Eukaryotic Expression System Pichia pastoris Affects the Lipase Catalytic Properties: A Monolayer Study

Recombinant DNA methods are being widely used to express proteins in both prokaryotic and eukaryotic cells for both fundamental and applied research purposes. Expressed protein must be well characterized to be sure that it retains the same properties as the native one, especially when expressed protein will be used in the pharmaceutical field. In this aim, interfacial and kinetic properties of native, untagged recombinant and tagged recombinant forms of a pancreatic lipase were compared using the monomolecular film technique. Turkey pancreatic lipase (TPL) was chosen as model. A kinetic study on the dependence of the stereoselectivity of these three forms on the surface pressure was performed using three dicaprin isomers spread in the form of monomolecular films at the air-water interface. The heterologous expression and the N-His-tag extension were found to modify the pressure preference and decrease the catalytic hydrolysis rate of three dicaprin isomers. Besides, the heterologous expression was found to change the TPL regioselectivity without affecting its stereospecificity contrary to the N-tag extension which retained that regioselectivity and changed the stereospecificity at high surface pressures. The study of parameters, termed Recombinant expression Effects on Catalysis (REC), N-Tag Effects on Catalysis (TEC), and N-Tag and Recombinant expression Effects on Catalysis (TREC) showed that the heterologous expression effects on the catalytic properties of the TPL were more deleterious than the presence of an N-terminal tag extension.


Introduction
Lipases (triacylglycerol acylhydrolases, EC 3.1.1.3) catalyze the hydrolysis of triacylglycerols at the interface between the insoluble substrate and water [1] and have been widely used in industrial fields including detergents, dairy products, diagnostics, oil processing and biotransformation due to substrate specificity, regiospecificity, enantiomeric selectivity, thermostability and alkaline stability [2].
Recombinant DNA methods are being widely used to express proteins in both prokaryotic and eukaryotic cells in order to obtain high levels of expression for both fundamental and applied research purposes. For large scale purification of recombinant proteins, affinity chromatography is usually preferred due to the specificity of the process and the relatively easy purification schemes. Chimeric proteins have been made with affinity tags like L-galactosidase, Escherichia colimaltose binding protein, FLAG peptide, glutathione-S-transferase and hexahistidine tag (His-tag) [3]. The His-tag, consisting of four to six consecutive histidine residues, functions as a predominant ligand in immobilized metal affinity chromatography (IMAC) [4]. The ability of the histidine residue complex to bind to metal ions with high affinity even in the presence of denaturing agents and the requirement of mild elution conditions has made the His-tag a versatile tool for protein purification and characterization [5,6]. However, few comparative kinetic studies have been performed between tagged and untagged eukaryotic enzymes.
Recent reports have shown that His-tag can alter the binding characteristics of recombinant proteins when compared to the wild-type native protein [7,8]. Hang et al. [8] showed that although the His-tagged subunits of the terminase enzyme from bacteriophage (lamda) formed holoenzymes with wild-type catalytic activity, one of the subunits showed altered interaction with DNA. The length, composition and location of His-tag can require further optimization depending upon the sequences of the native protein [7,9]. Nevertheless, there are several studies where 'affinity tags' have no adverse effect on the activity of the native proteins [10,11,12].
Recently, Horchani et al [13] studied the interfacial and kinetic properties of the wild type, untagged recombinant and tagged recombinant forms of three staphylococcal lipases (SSL, SXL and SAL3) which were compared using the monomolecular film technique. They showed that independently from the negative effects of the recombinant expression process per se (REC), the presence of an N-terminal tag extension would decrease the catalytic activities (TEC) of staphylococcal lipases by creating a steric hindrance during the interfacial binding step. The effects of the heterologous expression process on the catalytic properties of the staphylococcal lipases are three times more deleterious than the presence of an N-terminal tag extension [13].
No studies have been conducted to compare the kinetic properties of the wild type form of eukaryotic lipases with those of the tagged and untagged recombinant forms.
The turkey pancreatic lipase (TPL) was purified from delipidated pancreases. This avian pancreatic lipase contains 450 amino acids and presents an experimental mass of 49665.31 Da [14,15]. To investigate the structure-function relationships of this lipase, TPL was expressed in Pichia pastoris. The biochemical characterization of the rTPL, using an emulsified system (pH-Stat), and its comparison to the native TPL show that the rTPL seems to be identical to the native enzyme [16].
In this study, we report the expression of the tagged TPL (6His-rTPL) in Pichia pastoris. The recombinant tagged enzyme was purified and its kinetic properties are compared to the recombinant and the native forms using the monomolecular film technique in order to study the Recombinant expression Effects on Catalysis (REC), N-Tag Effects on Catalysis (TEC), and N-Tag and Recombinant expression Effects on Catalysis (TREC).

