jeanjean, D., Belleville, M., Rios, G., and Iborra, J.: Active membranes coated with immobilized Candida antarctica lipase B: preparation and application for continuous butyl butyrate synthesis in organic media, J. Membr. Sci., 201, 55–64 (2002). 14. Shakeri, M. and Kawakami, K.: Enhancement of Rhizopus oryzae lipase activity immobilized on alkyl-functionalized spherical mesocellular foam: Influence of alkyl chain length, Microporous Mesoporous Mater., 118, 115–120 (2009). 15. Palla, C. A., Pacheco, C., and Carrín, M. E.: Preparation and modification of chitosan particles for Rhizomucor miehei lipase immobilization, Biochem. Eng. J., 55, 199–207 (2011). 16. Tsai, Y. H., Wang, M. Y., and Suen, S. Y.: Purification of hepatocyte growth factor using polyvinyldiene fluoride-based immobilized metal affinity membranes: equilibrium adsorption study, J. Chromatogr. B Analyt. Technol. Biomed. Life Sci., 766, 133–143 (2002). 17. Zou, H., Luo, Q., and Zhou, D.: Affinity membrane chromatography for the analysis and purification of proteins, J. Biochem. Biophys. Methods, 49, 199–240 (2001). 18. Sirkar, K. K., Shanbhag, P. V., and Kovvali, A. S.: Membrane in a reactor: a functional perspective, Ind. Eng. Chem. Res., 38, 3715–3737 (1999). 19. Tsai, S. W., Liu, B. Y., and Chang, C. S.: Enhancement of (S) naproxen ester productivity from racemic naproxen by lipase in organic solvents, J. Chem. Technol. Biotechnol., 65, 156–162 (1996). 20. Hilal, N., Kochkodan, V., Nigmatullin, R., Goncharuk, V., and Al-Khatib, L.: Lipaseimmobilized biocatalytic membranes for enzymatic esterification: Comparison of various approaches to membrane preparation, J. Membr. Sci., 268, 198–207 (2006). 21. Sharma, R., Chisti, Y., and Banerjee, U. C.: Production, purification, characterization, and applications of lipases, Biotechnol. Adv., 19, 627–662 (2001). 22. Huang, X. J., Chen, P. C., Huang, F., Ou, Y., Chen, M. R., and Xu, Z. K.: Immobilization of Candida rugosa lipase on electrospun cellulose nanofiber membrane, J. Mol. Catal. B: Enzym., 70, 95–100 (2011).

172

CHEN ET AL.

23. Xu, J., Wang, Y., Hu, Y., Luo, G., and Dai, Y.: Immobilization of lipase by filtration into a specially designed microstructure in the CA/PTFE composite membrane, J. Mol. Catal. B: Enzym., 42, 55–63 (2006). 24. Liu, Y. C., Suen, S. Y., Huang, C. W., and ChangChien, C. C.: Effects of spacer arm on penicillin G acylase purification using immobilized metal affinity membranes, J. Membr. Sci., 251, 201–207 (2005). 25. Guo, W. and Ruckenstein, E.: Crosslinked glass fiber affinity membrane chromatography and its application to fibronectin separation, J. Chromatogr. B, 795, 61–72 (2003). 26. Lotrakul, P. and Dharmsthiti, S.: Lipase production by Aeromonas sobria LP004 in a medium containing whey and soybean meal, World J. Microbiol. Biotechnol., 13, 163–166 (1997). 27. Bradford, M. M.: A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem., 72, 248–254 (1976). 28. Fernandez, R. E., Bhattacharya, E., and Chadha, A.: Covalent immobilization of Pseudomonas cepacia lipase on semiconducting materials, Appl. Surf. Sci., 254, 4512–4519 (2008). 29. Zeng, H.-Y., Liao, K.-B., Deng, X., Jiang, H., and Zhang, F.: Characterization of the lipase immobilized on Mg-Al hydrotalcite for biodiesel, Process Biochem., 44, 791–798 (2009).

J. BIOSCI. BIOENG., 30. Bayramo lu, G., Hazer, B., Altintaş, B., and Arica, M. Y.: Covalent immobilization of lipase onto amine functionalized polypropylene membrane and its application in green apple flavor (ethyl valerate) synthesis, Process Biochem., 46, 372–378 (2011). 31. Wang, W. F., Tan, T. L., and Ong, P. P.: The v2 and v8+ v10 bands of CF2= CH2, J. Mol. Spectrosc., 181, 11–17 (1997). 32. Yang, G., Wu, J., Xu, G., and Yang, L.: Comparative study of properties of immobilized lipase onto glutaraldehyde-activated amino-silica gel via different methods, Colloids Surf. B, 78, 351–356 (2010). 33. Lu, P. and Hsieh, Y. L.: Lipase bound cellulose nanofibrous membrane via Cibacron Blue F3GA affinity ligand, J. Membr. Sci., 330, 288–296 (2009). 34. Arica, M. Y. and Bayramoglu, G.: Invertase reversibly immobilized onto polyethylenimine-grafted poly (GMA-MMA) beads for sucrose hydrolysis, J. Mol. Catal. B: Enzym., 38, 131–138 (2006). 35. Hiol, A., Jonzo, M. D., Rugani, N., Druet, D., Sarda, L., and Comeau, L. C.: Purification and characterization of an extracellular lipase from a thermophilic Rhizopus oryzae strain isolated from palm fruit, Enzyme Microb. Technol., 26, 421–430 (2000). 36. Talukder, M. M. R., Tamalampudy, S., Li, C. J., Yanglin, L., Wu, J., Kondo, A., and Fukuda, H.: An improved method of lipase preparation incorporating both solvent treatment and immobilization onto matrix, Biochem. Eng. J., 33, 60–65 (2007).