- Email: [email protected]
Expression in Pichia pastoris of Candida antarctica Lipase B and Lipase B Fused to a Cellulose-Binding Domain
Protein Expression and Purification 21, 386–392 (2001) doi:10.1006/prep.2000.1387, available online at http://www.idealibrary.com on
Expression in Pichia pastoris of Candida antarctica Lipase B and Lipase B Fused to a Cellulose-Binding Domain Johanna C. Rotticci-Mulder, Malin Gustavsson, Mats Holmquist, Karl Hult, and Mats Martinelle1 Department of Biotechnology, Royal Institute of Technology, SE-100 44 Stockholm, Sweden
Received September 7, 2000, and in revised form December 11, 2000
Candida antarctica lipase B (CALB) and C. antarctica lipase B fused to a cellulose-binding domain (CBD– CALB) were expressed functionally in the methylotrophic yeast Pichia pastoris. The cellulosebinding domain originates from cellulase A of the anaerobic rumen fungus Neocallimastix patriciarum. The genes were fused to the ␣-factor secretion signal sequence of Saccharomyces cerevisiae and placed under the control of the alcohol oxidase gene (AOX1) promoter. The recombinant proteins were secreted into the culture medium reaching levels of approximately 25 mg/L. The proteins were purified using hydrophobic interaction chromatography and gel filtration with an overall yield of 69%. Results from endoglycosidase H digestion of the proteins showed that CALB and CBD–CALB were N-glycosylated. The specific hydrolytic activities of recombinant CALB and CBD–CALB were identical to that reported for CALB isolated from its native source. The fusion of the CBD to the lipase resulted in a greatly enhanced binding toward cellulose for CBD–CALB compared with that for CALB. 䉷 2001 Academic Press
Triacylglycerol lipases (EC 3.1.1.3) are used for applied purposes as efficient catalysts in numerous industries, such as the detergent industry, oleochemistry, food industries, and fine chemical preparations (1). Lipases are carboxylic ester hydrolases that catalyze the hydrolysis of triglycerides. Most lipases display a large increase in activity when adsorbed to a water–lipid interface, a property referred to as interfacial activation (2). This characteristic has been shown to be associated with conformational changes in the protein, creating an To whom correspondence should be addressed. Fax: ⫹46 8 224601. E-mail: [email protected]. 1
386
open, substrate-accessible active site. However, some lipases show minor or no interfacial activation and display both esterase and lipase activity (3–6). All triacylglycerol lipases, with known three-dimensional crystal structure, have a similar general architecture and are members of the ␣/-hydrolase fold family of proteins (7). Candida antarctica lipase B (CALB)2 is a very efficient catalyst with many applications, for example, stereoselective transformations and polyester synthesis (8). This lipase has a solvent accessible active site (9) and displays no interfacial activation (6). The active site is a narrow funnel and therefore CALB displays higher activity toward carboxylic acid esters such as ethyl octanoate than toward triglycerides (10). In organic media CALB displays activities that are comparable with those found in water and the kinetic mechanism follows a bi-bi ping-pong mechanism with competitive substrate inhibition by the acyl acceptor (11). The enzyme has been functionally expressed in Aspergillus oryzae for large industrial scale purposes (12). For our research purposes we have sought a protein expression system that is convenient to work with and quick to transform and we have found this in Pichia pastoris. Lipases have a great potential in the pulp and paper industry for pitch removal and deinking processes as reviewed by Bajpaj (13). Unfortunately the use of enzymes adds a high cost to the papermaking process. This could be reduced by allowing the enzyme to react more efficiently; one way could be to anchor the enzyme to its substrate. Most enzymes acting on cellulose have two separate domains, a catalytic and a cellulose-binding domain 2 Abbreviations used: CALB, Candida antarctica lipase B; CBD, cellulose-binding domain; PCR, polymerase chain reaction; AOX1, alcohol oxidase gene.
1046-5928/01 $35.00 Copyright 䉷 2001 by Academic Press All rights of reproduction in any form reserved.
EXPRESSION OF Candida antarctica LIPASE B
(CBD) usually connected by a flexible linker (14). The main function of the binding domain is to anchor the catalytic domain close to its insoluble substrate, thereby increasing the local enzyme concentration. The family I CBD is a small wedge-shaped protein (33–40 amino acids) with one surface that binds to the cellulose. Several enzymes have been fused to CBDs for simplified and efficient immobilization onto cellulose (15, 16) or for more efficient catalysis at the cellulose surface (17, 18). The CBD used here comes from a cellulase isolated from the anaerobic rumen fungus Neocallimastix patriciarum (19). This CBD shows 57% sequence identity toward the CBD from Trichoderma reesei cellobiohydrolase I, for which the three-dimensional structure has been determined (20). In order to perform genetic engineering on CALB a simple and efficient expression system is needed. Lipases from Geotrichum candidum, Rhizopus oryzae, Candida rugosa, human pancreatic triglyceride lipase, and human bile salt stimulated lipase have been expressed in P. pastoris with good results (21–25). In this study we show that C. antarctica lipase B, wild type and when fused to a cellulose-binding domain (CBD– CALB), can be produced functionally in P. pastoris. The lipase functionality of CALB and CBD–CALB was analyzed using a standard lipase assay and an active-site titration method. The CBD was shown to be functional in a binding study using crystalline cellulose (Avicel).
387
amplification using the primers fCALBXhoI (5⬘CTCTCTCCTCGAGAAAAGAGAGGCTGAAGCTCTACCTTCCGGTTCGGACC) and rCALBNotI (5⬘CTCTCCGCGGCCGCTTAGGGGGTGACGATGCCG). This introduced a XhoI site at the 5⬘-end and a NotI-site at the 3⬘-end of the CALB gene. The PCR product was digested with XhoI and NotI and ligated to the vector YpDC541 (26) digested with the same restriction enzymes. This resulted in the expression vector YpCALB (Fig. 1). YpDC541 is the pPIC9 vector (Invitrogen) in which a PstI site in the ampicillin resistance gene is deleted and a kanamycin resistance gene and a f1 intergenic region are added as described by Holmquist et al. (26). All amplifications by PCR were performed by using Pfu polymerase (Stratagene, La Jolla, CA). For the YpCBDCALB vector construction (Fig. 1) the CBD was taken from cellulase A of the anaerobic fungus N. patriciarum using the vector pTCL. In this construct the CBD gene has a sequence coding for a FLAG peptide (FLAG is an affinity tail) and a linker, based on the T. reesei cellobiohydrolase II linker sequence, fused to the 5⬘- and 3⬘-end, respectively (19). The FLAG–CBDlinker fragment was PCR amplified using the primers fCBDXhoI (5⬘CTTCCTCGAGAAAAGAGAGGCTGAAGCTGATGATTACAAAGACGAT) and rCBDEcoRI (5⬘CTTCGAATTCACCATTGTTAACACG) for the introduction of a XhoI site at the 5⬘-end and an EcoRI site
MATERIALS AND METHODS Strains and Culture Media Escherichia coli strain TOP10F⬘ (Invitrogen, Carlsbad, CA) was used for vector construction work and P. pastoris strain SMD1168 (his4, pep4) (Invitrogen) was used for the expression of CALB and CBD–CALB. E. coli was grown in LB-Amp medium (10 g tryptone, 5 g yeast extract, 5 g NaCl, 1 mL 1 M NaOH, 50 mg ampicillin in 1 L water). P. pastoris was grown in YPD medium (10 g yeast extract, 20 g peptone, 20 g dextrose per liter) or BMGY medium (10 g yeast extract, 20 g peptone, 13.4 g yeast nitrogen base, 0.4 mg biotin, 10 mL glycerol, and 100 mL 1 M K2HPO4/KH2PO4, pH 6.0 per liter). BMMY medium (10 g yeast extract, 20 g peptone, 13.4 g yeast nitrogen base, 0.4 mg biotin, 5 mL methanol, and 100 mL 1 M K2HPO4/KH2PO4, pH 6.0 per liter) was used for induction. For selection of transformants, RD His⫺ agar plates were used (186 g sorbitol, 20 g agar, 20 g dextrose, 13.4 g yeast nitrogen base, 0.2 mg biotin, 50 mg amino acid mix without histidine per liter). Vector Construction and Transformation The C. antarctica lipase B gene (plasmid pMT1335) was isolated by polymerase chain reaction (PCR)
FIG. 1. Schematic overview of the constructed plasmids coding for CALB and CBD–CALB. 5⬘AOX1 promoter is the methanol-inducible promoter, ␣-factor is the secretion signal sequence from Saccharomyces cerevisiae, 3⬘AOX1 terminator is the transcription terminator region, His4 is the selection marker in Pichia pastoris, Kan res is the kanamycin resistance gene functional in P. pastoris, and ColE1 ori is the origin of replication and Amp res is the ampicilin resistance gene functional in E. coli. StuI is the restriction site used to linearize the plasmid before electroporation into P. pastoris.