Lipases
The native turkey pancreatic lipase (nTPL) was purified from delipidated pancreases as described by Sayari et al (2000) [15] and the recombinant TPL was expressed in Pichia pastoris (X33) and purified using adapted chromatographic methods, as previously described [16]. Starting with TPL full-length DNA (1353 pb) cloned into the pGAPZaA, which served as the template, the tagged TPL gene was obtained by PCR amplification using the following forward and reverse primers. Forward primer includes 6 Histidine codons (grey) and both primers including an EcoRI restriction site (underlined): Primer Fw: 59-GATCGAATTCCATCATCATCATCAT-CATTCTGAAGTTTGCTATGAC-39 Primer Rv: 59-GATC-GAATTCTTAGCAAGCAGTAAGGGT-39 The tagged recombinant TPL was expressed in Pichia pastoris (X33) and purified to homogeneity in a single step, using immobilized metal affinity chromatographic method ( Figure 1).

Protein concentration determinations
Protein concentration was determined as described by Bradford et al. [19], using BSA as standard. The lipase extinction coefficients were determined using the absorption UV spectroscopy [20]. The protein assays on the three lipases were performed simultaneously and in parallel, using the same chemical reagents.

Lipase activity and kinetic properties determination
The lipase activity was measured titrimetically at pH 8.5 and 37uC with a pH-Stat using olive oil emulsion or Tributyrin (TC 4 ) in presence of 4 mM NaDC and Turkey colipase. One lipase unit corresponds to 1 mmol of fatty acid liberated per min.
The kinetic constants (K m app, k cat , V max , and k cat /K m app.) calculated from the initial rate activity of native, recombinant and recombinant tagged TPL forms were determined using tributyrine. The effect of TC4 concentration on the hydrolysis rate was measured at various concentrations ranging from 0.68 to 68 mM.

SDS-PAGE
Analytical polyacrylamide gel electrophoresis of proteins in the presence of sodium dodecyl sulfate (SDS-PAGE) was performed by the method of Laemmli [21]. The proteins were stained with Coomassie brilliant blue.

Monomolecular film technique for kinetic measurements on lipase
Measurements were performed with KSV-2000 barostat equipment (KSV-Helsinki, Finland). The principle of the method was described previously by Verger and De Haas [22]. It involves the use of a ''zero-order'' trough with a reaction compartment and a reservoir compartment, which are connected to each other by a small surface channel. The enzyme solution was injected into the subphase of the reaction compartment when the lipid film covered both compartments. A mobile barrier, driven automatically by the barostat, moved back and forth over the reservoir to keep the surface pressure (p) constant, thus compensating for the substrate molecules removed from the film by enzyme hydrolysis. The surface pressure was measured in the reservoir compartment with a Wilhelmy plate (perimeter 3.94 cm) attached to an electromicrobalance, which was connected in turn to a microprocessor programmed to regulate the movement of the mobile barrier. As previously established [22], the sensitivity of the Wilhelmy plate was estimated at 0.1 mN.m 21 . Two 2-cm magnetic stirrers running at 250 rpm were placed in the reaction compartment. Kinetic measurements were performed at room temperature (25uC). The reaction compartment had an area of 130 cm 2 and a volume of 130 ml. The reservoir compartment was 148 mm wide and 249 mm long. The monolayers were formed by spreading a chloroform solution at the air/water interface, using a Hamilton microsyringe. Five to ten minutes were then allowed for the chloroform to evaporate and the film to form. Prior to each experiment, the Teflon trough used to form the monomolecular film was cleaned with water and then gently brushed in the presence of distilled ethanol, washed again with tap water, and finally rinsed with double-distilled water.

Statistical analysis
Experimental results were given as mean value6SD of three parallel measurements. Comparisons between values were analyzed by Student's test for unpaired data, and p,0.05 was considered significant. All statistical analysis were conducted using Microsoft Excel software.

Results and Discussion
1. Variations with surface pressure in the catalytic activities of nTPL, rTPL and 6His-rTPL using the three dicaprin isomers Kinetic experiments were performed at various surface pressures using three dicaprin isomers as substrates and three TPL forms. As can be seen in figure 2, the three dicaprin isomers were clearly differentiated by all the lipases studied, and this differentiation was more pronounced at high surface pressures.
When using the 1,2-sn-dicaprin, the nTPL and rTPL present their maximal activities at a surface pressure of 20 mN.m 21 , but the 6His-rTPL presents its optimal activity at 25 mN.m 21 . Up to those pressures, the three lipase form activities decreased (Figure 2A). Using 2,3-sn-dicaprin and 1,3-sn-dicaprin, the nTPL shows a maximum activity with bell-shaped curves at an optimal surface pressure of 20 mN.m 21 and 15 mN.m 21 respectively, contrary to the rTPL and 6His-rTPL which exhibit increased activity according to pressure showing high pressure preference (Figures 2B and 2C). When analyzing the maximum catalytic activities of the three TPL forms ( Figure 3A), one can easily state that nTPL maximum catalytic activities were always higher than those of the rTPL and 6His-rTPL ( Figure 3A).
The specific activities of the rTPL and 6His-rTPL are similar when 2,3-sn-dicaprin and 1,3-sn-dicaprin are used as substrates, but with the 1,2-sn-dicaprin, the maximum catalytic activity of the 6His-rTPL is comparable to that of nTPL ( Figure 3A).
Our results also show that nTPL hydrolyzes the three dicaprin isomers more efficiently than rTPL and 6His-rTPL do. So, the heterologous expression and N-tag extension change the pressure preference and decrease the specific activity of the native enzyme.

