Lipases (EC 184.108.40.206) are pivotal catalysts for organic synthesis. Due to their catalytic versatility, they are among the most important enzymes for industrial applications [1
]. In their natural environment they catalyze the hydrolysis of lipids in aqueous media, but under low water-content conditions they can promote the synthesis of esters by esterification, transesterification, alcoholysis or acidolysis, among other reactions.
However, several facts limit the use of lipases for some applications at an industrial scale. Their high cost and the possibility of enzyme inactivation by the acids and alcohols used as substrates are some of the most outstanding issues. For example, short-chain carboxylic acids tend to partition to the microenvironment around the lipase decreasing the local pH, which can affect catalytic activity [3
], and alcohols, which are competitive inhibitors of lipases, can irreversibly inactivate lipases [4
]. In this context, immobilization can be an excellent choice to improve the catalyst’s stability and recovery decreasing the production costs [1
]. The most common techniques used for protein immobilization involve their simple entrapment (encapsulation), their attachment to a support, or the preparation of cross-linked enzyme aggregates (CLEAs) with or without a carrier [5
]. Except in the first case, the protein establishes an interaction with the carrier, either by physical adsorption through hydrophobic or van der Waals interactions, by ionic binding or by covalent attachment or crosslinking. Materials that range from biopolymers to inorganic compounds can be used as carriers and among them nanomaterials offer several advantages. For example, they have larger specific surface area and less mass transfer limitations than other materials [9
]. In particular, magnetic nanoparticles (MNPs) have unique additional properties such as their superparamagnetic behavior and their easy separation under an external magnetic field. To protect the magnetic core and improve enzyme activity, the MNPs require coatings on their surface [6
] that can be further modified by specific functional groups (epoxy, amino, carboxylate, thiol, alkyl, etc.).
Non-covalent methods that rely in physical adsorption are cheap and simple, as the process is performed by direct contact of enzyme and carrier at mild temperature and do not require any chemical compounds. Since lipases are very hydrophobic proteins, they can be easily adsorbed on hydrophobic carriers functionalized with alkyl chains [10
However, for many applications, covalent attachment is preferred because it prevents from leaching of the enzyme [5
]. When the support is aminated (AMNP), bifunctional reagents such as glutaraldehyde (GA), can be used for activation. In the appropriate conditions, the free aldehyde group from the linker forms covalent imine bonds with reactive amino groups of the biomolecule [11
]. GA can also be used for covalent immobilization as magnetic cross-linked enzyme aggregates (mCLEAS), which involves the precipitation of soluble proteins in the presence of MNPs, and the subsequent cross-linking of the mixture with GA [7
Thus, the properties of the final catalyst depend on the way the protein is linked to the carrier, or in other words, on the type of carrier, the functional group in its surface and the immobilization conditions [10
]. In this work, we have assayed the immobilization on MNPs of the non-commercial lipase OPEr, a recombinant form of the enzyme naturally secreted by the fungus Ophiostoma piceae
(OPE) heterologously produced in Pichia pastoris
. This enzyme is a versatile lipase from the Candida rugosa
-like family, with high activity on triglycerides and sterol esters [18
]. OPE, and specially OPEr, have been tested in hydrolysis and synthesis reactions in their soluble form, revealing that they are versatile enzymes with catalytic efficiencies superior to those reported for other lipases [19
]. In view of its biotechnological potential, we tackled the immobilization of OPEr according to three strategies in order to compare their activity with that of the soluble enzyme. One of the approaches was non-covalent immobilization by hydrophobic interaction, and the other two procedures involved the use of GA-activated AMNPs for covalent attachment of the lipase either directly or by forming mCLEAs.
