Article pubs.acs.org/Biomac
Immobilization and Stabilization of Lipase (CaLB) through Hierarchical Interfacial Assembly Joey N. Talbert,† Li-Sheng Wang,‡ Bradley Duncan,‡ Youngdo Jeong,‡ Stephanie M. Andler,† Vincent M. Rotello,*,‡ and Julie M. Goddard*,† †
Department of Food Science, University of Massachusetts − Amherst, 102 Holdsworth Way, Amherst, Massachusetts 01003, United States ‡ Department of Chemistry, University of Massachusetts − Amherst, 379A LGRT, 710 North Pleasant Street, Amherst, Massachusetts 01003, United States ABSTRACT: Nanostructure-enabled hierarchical assembly holds promise for efficient biocatalyst immobilization for improved stability in bioprocessing. In this work we demonstrate the use of a hierarchical assembly immobilization strategy to enhance the physicochemical properties and stability of lipase B from Candida antarctica (CaLB). CaLB was complexed with iron oxide nanoparticles followed by interfacial assembly at the surface of an oil-in-water emulsion. Subsequent ring opening polymerization of the oil provided cross-linked microparticles that displayed an increase in catalytic efficiency when compared to the native enzyme and Novozym 435. The hierarchical immobilized enzyme assembly showed no leakage from the support in 50% acetonitrile and could be magnetically recovered across five cycles. Immobilized lipase exhibited enhanced thermal and pH stability, providing 72% activity retention after 24 h at 50 °C (pH 7.0) and 62% activity retention after 24 h at pH 3.0 (30 °C); conditions resulting in complete deactivation of the native lipase.
■
INTRODUCTION Lipases have been applied in the development of natural flavors, improved biofuels, biodegradable polyesters, structured lipids, and sugar esters.1−3 Compared to synthetic chemocatalysts, lipases enable transesterification, interesterification, ester synthesis, and triglyceride hydrolysis reactions to be performed with enhanced specificity and under milder conditions.4 These properties have made lipase an attractive catalyst for green processing. However, the native enzyme is limited in that it is not stable to environmental conditions and can be inactivated under processing conditions. To enhance the stability of lipases for bioprocessing, these enzymes have been immobilized by different means including adsorption, covalent and affinity attachment, cross-linking, and entrapment. Adsorption allows for one-pot immobilization of lipase and, when adsorbed to hydrophobic materials, has been used to promote interfacial activation of the enzyme at the material surface. This method has resulted in enhanced stability when compared to the native lipase.5 Protein desorption is the primary disadvantage to this methodleading to loss of enzyme during recovery and reuse.6,7 The addition of a coating to the enzyme-bound carrier has been utilized to prevent leakage, but results in loss of catalytic efficiency due to mass transfer limitations.6 Covalent and affinity attachment prevents protein leakage from the carrier and allows for surface active enzyme. However, conditions for covalent attachment may result in loss of enzymatic activity and the immobilized enzyme © XXXX American Chemical Society
can have limited stability compared to other forms of immobilization.8−10 Moreover, covalent and affinity attachment necessitates surface functionalizationrequiring additional steps in the immobilization process.11 Cross-linking in the form of cross-linked enzyme crystals and aggregates (i.e., CLECs and CLEAs) yield high protein loadings, but are limited by mass transfer and activity retention during modification.12,13 Entrapment of lipase in pores and or via a sol−gel process has shown to promote enzyme stabilization, but substrate diffusion limits catalytic efficiency and leakage from the support reduces recycling of the enzyme.14−23 An ideal immobilized enzyme system has been described as having no enzyme inactivation during immobilization, no leakage of the enzyme, enzyme immobilized in an optimum form for activity, and negligible mass transfer limitations after immobilization.1 In addition, it has been suggested that the stability of lipase is improved through macromolecular crowding, restricted molecular motion, and binding at hydrophobic interfaces.15,22 While traditional forms of lipase immobilization can be advantageous in meeting one or more of these criteria, they often do not fully meet the demands of an ideal enzyme system while also providing for optimal stabilization of the enzyme.4 As such, combinative approaches Received: July 3, 2014 Revised: September 11, 2014
A
dx.doi.org/10.1021/bm500970b | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
for enzyme immobilization have been pursued.24 These approaches include immobilization using chemical and physical methods, interfacial activation and bioimprinting, and chemical modification followed by immobilization.25−27 Recent work has suggested that hierarchical assembly can be used to rapidly immobilize an enzyme to the surface of oil microcapsules and that subsequent cross-linking of the microcapsules leads to a rigid support that traps the enzyme in such a way that improves storage stability when compared to the native enzyme.28 This strategy presents a promising approach to stabilizing lipase in such a way as to meet the demands of an ideal immobilized enzyme system. The objective of the current work was to apply the strategy of hierarchical assembly to immobilize lipase B from Candida antarctica (CaLB) and to evaluate the physicochemical properties and stability of the system.
