- Email: [email protected]
Fructo-oligosaccharides production by the Gluconacetobacter diazotrophicus levansucrase expressed in the methylotrophic yeast Pichia pastoris
Enzyme and Microbial Technology 28 (2001) 139 –144
www.elsevier.com/locate/enzmictec
Research papers
Fructo-oligosaccharides production by the Gluconacetobacter diazotrophicus levansucrase expressed in the methylotrophic yeast Pichia pastoris L.E. Trujilloa, J.G. Arrietaa,*, F. Dafhnisa, J. Garcı´ab, J. Valde´sb, Y. Tambarac, M. Pe´rezb, L. Herna´ndeza a
b
Plant Division, Centre for Genetic Engineering and Biotechnology, POBox 6162, Havana 10600, Cuba Research and Development Division, Centre for Genetic Engineering and Biotechnology, POBox 6162, Havana 10600, Cuba c Physical Chemistry Division, Centre for Genetic Engineering and Biotechnology, POBox 6162, Havana 10600, Cuba Received 9 June 1999; received in revised form 2 May 2000; accepted 28 June 2000
Abstract Levansucrase (LsdA) (EC 2.4.1.10) from Gluconacetobacter diazotrophicus (formerly Acetobacter diazotrophicus) yields high levels of fructo-oligosaccharides (FOS) from sucrose. A DNA fragment encoding the precursor LsdA lacking the first 57 amino acids was fused to the pho1 signal sequence under the control of the Pichia pastoris-alcohol oxidase 1 (AOX1) promoter. Methanol induction of a P. pastoris strain harboring a single copy of the lsdA expression cassette integrated in the genome resulted in the production of active levansucrase. After fermentation of the recombinant yeast, LsdA activity was detected in the periplasmic fraction (81%) and in the culture supernatant (18%) with an overall yield of 1% of total protein. The recombinant LsdA was glycosylated and displayed optimal pH and temperature for enzyme activity similar to those of the native enzyme, but thermal stability was increased. Neither fructosylpolymerase activity nor FOS production was affected. Incubation of recombinant LsdA in sucrose (500 g l⫺1) yielded 43% (w/w) of total sugar as 1-kestose, with a conversion efficiency about 70%. Intact recombinant yeast cells also converted sucrose to FOS although for a 30% efficiency. © 2001 Elsevier Science Inc. All rights reserved. Keywords: Recombinant expression; Pichia pastoris; Levansucrase; 1-kestose; Fructo-oligosaccharides; Gluconacetobacter diazotrophicus
1. Introduction Levansucrase (LsdA) (E.C 2.4.1.10) from Gluconacetobacter diazotrophicus (formerly Acetobacter diazotrophicus) catalyses the transfructosylation reaction from sucrose to a variety of acceptors including water (sucrose hydrolysis), glucose (exchange reaction), fructan (polymerase reaction) and sucrose (FOS synthesis) [1]. Marketed FOS consist essentially of 1-kestose, nystose and fructofuranosyl nystose, in which one to three fructosyl units are bound at the -2,1 position of sucrose respectively. These non-digestible sweet-tasting saccharides can be used as low-calorie edulcorants and dietetic fibers. Moreover, their ingestion promotes growth of beneficial bacteria in the gut improving humans and animals health [2]. * Corresponding author. Tel.: ⫹53-7-218-008; fax: ⫹53-7-218-070. E-mail addresses: [email protected] (J.G. Arrieta), luis.trujillo@ cigb.edu.cu (L.E. Trujillo).
LsdA is commercially attractive for the production of FOS, mainly 1-kestose, from sucrose [3]. However, the secretion levels of this enzyme in the wild type or a genetically modified G. diazotrophicus strain remained rather low after optimization of culture conditions [4]. In addition, the high viscosity of G. diazotrophicus cultures due to the presence of polysaccharides may hinder LsdA production at industrial-scale using the natural host [4]. The availability of this enzyme could be improved by using recombinant sources. LsdA has been expressed in an active form in Escherichia coli [5], but the protein accumulated intracellularly hampering its purification. The methylotrophic yeast Pichia pastoris constitutes an excellent recombinant expression system. Most important among the advantages of using this yeast as a host for the production of fructosyltransferases are the absence of sucrolytic activity, the availability of the strong methanolinduced alcohol oxidase 1 (AOX1) promoter, and the exis-
0141-0229/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved. PII: S 0 1 4 1 - 0 2 2 9 ( 0 0 ) 0 0 2 9 0 - 8
140
L.E. Trujillo et al. / Enzyme and Microbial Technology 28 (2001) 139 –144
Fig. 1. LsdA production at a fermenter scale. A 7.5 liters fermenter (B.E. Marubishi, Tokyo, Japan) containing 5 liters of fermentation medium was inoculated with clone PP15 at an initial OD530 ⫽ 0.2. Operation conditions were 28°C, pH 5.0, 700 rpm and aeration rate of 2 vvm. Upon depletion of glycerol as judged by raising pH and dissolved oxygen (21 h after inoculation), methanol was added to a final concentration of 1% (v/v). The methanol feeding rate was started at 2 g l⫺1 per h and gradually increased during the induction phase up to 3.5 g l⫺1 per h. Samples withdrawn at indicated time were analyzed for cellular mass (Œ) and levansucrase activity in the supernatant (F).
tence of an efficient protein secretion system which contrasts with low levels (0.5%) of secreted native proteins [6]. We report herein the recombinant expression and secretion of active LsdA in the methylotrophic yeast P. pastoris and the use of the recombinant enzyme for FOS production.
2. Materials and methods
ment from plasmid pALS5 [5], encoding the precursor LsdA lacking the first 57 residues was fused in frame to the pho1 signal sequence at the SmaI site of the expression vector pHILSI [8]. The resulting plasmid was named pALS105. Correct fusion of the lsdA gene with the pho1 signal peptide in pALS105 was confirmed by dideoxy-chain termination sequencing [11]. Genomic DNA isolation from yeast cells was performed as recommended in P. pastoris Expression Kit (Version 3.0) [8].