The magnetic nanobiocatalysts were tested in the synthesis of esters of short chain volatile fatty acids (VFA), because these compounds have potential biotechnological interest and offered a simple model to evaluate the influence of the chain-length of the substrate in the enzymatic activity. These esters, that contribute to the natural aroma and taste of fruits and vegetables, are profusely used as additives in the pharmacy, cosmetics, and food industries [25
] and can be extracted in very low concentration from natural sources [27
]. However, for industrial purposes, they are generally obtained by chemical transformation at high temperatures with non-selective catalysts. Under these conditions, unwanted secondary products are generated [2
], and the esters cannot be labelled as natural products [25
]. Now, the regulations promoting the production of natural ingredients, summed to the consumers’ preference for natural foodstuffs have fostered the importance of bio-based chemicals. The synthesis of esters by biocatalysis overcomes the above issues, as the reactions are specific, selective, clean, and developed under mild conditions. Hence, the esters enzymatically produced comply with the European and American regulations for natural compounds [25
] being the expected annual growth rate of their global market of around 6.4% for 2016–2021 (BBC, 2016). Lipases of the fungi C. rugosa
, Candida antarctica, Rhizopus oryzae
, Thermomyces lanuginosus
, or Rhizomucor miehei
are among the most frequently reported biocatalysts for the synthesis of aroma esters, generally by direct esterification of VFA and alcohols [1
]. These are well-known commercial catalysts used in a wide array of reactions. Apart from lipases, other enzymes from the α/β hydrolase family have demonstrated their ability to produce these aliphatic esters. For example, esterifications catalyzed by cutinases have shown to be very efficient with substrates with 4–7 carbon atoms, and several reports deal with their use to produce of short-chain esters [31
This study gathers the results of three approaches for immobilization of the lipase OPEr on MNPs. The simplest method was based on the non-covalent hydrophobic interaction of the lipase with a commercial magnetic carrier harboring hydrophobic octyl groups in the surface. On the other hand, the two methodologies applied for covalent immobilization of OPEr used GA-activated AMNPs as carrier. To form covalent mCLEAs, the protein solution is allowed to interact with GA-activated AMNPs for a short period before adding the precipitant and the crosslinker. Finally, the third protocol involved the formation of a covalent imine between GA-activated AMNPs and amino groups of the protein. The three magnetic preparations with immobilized OPEr were assayed as catalysts of the synthesis of the butyl esters of volatile fatty acids of different chain-length. The best nanobiocatalyst was selected to study the influence of several parameters in the esterification yields and the operational stability of the preparation with the C4-C7 fatty acid substrates.
4. Materials and Methods
4.1. Chemicals and Reagents
Butyric acid, valeric acid, isobutyric acid, isovaleric acid, hexanoic acid, heptanoic acid, 1-butanol, and p-nitrophenyl butyrate were purchased from Sigma-Aldrich (St. Louis, MO, USA). Other chemicals and solvents were of the purest available grade, provided by Sigma-Aldrich (St. Louis, MO, USA) or Merck (Darmstadt, Germany).
4.2. Strains, Culture Conditions, and Preparation of Enzyme Crudes
GS115 strain containing the ope
gene was maintained and cultivated to produce OPEr as previously reported [20
]. Cultures were then centrifuged (13,000 rpm, 4 °C) and fungal biomass discarded. Supernatants were concentrated by ultrafiltration in an YM3 Amicon device (Merck Millipore, Darmstadt, Germany) with a 50-kDa membrane. The crudes obtained were used without further purification.
4.3. Evaluation of Enzyme Activity and Protein Content
The standard assay to determine the activity of the catalyst was carried out monitoring at 410 nm the release of p
-nitrophenol from hydrolysis of 1.5 mM p
-nitrophenyl butyrate (p
NPB) in 20 mM Tris-HCl pH 7.0 at room temperature, using a Shimadzu UV-160A spectrophotometer. One unit of activity (1 U) is defined as the amount of enzyme used to release 1 µmol of p
= 15,200 M−1
) per minute under the defined conditions [66
]. Protein concentration was determined by the BCA assay, using bovine serum albumin as standard, and by measuring the absorbance at 280 nm in a Nanodrop (NanoDrop 2000, Thermo Scientific, city, state abbreviation if USA, country).
