Functional expression of Rhizopus oryzae lipase in Pichia pastoris: high-level production and some properties

Journal of Biotechnology 66 (1998) 147 – 156

Functional expression of Rhizopus oryzae lipase in Pichia pastoris: high-level production and some properties Stefan Minning, Claudia Schmidt-Dannert, Rolf D. Schmid * Institut fu¨r Technische Biochemie, Uni6ersita¨t Stuttgart, Allmandring 31, D-70569 Stuttgart, Germany Received 13 January 1998; received in revised form 14 August 1998; accepted 24 August 1998

Abstract The mature lipase of the fungus Rhizopus oryzae (ROL) was functionally expressed and secreted in the methylotrophic yeast Pichia pastoris. In a batch cultivation, where methanol feeding was linked to the dissolved oxygen content in the cultivation solution, a lipase activity of 500000 units per liter (60 mg active lipase per liter) of culture was achieved after initial glycerol feeding of the culture. Recombinant ROL lipase was purified to homogeneity by a simple two-step purification procedure and had a specific activity of 8571 U mg − 1 (triolein, 30°C, pH 8.1) which is comparable with the purified native enzyme. The properties of the recombinant lipase were similar to those reported both for the native lipase and for the enzyme expressed in Escherichia coli and refolded from inactive inclusion bodies. © 1998 Elsevier Science B.V. All rights reserved. Keywords: Lipase; Rhizopus oryzae; Pichia pastoris; Production; Purification; Properties

1. Introduction Lipases (EC 3.1.1.3) catalyze the hydrolysis of triacylglycerols at the interface between the insoluble substrate and water. Although the naturally occurring triacylglycerols are preferred substrates, lipases also catalyze the enantio- and regioselective hydrolysis or synthesis of a wide range of natural and unnatural esters (Bjo¨rkling et al., * Corresponding author. Tel.: + 49 711 6853193; fax: + 49 711 6854569; e-mail: [email protected]

1991; Bornscheuer, 1995; Schmid and Verger, 1998). In recent years, more than 30 lipases were isolated from Rhizopus strains and many of them were characterized (for a review, see Haas and Joerger, 1995). Rhizopus lipases as well as the closely related Rhizomucor miehei lipase (\ 55% homology) share a high 1,3-regiospecificity towards triacylglycerols, which makes them versatile enzymes in lipid modification. Non-recombinant crude enzyme preparations of Rhizopus lipases are commercially provided by several suppliers, e.g. Amano N, D and L lipases derived from Rhizopus

0168-1656/98/$ - see front matter © 1998 Elsevier Science B.V. All rights reserved. PII S0168-1656(98)00142-4

148

S. Minning et al. / Journal of Biotechnology 66 (1998) 147–156

Fig. 1. SDS-PAGE analysis of recombinant ROL expressed in E. coli and P. pastoris. Lane 1, molecular weight standards (kDa); lane 2, concentrated ROL supernatant expressed in P. pastoris; lane 3, ROL expressed in P. pastoris, treated with endo H; lane 4, ROL expressed in P. pastoris treated with endo H under denaturing conditions; lane 5, purified ROL expressed in E. coli; lane 6, purified ROL expressed in P. pastoris; line 7, molecular weight standards (kDa). (Lanes 1 and 2 were fused to the other five lanes to allow the comparison between the concentrated supernatant and the purified sample of ROL).

ni6eus (RNL), Rhizopus delemar (RDL) and Rhizopus ja6anicus (RJL) (for a review, see Bornscheuer et al., 1998). Recently, the taxonomic diversity of the genus Rhizopus was reinvestigated and, in spite of some minor variations, all hitherto diverse species were combined into Rhizopus oryzae (ROL) (Schipper, 1984). Consistent with this taxonomic re-classification, RNL, RDL and RJL have identical amino acid sequences, and the lipase from ROL differs by only two conservative substitutions (His134 is asparagine and Ile234 is leucine in ROL). Hence, the different properties of these lipases observed in commercial preparations may be due to either different production conditions affecting the glycosylation pattern of these lipases, or to proteolytic cleavage products arising from the mature, the pro- or the pre-proenzyme (Uyttenbroeck et al., 1993). Until now, the lipase genes from ROL ATCC 853 (Salleh et al., 1993; Ben Salah et al., 1994; Beer, 1995; Haalck et al., 1997), RNL IFO 9759 (Kohno et al., 1994) and RDL ATCC34612 (Haas et al., 1992) have been cloned. Both lipases from ROL and RDL have been overexpressed in E. coli (Joerger and Haas, 1993; Beer, 1995). In the case of ROL, even small amounts of mature active lipase proved to be toxic upon translocation in the