388
ROTTICCI-MULDER ET AL.
at the 3⬘-end of the gene. The lipase gene was amplified using the primers fCALBEcoRI (5⬘CTCTCTCCTCGAGAAAAGAGAGGCTGAAGCTCTACCTTCCGGTTCGGACC) and rCALBNotI for the introduction of an EcoRI and a NotI site at the 5⬘- and 3⬘-end, respectively. Each PCR product was digested with the corresponding restriction enzymes and ligated into the vector YpDC541 that was digested with the same enzymes. The resulting plasmid constructs, YpCALB and YpCBDCALB, were transformed into E. coli and transformants were selected on LB-Amp agar plates. The CALB and CBD– CALB genes were sequenced on both strands using the Sanger method (27) with dye terminators and were analyzed on an ABI prism 377 DNA sequencer (Perkin– Elmer, Wellesley, MA). The plasmids were linearized with StuI and electroporated (Bio-Rad Gene Pulser II, Hercules, CA) into electrocompetent P. pastoris cells (SMD1168), spread on RD His⫺ plates, and incubated at 30⬚C. The yeast cells were made electrocompetent according to the protocols supplied by Invitrogen. Screening of P. pastoris transformants for CALB and CBD–CALB secretion was performed using colony blotting following the procedure described by Holmquist et al. (21) using antibodies directed towards CALB. Cultivation of P. pastoris and Lipase Expression
method described by Patkar et al. (28). The pooled fractions containing the lipase were further purified by gel filtration on Superdex 75 (Amersham Pharmacia Biotech, Uppsala, Sweden) using 20 mM 3-(N-morpholino)propanesulfonic acid (Mops–KOH) buffer (pH 7.5). Dilution series of CALB and CBD–CALB (⑀(CALB) ⫽ 40,330 M⫺1 cm⫺1, ⑀(CBD–CALB) ⫽ 53,350 M⫺1 cm⫺1, calculated with tools at Swiss Prot Expasy, http://expasy.ncuge.cn/ cgi-bin/protparam) ranging from 5 to 20 M were prepared using 25 M stock solutions (20 mM Mops buffer, pH 7.5) of the enzymes. Deglycosylation Endoglycosidase H and endoglycosidase F (Boehringer–Mannheim, Mannheim, Germany) were used to cleave off N-linked carbohydrates from CALB and CBD–CALB produced in P. pastoris. Digestion was performed according to the manufacturer’s instructions under reducing conditions. The deglycosylated proteins were analyzed by means of SDS–PAGE and Western blot with CALB and anti-FLAG M2 antibodies (Sigma). Active-Site Titration of Recombinant Lipase Active-site titration of the purified lipase was performed using a methyl p-nitrophenyl n-hexylphosphonate inhibitor in order to determine the concentration of active enzyme (29). The active-site concentration was determined by measuring the concentration of released p-nitrophenolate spectrophotometrically at 25⬚C and 400 nm (⑀ ⫽ 16,620 M⫺1 cm⫺1).
Five-hundred milliliters of BMGY in a 5000-mL Eflask was inoculated with 1 mL of an overnight yeast culture in YPD and grown overnight at 28⬚C, 300 rpm. The medium was changed for 500 mL BMMY to induce for lipase expression. Methanol was added to the culture medium to a final concentration of 0.5%(v/v) every 24 h for the following 3 days. The sample was collected by separating the culture medium from the cells by centrifugation. One aliquot of the sample was taken and concentrated approximately 15 times using Macroseps (10K, Pall Filtron, Ann Arbor, MI). The concentrated sample was analyzed by SDS–PAGE (Bio-Rad, Protean II) on a 12% polyacrylamide gel. Western blot analysis was performed using CALB and anti-FLAG M2 antibodies (Sigma, St. Louis, MO).
The lipase hydrolytic activity toward tributyrin was measured using a pH-stat (TIM900 Titration Manager, Radiometer, Denmark) at 25⬚C and pH 7.5 by titration of the released fatty acid with 100 mM sodium hydroxide. The substrate solution (0.2 M tributyrin, 2% gum arabicum, 0.2 M CaCl2) was emulsified by sonication (Branson 250, 30 W) for 1 min. The reaction was started by the addition of enzyme to the substrate emulsion.
Purification of Recombinant Lipase
Cellulose-Binding Study
Ammonium acetate was slowly added to the culture medium to a final concentration of 0.8 M. This solution was loaded on a butyl–Sepharose Fast Flow column (Pharmacia Biotech, Uppsala, Sweden), previously equilibrated with phosphate buffer (50 mM, pH 6) containing ammonium acetate (0.8 M). The column was washed with 5 column volumes of the phosphate buffer containing ammonium acetate. The protein was eluted with a gradient from 100% phosphate buffer containing ammonium acetate to 100% water. The hydrophobic interaction chromatography step is based on the
Crystalline cellulose (Avicel) was washed with water before use to remove soluble components. Lipase dilutions were mixed with an equal volume of Avicel suspension [2% (w/v) in Mops buffer (pH 7.5) containing 1% bovine serum albumin] and incubated for 3 h at 4⬚C while mixing. Reference samples containing buffer instead of cellulose suspension were made and treated under the same conditions. Samples were centrifuged and the supernatants containing the unbound protein were separated from the cellulose pellet. The amount of free lipase present in a sample was determined by
Lipase Activity Assay
EXPRESSION OF Candida antarctica LIPASE B
389
comparing the lipase activity in the supernatant with the activity of the corresponding reference sample. RESULTS AND DISCUSSION Yeast expression vectors containing the genes for CALB or CBD–CALB were constructed and transformed into P. pastoris (SMD1168). The coding and noncoding strands of the CALB and CBD–CALB genes were sequenced and showed the expected sequences. The genes were fused to the ␣-factor secretion signal sequence of Saccharomyces cerevisiae and placed under the control of the methanol inducible alcohol oxidase gene (AOX1) promoter. Lipase-secreting P. pastoris transformants were identified by colony blotting using lipase antibodies and 80% of the transformants were found to secrete the recombinant lipase. The cell density and lipase activity were measured as a function of time after induction with methanol (Fig. 2). At the start of the induction phase a high increase of cell density was measured, probably due to the remaining glycerol from the cell pellet; after 20 h this was seen to stabilize in a small constant increase. The lipase activity increased steadily, even after 80 h of induction with methanol. Before induction and in the negative control no lipase activity was found. Concentrated sample from the culture with cells transformed with YpCALB contained CALB as a major 36-kDa protein as seen from SDS–PAGE and Western blot analyses (Fig. 3, lanes 1 and 5). The calculated molecular weight of CALB is 33 kDa; the higher molecular weight obtained could indicate glycosylation. SDS–PAGE analysis of the medium from YpCBDCALB transformants revealed a smear of high-molecular-weight proteins from 45 to 75 kDa and a 36-kDa protein. The high-molecular-weight smear resolves as two
FIG. 2. Time course study of CALB production. The cell density (䡩) measured as a function of cultivation time at 600 nm. The lipase activity (䡲) during cultivation in mol min⫺1 mL⫺1 using tributyrin as a substrate in the pH-stat.