4.4. Functionalization of Nude Magnetic Nanoparticles
Magnetic nanoparticles from Iolitec GmbH (Heilbronn, Germany) were functionalized with NH2
groups on their surface by treatment with (3-aminopropyl)triethoxysilane 99% (APTS, Sigma-Aldrich, city, state abbreviation if USA, country). MNPs (1 g, dry weight) were incubated with 10 mL of 130 mM APTS in methanol [67
] and mixed at 80 rpm and 28 °C. After 16 h, they were washed three times with ethanol 50% and sonicated in an ultrasonic bath (Selecta, Spain) between washes. Finally, the AMNPs were dried at 65 °C in an aeration oven.
4.5. Characterization of the Nanoparticles
The amount of amino groups bound to the magnetic support was calculated according to the procedure described by del Campo et al. [68
]. Morphological analysis was carried out by transmission electron microscopy (TEM) using a JEM HITACHI S-4800. Fourier Transform infrared (FTIR) spectra were collected using a FT/IR-4200 FTIR spectrometer (Jasco, Tokyo, Japan) in the spectral range 4000–400 cm−1
, with a spectral resolution of 4 cm−1
in transmittance mode. The samples were analyzed as KBr pellets. XRD patterns were recorded using a Siemens D5000 diffractometer equipped with a Cu anode (Cu Kα
radiation) and a LiF monochromator to study the structural and phase analysis. The Rietveld refinement of XRD patterns was performed using the TOPAS v4.2 software (Bruker AXS, Karlsruhe, Germany) and taking into account the crystallographic information for the different phases from Pearson’s crystal structure database for inorganic compounds [69
]. The average crystallite size of nude MNPs was also estimated by X-ray pattern using the Debye-Scherrer formula [70
= 0.89·λ/β·cosθ where Dhkl
is the average crystallite size, 0.89 is the shape factor (assuming spherical particles), λ is the X-ray wavelength used (1.5406 Å for Cu Kα
), β is the full-width at half-maximum (FWHM) of the experimental diffractions and θ is the Bragg’s angle. The magnetic measurements were performed on a Quantum Design XL-SQUID magnetometer (Quantum Design International, San Diego, CA, USA). Hysteresis measurements were taken with applied field range from 0 to 50,000 Oe (5T).
4.6. Immobilization of Crudes with the Recombinant Versatile Lipase from O. piceae
The crudes with OPEr were immobilized by three procedures. In all cases, the immobilization yield (%) was calculated from the difference between offered activity and residual activity in the supernatant at the end of the immobilization period. The activity recovery was determined by considering the OPEr activity initially offered for immobilization and the activity of the immobilized catalyst.
4.6.1. Covalent Immobilization on Amino-Functionalized Magnetic Nanoparticles Activated with Glutaraldehyde
First, AMNPs (1 g dry weight) were activated by incubation with 40 mL of 250 mM glutaraldehyde (Sigma-Aldrich) in water, for 3 h. Then, AMNPs modified with glutaraldehyde (AMNPs-GA), with surface aldehyde groups, were washed six times with water to remove the residual glutaraldehyde and one more time with 20 mM Tris-HCl buffer solution, pH 7.
AMNPs-GA (1 g) were incubated with OPEr crudes (0.07 mg protein/mg AMNP-GA) in a final volume of 30 mL (100 mM Tris-HCl buffer solution, pH 7) for 24 h at 28 °C with rotational mixing at 80 rpm (Multi Bio RS-24, Biosan). After immobilization, AMNP-GA-OPEr were thoroughly washed with 20 mM Tris-HCl buffer solution, pH 7 to remove the unbound proteins, and maintained in the same buffer at 4 °C until used.
4.6.2. Immobilization as Magnetic CLEAS
were prepared as reported by Kim et al. [12
]. In brief, 1 g of AMNPs was allowed to react with 0.5% GA in Tris-HCl 10 mM pH 8 for 3 h at 25 °C and 200 rpm. After three washes, the crude protein solution (0.05 mg protein/mg carrier) was added to the GA-activated AMNPs in a final volume of 10 mL of the buffer and maintained for 2 h at 25 °C and 50 rpm. Then, 10 mL of ammonium sulfate 4 M and 10 mL of the 0.5% GA solution were incorporated to the mixture and stirred at 250 rpm for 1 h at 25 °C and then at 4 °C for 24 h. The mCLEAs-OPEr were washed three times with the above buffer and four more with Tris-HCl 20 mM pH 7, and maintained in this buffer at 4 °C until used.