2.1. Microorganisms and media E. coli MC1061 [7] and P. pastoris strain GS115 (his4) [8] (Invitrogene, San Diego, CA) were used as expression and cloning hosts, respectively. MC1061 was grown at 37°C in LB medium [9] and ampicillin was added to final concentration of 50 g ml⫺1. P. pastoris was grown in YPD or MD medium prepared as described by Invitrogen [8]. The yeast fermentation medium consisted of: 5 g yeast extract l⫺1, 2 g NH4SO4 l⫺1, 14.2 g KH2PO4 l⫺1, 3.7 g MgSO4 䡠 7H2O l⫺1, 0.23 g CaCl2 䡠 2H2O l⫺1 and 25 g glycerol l⫺1, pH 5.0. Vitamins and traces were prepared and used as recommended by Cregg et al. [10]. 2.2. DNA manipulations All recombinant DNA techniques were performed according to Sambrook et al. [9]. The 1.9-kb SmaI-NruI frag-
2.3. Genetic transformation of Pichia pastoris P. pastoris GS115 spheroplasts were transformed with 10 g of plasmid pALS105, previously linearized at the StuI site so as to direct integration into the his4 locus by a single recombination event [8]. Transformants (His⫹ colonies) were selected through a prototrophy screening for histidinol dehydrogenase using minimal MD medium without histidine supplementation. 2.4. LsdA production in shaking batch cultures Baffled flasks (250 ml) containing 50 ml of YPD medium were inoculated from a single yeast colony and the cultures were grown with agitation at 200 rpm, during 18 h at 30°C (final OD600 ⫽ 2). Cells were harvested by centrif-
L.E. Trujillo et al. / Enzyme and Microbial Technology 28 (2001) 139 –144
Fig. 2. Levan synthesis by recombinant LsdA. Culture supernatant and cells from 1 ml of P. pastoris fermentation culture were separated by centrifugation. The cells were resuspended in 1 ml of disruption buffer (0.1% Triton X100, SDS 5 g l⫺1, 1 mM EDTA, 1 mM PMSF, 50 mM sodium phosphate pH 7.0) and milled with glass beads by vortexing. The cellular debris was spun down. Proteins from culture supernatant and cell lysate were separated by 12.5% SDS-PAGE. The gel was rinsed in distilled water and incubated overnight at 42°C in 100 g sucrose l⫺1, 100 mM sodium acetate, pH 5.5. Formed levan appeared as white bands. Lane 1: G. diazotrophicus levansucrase (20 ng), lane 2: GS115 cell lysate (50 g of total proteins); lanes 3,4,5: PP15 cell lysate 5, 25 and 50 g of total proteins respectively; lane 6: 10 l of GS115 10X-concentrated culture supernatant; lanes 7,8,9: PP15 10X-concentrated culture supernatant 1, 5 and 10 l, respectively.
ugation, washed twice with sterile distilled water and resuspended at an OD600 ⫽ 1 in 200 ml of YPD medium supplemented with 1% methanol instead of glucose for AOX1 promoter induction. The cultures were transferred to 2 liters baffled flasks and incubated at 30°C with agitation at 200 rpm. Methanol pulses to a final concentration of 0.5% were supplied every 24 h during culture growth. 2.5. Subcellular fractionation Spheroplasts were obtained as recommended by Invitrogen [8]: P. pastoris cells collected from 5 ml of broth were incubated with lyticase (Sigma) in 1 M Sorbitol, 1 mM EDTA, 5 g SDS l⫺1, 10 mM sodium citrate pH 7.5 for 45 min and centrifuged for 5 min at 10000⫻ g. Periplasmic proteins were recovered in the supernatant and the spheroplasts in the pellet were washed twice in 1.5 M Sorbitol, 1 mM EDTA, resuspended in 2 ml of lysis buffer containing 1 M Sorbitol, 10 mM MgCl2, 2 mM DTT, 1 mM EDTA, 1 mM PMSF, 50 mM NaCl, 50 mM sodium phosphate pH 6.0 and mechanically lysed by vortexing four times for 1 min in the presence of acid–washed 500-m glass beads (Sigma). Cytoplasmic activity was measured in the spheroplast extract.
141
on initial velocity measurements under the following conditions: 250 mM sucrose in 100 mM sodium acetate buffer pH 5.5, at 42°C. Levan formation on SDS-polyacrylamide gel was determined as reported previously by Herna´ndez et al. [1] Enzymatic studies were performed on recombinant LsdA partially purified from culture supernatants or cell extracts. Contaminants were precipitated with acetone 25% (v/v); after centrifugation, LsdA in the supernatant was precipitated by adding ethanol to a final concentration of 50% (v/v), and redissolved in 10 mM sodium acetate, pH 5.5. Activity dependence on temperature was studied in reaction mixtures of 50 l containing 0.3 units of LsdA, 250 mM sucrose and 100 mM sodium acetate buffer pH 5.5, incubated at several temperatures for 1 h. Enzyme activity dependence on pH was determined in reaction mixtures of 50 l containing 0.3 units of LsdA, 250 mM sucrose and buffered with 100 mM sodium acetate for pH values ranging between 3.0 and 6.0 or 100 mM sodium phosphate for pH range 6.0 – 8.0 incubated at 42°C for 1 h. Reactions were stopped by heating in a boiling water bath for 10 min and glucose release was measured as described above. 2.7. Polyacrylamide gel electrophoresis (PAGE), blotting and immuno-recognition SDS-PAGE in 12.5% gels was performed according to Laemmli [13]. For western blots, proteins were electrotransferred onto nitrocellulose membranes (Amersham Pharmacia Biotech, Uppsala, Sweden) using a Mini Trans-Blot Electrophoretic Transfer (BIORAD, Richmond, USA) at a constant current of 300 mA for 2 h. Dot blots were performed by applying protein samples directly onto the nitrocellulose membrane using a Bio-dotTM apparatus (BIORAD). LsdA in shaking batch cultures was detected by dot blotting 40 l supernatant samples taken after 72 h of methanol induction. For dot blot detection of LsdA in fermenter cultures, samples were withdrawn at different intervals and 20 l of culture supernatant or cell lysate were applied onto a nitrocellulose membrane. Immuno-detection of LsdA on nitrocellulose membranes was achieved with polyclonal antibodies generated against extracellular levansucrase from G. diazotrophicus SRT4 [5]. The amount of blotted recombinant protein was quantified by densitometry in an Imagine Densitometer GS-250 (BIORAD) using known amounts of natural LsdA as standard.