periplasmic space in E. coli (Beer et al., 1996). But, using a different expression system, the protein could be secreted into the periplasmic space as inactive inclusion bodies. Later, the expression of ROL in the cytoplasmic space as an inactive inclusion body could be performed (Haalck et al., 1997). Refolding of the active lipase requires the correct formation of three disulphide bridges. Although high expression levels of inclusion bodies were obtained, and an effective refolding procedure was developed for the ROL in the laboratory scale, resulting in active lipase with a specific activity of 10000 U mg − 1 (triolein, pH 8.3, 30°C) comparable with the native enzyme (10000 U mg − 1 olive oil, pH 8.5, 37°C) (Ben Salah et al., 1994; Beer, 1995), the large-scale production of active recombinant lipase was hampered by this cumbersome and expensive procedure. Somewhat better yields of recombinant pro-lipase in E. coli could be reached, as the pro-sequence was shown to suppress cell toxicity and enhance folding yields of the lipase (Beer et al., 1996). In this work, we present, for the first time, the functional expression of a mature Rhizopus lipase in a heterologous host. We used Pichia pastoris, a methylotrophic yeast that has become increasingly attractive for the large-scale production of het-

S. Minning et al. / Journal of Biotechnology 66 (1998) 147–156

erologous proteins (Sreekrishna and Kropp, 1996; Catoni et al., 1997). We further describe a cultivation process allowing the high-level secretion of recombinant ROL, and we compare the properties of the functionally expressed lipase with those of recombinant ROL refolded from inclusion bodies formed in E. coli.

2. Materials and methods

2.1. Methods Restriction enzymes, DNA-modifying enzymes, T4-DNA ligase and Taq polymerase were from MBI Fermentas (St. Leon-Rot, Germany), the Taq Dye Cycle Sequencing Kit was from Applied Biosystems (Weiterstadt, Germany), the DNA Gel-Extraction Kit, Midi Plasmid Kit and Prepspin Plasmid Kit were from Qiagen (Hilden, Germany), the different lipase substrates (triacylglycerols were from Fluka), peptone, yeast extract and yeast nitrogen base were from Difco (Augsburg, Germany), and zeocin was from Invitrogen. All reagents were of analytical grade unless otherwise stated. Table 1 Substrate specificity of ROL expressed in P. pastoris and E. coli Substrate

Triacetin (2:0) Tributyrin (4:0) Tricaproin (6:0) Tricaprylin (8:0) Tricaprin (10:0) Trilaurin (12:0) Trimyristin (14:0) Tripalmitin (16:0) Tristearin (18:0) Triolein (18:1)

Relative activity (%)a E. coli

P. pastoris

5.4 91.1 48.29 3.2 50.0 94.8 114.39 5.2 103.69 6.3 16.192.4 8.9 90.5 8.99 1.3 7.1 90.5 100.0 91.1

3.0 91.3 30.3 9 5.2 42.4 9 4.7 121.2 9 3.8 109.1 9 4.0 21.2 9 2.4 6.1 91.6 6.1 91.5 3.0 9 1.0 100.0 91.4