4.6.3. Immobilization by Adsorption on Commercial Magnetic Nanoparticles Functionalized with Hydrophobic Octyl Groups
A commercial solution containing 250 mg of SiMAG-Octyl (Chemicell, Germany) was carefully mixed for 2 min at 28 °C with OPEr crudes containing 0.07 mg protein per mg of carrier in 5 mL 100 mM Tris-HCl buffer pH 7. The biocatalyst was stored in 20 mM Tris-HCl buffer solution, pH 7 at 4 °C.
4.7. Activity of Immobilized Enzymes
The hydrolytic activity of immobilized enzymes was measured by the hydrolysis of p
-nitrophenyl butyrate as described in Section 4.3
. The amount of biocatalyst used was 0.25 mg. The results were expressed as mU per mg of biocatalyst.
4.8. Esterification of Volatile Fatty Acids in Isooctane
A suspension containing 11 U of the immobilized biocatalyst (AMNP-GA-OPEr, mCLEAs-OPEr or SiMAG-Octyl-OPEr) in 20 mM Tris-HCl buffer pH 7 was added to a clean vial. A magnet was used to attract the magnetic catalyst in order to remove the buffer. The reaction mixtures (500 µL) contained the substrates in a molar ratio 2:1 (alcohol:acid) in isooctane, at a concentration of 100 mM of each one of the acids (C4-C7 straight-chain volatile fatty acids), and 2% hexane as internal standard. Each mixture was deposited in a tube containing the catalyst to start the reactions. The reactions with the free (soluble) enzyme were performed using the same conditions except for the presence of 7% water in the medium. The standard experiments were performed under the conditions detailed above, with rotational mixing at 100 rpm (Multi Bio RS-24, Biosan), at 25 °C for 8 h. For studying their operational stability, the biocatalysts were washed with 1 mL of isooctane and 1 mL of 20 mM Tris-HCl buffer pH 7 and used in a new reaction cycle under the same conditions.
Some of the reaction parameters were individually modified to study their influence on the esterification yields and reaction rate. To test the effect of the molar ratio of the substrates, stoichiometric (1:1) and 3:1 molar proportions (alcohol:acid) were assayed. The influence of the acids’ concentration was analyzed in the range of 100 to 1000 mM. The esterification of two branched-chain volatile fatty acids, isobutyric acid and isovaleric acid, was also assessed under the standard conditions in reactions catalyzed by AMNP-GA-OPEr.
All these experiments were performed in duplicate, taking samples of 25 µL at 0, 2, 4, 6, and 8 hours to monitor the time-course of the reaction by gas chromatography/mass spectrometry, as described in Section 4.9
4.9. Monitoring Reactions by Gas Chromatography
As explained before, hexane was added to reactions as internal standard for gas chromatography/mass spectrometry analysis. The aliquots withdrawn from the reaction mixture, were deposited into a clean vial, and mixed with 25 µL of BSTFA (Sigma-Aldrich) for derivatization at 60 °C for 10 min. 1 µL of this solution was injected in a 7890A gas chromatograph coupled to a quadrupolar mass detector 5975C (Agilent, Palo Alto, CA). The injector and flame ionization detector were set up at 275 °C, and He (13 psi) was used as the carrier gas. The separation was carried out using a fused-silica capillary column DB5-HT (30 m × 250 µm × 0.1 µm, Agilent, Palo Alto, CA, USA). For analysis of volatile fatty acid esters, the oven was maintained at 80 °C for 1 min and then a temperature program was applied, with a 70 °C /min ramp rate to reach a final temperature of 240 °C. The peaks of substrates and products were identified from their retention time, compared to that of commercial standards. The esterification yields were calculated from a calibration curve of each product.