High-Level Production of RecombinantGeotrichum candidumLipases in YeastPichia pastoris

PROTEIN EXPRESSION AND PURIFICATION ARTICLE NO.

11, 35–40 (1997)

PT970747

High-Level Production of Recombinant Geotrichum candidum Lipases in Yeast Pichia pastoris Mats Holmquist,1,2 Daniel C. Tessier, and Miroslaw Cygler 2 Biotechnology Research Institute, National Research Council of Canada, 6100 Royalmount Avenue, Montreal, Quebec H4P 2R2, Canada

Received December 23, 1996, and in revised form March 20, 1997

We describe the heterologous high-level expression of the two Geotrichum candidum lipase (GCL) isoenzymes from strain ATCC 34614 in the methylotrophic yeast Pichia pastoris. The lipase cDNAs were placed under the control of the methanol-inducible alcohol oxidase promoter. The lipases expressed in P. pastoris were fused to the a-factor secretion signal peptide of Saccharomyces cerevisiae and were secreted into the culture medium. Cultures of P. pastoris expressing lipase accumulated active recombinant enzyme in the supernatant to levels of Ç60 mg/L virtually free from contaminating proteins. This yield exceeds that previously reported with S. cerevisiae by a factor of more than 60. Recombinant GCL I and GCL II had molecular masses of Ç63 and Ç66 kDa, respectively, as determined by SDS–PAGE. The result of endoglucosidase H digestion followed by Western blot analysis of the lipases suggested that the enzymes expressed in P. pastoris received N-linked high-mannose-type glycosylation to an extent, 6–8% (w/w), similar to that in G. candidum. The specific activities and substrate specificities of both recombinant lipases were determined and were found to agree with what has been reported for the enzymes isolated from the native source. q 1997 Academic Press

Lipases (EC 3.1.1.3) are enzymes that catalyze the cleavage of ester bonds and most become fully active only upon binding to the aggregates formed by their water-insoluble lipid substrates in aqueous emulsions. This property, known as interfacial activation, was early recognized (1) and distinguishes these lipases 1 Permanent address: Department of Biochemistry and Biotechnology, Royal Institute of Technology, SE-100 44 Stockholm, Sweden. 2 To whom correspondence should be addressed. E-mail: Mats. [email protected]/[email protected].

from classical esterases which exhibit full catalytic power toward monomeric substrate molecules. Interfacial activation was proposed (1) and later shown to be associated with conformational rearrangements in the lipase molecule (2–4) where access is given to the active site only in the open active state of the lipase. More than 3 decades ago it was realized that lipases from the fungus Geotrichum candidum have an unusually high substrate preference for neutral lipids containing cis (v-9) unsaturated fatty acid residues such as oleic acid (5,6). This property is valued and taken advantage of in industrial applications such as the enzymatic restructuring of lipids and oils into products with defined fatty acid composition (7). To this day, no explanation for this property has been presented. Many reports dealing with the production, isolation, and characterization of G. candidum lipases have appeared during the years [for a review see (8)]. However, contradictory results concerning their substrate specificty were reported. As lipases were further purified to apparent homogeneity the existence of several lipase isoforms differing in amino acid sequence and extent of glycosylation became apparent, some of which showed differences in substrate specificity (9–12). The identification of two lipase genes in G. candidum (13–15) and their individual expression and the subsequent characterization of both isoenzymes (16) revealed that GCL I has the unique substrate preference for lipids containing long-chain unsaturated fatty acids with a cis (v-9) double bond. The naturally occurring variant of the enzyme, GCL II, accepts a broader range of substrates despite the fact that 86% of the 544 amino acid residues are absolutely conserved. ˚ ) 3-D structure of native The high-resolution (1.8 A GCL II has been determined by X-ray crystallography and the catalytic Ser-His-Glu triad and the substratebinding site have been located (17,18). Yet the basis for the different substrate specificity of the GCL isoenzymes remained elusive. Both recombinant isoenzymes 35

1046-5928/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.