FIG. 3. SDS–PAGE (left) and Western blot (right) analysis of recombinant Candida antarctica lipase B (CALB) and CALB fused to a cellulose-binding domain (CBD–CALB) produced in P. pastoris. SDS– PAGE was performed on a 12% polyacrylamide gel and developed using Coomassie brilliant blue. The Western blot was developed using CALB antibodies and a secondary antibody conjugated with alkaline phosphatase. Lane Mw, molecular weight marker (kDa); lanes 1 and 5, concentrated sample CALB cultivation; lanes 2 and 6, CALB purified using hydrophobic interaction chromatography (HIC) and gel filtration (GF); lanes 3 and 7, concentrated sample CBD–CALB cultivation; lanes 4 and 8, CBD–CALB purified using HIC and GF.
major bands at lower protein concentration as shown in Fig. 3, lane 3. The 36-kDa protein corresponds in size to the recombinant CALB produced in P. pastoris. The theoretical molecular weight of CBD–CALB is 42 kDa; the higher molecular weight seen on SDS–PAGE indicates that the linker or CBD is glycosylated. Western blot analysis using antibodies directed toward CALB displayed a similar pattern as the SDS– PAGE analysis. The Western blot though revealed some breakdown of the lipase in the sample (Fig. 3, lanes 5 and 7). In the negative control, SMD1168 transformed with the plasmid YpDC541, not containing any insert, there was no reaction with the antibodies directed toward CALB. A 36-kDa protein indicating CALB without CBD was observed in the CBD–CALB preparations using CALB antibodies (Fig. 3, lane 5). Using anti-FLAG M2 antibodies, the high-molecular-weight smear was observed but not the 36-kDa band (Fig. 4, lane 3). This showed that the N-terminal of CBD–CALB was deleted in the 36-kDa protein. Presumably this was a result from proteolysis in the linker of CBD–CALB, this despite the use of the protease-deficient P. pastoris strain SMD 1168. It is known from literature that the strain can loose its protease deficiency, but this was not the reason here as checked using an agar-plate assay (30), which confirmed that our transformants were still protease deficient. Future research will focus on stabilizing this fusion protein construction. CALB and CBD–CALB were purified using hydrophobic interaction chromatography (28) and gel filtration yielding highly pure CALB and CBD–CALB as
390
ROTTICCI-MULDER ET AL.
FIG. 4. Western blot analysis of recombinant Candida antarctica lipase B (left) and CALB fused to a cellulose-binding domain (right) produced in Pichia pastoris. CALB was detected with CALB antibodies. CBD–CALB was detected using anti-FLAG M2 antibodies. Lane Mw, molecular weight marker (kDa); lane 1, CALB purified with hydrophobic interaction chromatography (HIC) and gel filtration (GF); lane 2, CALB purified with HIC and GF after endoglycosidase H treatment; lane 3, CBD–CALB purified with HIC and GF; lane 4, CBD–CALB purified with HIC and GF after treatment with endoglycosidase H.
observed after SDS–PAGE and Western blot analysis (Fig. 3, lanes 2 and 4). The 70-kDa protein present in the culture medium (Fig. 3, lane 1) could not be removed by hydrophobic interaction chromatography. This protein has a similar size to the commercially available alcohol oxidase from P. pastoris (Sigma). By the introduction of a subsequent gel filtration step we were able to remove the 70-kDa contaminant from the recombinant lipase. CALB was also separated from the CBD– CALB using gel filtration and only trace amounts of CALB were left as can be seen in Fig. 3, lanes 4 and 8. The protein levels of CALB and CBD–CALB after purification were measured spectrophotometrically at 280 nm, to give expression levels of approximately 25 mg/L. The yield from each of the purification steps can be seen in Table 1. One N-glycosylation site is present in CALB (9) and
FIG. 5. Adsorption isotherms of Candida antarctica lipase B (CALB) (䡩) and the lipase fused to a cellulose-binding domain (CBD– CALB) (䡲) on crystalline cellulose (Avicel) in 20 mM Mops buffer, pH 7.5, 4⬚C for 3 h.
one is predicted in the linker of CBD–CALB. Endoglycosidase H treatment of the purified recombinant proteins reduced the molecular weight of CALB (Fig. 4, lane 2) and of CBD–CALB (Fig., 4, lane 4). CALB was found to receive N-glycosylation when produced in A. oryzae and in the native host (12). For deglycosylation endoglycosidase F was used by Hoegh et al. We used endoglycosidase F and endoglycosidase H, giving the same decrease in molecular weight, confirming N-glycosylation on the lipase and/or the CBD-linker region. The electrophoretic migration on SDS–PAGE was the same for the lipase produced in A. oryzae as for the lipase produced in P. pastoris. From this we assume that the lipases produced in the different hosts receive similar extent of glycosylation. The enzymatic properties of the lipases were studied using a standard tributyrin assay and an activesite titration method (29). The active-site inhibitor methyl p-nitrophenyl n-hexylphosphonate was used
TABLE 2 TABLE 1 Activity from Produced Candida antarctica Lipase B Given in Micromoles of Substrate (Tributyrin) Hydrolyzed per Minute and per Milliliter of Lipase Solution
Medium Pooled HIC fractions Pooled GF fractions
Activity (mol min⫺1 mL⫺1)
Yield (%)
5.7 27 170
100 78 69
Note. The yield is calculated using the activity and volume of the lipase solutions.
Specific Hydrolytic Activity of Recombinant Candida antarctica Lipase B (CALB) and Lipase B Fused to a Cellulose-Binding Domain (CBD–CALB) toward Tributyrin Enzyme CALB CBD–CALB CALB-Novo
Specific activity (s⫺1) 300 ⫾ 14 290 ⫾ 12 300 ⫾ 18
Note. CALB-Novo is a lyophilized CALB preparation produced in Aspergillus oryzae by Novo-Nordisk, Denmark. The standard deviations are based on five measurements.