3. Results and discussion 2.6. Protein quantification and enzyme assays 3.1. Expression of the IsdA gene in P. pastoris Total proteins in culture supernatants and crude extracts were determined as described by Bradford [12]. Levansucrase activity was measured as glucose released by sucrose hydrolysis using a glucose oxidase-peroxidase-coupled colorimetric kit (Sigma). One unit of LsdA is defined as the amount of enzyme releasing 1 mol of glucose min⫺1 based
About 103 His⫹ colonies were obtained after GS115 spheroplast transformation with plasmid pALS105. Four His⫹ randomly selected clones were evaluated for LsdA expression in shaking batch cultures. In all cases, the enzyme accumulated active in the supernatant fluids although
142
L.E. Trujillo et al. / Enzyme and Microbial Technology 28 (2001) 139 –144
Fig. 3. Glycosylation analysis of P. pastoris-expressed LsdA. Proteins (40 g) from culture supernatant or cell lysate were denatured in 100 l of 0.5% SDS, 1% -mercaptoethanol at 100°C for 10 min. After addition of 1/10 volumen of 1 M-sodium citrate buffer pH 5.5 at 25°C, samples were reacted with Endoglycosidase H (New England Biolabs) at 0.25 units g⫺1 of total protein at 37°C for 10 h. For western blot, proteins (10 g) were separated by 12.5% SDS-PAGE and transferred to nitrocellulose membranes, then LsdA was immunodetected. Lane 1: G. diazotrophicus levansucrase (100 ng), lane 2: PP15 cell lysate, lane 3: Endo H-treated PP15 cell lysate; lane 4: PP15 culture supernatant; lane 5: Endo H-treated PP15 culture supernatant.
at different levels. These recombinant strains showed a wild-type methanol utilization phenotype (Mut⫹), indicating that the integration of the lsdA expression cassette in the host chromosome did not alter the functionality of the aox1 gene. Southern blot analysis of the highest expressor clone PP15 revealed that the lsdA expression cassette was integrated as a single copy at the chromosomal GS115 his4 locus by a single recombination event (data not shown). LsdA expression by clone PP15 was further studied at a fermenter scale. A typical 5 liters fermenter run is presented in Fig. 1. The induction phase lasted 64 h with a gradual increase of methanol feeding rate. During this phase the biomass increased from 114.0 to 356.7 wet-weight g l⫺1. The generation time of the recombinant strain was similar to that of the untransformed strain GS115 indicating that the recombinant production of LsdA was not toxic to the host cell. In supernatant samples withdrawn at different times, a progressive increase of levansucrase activity was observed (Fig. 1) that correlated with an increased production of the recombinant extracellular LsdA, as determined by immunoblotting (data not shown). Studies of the localization of LsdA activity in clone PP15 revealed that 18% (0.74 ⫻ 103 U l⫺1) of total LsdA activity was present in the culture medium, 81% (3.5 ⫻ 103 U l⫺1) was found, after cell fractionation, in the periplasmic fraction and a remaining 0.9% (0.037 ⫻ 103 U l⫺1) was associated with cellular debris. The final yield of recombinant LsdA, estimated by quantitative blotting procedure, was about 1% of the total protein. This yield is low in comparison to the expression of other proteins in the P. pastoris system [14]. Several strategies like optimization of fermentation conditions or construction of multi-copy integrant strains could be employed to increase LsdA expression levels in this host [14].
levan as a result of their fructosyl polymerase activity upon gel incubation in a sucrose-containing buffer (Fig. 2). Bands corresponding to both LsdA forms were fuzzy and migrated slower than the G. diazotrophicus enzyme used as control. The extracellular form migrated as a single band while the intracellular form appeared as a doublet suggesting the existence of different protein aggregates. As noted in further experiments (Fig. 3), these protein aggregates were dissociated and became in a single band when samples were prepared in -mercaptoethanol-containing buffer and heated to 95°C before subjecting to SDS-PAGE. The higher apparent molecular mass of the yeast-expressed LsdA (expected mass 58 kDa) in comparison to that of the natural enzyme (60.4 kDa) used as control, probably resulted from glycosylation. The presence of a single band suggests that no intermediates are accumulated, and points to a high efficiency of the glycosylation process. Three potential Nlinked glycosylation sites conforming the general rule N-XT/S, where X is not proline, are located at positions 202, 285 and 467 of the predicted mature recombinant LsdA sequence [5]. Digestion of both recombinant LsdA forms with endoglycosidase H altered their electrophoretic mobilities in denaturing SDS-PAGE, as revealed by western blotting (Fig. 3), confirming that the recombinant LsdA is glycosylated. The presence of N-linked sugars in both the intracellular and extracellular LsdA confirms that the protein entered the P. pastoris secretion pathway. Contrary to these results, Scotti et al. [15,16] reported that the Bacillus subtilis levansucrase driven by its own signal sequence or a yeast acid phosphatase signal did not enter the Saccharomyces cerevisiae secretory pathway, but accumulated in the cell in its precursor form associated with the cytoplasmic membrane. Fusion to the yeast invertase signal peptide allowed a small proportion of the enzyme to be translocated to the endoplasmatic reticulum (ER) lumen and being glycosylated. Fusion of the mature part of the B. subtillis levansucrase with the B. amyloliquefaciens ␣-amylase signal-pep-
3.2. Characterization of recombinant LsdA The cell-associated and extracellular forms of the recombinant LsdA, after being resolved in a 12.5% polyacrylamide gel under non-denaturing conditions, synthesized
Fig. 4. Thermostability of recombinant LsdA. Recombinant (Œ) or natural LsdA (■) (5 units) was incubated in 100 l of 50 mM sodium acetate pH 5.5 for 1 h at the indicated temperature and then residual activity was determined. Values represent means from 3 independent experiments.
L.E. Trujillo et al. / Enzyme and Microbial Technology 28 (2001) 139 –144
143
Fig. 5. A) FOS production by extracellular recombinant LsdA. Sucrose at 500 g l⫺1 was reacted with extracellular recombinant LsdA at 10 units g⫺1 of sucrose in 100 mM sodium acetate pH 5.5 at 40°C with agitation at 200 rpm, for the indicated time. Sugars in the reaction mixture were analyzed by HPLC in an Aminex HPX-87N column 0.78 ⫻ 30 cm (BIO-RAD, USA) coupled to a refractive index detector. The column temperature was kept at 85°C. A solution of 10 mM Na2SO4 was used as the mobile phase at a flow rate of 0.5 ml min⫺1. Pure solutions of fructose, glucose, sucrose and raffinose were used as standards. Values are the average of three different experiments. Symbols represent: Sucrose (}), glucose (䊐), trisaccharides (Œ), fructose (E). B) FOS production by intact P. pastoris cells expressing levansucrase. Recombinant yeast cells (0.5 g) were washed thrice with 100 mM sodium acetate pH 5.5 and incubated in 15 ml of sucrose 500 g l⫺1, 100 mM sodium acetate pH 5.5 at 37°C with agitation at 200 rpm, for the indicated time. HPLC analysis of the reaction mixtures and symbols are as described in note to Fig. 5A.
tide yielded an almost fully glycosylated protein. In neither case protein secretion occurred [15,16].