AID

PEP 0747

/

6q13$$$261

09-05-97 22:13:06

pepa

AP: PEP

36

HOLMQUIST, TESSIER, AND CYGLER

were previously expressed in Saccharomyces cerevisiae (16,19). That expression system allowed for the confirmation of the identity of the residues of the Ser-HisGlu catalytic triad (19) and an evolutionary positional shift of the acid member of the triad was investigated by site-directed mutagenesis (20). Furthermore, it was possible to show that the determinants for the high specificity for oleic acid-containing lipids are not located within the first 194 N-terminal amino acid residues encompassing the flap covering the active site of the lipase molecule in its closed conformation (16). This progress in our understanding of the mode of action of the G. candidum lipases was achieved despite the fact that the expression levels from S. cerevisiae were low (õ1 mg/L) and the lipase product was crude (Ç1% of total proteins in supernatant), leading to poor yields of pure lipase. As some mutants were expressed at even lower levels than the wild-type isoenzymes (19), the need for a high-level expression system became apparent in order to facilitate extensive mutational and structural analysis of the lipase isoenzymes. Such a system would allow for the production of lipase variants with altered structure in quantities sufficient not only for kinetic characterization but also for crystallization experiments and possibly X-ray diffraction analysis of the lipases free and in complex with substrate analogs. It would also provide a means to obtain pure G. candidum lipases in quantities adequate for applied catalysis. In that respect the access to the pure isoenzymes would permit the full exploitation of their different substrate specificities for synthetic purposes. In this communication we describe the heterologous high-level expression of pure and active G. candidum lipases in the methylotrophic yeast Pichia pastoris. Different culturing conditions and P. pastoris strains were investigated and a convenient assay for the identification of lipase-secreting P. pastoris transformants was developed. The produced recombinant lipases were characterized and their substrate specificities were determined with lipids reported to be diagnostic for the enzymes’ different substrate preferences. MATERIALS AND METHODS

Yeast and Escherichia coli strains. P. pastoris GS115 (his4) [(21); Invitrogen Corp., San Diego, CA.] and SMD1168 (his4, pep4) (Invitrogen Corp.) were used for the expression of G. candidum lipases. E. coli strain MC1061 (22) was used for all plasmid constructions. Yeast culture media. One liter of BMGY medium contained the following: yeast extract (10 g), peptone (20 g), glycerol (10 ml), biotin (400 mg), yeast nitrogen base with ammonium sulfate (13.4 g), and potassium phosphate buffer (100 ml, 1 M, pH 6.0). BMMY medium had the same composition as BMGY medium with the exception that 5 ml methanol per liter was added in-

AID

PEP 0747

/

6q13$$$261

09-05-97 22:13:06

stead of glycerol. MM medium contained only methanol, biotin, and nitrogen base with ammonium sulfate in concentrations as above. To prepare plates, agar [1.5% (w/v)] was added to the medium. Construction of lipase expression vectors. The PstI site in the ampicillin resistance gene of the P. pastoris expression vector pPIC9 (Invitrogen Corp.) was deleted by substituting the 1.4-kb AflIII–ScaI DNA fragment of pPIC9 with that of pUC-8 digested with the same restriction enzymes. This resulted in the expression vector YpDC519. The gene coding for lipase I from G. candidum (strain ATCC34614) was excised as a 2-kb PstI–SnaBI fragment from the S. cerevisiae expression plasmid YpDC420 [pVT100-U (23) deleted of its ADH1 promoter and expressing GCL I under the control of the a-factor promoter] and cloned into plasmid YpDC519, yielding plasmid YpDC521. The GCL I gene included a segment coding for an N-terminal extension of the lipase consisting of a (His)8-tag followed by a LeuValProArg thrombin-cleavage site. This construct brought the GCL I gene under the control of the methanolinducible alcohol oxidase promoter (AOX1)3 and inframe with the a-factor secretion signal peptide of S. cerevisiae. The gene coding for GCL II from G. candidum (strain ATCC34614) was isolated as a 1.8-kb SfiI–PvuII DNA fragment from plasmid YpDC240 (19) and cloned into the SfiI–SnaBI sites of plasmid YpDC521. The resulting construct YpDC523 of GCL II received an N-terminal extension identical to that of the GCL I construct YpDC521. Transformation of plasmid DNAs into P. pastoris. Plasmid DNA (10 mg YpDC521 or YpDC523 for GCL I or GCL II, respectively) harboring the lipase gene was digested with StuI for 2 h in a total volume of 20 ml. The digested DNA was ethanol precipitated and resuspended in TE, pH 7.5 (10 ml). Linearized plasmid DNA (5 ml) was electroporated into 80 ml of competent P. pastoris cells using a Bio-Rad Gene Pulser instrument and a 0.2-cm-width cuvette. The charging voltage, resistance, and capacitance were set at 1500 V, 400 V, and 25 mF, respectively. This resulted in a pulse length of approximately 8 ms and a field strength of about 7500 V/m. After electroporation 0.5 ml of ice-cold 1 M sorbitol was added to the sample and aliquots were plated on dextrose-based medium without histidine supplementation (SD-His) and incubated at 307C. As a 3 Abbreviations used: AOX1, alcohol oxidase promoter; BMGY, buffered glycerol complex medium; BMMY, buffered methanol complex medium; E-64, N-[N-(L-3-trans-carboxyoxirane-2-carbonyl)-Lleucyl]agmatine; GCL, Geotrichum candidum lipase; MM, minimal methanol; MOPS, 3-[N-morpholino]propanesulfonic acid; Ni2/-NTA, nickel nitrilotriacetic acid; pepstatin A, isovaleryl-Val-Val-StaAla-Sta, Sta Å statine Å (3S,4S)amino-3-hydroxy-6-methylheptanoic acid; PMSF, phenylmethylsulfonyl fluoride; SD-His, dextrose-based medium without histidine supplementation.