149

2.2. Strains, plasmids and media E. coli DH5a (F − end A1 hsdR17 (rk − , mk + ) sup E44 thi1 lgyrA96 relA1 D(argF laczya) U169) was used for the cloning of the plasmid and P. pastoris GS115 (his4) (Invitrogen) for lipase expression. Plasmid pT1-OmpAROL containing the cDNA of ROL cloned in vector pCYTEXP1 has been described elsewhere (Beer, 1995). The vector pPICZaA was supplied from Invitrogen. E. coli was grown at 37°C in Luria–Bertani (LB) medium containing 25 mg ml − 1 zeocin for selection of clones transformed with the vector pPICZaA. P. pastoris was routinely grown in shaking flasks at 30°C, in a rich standard medium containing 1% (w/v) yeast extract, 2% (w/v) peptone, 100 mM potassium phosphate buffer, pH 6.0, 1.34% (w/v) yeast nitrogen base, 4× 10 − 5% (w/v) biotin, 1% (v/v) glycerol (BMGY) before induction, or 0.5% (v/v) methanol (BMMY) for induction. For maintaining cultures and plates, 1% (w/v) yeast extract, 2% (w/v) peptone, 2% (w/v) dextrose (YEPD) medium was used, and for selection of transformants, YEPD plates containing zeocin (100 mg ml − 1) were used.

2.3. Recombinant DNA techniques and DNA-sequencing Standard recombinant DNA methods were carried out according to the methods described in Ausubel et al. (1987) and Sambroock et al. (1989). For sequence determination, the fluorescencebased dideoxy DNA cycle sequencing method was used. DNA sequencing was carried out with the Taq Dye Deoxy™ Cycle Sequencing Kit (Applied Biosystems) and with a 373A DNA Sequencing System (Applied Biosystems) according to the manufacturer’s instructions.

2.4. Construction of the plasmid pPICZaA-ROL

a

Relative activity towards various triacylglycerols (20 mM) was determined by pH-stat assay at 30°C and pH 8.1. The relative activity on each triacylglycerol is expressed as the percentage of that with triolein as the substrate, set as 100% (100% complies to 200 U in the case of ROL from P. pastoris and 160 U in the case of ROL from E. coli ).

The E. coli expression vector pT1-OmpAROL containing the ROL gene served as a template for polymerase chain reaction (PCR). In order to fuse the sequence encoding the mature lipase directly

150

S. Minning et al. / Journal of Biotechnology 66 (1998) 147–156

Fig. 2. Effect of temperature on lipase activity and stability of ROL expressed in E. coli and P. pastoris. Effect of temperature on activity was determined by the pH-stat assay at pH 8.1. The effect of temperature on lipase stability was determined by incubating lipase aliquots at different temperatures in 100 mM phosphate buffer, pH 7. The residual activity was measured after 30 min of incubation.

in frame to the a-factor prepro signal sequence from Saccharomyces cere6isiae to target the protein to the secretory pathway (Cregg et al., 1993; Scorer et al., 1993), a PCR was performed with one oligonucleotide (P – Eco, 5%-CGACCGTAGCGCAGGAATTCTCTGATGGTGGTAAGG-3%) complementary to the 5%-end of the ROL gene introducing an EcoRI-site, and the second (P – Not, 5%-CTTAGGATCCACGTGCGGCCGCTTATTACAAACAAGC-3%) complementary to the 3%-end introducing a NotI-site. Next, the obtained PCR fragment was digested with NotI and EcoRI and ligated into pPICZaA, linearized with the same enzymes to give pPICZaA-ROL. After transformation of E. coli with pPICZaAROL, zeocin-resistant clones were selected and analyzed by restriction digestion. The ROL coding gene and the fused a-factor of clones with correct size and digestion pattern were sequenced completely to confirm the sequence and the correct fusion of the lipase gene to the a-factor secretion signal, located on the vector.