The dependence of the extracellular recombinant LsdA activity on pH and temperature was similar to that of the
144
L.E. Trujillo et al. / Enzyme and Microbial Technology 28 (2001) 139 –144
natural levansucrase [1,3] with the highest activity for splitting of sucrose at pH 5.0 and 60°C (data not shown). According to this observation, glycosylation did not alter optimal conditions for enzyme activity, but it did increase LsdA thermal stability (Fig. 4). 3.3. FOS production by cell-associated and extracellular LsdA As shown in Fig. 5A, the behavior of the extracellular form of the recombinant LsdA in the presence of sucrose was similar to that of the natural LsdA [3]. The enzyme converted sucrose mostly to FOS, mainly 1-kestose and nystose, liberating glucose. Small amounts of free fructose and levan were also formed as a consequence of sucrose hydrolysis and fructan polymerase reactions, respectively. The time course of FOS production (Fig. 5A) revealed that after 24 h more than 85% of the initial sucrose was consumed by the action of LsdA. The trisaccharide 1-kestose, which is the FOS of the highest commercial interest reached a maximal value of 217 g l⫺1. The efficiency of 1-kestose synthesis, expressed as the fraction of fructosyl residues transferred to sucrose was about 70% (w/w), as similarly reported for the natural LsdA [3]. Since most of LsdA was detected in the yeast periplasm, we investigated the possibility of producing FOS by incubating intact PP15 cells with sucrose. The reaction products were analyzed at different time points (Fig. 5B). After cell incubation for 30 h, 85% of the sucrose was consumed. FOS reached a maximal value of 100 g l⫺1 and represented about 27% of total sugars present in the reaction mixture. The efficiency of FOS synthesis was about 30% (w/w). This value, lower than that obtained with the free recombinant LsdA, could be explained by a limited sucrose diffusion to the periplasm resulting in suboptimal concentrations of the substrate for FOS synthesis. Ramos et al. [17] reported direct action on sucrose of intact recombinant P. pastoris cells accumulating the S. cerevisiae invertase in the periplasm. To date, industrial production of FOS depends chiefly on fungal enzymes [2], although a new promising perspective is their direct production in transgenic plants [18,19]. The availability of recombinant P. pastoris as a new source of a FOS-producing enzyme might result of biotechnological interest.
Acknowledgments We thank Dr. Pablo Vera and Ismael Rodrigo for critical and constructive review of the manuscript. This work was supported partially by Grant f/2770 –1 from The International Foundation for Sciences.
References [1] Herna´ndez L, Arrieta J, Menendez C, Vazquez R, Coego A, Suarez V, Selman G, Petit-Glatron MF, Chambert R. Isolation and enzymatic properties of levansucrase secreted by Acetobacter diazotrophicus SRT4, a bacterium associated with sugar cane. Biochem J 1995;309:113– 8. [2] Yun JW. Fructooligosaccharides-ocurrence, preparation and application. Enzyme and Microbial Technology 1996;19:107–17. [3] Ta´mbara Y, Hormaza JV, Pe´rez C, Leo´n A, Arrieta J, Herna´ndez L. Structural analysis and optimised production of fructo-oligosaccharides by levansucrase from Acetobacter diazotrophicus SRT4. Biotechnology Letters 1999;21:117–21. [4] Herna´ndez L, Ramı´rez R, Hormaza JV, Madrazo J, Arrieta JG. Increased levansucrase production by a genetically modified Acetobacter diazotrophicus strain in shaking batch cultures. Letters in Applied Microbiology 1999;28:41– 4. [5] Arrieta JG, Herna´ndez L, Coego A, Sua´rez V, Balmori E, Mene´ndez C, Petit-Glatron MF, Chambert R, Selman-Housein G. Molecular characterization of the levansucrase gene from the endophytic sugarcane bacterium Acetobacter diazotrophicus SRT4. Microbiology 1996;142:1077– 85. [6] Romanos MA, Scorer CA, Clare JJ. Foreign gene expression in yeast: a review. Yeast 1992;8:423– 88. [7] Meissner PS, Sisk WP, Berman ML. Bacteriophage cloning system for the construction of directional cDNA libraries. Proc Natl Acad Sci USA 1987;84:4171. [8] Pichia Expression kit. A manual of methods for expression of recombinant proteins in P. pastoris, Version 3.0. San Diego, USA: Invitrogen Corporation. [9] Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1989. [10] Cregg JM, Tschopp JF, Stillman C, Siegel R, Akong M, Craig WS, Buckholz RG, Madden KR, Kellaris PA, Davies GR, Smiley BL, Cruze J, Torregrossa R, Velicelebi G, Thill GP. High-level expression and efficient assembly of hepatitis B surface antigen in the methylotrophic yeast Pichia pastoris. Bio/Technology 1987;5:479 – 85. [11] Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain terminating inhibitors. Proc Natl Acad Sci USA 1977;74:5463–7. [12] Bradford MM. A rapid and sensitive method for quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976;72:248 –54. [13] Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680 –5. [14] Cregg JM, Vedvick TS, Rashke WC. Recent advances in the expression of foreign genes in Pichia pastoris. Bio/Technology 1993;11:905–10. [15] Scotti PA, Chambert R, Petit-Glatron MF. Extracellular levansucrase of Bacillus subtilis produced in yeast remains in the cell in its precursor form. Yeast 1994;10:29 –38. [16] Scotti PA, Praestegaard M, Chambert R, Petit-Glatron MF. The targeting of Bacillus subtilis levansucrase in yeast is correlated to both the hydrophobicity of the signal peptide and the net charge of the N-terminus mature part. Yeast 1996;12:953– 63. [17] Ramos R, Cruz A, Figueroa N. Proceeding for industrial scale production of glucose-fructose liquors from sucrose. Cuban Patent No. 92–91, 1991. [18] Sevenier R, Hall RD, Van der Meer I, Hakkert HJC, van Tunen AJ, Koops AJ. High level fructan accumulation in a transgenic sugar beet. Nature Biotechnology 1998;16:843– 6. [19] Hellwege EM, Gritscher D, Willmitzer L, Heyer AG. Transgenic potato tubers accumulate high levels of 1-kestose and nystose: functional identification of a sucrose-sucrose 1-fructosyltransferase of artichoke (Cynara scolymus) blossom discs. The Plant Journal 1997; 12:1057– 65.