EXPRESSION OF Candida antarctica LIPASE B
to quantify the concentration of active enzyme, enabling a good comparison between the different preparations. CALB and CBD–CALB produced in P. pastoris and CALB obtained from Novo-Nordisk displayed the same specific hydrolytic activities (Table 2). These activities also agreed with the activity displayed by CALB isolated from the original source (28). These results showed that the CBD did not interfere with the activity of the lipase. The cellulose-binding ability of CBD–CALB was studied in an adsorption isotherm using crystalline cellulose (Avicel) (Fig. 5). The binding toward Avicel was enhanced for CBD–CALB compared with CALB, showing that the binding domain was functional. Bound CALB, but not CBD–CALB, was possible to remove from the crystalline cellulose (Avicel) by washing with water. Thus, the fusion of a CBD to the lipase appears to be an efficient way to anchor the lipase to the cellulose. ACKNOWLEDGMENTS Financial support from the NUTEK Centre for Bioprocess Technology (CBioPT) is gratefully acknowledged. Novo-Nordisk A/S, Bagsvaerd, Denmark, is acknowledged for providing the Candida antarctica lipase B gene and CALB antibodies. CSIRO, Brisbane, Australia is acknowledged for their kind gift of the plasmid containing the cellulose-binding domain. We thank Stuart Denman for valuable discussions and comments on the manuscript. REFERENCES 1. Schmid, R. D., and Verger, R. (1998) Lipases: Interfacial enzymes with attractive applications. Angew. Chem. Int. Ed. 37, 1608– 1633. 2. Sarda, L., and Desnuelle, P. (1958) Action de la lipase pancreatique sur les esters en emulsion. Biochim. Biophys. Acta 30, 513–521. 3. Lesuisse, E., Schanck, K., and Colson, C. (1993) Purification and preliminary characterization of the extracellular lipase of Bacillus subtilis 168, an extremely basic pH-tolerant enzyme. Eur. J. Biochem. 216, 155–160. 4. Jaeger, K.-E., Ransac, S., Koch, H., Ferrato, F., and Dijkstra, B. W. (1993) Topological characterization and modeling of the 3D structure of lipase from Pseudomonas aeruginosa. FEBS Lett. 332, 143–149. 5. Hjorth, A., Carriere, F., Cudrey, C., Wo¨ldike, H., Boel, E., Lawason, D. M., Ferrato, F., Cambillau, C., Dodson, G. G., Thim, L., and Verger, R. (1993) A structural domain (the lid) found in pancreatic lipases is absent in the guinea pig (phospho)lipase. Biochemistry 32, 4702–4707. 6. Martinelle, M., Holmquist, M., and Hult, K. (1995) On the interfacial activation of Candida antarctica lipase A and B as compared with Humicola lanuginosa lipase. Biochim. Biophys. Acta 1258, 272–276. 7. Ollis, D. L., Cheah, E., Cygler, M., Dijkstra, B., Frolow, F., Franken, S. M., Harel, M., Remington, S. J., Silman, I., Schrag, J., Sussman, J. L., Verschueren, K. H. G., and Goldman, A. (1992) The ␣/ hydrolase fold. Protein Eng. 5, 197–211. 8. Anderson, E. M., Larsson, K. M., and Kirk, O. (1998) One biocatalyst—Many applications: The use of Candida antarctica B-lipase in organic synthesis. Biocatal. Biotransform. 16, 181–204.
391
9. Uppenberg, J., Hansen, M. T., Patkar, S., and Jones, A. (1994) The sequence, crystal structure determination and refinement of two crystal forms of lipase B from Candida antarctica. Structure 2, 293–308. 10. Martinelle, M. (1995) “Lipases from Candida antarctica and Humicola lanuginosa: Structure–Activity Relationships and Applied Catalysis,” Ph.D. thesis, Royal Institute of Technology, Stockholm. ISBN 91-7170-720-4. 11. Martinelle, M., and Hult, K. (1995) Kinetics of acyl transfer reactions in organic media catalysed by Candida antarctica lipase B. Biochim. Biophys. Acta 1251, 191–197. 12. Hoegh, I., Patkar, S., Halkier, T., and Hansen, M. T. (1995) Two lipases from Candida antarctica: Cloning and expression in Aspergillus oryzae. Can. J. Bot. 73, S869–S875. 13. Bajpaj, P. (1999) Application of enzymes in the pulp and paper industry. Biotechnol. Prog. 15, 147–157. 14. Linder, M., and Teeri, T. (1997) The roles and function of cellulosebinding domains. J. Biotechnol. 57, 15–28. 15. Shpigel, E., Goldlust, A., Efroni, G., Avraham, A., Eshel, A., Dekel, M., and Shoseyov, O. (1999) Immobilization of recombinant heparinase I fused to cellulose-binding domain. Biotechnol. Bioeng. 65, 17–23. 16. Ong, E., Gilkes, N. R., Miller, R. C., Warren, R. A. J., and Kilburn, D. G. (1991) Enzyme immobilization using a cellulose-binding domain: properties of a -glucosidase fusion protein. Enzyme Microb. Technol. 13, 59–65. 17. Maglione, G., Matsushita, O., Russel, J. B., and Wilson, D. B. (1992) Properties of a genetically reconstructed Prevotella ruminicola endoglucanase. Appl. Environ. Microb. 58, 3593–3597. 18. Karita, S., Sakka, K., and Ohmiya, K. (1996) Cellulose-binding domains confer an enhanced activity against insoluble cellulose to Ruminococcus albus endoglucanase IV. J. Ferment. Bioeng. 81, 553–556. 19. Denman, S., Xue, G. P., and Patel, B. (1996) Characterization of a Neocallimastix patriciarum cellulase cDNA (celA) homologous to Trichoderma reesei cellobiohydrolase II. Appl. Environ. Microb. 62, 1889–1896. 20. Kraulis, P. J., Clore, G. M., Nilges, M., Jones, T. A., Petterson, G., Knowles, J., and Gronenborn, A. M. (1989) Determination of the three-dimensional solution structure of the C-terminal domain of cellobiohydrolase I from Trichoderma reesei: A study using nuclear magnetic resonance and hybrid distance geometry– dynamical simulated annealing. Biochemistry 28, 7241–7257. 21. Holmquist, M., Tessier, D. C., and Cygler, M. (1997) High-level production of recombinant Geotrichum candidum lipases in yeast Pichia pastoris. Protein Express. Purif. 11, 35–40. 22. Minning, S., Schmidt-Dannert, C., and Schmid, R. D. (1998) Functional expression of Rhizopus oryzae lipase in Pichia pastoris: High-level production and some properties. J. Biotechnol. 66, 147–156. 23. Brocca, S., Schmidt-Dannert, C., Lotti, M., Alberghina, L., and Schmid R. D. (1998) Design, total synthesis, and functional overexpression of the Candida rugosa lip1 gene coding for a major industrial lipase. Protein Sci. 7, 1415–1422. 24. Yang, Y., and Lowe, M. E. (1998) Human pancreatic triglyceride lipase expressed in yeast cells: Purification and characterization. Protein Express. Purif. 13, 36–40. 25. Sahasrabudhe, A. V., Solapure S. M., Khurana, R., Suryanareayan, V., Ravishankar, S., deSousa, S. M., and Das, G. (1998) Production of recombinant human bile salt stimulated lipase and its variant in Pichia pastoris. Protein Express. Purif. 3, 425–433. 26. Holmquist, M., Tessier, D. C., and Cygler, M. (1997) Identification of residues essential for differential fatty acid specificity of
392
ROTTICCI-MULDER ET AL.
Geotrichum candidum lipases I and II. Biochemistry 36, 15019– 15025. 27. Sanger, F., Nicklen, S, and Caulson, A. R. (1977) DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463–5467. 28. Patkar, S. A., Bjo¨rkling, F., Zundel, M., Schulein, M., Svendsen, A., Heldt-Hansen, H. P., and Gormsen, E. (1993) Purification
of two lipases from Candida antarctica and their inhibition by various inhibitors. Ind. J. Chem. 32B, 76–80. 29. Rotticci, D., Norin, T., Hult, K., and Martinelle, M. (2000) An active-site titration method for lipases. Biochim. Biophys. Acta 1483, 132–140. 30. Jones, E. W. (1977) Proteinase mutants of Saccharomyces cerevisiae. Genetics 85, 23–33.