2.5. Culti6ation All cultivations were carried out in a 5 l bioreactor (Infors) at 30°C. The bioreactor was inoculated with 250 ml of a flask culture grown overnight to an optical density (OD) of 2–3 in rich standard medium. The centrifuged culture was resuspended in the same amount of fresh medium. The cultivation was maintained at constant pH with the aid of 100 mM potassium phosphate buffer, which was provided during the cultivation if needed. The agitation rate was varied from 150 to 585 rpm depending on the oxygen content. The aeration rate was 3 l min − 1. The lipolytic activity of the supernatant, OD, and cell wet weight were monitored throughout the cultivation. The methanol content was monitored with a YSI 2700 Select (Yellow Springs). Batch cultivations were performed in either a complex or a minimal medium at pH 5.5 and 4.5, respectively. The complex medium used was similar to the standard rich medium described, but excluding the expensive yeast nitrogen base. The minimal medium contained FM21 (Storme et al.,

S. Minning et al. / Journal of Biotechnology 66 (1998) 147–156

151

Fig. 3. Effect of pH on lipase activity and stability of ROL expressed in E. coli and P. pastoris. Effect of pH stability was determined by incubating lipase aliquots at different pH in 100 mM phosphate buffer at 30°C. The residual activity was measured after 18 h.

1994) (5 g l − 1 KH2PO4, 6 g l − 1 (NH4)2SO4, 2 g l − 1 MgSO4 · 7H2O and 3 g l − 1 CaSO4 · 2H2O), PTM1 (6 g l − 1 CuSO4 · 5H2O, 0.8 g l − 1 KI, 3 g l − 1 MnSO4 · H2O, 0.2 g l − 1 NaMoO4 · 2H2O, 0.03 g l − 1 H3BO3, 0.5 g l − 1 CoCl2 · 6H2O, 20 g l − 1 ZnSO4, 0.5% (v/v) H2SO4, 65 g l − 1 FeSO4 · 7H2O), 2.5× 10 − 5% biotin and 1× 10 − 4% histidine. Methanol (10–50 ml) was added daily or linked to the carbon dioxide content in the off-gas as well as to the dissolved oxygen content (DO) in the culture, as described, to keep its concentration almost stable during the cultivation time.

2.6. Isolation of acti6e ROL from inclusion bodies expressed in E. coli The E. coli derived recombinant ROL was expressed, refolded and purified as described elsewhere (Joerger and Haas, 1993; Beer et al., 1996; Haalck et al., 1997). Finally, from 500 ml of culture, 5–8 mg of active lipase were refolded and after ultrafiltration followed by ion exchange chromatography on S-sepharose, a homogeneous lipase preparation with a specific activity of 6667 U mg − 1 was obtained and used together with the

P. pastoris derived characterization.

ROL

for

comparative

2.7. Enzyme assay Lipase activity was routinely measured using triolein as substrate. Triolein (20 mM) was emulsified in distilled water containing 0.5% CaCl2 · 2H2O (w/v) and 2% (w/v) gum arabic as stabilizer using a homogeniser (Ultraturrax T25, Janke and Kunkel) for 7 min at maximum speed. Twenty milliliters of the substrate solution were heated to 30°C and adjusted to pH 8.1. After addition of 25–50 ml of the enzyme solution, the activity was measured with a pH-stat (Metrohm). Liberated fatty acids were titrated automatically with 0.01 N NaOH to maintain the pH constant at 8.1. One unit (U) of lipase activity was defined as the amount of enzyme that liberates 1 mmol fatty acid per minute under assay conditions. For the determination of substrate specificity, 20 mM of triacetin, tributyrin, tricaproin, tricaprylin, tricaprin, trilaurin, trimyristin, and tristearin were used as substrates in the pH-stat assay.

152

S. Minning et al. / Journal of Biotechnology 66 (1998) 147–156

Fig. 4. ROL production and cell growth under different fermentation conditions with and without off-gas analysis. Cultivation A, simplified complex medium, pH 5.5 methanol added every 24 h without off-gas analysis; cultivation B, simplified complex medium, pH 5.5, with methanol addition linked to off-gas analysis; cultivation C, synthetic medium, pH 4.5, as described in the text, methanol added every 24 h without off-gas analysis.

2.8. N-terminal sequencing Amino terminal sequence analysis was performed in a gas-phase sequencer 470A (Applied Biosystems) following the manufacturer’s instructions. After blotting, the PVDF membranes were stained with Coomassie Brilliant Blue R-250, and the lipase bands were cut out and used for N-terminal sequence determination.

protein) at 30°C in a 50 mM potassium acetate buffer, pH 5.5, containing 0.5 mM phenylmethylsulphonyl fluoride to prevent proteolysis. For the deglycosylation of denatured samples, the protein was first incubated with 0.01% (w/v) sodium dodecyl sulphonate (SDS) at 95°C for 3 min.