www.elsevier.com/locate/enzmictec
Research papers
Fructo-oligosaccharides production by the Gluconacetobacter diazotrophicus levansucrase expressed in the methylotrophic yeast Pichia pastoris L.E. Trujilloa, J.G. Arrietaa,*, F. Dafhnisa, J. Garcı´ab, J. Valde´sb, Y. Tambarac, M. Pe´rezb, L. Herna´ndeza a
b
Plant Division, Centre for Genetic Engineering and Biotechnology, POBox 6162, Havana 10600, Cuba Research and Development Division, Centre for Genetic Engineering and Biotechnology, POBox 6162, Havana 10600, Cuba c Physical Chemistry Division, Centre for Genetic Engineering and Biotechnology, POBox 6162, Havana 10600, Cuba Received 9 June 1999; received in revised form 2 May 2000; accepted 28 June 2000
Abstract Levansucrase (LsdA) (EC 2.4.1.10) from Gluconacetobacter diazotrophicus (formerly Acetobacter diazotrophicus) yields high levels of fructo-oligosaccharides (FOS) from sucrose. A DNA fragment encoding the precursor LsdA lacking the first 57 amino acids was fused to the pho1 signal sequence under the control of the Pichia pastoris-alcohol oxidase 1 (AOX1) promoter. Methanol induction of a P. pastoris strain harboring a single copy of the lsdA expression cassette integrated in the genome resulted in the production of active levansucrase. After fermentation of the recombinant yeast, LsdA activity was detected in the periplasmic fraction (81%) and in the culture supernatant (18%) with an overall yield of 1% of total protein. The recombinant LsdA was glycosylated and displayed optimal pH and temperature for enzyme activity similar to those of the native enzyme, but thermal stability was increased. Neither fructosylpolymerase activity nor FOS production was affected. Incubation of recombinant LsdA in sucrose (500 g l⫺1) yielded 43% (w/w) of total sugar as 1-kestose, with a conversion efficiency about 70%. Intact recombinant yeast cells also converted sucrose to FOS although for a 30% efficiency. © 2001 Elsevier Science Inc. All rights reserved. Keywords: Recombinant expression; Pichia pastoris; Levansucrase; 1-kestose; Fructo-oligosaccharides; Gluconacetobacter diazotrophicus
1. Introduction Levansucrase (LsdA) (E.C 2.4.1.10) from Gluconacetobacter diazotrophicus (formerly Acetobacter diazotrophicus) catalyses the transfructosylation reaction from sucrose to a variety of acceptors including water (sucrose hydrolysis), glucose (exchange reaction), fructan (polymerase reaction) and sucrose (FOS synthesis) [1]. Marketed FOS consist essentially of 1-kestose, nystose and fructofuranosyl nystose, in which one to three fructosyl units are bound at the -2,1 position of sucrose respectively. These non-digestible sweet-tasting saccharides can be used as low-calorie edulcorants and dietetic fibers. Moreover, their ingestion promotes growth of beneficial bacteria in the gut improving humans and animals health [2]. * Corresponding author. Tel.: ⫹53-7-218-008; fax: ⫹53-7-218-070. E-mail addresses: [email protected] (J.G. Arrieta), luis.trujillo@ cigb.edu.cu (L.E. Trujillo).
LsdA is commercially attractive for the production of FOS, mainly 1-kestose, from sucrose [3]. However, the secretion levels of this enzyme in the wild type or a genetically modified G. diazotrophicus strain remained rather low after optimization of culture conditions [4]. In addition, the high viscosity of G. diazotrophicus cultures due to the presence of polysaccharides may hinder LsdA production at industrial-scale using the natural host [4]. The availability of this enzyme could be improved by using recombinant sources. LsdA has been expressed in an active form in Escherichia coli [5], but the protein accumulated intracellularly hampering its purification. The methylotrophic yeast Pichia pastoris constitutes an excellent recombinant expression system. Most important among the advantages of using this yeast as a host for the production of fructosyltransferases are the absence of sucrolytic activity, the availability of the strong methanolinduced alcohol oxidase 1 (AOX1) promoter, and the exis-
0141-0229/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved. PII: S 0 1 4 1 - 0 2 2 9 ( 0 0 ) 0 0 2 9 0 - 8
140
L.E. Trujillo et al. / Enzyme and Microbial Technology 28 (2001) 139 –144
Fig. 1. LsdA production at a fermenter scale. A 7.5 liters fermenter (B.E. Marubishi, Tokyo, Japan) containing 5 liters of fermentation medium was inoculated with clone PP15 at an initial OD530 ⫽ 0.2. Operation conditions were 28°C, pH 5.0, 700 rpm and aeration rate of 2 vvm. Upon depletion of glycerol as judged by raising pH and dissolved oxygen (21 h after inoculation), methanol was added to a final concentration of 1% (v/v). The methanol feeding rate was started at 2 g l⫺1 per h and gradually increased during the induction phase up to 3.5 g l⫺1 per h. Samples withdrawn at indicated time were analyzed for cellular mass (Œ) and levansucrase activity in the supernatant (F).
tence of an efficient protein secretion system which contrasts with low levels (0.5%) of secreted native proteins [6]. We report herein the recombinant expression and secretion of active LsdA in the methylotrophic yeast P. pastoris and the use of the recombinant enzyme for FOS production.
2. Materials and methods
ment from plasmid pALS5 [5], encoding the precursor LsdA lacking the first 57 residues was fused in frame to the pho1 signal sequence at the SmaI site of the expression vector pHILSI [8]. The resulting plasmid was named pALS105. Correct fusion of the lsdA gene with the pho1 signal peptide in pALS105 was confirmed by dideoxy-chain termination sequencing [11]. Genomic DNA isolation from yeast cells was performed as recommended in P. pastoris Expression Kit (Version 3.0) [8].