Expression in Pichia pastoris of Candida antarctica Lipase B and Lipase B Fused to a Cellulose-Binding Domain Johanna C. Rotticci-Mulder, Malin Gustavsson, Mats Holmquist, Karl Hult, and Mats Martinelle1 Department of Biotechnology, Royal Institute of Technology, SE-100 44 Stockholm, Sweden
Received September 7, 2000, and in revised form December 11, 2000
Candida antarctica lipase B (CALB) and C. antarctica lipase B fused to a cellulose-binding domain (CBD– CALB) were expressed functionally in the methylotrophic yeast Pichia pastoris. The cellulosebinding domain originates from cellulase A of the anaerobic rumen fungus Neocallimastix patriciarum. The genes were fused to the ␣-factor secretion signal sequence of Saccharomyces cerevisiae and placed under the control of the alcohol oxidase gene (AOX1) promoter. The recombinant proteins were secreted into the culture medium reaching levels of approximately 25 mg/L. The proteins were purified using hydrophobic interaction chromatography and gel filtration with an overall yield of 69%. Results from endoglycosidase H digestion of the proteins showed that CALB and CBD–CALB were N-glycosylated. The specific hydrolytic activities of recombinant CALB and CBD–CALB were identical to that reported for CALB isolated from its native source. The fusion of the CBD to the lipase resulted in a greatly enhanced binding toward cellulose for CBD–CALB compared with that for CALB. 䉷 2001 Academic Press
Triacylglycerol lipases (EC 3.1.1.3) are used for applied purposes as efficient catalysts in numerous industries, such as the detergent industry, oleochemistry, food industries, and fine chemical preparations (1). Lipases are carboxylic ester hydrolases that catalyze the hydrolysis of triglycerides. Most lipases display a large increase in activity when adsorbed to a water–lipid interface, a property referred to as interfacial activation (2). This characteristic has been shown to be associated with conformational changes in the protein, creating an To whom correspondence should be addressed. Fax: ⫹46 8 224601. E-mail: [email protected]. 1
386
open, substrate-accessible active site. However, some lipases show minor or no interfacial activation and display both esterase and lipase activity (3–6). All triacylglycerol lipases, with known three-dimensional crystal structure, have a similar general architecture and are members of the ␣/-hydrolase fold family of proteins (7). Candida antarctica lipase B (CALB)2 is a very efficient catalyst with many applications, for example, stereoselective transformations and polyester synthesis (8). This lipase has a solvent accessible active site (9) and displays no interfacial activation (6). The active site is a narrow funnel and therefore CALB displays higher activity toward carboxylic acid esters such as ethyl octanoate than toward triglycerides (10). In organic media CALB displays activities that are comparable with those found in water and the kinetic mechanism follows a bi-bi ping-pong mechanism with competitive substrate inhibition by the acyl acceptor (11). The enzyme has been functionally expressed in Aspergillus oryzae for large industrial scale purposes (12). For our research purposes we have sought a protein expression system that is convenient to work with and quick to transform and we have found this in Pichia pastoris. Lipases have a great potential in the pulp and paper industry for pitch removal and deinking processes as reviewed by Bajpaj (13). Unfortunately the use of enzymes adds a high cost to the papermaking process. This could be reduced by allowing the enzyme to react more efficiently; one way could be to anchor the enzyme to its substrate. Most enzymes acting on cellulose have two separate domains, a catalytic and a cellulose-binding domain 2 Abbreviations used: CALB, Candida antarctica lipase B; CBD, cellulose-binding domain; PCR, polymerase chain reaction; AOX1, alcohol oxidase gene.
1046-5928/01 $35.00 Copyright 䉷 2001 by Academic Press All rights of reproduction in any form reserved.
EXPRESSION OF Candida antarctica LIPASE B
(CBD) usually connected by a flexible linker (14). The main function of the binding domain is to anchor the catalytic domain close to its insoluble substrate, thereby increasing the local enzyme concentration. The family I CBD is a small wedge-shaped protein (33–40 amino acids) with one surface that binds to the cellulose. Several enzymes have been fused to CBDs for simplified and efficient immobilization onto cellulose (15, 16) or for more efficient catalysis at the cellulose surface (17, 18). The CBD used here comes from a cellulase isolated from the anaerobic rumen fungus Neocallimastix patriciarum (19). This CBD shows 57% sequence identity toward the CBD from Trichoderma reesei cellobiohydrolase I, for which the three-dimensional structure has been determined (20). In order to perform genetic engineering on CALB a simple and efficient expression system is needed. Lipases from Geotrichum candidum, Rhizopus oryzae, Candida rugosa, human pancreatic triglyceride lipase, and human bile salt stimulated lipase have been expressed in P. pastoris with good results (21–25). In this study we show that C. antarctica lipase B, wild type and when fused to a cellulose-binding domain (CBD– CALB), can be produced functionally in P. pastoris. The lipase functionality of CALB and CBD–CALB was analyzed using a standard lipase assay and an active-site titration method. The CBD was shown to be functional in a binding study using crystalline cellulose (Avicel).
387
amplification using the primers fCALBXhoI (5⬘CTCTCTCCTCGAGAAAAGAGAGGCTGAAGCTCTACCTTCCGGTTCGGACC) and rCALBNotI (5⬘CTCTCCGCGGCCGCTTAGGGGGTGACGATGCCG). This introduced a XhoI site at the 5⬘-end and a NotI-site at the 3⬘-end of the CALB gene. The PCR product was digested with XhoI and NotI and ligated to the vector YpDC541 (26) digested with the same restriction enzymes. This resulted in the expression vector YpCALB (Fig. 1). YpDC541 is the pPIC9 vector (Invitrogen) in which a PstI site in the ampicillin resistance gene is deleted and a kanamycin resistance gene and a f1 intergenic region are added as described by Holmquist et al. (26). All amplifications by PCR were performed by using Pfu polymerase (Stratagene, La Jolla, CA). For the YpCBDCALB vector construction (Fig. 1) the CBD was taken from cellulase A of the anaerobic fungus N. patriciarum using the vector pTCL. In this construct the CBD gene has a sequence coding for a FLAG peptide (FLAG is an affinity tail) and a linker, based on the T. reesei cellobiohydrolase II linker sequence, fused to the 5⬘- and 3⬘-end, respectively (19). The FLAG–CBDlinker fragment was PCR amplified using the primers fCBDXhoI (5⬘CTTCCTCGAGAAAAGAGAGGCTGAAGCTGATGATTACAAAGACGAT) and rCBDEcoRI (5⬘CTTCGAATTCACCATTGTTAACACG) for the introduction of a XhoI site at the 5⬘-end and an EcoRI site
MATERIALS AND METHODS Strains and Culture Media Escherichia coli strain TOP10F⬘ (Invitrogen, Carlsbad, CA) was used for vector construction work and P. pastoris strain SMD1168 (his4, pep4) (Invitrogen) was used for the expression of CALB and CBD–CALB. E. coli was grown in LB-Amp medium (10 g tryptone, 5 g yeast extract, 5 g NaCl, 1 mL 1 M NaOH, 50 mg ampicillin in 1 L water). P. pastoris was grown in YPD medium (10 g yeast extract, 20 g peptone, 20 g dextrose per liter) or BMGY medium (10 g yeast extract, 20 g peptone, 13.4 g yeast nitrogen base, 0.4 mg biotin, 10 mL glycerol, and 100 mL 1 M K2HPO4/KH2PO4, pH 6.0 per liter). BMMY medium (10 g yeast extract, 20 g peptone, 13.4 g yeast nitrogen base, 0.4 mg biotin, 5 mL methanol, and 100 mL 1 M K2HPO4/KH2PO4, pH 6.0 per liter) was used for induction. For selection of transformants, RD His⫺ agar plates were used (186 g sorbitol, 20 g agar, 20 g dextrose, 13.4 g yeast nitrogen base, 0.2 mg biotin, 50 mg amino acid mix without histidine per liter). Vector Construction and Transformation The C. antarctica lipase B gene (plasmid pMT1335) was isolated by polymerase chain reaction (PCR)
FIG. 1. Schematic overview of the constructed plasmids coding for CALB and CBD–CALB. 5⬘AOX1 promoter is the methanol-inducible promoter, ␣-factor is the secretion signal sequence from Saccharomyces cerevisiae, 3⬘AOX1 terminator is the transcription terminator region, His4 is the selection marker in Pichia pastoris, Kan res is the kanamycin resistance gene functional in P. pastoris, and ColE1 ori is the origin of replication and Amp res is the ampicilin resistance gene functional in E. coli. StuI is the restriction site used to linearize the plasmid before electroporation into P. pastoris.