2.11. Isoelectric focusing and SDS-polyacrylamide gel electrophoresis

2.9. Protein determination Protein concentration was determined with the bicinchoninic acid protein assay kit (Pierce) using the enhanced method according to the manufacturer’s instruction (Pierce Instructions 23220/ 23225) and bovine serum albumin as standard.

2.10. Endo-b-N-acetylglycosamidase H digestion Protein samples were incubated for 12 h with endo-b-N-acetylglycosamidase H (25 mU mg − 1

Analytical SDS-polyacrylamide gel electrophoresis (PAGE) (8–25%) and isoelectric focusing (IEF) (pH 3–9) were performed with a Pharmacia Phast System according to the manufacturer’s recommendations. The gels were stained for protein detection by a silver staining procedure (Butcher and Tomkins, 1986). Preparative gel electrophoresis for the N-terminal sequencing was carried out in a 12.5% polyacrylamide gel and proteins were stained with Coomassie Brilliant Blue R-250.

S. Minning et al. / Journal of Biotechnology 66 (1998) 147–156

153

Fig. 5. ROL production during cultivation D. Triangles indicate batch feeding of 1% glycerol (underlined) or 0.5% (v/v) methanol, respectively, depending on an increase of DO in the cultivation solution.

2.12. Effect of temperature and pH on enzyme stability and acti6ity Residual activity was measured by pH-stat assay at 30°C, using triolein as the substrate. The effect of the temperature on lipase stability was determined by incubating aliquots of lipase solution for 30 min in 100 mM phosphate buffer, pH 8.1, at different temperatures. Residual activity was measured by pH-stat assay at pH 8.1, using triolein as substrate.

3. Results and discussion

3.1. Transformation in P. pastoris and selection of positi6e transformants Two zeocin resistant E. coli colonies harboring the plasmid pPICZaAROL were selected and the gene coding for the ROL together with the a-factor was sequenced completely, proving that no mutation occurred during the PCR. For the transformation of P. pastoris GS115, pPICZaA-ROL was linearized with SacI and used for electropora-

tion of competent yeast cells according to Invitrogen’s recommendations. Ten positive transformants were selected on solid selective medium (YEPD containing zeocin) and plated on solid minimal medium containing 1% (w/v) of emulsified tributyrin. Lipase expression was induced by 0.1 ml methanol added onto the lid covering the plates. Transformants expressing lipolytic activity, as indicated by the formation of a clear halo on tributyrin plates, were selected and grown in shaking flasks containing 50 ml BMGY medium. After the cultures reached an OD of 2–6 they were centrifuged and resuspended in the same volume of BMMY medium for induction of lipase expression. After 48 h of cultivation under inducing conditions, the lipolytic activity of the supernatant was determined. The lipase activity of different clones varied significantly between 30 and 110 U ml − 1. For further investigations, the best clone was chosen for lipase production in 2 l flasks containing 500 ml medium. After 2 days of induction, the activity of the supernatant was 110 U ml − 1 and the specific activity 13.8 U mg − 1. The cells were separated by centrifugation, and the supernatant (500 ml) was diafiltered against

154

S. Minning et al. / Journal of Biotechnology 66 (1998) 147–156

water through a 10 kDa membrane, yielding a lipase solution in the retentate (100 ml with 420 U ml − 1) with a 1.7-fold increased specific activity of 23.1 U mg − 1. For further purification, 25 ml of the lipase solution were applied on a SP-Sepharose column (5 ×30 cm, Pharmacia) equilibrated with 10 mM CaCl2 in water, the lipase was eluted with a 1 M NaCl in water. Finally, a lipase solution (28 ml with 150 U ml − 1) with a further 363-fold increased specific activity of 8571 U mg − 1 was obtained by a simple two-step purification protocol. Specific activities of 10000 U mg − 1 (ROL) and 5094 U mg − 1 (RDL) have been reported for recombinant lipases expressed in E. coli and isolated by a cumbersome multi-step refolding and purification protocol (Joerger and Haas, 1993; Beer, 1995).