pepa

AP: PEP

RECOMBINANT Geotrichum candidum LIPASES

negative control P. pastoris cells were also transformed with the plasmid YpDC519 not carrying a lipase gene. After 3 days of incubation colonies of transformed P. pastoris appeared on the plates. Subsequently the colonies were screened to identify lipase-secreting transformants. Screening for lipase-secreting P. pastoris transformants: Yeastern method. To identify lipase-secreting P. pastoris transformants, individual colonies were picked and patched on SD-His plates. After incubation at 307C for 24 h the plate was replicated on an MM plate containing 0.5% (v/v) methanol to induce lipase production. The MM plate was incubated inverted at 307C for 48 h with the addition of 200 ml 80% (v/v) methanol to the lid of the plate after 24 h to maintain induction. A nitrocellulose filter was placed on the colonies of the MM plate and 200 ml 80% (v/v) methanol was added to the lid. After incubation at 307C for 24 h the filter was removed and rinsed thoroughly with water. The filter was incubated in 9% (w/v) skim milk for 15 min and subsequently for 4 h with a polyclonal rabbit antiserum raised against denatured GCL II (19). GCL – antibody complexes were visualized with an alkaline phosphatase-conjugated goat anti-rabbit antiserum (BioRad). This enabled the identification of lipase-secreting transformants by the purple color developed on the filter at the position of lipase-secreting clones. Positive transformants were selected for further culturing. Cultivation of P. pastoris. P. pastoris transformants were grown at 307C in 25 ml BMGY medium in a 50-ml conical tube in a shaking incubator until a cell density of OD600 4 was reached. The culture was centrifuged and the supernatant discarded. Subsequently the cells were resuspended in 100 ml BMMY medium in a 250-ml Erlenmeyer flask to an OD600 1 to start induction. Protease inhibitors PMSF, pepstatin A, and E-64 were added to final concentrations of 62, 1.5, and 2.8 mM, respectively. The culture was kept in a shaking incubator at 307C for 4 days (250–300 rpm) with the addition of 0.5 ml methanol once a day to maintain induction. The kinetics of lipase secretion in the culture medium was determined by the analysis of daily samples. Polyacrylamide gel electrophoresis and Western blot analysis. The culture medium (10 ml) containing the produced lipase was analyzed by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) on an 8% gel (24). Proteins were stained with Coomassie brilliant blue. Western blot analysis was performed by transferring the SDS–PAGE-separated sample to a nitrocellulose filter using a Bio-Rad transblot apparatus according to the company’s specifications. Samples of the recombinant lipases were deglycosylated using