2.1. Microorganisms and media E. coli MC1061 [7] and P. pastoris strain GS115 (his4) [8] (Invitrogene, San Diego, CA) were used as expression and cloning hosts, respectively. MC1061 was grown at 37°C in LB medium [9] and ampicillin was added to final concentration of 50 g ml⫺1. P. pastoris was grown in YPD or MD medium prepared as described by Invitrogen [8]. The yeast fermentation medium consisted of: 5 g yeast extract l⫺1, 2 g NH4SO4 l⫺1, 14.2 g KH2PO4 l⫺1, 3.7 g MgSO4 䡠 7H2O l⫺1, 0.23 g CaCl2 䡠 2H2O l⫺1 and 25 g glycerol l⫺1, pH 5.0. Vitamins and traces were prepared and used as recommended by Cregg et al. [10]. 2.2. DNA manipulations All recombinant DNA techniques were performed according to Sambrook et al. [9]. The 1.9-kb SmaI-NruI frag-
2.3. Genetic transformation of Pichia pastoris P. pastoris GS115 spheroplasts were transformed with 10 g of plasmid pALS105, previously linearized at the StuI site so as to direct integration into the his4 locus by a single recombination event [8]. Transformants (His⫹ colonies) were selected through a prototrophy screening for histidinol dehydrogenase using minimal MD medium without histidine supplementation. 2.4. LsdA production in shaking batch cultures Baffled flasks (250 ml) containing 50 ml of YPD medium were inoculated from a single yeast colony and the cultures were grown with agitation at 200 rpm, during 18 h at 30°C (final OD600 ⫽ 2). Cells were harvested by centrif-
L.E. Trujillo et al. / Enzyme and Microbial Technology 28 (2001) 139 –144
Fig. 2. Levan synthesis by recombinant LsdA. Culture supernatant and cells from 1 ml of P. pastoris fermentation culture were separated by centrifugation. The cells were resuspended in 1 ml of disruption buffer (0.1% Triton X100, SDS 5 g l⫺1, 1 mM EDTA, 1 mM PMSF, 50 mM sodium phosphate pH 7.0) and milled with glass beads by vortexing. The cellular debris was spun down. Proteins from culture supernatant and cell lysate were separated by 12.5% SDS-PAGE. The gel was rinsed in distilled water and incubated overnight at 42°C in 100 g sucrose l⫺1, 100 mM sodium acetate, pH 5.5. Formed levan appeared as white bands. Lane 1: G. diazotrophicus levansucrase (20 ng), lane 2: GS115 cell lysate (50 g of total proteins); lanes 3,4,5: PP15 cell lysate 5, 25 and 50 g of total proteins respectively; lane 6: 10 l of GS115 10X-concentrated culture supernatant; lanes 7,8,9: PP15 10X-concentrated culture supernatant 1, 5 and 10 l, respectively.
ugation, washed twice with sterile distilled water and resuspended at an OD600 ⫽ 1 in 200 ml of YPD medium supplemented with 1% methanol instead of glucose for AOX1 promoter induction. The cultures were transferred to 2 liters baffled flasks and incubated at 30°C with agitation at 200 rpm. Methanol pulses to a final concentration of 0.5% were supplied every 24 h during culture growth. 2.5. Subcellular fractionation Spheroplasts were obtained as recommended by Invitrogen [8]: P. pastoris cells collected from 5 ml of broth were incubated with lyticase (Sigma) in 1 M Sorbitol, 1 mM EDTA, 5 g SDS l⫺1, 10 mM sodium citrate pH 7.5 for 45 min and centrifuged for 5 min at 10000⫻ g. Periplasmic proteins were recovered in the supernatant and the spheroplasts in the pellet were washed twice in 1.5 M Sorbitol, 1 mM EDTA, resuspended in 2 ml of lysis buffer containing 1 M Sorbitol, 10 mM MgCl2, 2 mM DTT, 1 mM EDTA, 1 mM PMSF, 50 mM NaCl, 50 mM sodium phosphate pH 6.0 and mechanically lysed by vortexing four times for 1 min in the presence of acid–washed 500-m glass beads (Sigma). Cytoplasmic activity was measured in the spheroplast extract.
141
on initial velocity measurements under the following conditions: 250 mM sucrose in 100 mM sodium acetate buffer pH 5.5, at 42°C. Levan formation on SDS-polyacrylamide gel was determined as reported previously by Herna´ndez et al. [1] Enzymatic studies were performed on recombinant LsdA partially purified from culture supernatants or cell extracts. Contaminants were precipitated with acetone 25% (v/v); after centrifugation, LsdA in the supernatant was precipitated by adding ethanol to a final concentration of 50% (v/v), and redissolved in 10 mM sodium acetate, pH 5.5. Activity dependence on temperature was studied in reaction mixtures of 50 l containing 0.3 units of LsdA, 250 mM sucrose and 100 mM sodium acetate buffer pH 5.5, incubated at several temperatures for 1 h. Enzyme activity dependence on pH was determined in reaction mixtures of 50 l containing 0.3 units of LsdA, 250 mM sucrose and buffered with 100 mM sodium acetate for pH values ranging between 3.0 and 6.0 or 100 mM sodium phosphate for pH range 6.0 – 8.0 incubated at 42°C for 1 h. Reactions were stopped by heating in a boiling water bath for 10 min and glucose release was measured as described above. 2.7. Polyacrylamide gel electrophoresis (PAGE), blotting and immuno-recognition SDS-PAGE in 12.5% gels was performed according to Laemmli [13]. For western blots, proteins were electrotransferred onto nitrocellulose membranes (Amersham Pharmacia Biotech, Uppsala, Sweden) using a Mini Trans-Blot Electrophoretic Transfer (BIORAD, Richmond, USA) at a constant current of 300 mA for 2 h. Dot blots were performed by applying protein samples directly onto the nitrocellulose membrane using a Bio-dotTM apparatus (BIORAD). LsdA in shaking batch cultures was detected by dot blotting 40 l supernatant samples taken after 72 h of methanol induction. For dot blot detection of LsdA in fermenter cultures, samples were withdrawn at different intervals and 20 l of culture supernatant or cell lysate were applied onto a nitrocellulose membrane. Immuno-detection of LsdA on nitrocellulose membranes was achieved with polyclonal antibodies generated against extracellular levansucrase from G. diazotrophicus SRT4 [5]. The amount of blotted recombinant protein was quantified by densitometry in an Imagine Densitometer GS-250 (BIORAD) using known amounts of natural LsdA as standard.