3.2. Characterization of recombinant ROL The purified lipase solution was used for characterization. Some properties of the recombinant ROL expressed in P. pastoris were compared to those of the recombinant ROL refolded from IBs expressed in E. coli. SDS-PAGE analysis of the purified lipase showed a single protein band of 30 kDa similar to the size of the E. coli derived recombinant ROL. As the ROL protein contains four common Nglycosylation sites of the sequence Asn-X-Ser/Thr, we investigated glycosylation of the recombinant enzyme by treatment with endo-b-N-acetylglycosamidase H. No differences in the molecular weight of the recombinant protein before and after incubation with endo-b-N-acetylglycosamidase could be observed, suggesting that no glycosylation had occurred (Fig. 1). Amino terminal sequencing of the recombinant ROL expressed in P. pastoris proved the correct cleavage of the a-factor signal sequence at the Kex2 cleavage site. When compared to the mature ROL expressed in E. coli, the N-terminal sequence of the Kex2 processed lipase bears two additional amino acids (glutamic acid and phenylalanine) derived from the in-frame fusion of the ROL gene to the a-factor prepro-signal sequence in pPICZaA.

A comparative study of the properties of native ROL and the recombinant enzyme expressed in E. coli has been limited to pH and temperature stability (Beer, 1995). We thus choose to study substrate specificity of the recombinant enzymes towards various triacylglycerols differing in chain length of the acyl moiety. As shown in Table 1, both recombinant lipases (refolded from inclusion bodies expressed in E. coli and functionally expressed in P. pastoris) preferred middle chain triacylglycerols (C8 and C10), but also showed high activity towards triolein (C18:1). In contrast, saturated long-chain triacylglycerols (\C12) were only poor substrates, possibly due to their reduced emulsification under our reaction conditions. A similar substrate specificity towards various triglyceroles was reported for RDL (Joerger and Haas, 1994; Shimada et al., 1997). Both recombinant ROLs expressed in P. pastoris and E. coli exhibited a rather alkaline pI] 9.3 as determined by IEF, where the protein band appeared just outside the pH gradient of 3–9.3 of a standard IEF gel. The pI deduced from the amino acid sequence is 8.3 (Bornscheuer et al., 1998) and that of the native RDL is 8.6 (Haas et al., 1992). For the determination of the temperature and pH optima, lipase activities of both types of recombinant ROLs were measured at different temperatures at pH 8.1 and at different pH values at 30°C, respectively (Figs. 2 and 3). Similarly to the ROL expressed in E. coli (Beer, 1995) and to the native as well as recombinant RDL (Haas et al., 1992; Joerger and Haas, 1993), the recombinant ROL secreted by P. pastoris was most active at 30°C. However, thermal stability of ROL functionally expressed in P. pastoris was higher than that of the ROL refolded from inclusion bodies: while the functionally expressed ROL showed 70% residual activity after 30 min incubation at 50°C, no residual activity remained under the same conditions with refolded ROL. The reason for this observation remains unclear, but is in line with studies on recombinant RDL which showed a lower temperature stability as compared to the native enzyme (no residual activity after 15 min at 50°C (Haas et al., 1992)), comparable to that of ROL expressed in E. coli.