AID

PEP 0747

/

6q13$$$261

09-05-97 22:13:06

37

endoglucosidase H (Endo H; Boehringer Mannheim) according to the supplier’s instructions. Metal-affinity purification of lipases. Culture supernatants were concentrated with a Centriprep-30 (Amicon, Beverly, MA) into 40 mM potassium phosphate buffer, pH 7.0. NaCl was added to a final concentration of 0.5 M. The samples were applied to 2.5 ml nickel nitrilotriacetic acid (Ni2/-NTA) agarose column (Qiagen, Chatsworth, CA), previously equilibrated with 40 mM potassium phosphate buffer, pH 7.0, containing 0.5 M NaCl (buffer A). The column was washed with buffer A until the absorbance reached the baseline. Adsorbed proteins were eluted from the column by a linear gradient (30 ml) of 0–200 mM imidazole in buffer A at a flow rate of 0.3 ml/min. Determination of lipase concentration. The culture medium containing recombinant lipase was concentrated five times by means of a Centriprep-30 (Amicon) into 20 mM MOPS buffer, pH 7.0. Lipase concentration was determined spectrophotometrically according to the method of Bradford (25) (Bio-Rad) using BSA as a standard. Lipase activity determination. Lipase activity was determined titrimetrically. The substrate emulsion was prepared by mixing substrate ester (0.1 g), gum arabic [19 ml, 2% (w/v)] and calcium chloride (1 ml, 2 M). The solution was emulsified in a sonicator (Heat System, Ultrasonics, Inc.) for 2 min using a macroprobe (1.3-cm diameter) with the power output set at 50%. Lipase (1–15 mg) was added to substrate emulsion (2 ml) and the enzymatic activity was determined with a pH-stat (RTS822 Recording Titration System; Radiometer, Copenhagen, Denmark) at pH 7.0 and 257C by the titration of the released fatty acid with 50 mM sodium hydroxide. Lipase activity was expressed in mmol released fatty acidrmin01rmg01 protein. RESULTS AND DISCUSSION

Details of the lipase-encoding segments in the constructed expression vectors are shown in Fig. 1. These vectors allowed for the production of the two G. candidum lipases in P. pastoris. The lipase cDNAs were placed under the control of the methanol-inducible AOX1 promoter. The plasmids were designed so that the mature lipases will be left with an N-terminal (His)8-tag after cleavage of the a-factor secretion signal peptide. The presence of the (His)8-tag extension does not affect the catalytic properties of the lipase (20). The secretion of lipase relies upon efficient cleavage by the dibasic specific protease Kex2p (26). The Glu-Ala repeats are further removed by the dipeptidyl aminopeptidase Ste13p (27) (Fig. 1). Integration of either of the constructed expression vectors into the P. pastoris genome resulted in transformants that efficiently expressed and secreted active G. candidum lipase. Yeast

pepa

AP: PEP

38

HOLMQUIST, TESSIER, AND CYGLER

FIG. 1. Schematic representation of the lipase-encoding segments in the constructed plasmids for the expression of G. candidum lipases (YpDC521, GCL I; YpDC523, GCL II) in P. pastoris. Restriction sites used for cloning of the lipase genes are shown. The lipase genes are under the control of the methanol-inducible AOX1 promoter. The amino acid sequence at the junction between the a-factor secretion signal peptide (PRE–PRO) and the lipase is shown in one-letter code. The secretion of mature GCL relies on cleavage by the dibasic specific protease Kex2p (26) at the site indicated and the removal of the EA repeats (underlined) by the dipeptidyl aminopeptidase Ste13p (27). The N-terminal (H)8-tag followed by the LVPR thrombin-cleavage site in the secreted GCL product is shown. The positions of putative NÉ ). linked glycosylation sites in the lipases are indicated (•

transformants were screened for lipase secretion with the plate/filter assay which we call Yeastern. The intensity of the color developed in the assay enabled the easy identification of transformants that efficiently secreted recombinant lipase. More than 80% of all transformants were found to secrete lipase. In cultures of P. pastoris transformants in strain GS115 grown in BMMY medium, lipase production was detected in the culture medium 1 day after induction with methanol. The lipase concentration kept increasing during the next 3 days of induction (Figs. 2A and 2B) and resulted in the accumulation of 56 { 13 and 62 { 9 mg (n Å 3) active GCL I and GCL II, respectively, per liter culture. In the negative control cells transformed with YpDC519 lacking the lipase gene, no lipase product could be detected (data not shown). The lipase quantities obtained here represent a dramatic increase in the levels of expression compared to our

previously reported yields using S. cerevisiae (õ1 mg/ L) (16). High expression levels have been reported recently for recombinant Candida rugosa lipase, related to GCL, which, however, accumulated intracellularly in S. cerevisiae (28). Due to the nonstandard codon usage of C. rugosa (29) the heterologously expressed lipase does not have the native amino acid sequence and is not active. The GCL product expressed in P. pastoris represents a single species which is immunoreactive toward antibodies raised against denatured lipase (Fig. 2B). The secreted lipase appeared to be very stable under the given culturing conditions, as no signs of proteolytic degradation could be detected either by SDS–PAGE or by Western blot analysis (Figs. 2A and 2B). To determine if the presence of protease inhibitors during cultivation would inactivate the secreted lipase, P. pastoris cells were cultured under four different reaction conditions: in the absence of protease inhibitors