3. Results and discussion 2.6. Protein quantification and enzyme assays 3.1. Expression of the IsdA gene in P. pastoris Total proteins in culture supernatants and crude extracts were determined as described by Bradford [12]. Levansucrase activity was measured as glucose released by sucrose hydrolysis using a glucose oxidase-peroxidase-coupled colorimetric kit (Sigma). One unit of LsdA is defined as the amount of enzyme releasing 1 mol of glucose min⫺1 based
About 103 His⫹ colonies were obtained after GS115 spheroplast transformation with plasmid pALS105. Four His⫹ randomly selected clones were evaluated for LsdA expression in shaking batch cultures. In all cases, the enzyme accumulated active in the supernatant fluids although
142
L.E. Trujillo et al. / Enzyme and Microbial Technology 28 (2001) 139 –144
Fig. 3. Glycosylation analysis of P. pastoris-expressed LsdA. Proteins (40 g) from culture supernatant or cell lysate were denatured in 100 l of 0.5% SDS, 1% -mercaptoethanol at 100°C for 10 min. After addition of 1/10 volumen of 1 M-sodium citrate buffer pH 5.5 at 25°C, samples were reacted with Endoglycosidase H (New England Biolabs) at 0.25 units g⫺1 of total protein at 37°C for 10 h. For western blot, proteins (10 g) were separated by 12.5% SDS-PAGE and transferred to nitrocellulose membranes, then LsdA was immunodetected. Lane 1: G. diazotrophicus levansucrase (100 ng), lane 2: PP15 cell lysate, lane 3: Endo H-treated PP15 cell lysate; lane 4: PP15 culture supernatant; lane 5: Endo H-treated PP15 culture supernatant.
at different levels. These recombinant strains showed a wild-type methanol utilization phenotype (Mut⫹), indicating that the integration of the lsdA expression cassette in the host chromosome did not alter the functionality of the aox1 gene. Southern blot analysis of the highest expressor clone PP15 revealed that the lsdA expression cassette was integrated as a single copy at the chromosomal GS115 his4 locus by a single recombination event (data not shown). LsdA expression by clone PP15 was further studied at a fermenter scale. A typical 5 liters fermenter run is presented in Fig. 1. The induction phase lasted 64 h with a gradual increase of methanol feeding rate. During this phase the biomass increased from 114.0 to 356.7 wet-weight g l⫺1. The generation time of the recombinant strain was similar to that of the untransformed strain GS115 indicating that the recombinant production of LsdA was not toxic to the host cell. In supernatant samples withdrawn at different times, a progressive increase of levansucrase activity was observed (Fig. 1) that correlated with an increased production of the recombinant extracellular LsdA, as determined by immunoblotting (data not shown). Studies of the localization of LsdA activity in clone PP15 revealed that 18% (0.74 ⫻ 103 U l⫺1) of total LsdA activity was present in the culture medium, 81% (3.5 ⫻ 103 U l⫺1) was found, after cell fractionation, in the periplasmic fraction and a remaining 0.9% (0.037 ⫻ 103 U l⫺1) was associated with cellular debris. The final yield of recombinant LsdA, estimated by quantitative blotting procedure, was about 1% of the total protein. This yield is low in comparison to the expression of other proteins in the P. pastoris system [14]. Several strategies like optimization of fermentation conditions or construction of multi-copy integrant strains could be employed to increase LsdA expression levels in this host [14].
levan as a result of their fructosyl polymerase activity upon gel incubation in a sucrose-containing buffer (Fig. 2). Bands corresponding to both LsdA forms were fuzzy and migrated slower than the G. diazotrophicus enzyme used as control. The extracellular form migrated as a single band while the intracellular form appeared as a doublet suggesting the existence of different protein aggregates. As noted in further experiments (Fig. 3), these protein aggregates were dissociated and became in a single band when samples were prepared in -mercaptoethanol-containing buffer and heated to 95°C before subjecting to SDS-PAGE. The higher apparent molecular mass of the yeast-expressed LsdA (expected mass 58 kDa) in comparison to that of the natural enzyme (60.4 kDa) used as control, probably resulted from glycosylation. The presence of a single band suggests that no intermediates are accumulated, and points to a high efficiency of the glycosylation process. Three potential Nlinked glycosylation sites conforming the general rule N-XT/S, where X is not proline, are located at positions 202, 285 and 467 of the predicted mature recombinant LsdA sequence [5]. Digestion of both recombinant LsdA forms with endoglycosidase H altered their electrophoretic mobilities in denaturing SDS-PAGE, as revealed by western blotting (Fig. 3), confirming that the recombinant LsdA is glycosylated. The presence of N-linked sugars in both the intracellular and extracellular LsdA confirms that the protein entered the P. pastoris secretion pathway. Contrary to these results, Scotti et al. [15,16] reported that the Bacillus subtilis levansucrase driven by its own signal sequence or a yeast acid phosphatase signal did not enter the Saccharomyces cerevisiae secretory pathway, but accumulated in the cell in its precursor form associated with the cytoplasmic membrane. Fusion to the yeast invertase signal peptide allowed a small proportion of the enzyme to be translocated to the endoplasmatic reticulum (ER) lumen and being glycosylated. Fusion of the mature part of the B. subtillis levansucrase with the B. amyloliquefaciens ␣-amylase signal-pep-
3.2. Characterization of recombinant LsdA The cell-associated and extracellular forms of the recombinant LsdA, after being resolved in a 12.5% polyacrylamide gel under non-denaturing conditions, synthesized
Fig. 4. Thermostability of recombinant LsdA. Recombinant (Œ) or natural LsdA (■) (5 units) was incubated in 100 l of 50 mM sodium acetate pH 5.5 for 1 h at the indicated temperature and then residual activity was determined. Values represent means from 3 independent experiments.
L.E. Trujillo et al. / Enzyme and Microbial Technology 28 (2001) 139 –144
143
Fig. 5. A) FOS production by extracellular recombinant LsdA. Sucrose at 500 g l⫺1 was reacted with extracellular recombinant LsdA at 10 units g⫺1 of sucrose in 100 mM sodium acetate pH 5.5 at 40°C with agitation at 200 rpm, for the indicated time. Sugars in the reaction mixture were analyzed by HPLC in an Aminex HPX-87N column 0.78 ⫻ 30 cm (BIO-RAD, USA) coupled to a refractive index detector. The column temperature was kept at 85°C. A solution of 10 mM Na2SO4 was used as the mobile phase at a flow rate of 0.5 ml min⫺1. Pure solutions of fructose, glucose, sucrose and raffinose were used as standards. Values are the average of three different experiments. Symbols represent: Sucrose (}), glucose (䊐), trisaccharides (Œ), fructose (E). B) FOS production by intact P. pastoris cells expressing levansucrase. Recombinant yeast cells (0.5 g) were washed thrice with 100 mM sodium acetate pH 5.5 and incubated in 15 ml of sucrose 500 g l⫺1, 100 mM sodium acetate pH 5.5 at 37°C with agitation at 200 rpm, for the indicated time. HPLC analysis of the reaction mixtures and symbols are as described in note to Fig. 5A.
tide yielded an almost fully glycosylated protein. In neither case protein secretion occurred [15,16].