S. Minning et al. / Journal of Biotechnology 66 (1998) 147–156

3.3. Culti6ations Batch cultivations using either a simple complex medium (Catoni et al., 1997) or a synthetic medium were investigated for further improvement of recombinant lipase production in a larger scale. Fermentation A was performed in simplified complex medium at pH 5.5 containing 1% (w/v) yeast extract, 2% (w/v) peptone, 4×10 − 5% (w/v) biotin and 0.5% (w/v) methanol as the sole carbon source. Here, 25–50 ml methanol were added every 24 h to keep its concentration almost constant. After 98 h of fermentation, the lipase activity in the supernatant reached 100 U ml − 1 (Fig. 4). In fermentation B, methanol feeding was linked to the carbon dioxide content measured in the off-gas: as soon as the carbon dioxide decreased significantly, indicating starvation of the cells, 25 – 50 ml methanol were fed. By this strategy, lipase productivity could be significantly increased from 1020 to 1837 U l − 1 h − 1, resulting in 180 U ml − 1 of lipase activity in the culture supernatant after 98 h of cultivation. To increase cell mass, fermentation D was run with glycerol feeding the first 24 h. After this, the induction with methanol started regarding the DO of the culture. Every time the DO increased, we fed 25 ml of methanol, resulting in a sudden increase of the DO (Fig. 5). In this case, we could increase the lipase productivity up to 5435 U l − 1 h − 1 after 92 h of cultivation. With more advanced analytic procedures and feeding strategies, increased yields of active lipase should be possible (Jime´nez et al., 1997). In fermentation C, the complex medium was exchanged for a less expensive synthetic medium. Methanol feeding was performed as described for fermentation B. However, lipase productivity under these fermentation conditions was low, yielding only 30–40 U ml − 1 in the supernatant even after prolonged cultivation of 160 h (Fig. 4).

4. Conclusion We have found that the methylotrophic yeast P. pastoris constitutes an efficient expression sys-

155

tem for the production of recombinant mature Rhizopus lipase in a functional form. A batch fermentation with glycerol feeding, followed by controlled methanol feeding, allowed the production of 500000 U ROL per liter of culture, which is comparable to the amount of active ROL or RDL obtained by a refolding protocol from inclusion bodies of this enzyme produced in recombinant E. coli. Since P. pastoris did not secrete any significant amount of proteins other than the recombinant lipase, pure lipase with a specific activity of 8571 U mg − 1 could be obtained by a simple two-step purification protocol. The activity of the recombinant enzyme is comparable to the specific activities reported for native and recombinant ROL and RDL. The properties of ROL produced with P. pastoris were comparable to those of native and recombinant ROL and RDL, but the temperature stability of the enzyme expressed in P. pastoris was found to be higher.

Acknowledgements We are grateful to Volker Noedinger for performing the protein sequencing, and we thank the European Community for financial support under contract Bio4-CT96-00005.

References Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Smith, J.A., Struhlo, K., 1987. Current Protocols in Molecular Biology. New York, Greene Publishing Associates. Beer, H.D., 1995. Untersuchung der Lipase aus Rhizopus oryzae auf molekularer Ebene: Das Gen, der Reaktionsmechanismus und die Funktion der Prosequenz, Ph.D. Thesis, University of Hannover. Beer, H.D., Wohlfahrt, G., Schmid, R.D., McCarthy, J.E., 1996. The folding and activity of the extracellular lipase of Rhizopus oryzae are modulated by a prosequence. Biochem. J. 319, 351 – 359. Ben Salah, A., Fendri, K., Gargoury, Y., 1994. La lipase de Rhizopus oryzae: production, purification et characte´ristiques biochimiques. Rev. Fr. Corps Gras 41, 133 – 137. Bjo¨rkling, F., Godtfredsen, S.E., Kirk, O., 1991. The future impact of industrial lipases. Trends Biotechnol. 9, 360 – 363.