■
Scheme 1. Assembly of Lipase Crosslinked Microparticles
EXPERIMENTAL SECTION
Materials. Lipase CV-CALBY (Candida antarctica lipase B) was purchased from Chiralvision (Leiden, Netherlands). Resorufin butyrate was purchased from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA). Bicinchoninic acid (BCA) assay reagents, Coomassie G250 (Bradford) dye, and bovine serum albumin were purchased from Thermo Scientific (Rockford, IL, U.S.A.). Amicon Ultra centrifugal filter devices (10 K MWCO) were purchased from Millipore (Carrigtwohill, Co. Cork, Ireland). Diethylene glycol (DEG), sodium hydroxide (NaOH), iron(III) chloride hexahydrate (FeCl3·6H2O), hydroxytyramine hydrochloride (Dopamine), sand (40−100 mesh), and syringe filters (0.22 μm) were purchased from Fisher Scientific (Fairlawn, NJ, U.S.A.). Iron(II) chloride tetrahydrate (FeCl2·4H2O), dicyclopentadiene (DCPD), ruthenium-based first generation Grubbs’ catalyst, trichlorobenzene (TCB), and lipase acrylic resin from Candida antarctica (Novozym 435) were purchased from SigmaAldrich (Saint Louis, MO, USA). Enzyme Purification. A 10 wt %/vol solution of lipase in 10 mM MES buffer pH 7.0 was purified through a 0.22 μm cellulose filter followed by centrifugation through a 10K MWCO centrifugal filter (25 °C; 5000g; 20 min). Samples were diluted to the desired concentration (typically 1 mg/mL) in nanopure water or 10 mM MES buffer pH 7.0 and refrigerated. Purified enzyme was used within 5 h of purification. Nanoparticles. Iron oxide nanoparticles (NPs) were made by onepot procedure using diethylene glycol as solvent.29 FeCl3 and FeCl2 were mixed with NaOH (as oxidant), and then heated to 220 °C for 2 h to form Fe3O4 NPs. In the end of reaction, dopamine was introduced as the capping agent. After cooling down to room temperature, the NPs were separated by magnet, and washed with ethanol 5 times. The washed NPs were dried by nitrogen gas flow. The dopamine functionalized Fe3O4 NPs, at a concentration of 2 mg/mL, were suspended in 5 mM sodium acetate buffer (pH 5.0) for further experiments. Cross-Linked Microparticles (CLMPs) Fabrication. Microparticles were produced from a liquid biphasic system consisting of an aqueous phase and an oil phase using a modification of the procedure described by Jeong (Scheme 1).28 The oil phase was comprised of 6.67 M dicyclopentadiene, 0.40 M trichlorobenzene, 6 mM Grubb’s catalyst (1st generation), and 0.42 M toluene. Grubb’s catalyst was solubilized by sonication to 0.12 M in the toluene prior to addition to the rest of the oil components. The aqueous phase was composed of 55.5 μg/mL purified lipase and 100 μg/mL dopaminecapped iron oxide nanoparticles. The enzyme and nanoparticles were allowed to mix at room temperature (ca. 22 °C) for 5 min prior to the addition of the oil phase. The oil phase was added to aqueous phase to give a 10% (v/v) oil phase and a 90% (v/v) aqueous phase in the final mixture. The mixture was emulsified for 30 s at 4200 cpm using an amalgamator (Wykle Research). The emulsified solution was allowed to sit for 5 h at room temperature (ca. 22 °C), then separated by centrifugation (2000g; 30 s; 3.8 cm tube height), washed threee times
with nanopure water, and stored in 10 mM MES buffer (pH 7.0) or nanopure water at 4 °C for further experiments. Enzyme Kinetics. Purified native lipase and CLMP samples at a concentration of 1.67 μg of protein/mL were tested for activity using 0.1−100 μM resorufin butyrate at 50 °C in 10 mM MES buffer (pH 7.0) under constant shaking (Biotek Synergy; medium speed).30 Fluorescence values were measured at 1 min intervals for 10 min and read at 528/20 nm excitation and 590/20 nm emission (Biotek Synergy). Novozym 435 samples, blended with 40−100 mesh sand to facilitate handling, were diluted to a concentration 4.90 μg protein/mL and tested for activity using 0.1−100 μM resorufin butyrate at 50 °C in 10 mM MES buffer (pH 7.0) under constant rotation (20 rpm). Fluorescence