388
ROTTICCI-MULDER ET AL.
at the 3⬘-end of the gene. The lipase gene was amplified using the primers fCALBEcoRI (5⬘CTCTCTCCTCGAGAAAAGAGAGGCTGAAGCTCTACCTTCCGGTTCGGACC) and rCALBNotI for the introduction of an EcoRI and a NotI site at the 5⬘- and 3⬘-end, respectively. Each PCR product was digested with the corresponding restriction enzymes and ligated into the vector YpDC541 that was digested with the same enzymes. The resulting plasmid constructs, YpCALB and YpCBDCALB, were transformed into E. coli and transformants were selected on LB-Amp agar plates. The CALB and CBD– CALB genes were sequenced on both strands using the Sanger method (27) with dye terminators and were analyzed on an ABI prism 377 DNA sequencer (Perkin– Elmer, Wellesley, MA). The plasmids were linearized with StuI and electroporated (Bio-Rad Gene Pulser II, Hercules, CA) into electrocompetent P. pastoris cells (SMD1168), spread on RD His⫺ plates, and incubated at 30⬚C. The yeast cells were made electrocompetent according to the protocols supplied by Invitrogen. Screening of P. pastoris transformants for CALB and CBD–CALB secretion was performed using colony blotting following the procedure described by Holmquist et al. (21) using antibodies directed towards CALB. Cultivation of P. pastoris and Lipase Expression
method described by Patkar et al. (28). The pooled fractions containing the lipase were further purified by gel filtration on Superdex 75 (Amersham Pharmacia Biotech, Uppsala, Sweden) using 20 mM 3-(N-morpholino)propanesulfonic acid (Mops–KOH) buffer (pH 7.5). Dilution series of CALB and CBD–CALB (⑀(CALB) ⫽ 40,330 M⫺1 cm⫺1, ⑀(CBD–CALB) ⫽ 53,350 M⫺1 cm⫺1, calculated with tools at Swiss Prot Expasy, http://expasy.ncuge.cn/ cgi-bin/protparam) ranging from 5 to 20 M were prepared using 25 M stock solutions (20 mM Mops buffer, pH 7.5) of the enzymes. Deglycosylation Endoglycosidase H and endoglycosidase F (Boehringer–Mannheim, Mannheim, Germany) were used to cleave off N-linked carbohydrates from CALB and CBD–CALB produced in P. pastoris. Digestion was performed according to the manufacturer’s instructions under reducing conditions. The deglycosylated proteins were analyzed by means of SDS–PAGE and Western blot with CALB and anti-FLAG M2 antibodies (Sigma). Active-Site Titration of Recombinant Lipase Active-site titration of the purified lipase was performed using a methyl p-nitrophenyl n-hexylphosphonate inhibitor in order to determine the concentration of active enzyme (29). The active-site concentration was determined by measuring the concentration of released p-nitrophenolate spectrophotometrically at 25⬚C and 400 nm (⑀ ⫽ 16,620 M⫺1 cm⫺1).
Five-hundred milliliters of BMGY in a 5000-mL Eflask was inoculated with 1 mL of an overnight yeast culture in YPD and grown overnight at 28⬚C, 300 rpm. The medium was changed for 500 mL BMMY to induce for lipase expression. Methanol was added to the culture medium to a final concentration of 0.5%(v/v) every 24 h for the following 3 days. The sample was collected by separating the culture medium from the cells by centrifugation. One aliquot of the sample was taken and concentrated approximately 15 times using Macroseps (10K, Pall Filtron, Ann Arbor, MI). The concentrated sample was analyzed by SDS–PAGE (Bio-Rad, Protean II) on a 12% polyacrylamide gel. Western blot analysis was performed using CALB and anti-FLAG M2 antibodies (Sigma, St. Louis, MO).
The lipase hydrolytic activity toward tributyrin was measured using a pH-stat (TIM900 Titration Manager, Radiometer, Denmark) at 25⬚C and pH 7.5 by titration of the released fatty acid with 100 mM sodium hydroxide. The substrate solution (0.2 M tributyrin, 2% gum arabicum, 0.2 M CaCl2) was emulsified by sonication (Branson 250, 30 W) for 1 min. The reaction was started by the addition of enzyme to the substrate emulsion.
Purification of Recombinant Lipase
Cellulose-Binding Study
Ammonium acetate was slowly added to the culture medium to a final concentration of 0.8 M. This solution was loaded on a butyl–Sepharose Fast Flow column (Pharmacia Biotech, Uppsala, Sweden), previously equilibrated with phosphate buffer (50 mM, pH 6) containing ammonium acetate (0.8 M). The column was washed with 5 column volumes of the phosphate buffer containing ammonium acetate. The protein was eluted with a gradient from 100% phosphate buffer containing ammonium acetate to 100% water. The hydrophobic interaction chromatography step is based on the
Crystalline cellulose (Avicel) was washed with water before use to remove soluble components. Lipase dilutions were mixed with an equal volume of Avicel suspension [2% (w/v) in Mops buffer (pH 7.5) containing 1% bovine serum albumin] and incubated for 3 h at 4⬚C while mixing. Reference samples containing buffer instead of cellulose suspension were made and treated under the same conditions. Samples were centrifuged and the supernatants containing the unbound protein were separated from the cellulose pellet. The amount of free lipase present in a sample was determined by
Lipase Activity Assay
EXPRESSION OF Candida antarctica LIPASE B
389
comparing the lipase activity in the supernatant with the activity of the corresponding reference sample. RESULTS AND DISCUSSION Yeast expression vectors containing the genes for CALB or CBD–CALB were constructed and transformed into P. pastoris (SMD1168). The coding and noncoding strands of the CALB and CBD–CALB genes were sequenced and showed the expected sequences. The genes were fused to the ␣-factor secretion signal sequence of Saccharomyces cerevisiae and placed under the control of the methanol inducible alcohol oxidase gene (AOX1) promoter. Lipase-secreting P. pastoris transformants were identified by colony blotting using lipase antibodies and 80% of the transformants were found to secrete the recombinant lipase. The cell density and lipase activity were measured as a function of time after induction with methanol (Fig. 2). At the start of the induction phase a high increase of cell density was measured, probably due to the remaining glycerol from the cell pellet; after 20 h this was seen to stabilize in a small constant increase. The lipase activity increased steadily, even after 80 h of induction with methanol. Before induction and in the negative control no lipase activity was found. Concentrated sample from the culture with cells transformed with YpCALB contained CALB as a major 36-kDa protein as seen from SDS–PAGE and Western blot analyses (Fig. 3, lanes 1 and 5). The calculated molecular weight of CALB is 33 kDa; the higher molecular weight obtained could indicate glycosylation. SDS–PAGE analysis of the medium from YpCBDCALB transformants revealed a smear of high-molecular-weight proteins from 45 to 75 kDa and a 36-kDa protein. The high-molecular-weight smear resolves as two
FIG. 2. Time course study of CALB production. The cell density (䡩) measured as a function of cultivation time at 600 nm. The lipase activity (䡲) during cultivation in mol min⫺1 mL⫺1 using tributyrin as a substrate in the pH-stat.