156

S. Minning et al. / Journal of Biotechnology 66 (1998) 147–156

Bornscheuer, U.T., Pleiss, J., Schmidt-Dannert, C., Schmid, R.D., 1998. Lipases from Rhizopus species: genetics, structure and applications. In: Alberghina, H. (Ed.), Protein Engineering in Industrial Biotechnology, Harwood, London, in press. Bornscheuer, U.T., 1995. Lipase-catalyzed synthesis of monoacylglycerols. Enzyme. Microb. Technol. 17, 578– 586. Butcher, L.A., Tomkins, J.H., 1986. Silver staining method useful for ultrathin polyacrylamid gels on support films. Anal. Biochem. 148, 384–388. Catoni, E., Schmidt-Dannert, C., Brocca, S., Schmid, R.D., 1997. Overexpression of lipase A and B of Geotrichum candidum in Pichia pastoris: High-level production and some properties of functional expressed lipase B. Biotechniques 11, 689 – 695. Cregg, J.M., Vedvick, T.S., Raschke, W.C., 1993. Recent advances in the expression of foreign genes in Pichia pastoris. Biotechnology 11, 905–910. Haalck, L., Paltauf, F., Pleiss, J., Schmid, R.D., Spener, F., Stadler, P., 1997. Stereoselectivity of lipase from Rhizopus oryzae towards triacylglycerols and analogs: computeraided modeling and experimential validation. Methods Enzymol. 184, 353 – 376. Haas, M.J., Cichowicz, D.J., Bailey, D.G., 1992. Purification and characterization of an extracellular lipase from the fungus Rhizopus delemar. Lipids 27, 571–576. Haas, M.J., Joerger, R.D., 1995. Lipases of the genera Rhizopus and Rhizomucor: versatile catalysts in nature and the laboratory. In: Khachatourians, G.G., Hui, Y.H. (Eds.), Food Biotechnology: Microorganisms. VCH, Weinheim, pp. 549 – 588. Jime´nez, R.E., Sa´nchez, K., Roca, H., Delgado, J.M., 1997. Different methanol feeding strategies to recombinant Pichia pastoris cultures producing high level of dextranase. Biotechniques 11, 461–466. Joerger, R.D., Haas, M.J., 1993. Overexpression of a Rhizopus delemar lipase gene in Escherichia coli. Lipids 28, 81– 88.

.

Joerger, R.D., Haas, M.J., 1994. Alteration of chain length selectivity of a Rhizopus delemar lipase through site-directed mutagenesis. Lipids 29, 377 – 384. Kohno, M., Kugimiya, W., Hashimoto, Y., Morita, Y., 1994. Purification, characterization, and crystallization of two types of lipase from Rhizopus ni6eus. Biosci. Biotechnol. Biochem. 58, 1007 – 1012. Salleh, A.B., Musani, M., Yunus, K., Ampon, W.M.Z., Razak, C.N.A., 1993. Extra- and intra-cellular lipases from a thermophilic Rhizopus oryzae and factors affecting their production. Can. J. Microbiol. 39, 978 – 981. Sambroock, J., Fritsch, E.F., Maniatis, T., 1989. Molecular cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Scorer, C.A., Buckholz, R.G., Clare, J.J., Romanos, M.A., 1993. The intracellular production and secretion of HIV-1 envelope protein in the methylotrophic yeast Pichia pastoris. Gene 136, 111 – 119. Schipper, M.A.A., 1984. A revision of the genus Rhizopus II. The Rhizopus stolonifer-group and Rhizopus oryzae. Stud. Mycol. 25, 1 – 19. Schmid, R.D., Verger, R., 1998. Lipases, interfacial enzymes with attractive applications. Agnew. Chem. Int. Ed. Engl. 37, 1608 – 1633. Shimada, Y., Sugihara, A., Nakano, H., Nagao, T., Suenaga, M., Nakai, S., Tominaga, Y., 1997. Fatty acid specificity of Rhizopus delemar lipase in acidolysis. J. Ferment. Bioeng. 83, 321 – 327. Sreekrishna, K., Kropp, K.E., 1996. Pichia pastoris. In: Wolf, K. (Ed.), Nonconventional Yeasts in Biotechnology— A Handbook. Springer Verlag, Berlin, pp. 203 – 250. Storme, V., de Meutter, J., Messens, J., Lauwereys, M., Laroche, Y., 1994. High-level secretion and very efficient isotopic labeling of tick anticoagulant peptide (TAP) expressed in the methylotropic yeast Pichia pastoris. Biotechnology 12, 1119 – 1124. Uyttenbroeck, W., Hendriks, D., Vriend, G., de Baere, I., Moens, L., Scharpe´, S., 1993. Molecular characterization of an extra acid-resistant lipase produced by Rhizopus ja6anicus. Biol. Chem. Hoppe-Seyler 374, 245 – 254.