values were obtained every 3 min for 12 min and measured at 528/20 nm excitation and 590/20 nm emission (Biotek Synergy). Substrate conversion and subsequent rates were determined from a standard curve of resorufin sodium salt. The Michaelis constant (Km) and maximum velocity (Vmax) were extrapolated from nonlinear regression of a plot of velocity versus substrate concentration using Michaelis−Menten enzyme kinetics (v. 5.04, Graphpad Software, Inc., La Jolla, CA, U.S.A.). The turnover number (kcat) was determined by dividing the Vmax by the initial enzyme concentration assuming a molecular weight of 33 000 Da for the enzyme.31 Thermal Stability. Purified native enzyme, CLMPs, and Novozym 435 were diluted to a concentration of 5 μg of protein/mL in 10 mM MES buffer (pH 7.0) and incubated at 50 °C, 60 °C, 70 °C, and 80 °C. Native enzyme samples were pulled every 10 min for 1 h. Novozym 435 and CLMP samples were pulled at 2.5, 5, 10, and 24 h and immediately cooled on ice. At each time interval, native lipase and CLMP samples were diluted to a concentration of 1.67 μg/mL, and Novozym 435 samples were diluted to a concentration of 4.90 μg/mL, in 10 mM MES buffer (pH 7.0) and tested for activity using 10 μM resorufin butyrate at 50 °C. Fluorescence values, substrate conversion, and subsequent rates were determined as described, previously. Activity was expressed as a percent of retained activity relative to the initial activity of each sample. pH Stability. Purified native enzyme and CLMPs were diluted to a concentration of 5 μg protein/mL in 10 mM buffer at the pH values ranging from 3.0 to 10.0 and incubated at 30 °C for up to 24 h. Buffers included citric (pH 3.0−6.0), phosphate (pH 7.0−8.0), and carbonate (pH 9.0−10.0). At each time interval, native lipase and CLMP samples were diluted to a concentration of 1.67 μg of protein/mL in 0.1 M MES buffer (pH 7.0) and tested for activity as described above. Novozym 435 control were diluted to a concentration of 14.7 μg protein/mL in 10 mM buffer at the pH values and incubated at 30 °C B
dx.doi.org/10.1021/bm500970b | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
for up to 24 h under constant rotation (0.1g) prior to dilution to 4.90 μg of protein/mL in 0.1 M MES buffer (pH 7.0). Substrate conversion and subsequent rates were determined as described, previously. Protein Concentration. The protein concentration of purified native lipase was determined using bicinchoninic acid (BCA).32 Absorbance values were read at 562 nm and compared to a bovine serum albumin standard curve. Based on the results of the BCA assay, absorbance curves of lipase were developed and measured using the Bradford method at 595 nm and ultraviolet light absorbance at 280 nm.33 The BCA method was employed to measure the protein concentration of Novozym 435. Absorbance values at 280 nm were used for routine analysis of purified protein concentration. Protein Leaching. Protein leaching was determined for CLMP and Novozym 435 samples using 50% (v/v) acetonitrile in water.6 Samples were prepared to a final concentration of 150 μg protein/mL in 50% acetonitrile, rotated at 20 rpm for 2 h at 50 °C, then centrifuged (2000g; 60 s) prior to the removal of the supernatant. The supernatant was diluted 10-fold with nanopure water and centrifuged through a 10K MWCO centrifugal filter (25 °C; 5000g; 20 min). Bradford protein assay was performed on the concentrate and compared to a standard curve of native lipase at 595 nm. Scanning Electron Microscopy. Scanning electron microscope images were acquired using an FEI Magellan 400 Field-Emission Scanning Electron Microscope with a 1.0 kV acceleration voltage. Transmission electron microscopy. Samples on CF-400-Cu carbon film square mesh copper grid (Electron Microscopy Science; Hatfield, PA) were analyzed using a JEOL CX-100 kV TEM. Particle size was determined using ImageJ software. Confocal and Light Microscopy. Fluorescent dye encapsulated lipase