The dependence of the extracellular recombinant LsdA activity on pH and temperature was similar to that of the
144
L.E. Trujillo et al. / Enzyme and Microbial Technology 28 (2001) 139 –144
natural levansucrase [1,3] with the highest activity for splitting of sucrose at pH 5.0 and 60°C (data not shown). According to this observation, glycosylation did not alter optimal conditions for enzyme activity, but it did increase LsdA thermal stability (Fig. 4). 3.3. FOS production by cell-associated and extracellular LsdA As shown in Fig. 5A, the behavior of the extracellular form of the recombinant LsdA in the presence of sucrose was similar to that of the natural LsdA [3]. The enzyme converted sucrose mostly to FOS, mainly 1-kestose and nystose, liberating glucose. Small amounts of free fructose and levan were also formed as a consequence of sucrose hydrolysis and fructan polymerase reactions, respectively. The time course of FOS production (Fig. 5A) revealed that after 24 h more than 85% of the initial sucrose was consumed by the action of LsdA. The trisaccharide 1-kestose, which is the FOS of the highest commercial interest reached a maximal value of 217 g l⫺1. The efficiency of 1-kestose synthesis, expressed as the fraction of fructosyl residues transferred to sucrose was about 70% (w/w), as similarly reported for the natural LsdA [3]. Since most of LsdA was detected in the yeast periplasm, we investigated the possibility of producing FOS by incubating intact PP15 cells with sucrose. The reaction products were analyzed at different time points (Fig. 5B). After cell incubation for 30 h, 85% of the sucrose was consumed. FOS reached a maximal value of 100 g l⫺1 and represented about 27% of total sugars present in the reaction mixture. The efficiency of FOS synthesis was about 30% (w/w). This value, lower than that obtained with the free recombinant LsdA, could be explained by a limited sucrose diffusion to the periplasm resulting in suboptimal concentrations of the substrate for FOS synthesis. Ramos et al. [17] reported direct action on sucrose of intact recombinant P. pastoris cells accumulating the S. cerevisiae invertase in the periplasm. To date, industrial production of FOS depends chiefly on fungal enzymes [2], although a new promising perspective is their direct production in transgenic plants [18,19]. The availability of recombinant P. pastoris as a new source of a FOS-producing enzyme might result of biotechnological interest.
Acknowledgments We thank Dr. Pablo Vera and Ismael Rodrigo for critical and constructive review of the manuscript. This work was supported partially by Grant f/2770 –1 from The International Foundation for Sciences.
References [1] Herna´ndez L, Arrieta J, Menendez C, Vazquez R, Coego A, Suarez V, Selman G, Petit-Glatron MF, Chambert R. Isolation and enzymatic properties of levansucrase secreted by Acetobacter diazotrophicus SRT4, a bacterium associated with sugar cane. Biochem J 1995;309:113– 8. [2] Yun JW. Fructooligosaccharides-ocurrence, preparation and application. Enzyme and Microbial Technology 1996;19:107–17. [3] Ta´mbara Y, Hormaza JV, Pe´rez C, Leo´n A, Arrieta J, Herna´ndez L. Structural analysis and optimised production of fructo-oligosaccharides by levansucrase from Acetobacter diazotrophicus SRT4. Biotechnology Letters 1999;21:117–21. [4] Herna´ndez L, Ramı´rez R, Hormaza JV, Madrazo J, Arrieta JG. Increased levansucrase production by a genetically modified Acetobacter diazotrophicus strain in shaking batch cultures. Letters in Applied Microbiology 1999;28:41– 4. [5] Arrieta JG, Herna´ndez L, Coego A, Sua´rez V, Balmori E, Mene´ndez C, Petit-Glatron MF, Chambert R, Selman-Housein G. Molecular characterization of the levansucrase gene from the endophytic sugarcane bacterium Acetobacter diazotrophicus SRT4. Microbiology 1996;142:1077– 85. [6] Romanos MA, Scorer CA, Clare JJ. Foreign gene expression in yeast: a review. Yeast 1992;8:423– 88. [7] Meissner PS, Sisk WP, Berman ML. Bacteriophage cloning system for the construction of directional cDNA libraries. Proc Natl Acad Sci USA 1987;84:4171. [8] Pichia Expression kit. A manual of methods for expression of recombinant proteins in P. pastoris, Version 3.0. San Diego, USA: Invitrogen Corporation. [9] Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1989. [10] Cregg JM, Tschopp JF, Stillman C, Siegel R, Akong M, Craig WS, Buckholz RG, Madden KR, Kellaris PA, Davies GR, Smiley BL, Cruze J, Torregrossa R, Velicelebi G, Thill GP. High-level expression and efficient assembly of hepatitis B surface antigen in the methylotrophic yeast Pichia pastoris. Bio/Technology 1987;5:479 – 85. [11] Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain terminating inhibitors. Proc Natl Acad Sci USA 1977;74:5463–7. [12] Bradford MM. A rapid and sensitive method for quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976;72:248 –54. [13] Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680 –5. [14] Cregg JM, Vedvick TS, Rashke WC. Recent advances in the expression of foreign genes in Pichia pastoris. Bio/Technology 1993;11:905–10. [15] Scotti PA, Chambert R, Petit-Glatron MF. Extracellular levansucrase of Bacillus subtilis produced in yeast remains in the cell in its precursor form. Yeast 1994;10:29 –38. [16] Scotti PA, Praestegaard M, Chambert R, Petit-Glatron MF. The targeting of Bacillus subtilis levansucrase in yeast is correlated to both the hydrophobicity of the signal peptide and the net charge of the N-terminus mature part. Yeast 1996;12:953– 63. [17] Ramos R, Cruz A, Figueroa N. Proceeding for industrial scale production of glucose-fructose liquors from sucrose. Cuban Patent No. 92–91, 1991. [18] Sevenier R, Hall RD, Van der Meer I, Hakkert HJC, van Tunen AJ, Koops AJ. High level fructan accumulation in a transgenic sugar beet. Nature Biotechnology 1998;16:843– 6. [19] Hellwege EM, Gritscher D, Willmitzer L, Heyer AG. Transgenic potato tubers accumulate high levels of 1-kestose and nystose: functional identification of a sucrose-sucrose 1-fructosyltransferase of artichoke (Cynara scolymus) blossom discs. The Plant Journal 1997; 12:1057– 65.