FIG. 3. SDS–PAGE (left) and Western blot (right) analysis of recombinant Candida antarctica lipase B (CALB) and CALB fused to a cellulose-binding domain (CBD–CALB) produced in P. pastoris. SDS– PAGE was performed on a 12% polyacrylamide gel and developed using Coomassie brilliant blue. The Western blot was developed using CALB antibodies and a secondary antibody conjugated with alkaline phosphatase. Lane Mw, molecular weight marker (kDa); lanes 1 and 5, concentrated sample CALB cultivation; lanes 2 and 6, CALB purified using hydrophobic interaction chromatography (HIC) and gel filtration (GF); lanes 3 and 7, concentrated sample CBD–CALB cultivation; lanes 4 and 8, CBD–CALB purified using HIC and GF.
major bands at lower protein concentration as shown in Fig. 3, lane 3. The 36-kDa protein corresponds in size to the recombinant CALB produced in P. pastoris. The theoretical molecular weight of CBD–CALB is 42 kDa; the higher molecular weight seen on SDS–PAGE indicates that the linker or CBD is glycosylated. Western blot analysis using antibodies directed toward CALB displayed a similar pattern as the SDS– PAGE analysis. The Western blot though revealed some breakdown of the lipase in the sample (Fig. 3, lanes 5 and 7). In the negative control, SMD1168 transformed with the plasmid YpDC541, not containing any insert, there was no reaction with the antibodies directed toward CALB. A 36-kDa protein indicating CALB without CBD was observed in the CBD–CALB preparations using CALB antibodies (Fig. 3, lane 5). Using anti-FLAG M2 antibodies, the high-molecular-weight smear was observed but not the 36-kDa band (Fig. 4, lane 3). This showed that the N-terminal of CBD–CALB was deleted in the 36-kDa protein. Presumably this was a result from proteolysis in the linker of CBD–CALB, this despite the use of the protease-deficient P. pastoris strain SMD 1168. It is known from literature that the strain can loose its protease deficiency, but this was not the reason here as checked using an agar-plate assay (30), which confirmed that our transformants were still protease deficient. Future research will focus on stabilizing this fusion protein construction. CALB and CBD–CALB were purified using hydrophobic interaction chromatography (28) and gel filtration yielding highly pure CALB and CBD–CALB as
390
ROTTICCI-MULDER ET AL.
FIG. 4. Western blot analysis of recombinant Candida antarctica lipase B (left) and CALB fused to a cellulose-binding domain (right) produced in Pichia pastoris. CALB was detected with CALB antibodies. CBD–CALB was detected using anti-FLAG M2 antibodies. Lane Mw, molecular weight marker (kDa); lane 1, CALB purified with hydrophobic interaction chromatography (HIC) and gel filtration (GF); lane 2, CALB purified with HIC and GF after endoglycosidase H treatment; lane 3, CBD–CALB purified with HIC and GF; lane 4, CBD–CALB purified with HIC and GF after treatment with endoglycosidase H.
observed after SDS–PAGE and Western blot analysis (Fig. 3, lanes 2 and 4). The 70-kDa protein present in the culture medium (Fig. 3, lane 1) could not be removed by hydrophobic interaction chromatography. This protein has a similar size to the commercially available alcohol oxidase from P. pastoris (Sigma). By the introduction of a subsequent gel filtration step we were able to remove the 70-kDa contaminant from the recombinant lipase. CALB was also separated from the CBD– CALB using gel filtration and only trace amounts of CALB were left as can be seen in Fig. 3, lanes 4 and 8. The protein levels of CALB and CBD–CALB after purification were measured spectrophotometrically at 280 nm, to give expression levels of approximately 25 mg/L. The yield from each of the purification steps can be seen in Table 1. One N-glycosylation site is present in CALB (9) and
FIG. 5. Adsorption isotherms of Candida antarctica lipase B (CALB) (䡩) and the lipase fused to a cellulose-binding domain (CBD– CALB) (䡲) on crystalline cellulose (Avicel) in 20 mM Mops buffer, pH 7.5, 4⬚C for 3 h.
one is predicted in the linker of CBD–CALB. Endoglycosidase H treatment of the purified recombinant proteins reduced the molecular weight of CALB (Fig. 4, lane 2) and of CBD–CALB (Fig., 4, lane 4). CALB was found to receive N-glycosylation when produced in A. oryzae and in the native host (12). For deglycosylation endoglycosidase F was used by Hoegh et al. We used endoglycosidase F and endoglycosidase H, giving the same decrease in molecular weight, confirming N-glycosylation on the lipase and/or the CBD-linker region. The electrophoretic migration on SDS–PAGE was the same for the lipase produced in A. oryzae as for the lipase produced in P. pastoris. From this we assume that the lipases produced in the different hosts receive similar extent of glycosylation. The enzymatic properties of the lipases were studied using a standard tributyrin assay and an activesite titration method (29). The active-site inhibitor methyl p-nitrophenyl n-hexylphosphonate was used
TABLE 2 TABLE 1 Activity from Produced Candida antarctica Lipase B Given in Micromoles of Substrate (Tributyrin) Hydrolyzed per Minute and per Milliliter of Lipase Solution
Medium Pooled HIC fractions Pooled GF fractions
Activity (mol min⫺1 mL⫺1)
Yield (%)
5.7 27 170
100 78 69
Note. The yield is calculated using the activity and volume of the lipase solutions.
Specific Hydrolytic Activity of Recombinant Candida antarctica Lipase B (CALB) and Lipase B Fused to a Cellulose-Binding Domain (CBD–CALB) toward Tributyrin Enzyme CALB CBD–CALB CALB-Novo
Specific activity (s⫺1) 300 ⫾ 14 290 ⫾ 12 300 ⫾ 18
Note. CALB-Novo is a lyophilized CALB preparation produced in Aspergillus oryzae by Novo-Nordisk, Denmark. The standard deviations are based on five measurements.