CLMPs were produced by a modification of the method described above, in which Nile Red was incorporated into the trichlorobenzene phase prior to cross-linking. A light/confocal microscope (Nikon D-Eclipse C1 80i, Nikon, Melville, NY) was used to for light and confocal images. Confocal images were captured using 488 nm excitation and 515 nm emission of Nile Red encapsulated CLMPs. Images were collected and analyzed using the instrument software (EZ-CS1 version 3.8, Nikon, Melville, NY). Zeta Potential. The electrical charge (ζ-potential) of nanoparticles associated with the enzyme was measured using a particle electrophoresis instrument (Zetasizer Nano ZS, Malvern Instruments, Worcestershire, U.K.). Nanoparticles and enzyme were diluted in 10 mM MES buffer, pH 7.0. The solutions were mixed to give final concentrations of 100 μg of nanoparticles/mL and 0−132 μg of lipase/mL. Measurements were obtained after 5 min of mixing at room temperature (ca. 22 °C). The zeta potential of native lipase without added nanoparticles was measured at a concentration of 3.6 mg/mL in 10 mM MES buffer pH 7.0. The Smoluchowsky model was used by the instrument’s software program to convert the electrophoretic mobility measurements into ζ-potential values. Recovery and Recycling. Activity retention of lipase CLMPs during repeated cycles of recovery and reuse were assessed by diluting lipase CLMPs to a protein concentration of 15 μg/mL in 10 mM MES buffer (pH 7.0) and measuring activity as previously described. Lipase CLMP recovery was performed by exposure to a magnetic separator (Ocean Nanotech; Springdale, AR) for 2 h at room temperature (ca. 22 °C). After separation, buffer was removed, lipase CLMPs were diluted, and activity was tested as described previously. Subsequent cycles (five total) were prepared and completed in the same manner. CLMPs held at room temperature at the initial concentration served as controls. Statistical Analysis. Curve fitting and one-way analysis of variance (ANOVA) with Tukey’s pairwise comparison for the identification of statistical differences at p < 0.05 were performed using Graphpad Prism software (v. 5.04, Graphpad Software, La Jolla, CA). Reported data are representative of repeated independent experiments.
effectiveness of an immobilized enzyme system that met the criteria described by Adlercreutz.4 The criteria include: (1) no enzyme inactivation during immobilization, (2) no enzyme leakage after immobilization, (3) the enzyme should be present in fully activated form, and (4) mass transfer limitations should be negligible. Cross-linked microparticles were developed by applying principles of hierarchical assembly of enzymes and materials to form nanocomposites (Scheme 1). The nanocomposites consisted of three primary components: nanoparticles, lipase, and the microcapsule. Iron oxide nanoparticles were formed with a dopamine capped functional group at an average particle size of 2.34 nm (Figure 1). Iron oxide was employed to enable magnetic separation of the CLMPs for recovery and reuse during processing applications.
RESULTS AND DISCUSSION CLMP Characterization. The overall objectives of this work were to rationally design and demonstrate the
Figure 2. Zeta potential of lipase (CaLB) with 0.1 mg/mL dopaminecapped iron oxide nanoparticles at 25 °C, pH 7.0 (10 mM MES buffer). Values represent average ± standard deviation of n = 3 determinations.
Figure 1. TEM image of dopamine-capped iron oxide nanoparticles.
The nanoparticles were capped with dopamine to facilitate ionic interaction of the particles with lipase, which has a net negative charge (pI = 6.0) at its optimum pH value of 7.0.31,34 The developed dopamine-capped nanoparticles exhibited a net positive charge under conjugation conditions (zeta potential = +34.4 ± 1.9 mV) (Figure 2). In the presence of the native lipase, which has a zeta potential of −10.8 ± 2.3 mV under conjugation conditions, the nanoparticles complex rapidly (