Effects of recombinant expression and N-tag extension on catalysis (REC, TEC and TREC)
To quantify the effects of the recombinant expression process and/or the presence of an N-terminal extension on the catalytic hydrolysis of the three dicaprin isomers, we coined the terms ''Recombinant expression Effects on Catalysis'' (REC), ''N-Tag Effects on Catalysis'' (TEC) and ''N-Tag and Recombinant expression Effects on Catalysis'' (TREC), which were defined as follows: Table 1. REC, TEC and TREC values of TPL obtained using 1,2-sn dicaprin, 2,3-sn-dicaprin and 1,3-sn-dicaprin substrates, spread at the air/water interface forming monomolecular films at 20, 25 and 30 mN.  Table 1). The positive effect of the N-terminal tag extension was found when 1,2-sn-dicaprin was used as substrat, but when 2,3-sn-dicaprin and 1,3-sn-dicaprin were used, negative effects on catalytic properties of TPL were recorded.
However, the heterologous expression changes the TPL regioselectivity without affecting the stereospecificity contrary to the N-tag extension which retains the TPL regioselectivity and changes the stereospecificity at high surface pressures, so the 6His-rTPL remains sn1 specific regardless of the surface pressure. This result can be explained by the large difference between rTPL and 6His-rTPL activities at high surface pressure when 1,2-sn-dicaprin was used as substrat ( Figure 2A) and the difference between rTPL and 6His-rTPL regioselectivity at high surface pressures where the 6His-rTPL would be specific to the adjacent ester of the diglyceride isomers ( Figure 4A).

General Discussion
Nowadays, Pichia pastoris is the most frequently used yeast system for heterologous protein production. Over the last few years, several products based on this platform have gained approval as biopharmaceutical drugs. Protein folding and secretion have been identified as significant bottlenecks in yeast expression systems, pinpointing a major target for strain optimi- and PPL are reproduced from previous work [17]. The vicinity index (V.I.) is calculated with the following formula: VI~A zation. However, few studies have investigated the heterologous expression and the N-terminal tag extension effects on the expressed proteins properties. To our knowledge, this is the first attempt to discriminate between the negative effects of the heterologous expression process and those of the N-terminal tag extension process on pancreatic lipase catalysis. One of the main features of lipolytic enzymes is the fact that, although they are water-soluble, they catalyse substrate hydrolysis at lipid-water interfaces. In the simplest kinetic model developed by Verger and co-workers [22], the coupling between the classical Michaelis-Menten step and various interfacial processes such as the penetration and activation of the lipase and the solubilization of the reaction products were described. It is worth noting that the negative effects of the heterologous expression and N-terminal tag extension processes are not very marked when a partly water-soluble (TC4) and an insoluble (Olive oil) substrates are used (Table 3). When k cat and catalytic efficiency (k cat /k m app ) were calculated, we can see than there are not differences between native, recombinant and recombinant tagged TPL confirming that there are not significant differences between the three TPL forms when classical analysis techniques were used ( Figure 3B).
The rTPL and 6His-rTPL seem to share the same biochemical and kinetic properties with the native enzyme under the same experimental conditions (pH stat technique). But when sensitive method such as the mono-molecular film technique was used, we found significant differences in the interfacial properties of the nTPL, rTPL and 6His-rTPL showing the negative effects of the heterologous expression and the N-terminal tag extension on the TPL properties. The most pronounced difference was the significant decrease in the specific activity of the recombinant and the tagged recombinant enzyme in comparison to the native TPL ( Figure 3A). We also found that heterologous expression changed the regioselectivity of the TPL without affecting the stereospecificity contrary to the N-tag extension which retained the regioselectivity of the TPL and changed the stereospecificity at high surface pressures. These results show that the recombinant expression and the His tag extension could affect the interfacial properties of expressed lipases.
All in all, expressed lipolytic enzymes must be well characterized to be sure that they retain the same properties as the native ones, especially when they have a therapeutic purpose.