FIG. 2. Secretion of G. candidum lipases expressed in P. pastoris strain GS115. (A) SDS–PAGE and (B) Western blot analysis using rabbit anti-GCL antiserum. Ten microliters of culture supernatant was analyzed at Days 1, 2, 3, and 4 after induction. (C) Determination of carbohydrate content and detection of N-terminal (H)8-tag in the recombinant GCL. Western blot using a Ni2/-NTA alkaline phosphatase conjugate (Qiagen) detection of GCL I and GCL II prior to (0) and after (/) endoglucosidase H treatment. Molecular masses of standards (kDa) are indicated.

AID

PEP 0747

/

6q13$$$261

09-05-97 22:13:06

pepa

AP: PEP

39

RECOMBINANT Geotrichum candidum LIPASES

and in the presence of PMSF, pepstatin A, or E-64. The lipase products obtained under these conditions were identical in terms of purity and enzymatic activity as determined by SDS–PAGE, Western blot analysis, and rate of triolein hydrolysis (data not shown). The secreted lipase was thus not inactivated by the protease inhibitors present in the culture medium. As a precaution we recommend the addition of the inhibitors to the culture medium. In the absence of inhibitors we have once observed truncated lipase products, possibly generated by the action of intracellular proteases released as a result of minor cell lysis during cultivation. Experiments with the protease-deficient strain SMD1168 of P. pastoris resulted in expression levels and quality of the recombinant lipases similar to those seen with strain GS115. The molecular masses of produced GCL I and GCL II were found to be Ç63 and Ç66 kDa, respectively, as determined by SDS–PAGE analysis (Fig. 2). A lower apparent molecular mass of the GCL I has previously been observed (10,11). Treatment with endoglucosidase H lowered the apparent molecular masses of GCL I and GCL II by Ç5 and Ç4 kDa, respectively (Fig. 2C). The extent of N-linked high-mannose glycosylation of the lipases was thus in a range similar to that seen with the native lipases from G. candidum or to that produced in S. cerevisiae strain YE410 (18) deficient in the processing steps of carbohydrate extension [mnn9; (30)]. In contrast to the high expression levels from P. pastoris transformants cultured in BMMY medium, their growth in unbuffered MM medium, to avoid potential proteolysis of the secreted product, yielded only trace amounts of lipase. To facilitate purification of the recombinant lipases they were produced with an N-terminal (His)8-tag extension. The presence of the (His)8-tag in the produced GCL was confirmed by Western blot analysis using a Ni2/-NTA alkaline phosphatase conjugate (Qiagen) (Fig. 2C). The (His)8-tag extension allows for singlestep purification of the protein by means of metal-affinity chromatography on a Ni2/-NTA agarose column (Qiagen). This approach was previously demonstrated to be feasible, but less than 15% of the secreted lipase produced in S. cerevisiae could be purified (16). Here, using the same method, we were able to isolate 50% of the produced lipase despite very conservative pooling of fractions containing lipase. Because the lipase secreted by P. pastoris is virtually pure from contaminating proteins (Fig. 2), further purification on a Ni2/-NTA column is only necessary if a very high purity lipase is desired. This contrasts the results in S. cerevisiae where the lipase product corresponded to 1% of the total protein content of the culture broth at the time of harvest (19). For applied G. candidum lipase catalysis or kinetic investigations the cultured P. pastoris cells may simply be separated from the culture broth

AID

PEP 0747

/

6q13$$$261

09-05-97 22:13:06

TABLE 1

Specific Activities and Substrate Specificities of G. candidum Lipase Isoenzymes with Triacylglycerol and Ethyl Ester Substrates of Oleic (C18:1) and Octanoic (C8) Acids Specific activity (mmolrmin01rmg01) Substrate Triglyceride GCL I GCL II Ethyl ester GCL I GCL II

C18:1

C8

1200 770

380 770

380 400

2.4 30

Specificity C18:1/C8

3.2 1.0 160 13

Note. The recombinant lipases were expressed in P. pastoris strain GS115. The relative error in the specific activity values was £10%, based on conducting each experiment in duplicate or triplicate.