EXPRESSION OF Candida antarctica LIPASE B
to quantify the concentration of active enzyme, enabling a good comparison between the different preparations. CALB and CBD–CALB produced in P. pastoris and CALB obtained from Novo-Nordisk displayed the same specific hydrolytic activities (Table 2). These activities also agreed with the activity displayed by CALB isolated from the original source (28). These results showed that the CBD did not interfere with the activity of the lipase. The cellulose-binding ability of CBD–CALB was studied in an adsorption isotherm using crystalline cellulose (Avicel) (Fig. 5). The binding toward Avicel was enhanced for CBD–CALB compared with CALB, showing that the binding domain was functional. Bound CALB, but not CBD–CALB, was possible to remove from the crystalline cellulose (Avicel) by washing with water. Thus, the fusion of a CBD to the lipase appears to be an efficient way to anchor the lipase to the cellulose. ACKNOWLEDGMENTS Financial support from the NUTEK Centre for Bioprocess Technology (CBioPT) is gratefully acknowledged. Novo-Nordisk A/S, Bagsvaerd, Denmark, is acknowledged for providing the Candida antarctica lipase B gene and CALB antibodies. CSIRO, Brisbane, Australia is acknowledged for their kind gift of the plasmid containing the cellulose-binding domain. We thank Stuart Denman for valuable discussions and comments on the manuscript. REFERENCES 1. Schmid, R. D., and Verger, R. (1998) Lipases: Interfacial enzymes with attractive applications. Angew. Chem. Int. Ed. 37, 1608– 1633. 2. Sarda, L., and Desnuelle, P. (1958) Action de la lipase pancreatique sur les esters en emulsion. Biochim. Biophys. Acta 30, 513–521. 3. Lesuisse, E., Schanck, K., and Colson, C. (1993) Purification and preliminary characterization of the extracellular lipase of Bacillus subtilis 168, an extremely basic pH-tolerant enzyme. Eur. J. Biochem. 216, 155–160. 4. Jaeger, K.-E., Ransac, S., Koch, H., Ferrato, F., and Dijkstra, B. W. (1993) Topological characterization and modeling of the 3D structure of lipase from Pseudomonas aeruginosa. FEBS Lett. 332, 143–149. 5. Hjorth, A., Carriere, F., Cudrey, C., Wo¨ldike, H., Boel, E., Lawason, D. M., Ferrato, F., Cambillau, C., Dodson, G. G., Thim, L., and Verger, R. (1993) A structural domain (the lid) found in pancreatic lipases is absent in the guinea pig (phospho)lipase. Biochemistry 32, 4702–4707. 6. Martinelle, M., Holmquist, M., and Hult, K. (1995) On the interfacial activation of Candida antarctica lipase A and B as compared with Humicola lanuginosa lipase. Biochim. Biophys. Acta 1258, 272–276. 7. Ollis, D. L., Cheah, E., Cygler, M., Dijkstra, B., Frolow, F., Franken, S. M., Harel, M., Remington, S. J., Silman, I., Schrag, J., Sussman, J. L., Verschueren, K. H. G., and Goldman, A. (1992) The ␣/ hydrolase fold. Protein Eng. 5, 197–211. 8. Anderson, E. M., Larsson, K. M., and Kirk, O. (1998) One biocatalyst—Many applications: The use of Candida antarctica B-lipase in organic synthesis. Biocatal. Biotransform. 16, 181–204.
391
9. Uppenberg, J., Hansen, M. T., Patkar, S., and Jones, A. (1994) The sequence, crystal structure determination and refinement of two crystal forms of lipase B from Candida antarctica. Structure 2, 293–308. 10. Martinelle, M. (1995) “Lipases from Candida antarctica and Humicola lanuginosa: Structure–Activity Relationships and Applied Catalysis,” Ph.D. thesis, Royal Institute of Technology, Stockholm. ISBN 91-7170-720-4. 11. Martinelle, M., and Hult, K. (1995) Kinetics of acyl transfer reactions in organic media catalysed by Candida antarctica lipase B. Biochim. Biophys. Acta 1251, 191–197. 12. Hoegh, I., Patkar, S., Halkier, T., and Hansen, M. T. (1995) Two lipases from Candida antarctica: Cloning and expression in Aspergillus oryzae. Can. J. Bot. 73, S869–S875. 13. Bajpaj, P. (1999) Application of enzymes in the pulp and paper industry. Biotechnol. Prog. 15, 147–157. 14. Linder, M., and Teeri, T. (1997) The roles and function of cellulosebinding domains. J. Biotechnol. 57, 15–28. 15. Shpigel, E., Goldlust, A., Efroni, G., Avraham, A., Eshel, A., Dekel, M., and Shoseyov, O. (1999) Immobilization of recombinant heparinase I fused to cellulose-binding domain. Biotechnol. Bioeng. 65, 17–23. 16. Ong, E., Gilkes, N. R., Miller, R. C., Warren, R. A. J., and Kilburn, D. G. (1991) Enzyme immobilization using a cellulose-binding domain: properties of a -glucosidase fusion protein. Enzyme Microb. Technol. 13, 59–65. 17. Maglione, G., Matsushita, O., Russel, J. B., and Wilson, D. B. (1992) Properties of a genetically reconstructed Prevotella ruminicola endoglucanase. Appl. Environ. Microb. 58, 3593–3597. 18. Karita, S., Sakka, K., and Ohmiya, K. (1996) Cellulose-binding domains confer an enhanced activity against insoluble cellulose to Ruminococcus albus endoglucanase IV. J. Ferment. Bioeng. 81, 553–556. 19. Denman, S., Xue, G. P., and Patel, B. (1996) Characterization of a Neocallimastix patriciarum cellulase cDNA (celA) homologous to Trichoderma reesei cellobiohydrolase II. Appl. Environ. Microb. 62, 1889–1896. 20. Kraulis, P. J., Clore, G. M., Nilges, M., Jones, T. A., Petterson, G., Knowles, J., and Gronenborn, A. M. (1989) Determination of the three-dimensional solution structure of the C-terminal domain of cellobiohydrolase I from Trichoderma reesei: A study using nuclear magnetic resonance and hybrid distance geometry– dynamical simulated annealing. Biochemistry 28, 7241–7257. 21. Holmquist, M., Tessier, D. C., and Cygler, M. (1997) High-level production of recombinant Geotrichum candidum lipases in yeast Pichia pastoris. Protein Express. Purif. 11, 35–40. 22. Minning, S., Schmidt-Dannert, C., and Schmid, R. D. (1998) Functional expression of Rhizopus oryzae lipase in Pichia pastoris: High-level production and some properties. J. Biotechnol. 66, 147–156. 23. Brocca, S., Schmidt-Dannert, C., Lotti, M., Alberghina, L., and Schmid R. D. (1998) Design, total synthesis, and functional overexpression of the Candida rugosa lip1 gene coding for a major industrial lipase. Protein Sci. 7, 1415–1422. 24. Yang, Y., and Lowe, M. E. (1998) Human pancreatic triglyceride lipase expressed in yeast cells: Purification and characterization. Protein Express. Purif. 13, 36–40. 25. Sahasrabudhe, A. V., Solapure S. M., Khurana, R., Suryanareayan, V., Ravishankar, S., deSousa, S. M., and Das, G. (1998) Production of recombinant human bile salt stimulated lipase and its variant in Pichia pastoris. Protein Express. Purif. 3, 425–433. 26. Holmquist, M., Tessier, D. C., and Cygler, M. (1997) Identification of residues essential for differential fatty acid specificity of
392
ROTTICCI-MULDER ET AL.
Geotrichum candidum lipases I and II. Biochemistry 36, 15019– 15025. 27. Sanger, F., Nicklen, S, and Caulson, A. R. (1977) DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463–5467. 28. Patkar, S. A., Bjo¨rkling, F., Zundel, M., Schulein, M., Svendsen, A., Heldt-Hansen, H. P., and Gormsen, E. (1993) Purification
of two lipases from Candida antarctica and their inhibition by various inhibitors. Ind. J. Chem. 32B, 76–80. 29. Rotticci, D., Norin, T., Hult, K., and Martinelle, M. (2000) An active-site titration method for lipases. Biochim. Biophys. Acta 1483, 132–140. 30. Jones, E. W. (1977) Proteinase mutants of Saccharomyces cerevisiae. Genetics 85, 23–33.