by centrifugation to yield a solution that only needs to be desalted and concentrated to give a lipase product of high purity (Fig. 2). To assess that the recombinant G. candidum lipase isoenzymes were correctly folded and exhibit the expected enzymatic properties, the specific activity and substrate specificities of recombinant GCL I and GCL II were determined with triacylglycerol and ethyl esters of oleic and octanoic acids. As seen in Table 1, GCL I has a higher preference for oleic acid esters than GCL II which agrees with previous findings (11,12,16). The specific activities measured toward triolein indicate that the lipases were fully active. CONCLUSIONS

The reason for the unusual substrate preference of a lipase from G. candidum for lipids containing cis (v9) unsaturated fatty acids has remained a mystery for many years, partly due to the difficulties in separating the existing isoforms of the lipase from the natural source. Recently this property was unambiguously traced to a separate gene product. The existence of two genes coding for lipase isoenzymes with distinct differences in substrate specificity provides a unique opportunity to determine which parts of the lipase molecule are important for differentiating between substrates. However, to take advantage of this, a high-level expression system for the production of recombinant lipases with altered structures is necessary. Both of the identified lipase genes of G. candidum have previously been expressed in S. cerevisiae. However, the low yields and crude quality of the recombinant lipases made that expression system poorly suited for extensive mutational analysis and further characterization of the regions involved in substrate differentiation. We have shown that the two naturally occurring GCL isoenzymes can

pepa

AP: PEP

40

HOLMQUIST, TESSIER, AND CYGLER

be expressed in P. pastoris in large amounts. The proteins are very pure, stable and catalytically active. Having established a new means for the production of recombinant GCL creates the platform for the experimental exploration of the structural basis of lipase substrate specificities. ACKNOWLEDGMENTS Mats Holmquist is a recipient of a fellowship from the Swedish Research Council for Engineering Sciences. NRCC publication No. 39966.

REFERENCES 1. Sarda, L., and Desnuelle, P. (1958) Action de la lipase pancre´atique sur les esters en e´mulsion. Biochim. Biophys. Acta 30, 513– 521. 2. Brzozowski, A. M., Derewenda, U., Derewenda, Z. S., Dodson, G. G., Lawson, D. M., Turkenburg, J. P., Bjo¨rkling, F., HugeJensen, B., Patkar, S. A., and Thim, L. (1991) A model for interfacial activation in lipases from the structure of a fungal lipase– inhibitor complex. Nature 351, 491–494. 3. Grochulski, P., Li, Y., Schrag, J. D., Bouthillier, F., Smith, P., Harrison, D., Rubin, B., and Cygler, M. (1993) Insights into interfacial activation from an open structure of Candida rugosa lipase. J. Biol. Chem. 268, 12843–12847. 4. van Tilbeurgh, H., Egloff, M-P., Martinez, C., Rugani, N., Verger, R., and Cambillau, C. (1993) Interfacial activation of the lipase– procolipase complex by mixed micelles revealed by X-ray crystallography. Nature 362, 814–820. 5. Alford, J. A., and Pierce, D. A. (1961) Lipolytic activity of microorganisms at low and intermediate temperatures. J. Food Sci. 26, 518–524. 6. Jensen, R. G., Sampugna, J., Guinn, J. G., Carpenter, D. L., and Marks, T. A. (1965) Specificity of a lipase from Geotrichum candidum for cis-octadecenoic acid. J. Am. Chem. Soc. 42, 1029– 1032. 7. Macrae, A. R. (1985) in ‘‘Biocatalysts in Organic Synthesis’’ (Tramper, J., van der Plas, H. C., and Linko, P., Eds.). Stud. Org. Chem. 22, 195–208. 8. Charton, E. (1991) Purification et characterisation de deux lipases de Geotrichum candidum CMICC 335426. Ph.D. thesis, Institut National Agronomique Paris–Grignon, France. 9. Baillargeon, M. W. (1990) Purification and specificity of lipases from Geotrichum candidum. Lipids 25, 841–848. 10. Sidebottom, C. M., Charton, E., Dunn, P. P. J., Mycock, G., Davies, C., Sutton, J. L., Macrae, A. R., and Slabas, A. R. (1991) Geotrichum candidum produces several lipases with markedly different substrate specificities. Eur. J. Biochem. 202, 485–491. 11. Charton, E., and Macrae, A. R. (1992) Substrate specificities of lipase A and B from Geotrichum candidum CMICC 335426. Biochim. Biophys. Acta 1123, 59–64. 12. Sugihara, A., Hata, S., Shimada, Y., Goto, K., Tsunasawa, S., and Tominaga, Y. (1993) Characterization of Geotrichum candidum lipase III with some preference for the inside bond of triglyceride. Appl. Microbiol. Biotechnol. 40, 279–283.

AID

PEP 0747

/

6q13$$$261

09-05-97 22:13:06

13. Shimada, Y., Sugihara, A., Tominaga, Y., Iizumi, T., and Tsunasawa, S. (1989) cDNA molecular cloning of Geotrichum candidum lipase. J. Biochem. 106, 383–388. 14. Shimada, Y., Sugihara, A., Iizumi, T., and Tominaga, Y. (1990) cDNA cloning and characterization of Geotrichum candidum lipase II. J. Biochem. 107, 703–707. 15. Bertolini, M. C., Larame´e, L., Thomas, D. Y., Cygler, M., Schrag, J. D., and Vernet, T. (1994) Polymorphism in the lipase genes of Geotrichum candidum strains. Eur. J. Biochem. 219, 119–125. 16. Bertolini, M. C., Schrag, J. D., Cygler, M., Ziomek, E., Thomas, D. Y., and Vernet, T. (1995) Expression and characterization of Geotrichum candidum lipase. I. Comparison of specificity profile with lipase II. Eur. J. Biochem. 228, 863–869. 17. Schrag, J. D., Li, Y., Wu, S., and Cygler, M. (1991) Ser-His-Glu triad forms the catalytic site of the lipase from Geotrichum candidum. Nature 351, 761–764. ˚ refined structure of 18. Schrag, J. D., and Cygler, M. (1993) 1.8 A the lipase from Geotrichum candidum. J. Mol. Biol. 230, 575– 591. 19. Vernet, T., Ziomek, E., Recktenwald, A., Schrag, J. D., de Montigny, C., Tessier, D. C., Thomas, D. Y., and Cygler, M. (1993) Cloning and expression of Geotrichum candidum lipase II gene in yeast: Probing the enzyme active site by site-directed mutagenesis. J. Biol. Chem. 268, 26212–26219. 20. Schrag, J. D., Vernet, T., Larame´e, L., Thomas, D. Y., Recktenwald, A., Okoniewska, M., Ziomek, E., and Cygler, M. (1995) Redesigning the active site of Geotrichum candidum lipase. Protein Eng. 8, 835–842. 21. Cregg, J. M., Barringer, K. J., Heseler, A. Y., and Madden, K. R. (1985) Pichia pastoris as a host system for transformations. Mol. Cell Biol. 5, 3376–3385. 22. Casadaban, M., and Cohen, S. N. (1980) Analysis of gene control signals by DNA fusion and cloning in Escherichia coli. J. Mol. Biol. 138, 179–207. 23. Vernet, T., Dignard, D., and Thomas, D. Y (1987) A family of yeast expression vectors containing the phage f1 intergenic region. Gene 52, 225–233. 24. Laemmli, U. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. 25. Bradford, M. M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254. 26. Julius, D., Brake, A., Blair, L., Kunisawa, R., and Thorner, J. (1984) Isolation of the putative structural gene for the lysine– arginine-cleaving endopeptidase required for processing of yeast prepro-a-factor. Cell 37, 1075–1089. 27. Julius, D., Blair, L., Brake, A., Sprague, G., and Thorner, J. (1983) Yeast a-factor is processed from a larger precursor polypeptide: The essential role of a membrane-bound dipeptidyl aminopeptidase. Cell 32, 839–852. 28. Fusetti, F., Brocca, S., Porro, D., and Lotti, M. (1996) Effect of the leader sequence on the expression of recombinant C. rugosa lipase by S. cerevisiae cells. Biotechnol. Lett. 18, 281–286. 29. Kawaguchi, Y., Honda, H., Taniguchi-Morimura, J., and Iwasaki, S. (1989) The codon CUG is read as serine in an asporogenic yeast Candida cylindracea. Nature 341, 164–166. 30. Tsai, P. K., Frevert, J., and Ballou, C. E. (1984) Carbohydrate structure of Saccharomyces cerevisiae mnn9 mannoprotein. J. Biol. Chem. 259, 3805–3811.

pepa

AP: PEP