High-level expression of Aliciclobacillus acidocaldarius thioredoxin in Pichia pastoris and Bacillus subtilis

High-level expression of Aliciclobacillus acidocaldarius thioredoxin in Pichia pastoris and Bacillus subtilis

Protein Expression and Purification 30 (2003) 179–184 www.elsevier.com/locate/yprep High-level expression of Aliciclobacillus acidocaldarius thioredox...

157KB Sizes 0 Downloads 28 Views

Protein Expression and Purification 30 (2003) 179–184 www.elsevier.com/locate/yprep

High-level expression of Aliciclobacillus acidocaldarius thioredoxin in Pichia pastoris and Bacillus subtilis Digilio Filomena Anna,a Morra Rosa,a Pedone Emilia,c Bartolucci Simonetta,b and Rossi Mosea,* b

a Institute of Protein Biochemistry, CNR, Via Marconi 10, 80125 Naples, Italy Department of Biological Chemistry, University of Naples Federico II, Via Mezzocannone 16, 80134 Naples, Italy c Istituto di Biostrutture e Bioimmagini, Via Mezzocannone 8, 80134 Naples, Italy

Received 12 December 2002, and in revised form 4 March 2003

Abstract Thioredoxins are ubiquitous proteins which catalyze the reduction of disulfide bridges on target proteins and are involved in many cellular reactions. In a previous work, a thioredoxin from the thermophilic organism Aliciclobacillus acidocaldarius (Alitrx) was purified, characterized, and its gene expressed in Escherichia coli. In order to produce larger quantities of Alitrx, the protein has been expressed in the methylotrophic yeast Pichia pastoris and in the gram positive bacteria Bacillus subtilis. The growth conditions of strains showing high-level expression of Alitrx were optimized for both systems in shake-flask cultures. Active proteins were secreted in the culture media at a level of approximately 0.9 and 0.5 g/L, respectively, for P. pastoris and B. subtilis. The proteins were purified almost to homogeneity by a thermal precipitation procedure, with a 90-fold and 50-fold higher total yield with respect to that obtained with the same protein expressed in E. coli. The results indicate that either of these two systems could be utilized as a host for large-scale production of recombinant Alitrx. Ó 2003 Elsevier Science (USA). All rights reserved.

Thioredoxins (Trxs) are small proteins that have a redox-active disulfide/dithiol group within the conserved active site sequence Cys–Gly–Pro–Cys. Reduced Trx catalyzes the reduction of disulfide bonds in multiple substrate proteins, and oxidized Trx is reversibly reduced by the action of Trx reductase and NADPH [1,2]. Trx is widely present in prokaryotes and eukaryotes, and appears to be ubiquitous in almost all living cells [1,3]. It is a multifunctional protein that has regulatory roles in cellular signalling and gene transcription [3] in addition to cytoprotective activities due to its quenching of reactive oxygen species [4,5]. Thioredoxin systems serve as hydrogen donors, for example, for ribonucleotide reductase, phosphoadenosyl phosphosulfate reductase, and methionine sulfoxide reductase [1]. Moreover, thioredoxin has been involved in thiol-disulfide exchange and disulfide bond formation [6]. The multiple roles played by Trx as well as its ability to reduce di* Corresponding author. Fax: +39-081-6132277. E-mail address: [email protected] (R. Mose).

sulfide bridges that do not respond to other cellular reductants have potential applications for the solution of certain problems, especially in nutrition and medicine [7–9]. Recently, the cloning and characterization of thioredoxin from the thermophilic organism Aliciclobacillus acidocaldarius (Alitrx) and the expression of the biologically active protein in Escherichia coli was reported [10]. This protein shows a higher resistance to temperature relative to its mesophilic counterparts [11,12], and this feature offers several interesting advantages in terms of biotechnological applications. Although protein obtained in E. coli was sufficient to perform a variety of experiments, the efficiency of Alitrx production (about 5.4% of total cytosolic amount) and the complex purification procedures proved unsatisfactory. In an attempt to establish a more efficient system for the production of this protein, Pichia pastoris and Bacillus subtilis were investigated as potential expression systems. Both of these expression systems can be scaled-up from shakeflask to large fermenter cultures and both are able to

1046-5928/03/$ - see front matter Ó 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S1046-5928(03)00095-0

180

D.F. Anna et al. / Protein Expression and Purification 30 (2003) 179–184

efficiently secrete heterologous proteins [13–16]. Secretion serves as a major purification step by separating the foreign protein product from the bulk of native proteins and other molecules contained in the cell. In addition, the thermostability of the thermophilic proteins was exploited for its purification by a thermal precipitation procedure. This paper reports the cloning and expression of Alitrx from A. acidocaldarius in P. pastoris and B. subtilis and its purification. The level of purified recombinant protein reached approximately 900 mg/L of culture medium in P. pastoris, and approximately 500 mg/L of culture medium in B. subtilis.

Materials and methods Analytical methods for DNA and protein DNA manipulations were done according to Sambrook et al. [17]. DNA sequencing was carried out by the Sanger method, using the Sequenase version 2.0 sequencing kit (Amersham) on both strands. PCRs were done by using Fisher Red Taq. Protein concentration was determined by the Pierce method [18], using BSA as standard. SDS–PAGE was performed essentially, as described by Laemmli [19]. Strains and vectors Escherichia coli DH5a strain was used for yeast and bacillus vectors propagation, while the P. pastoris GS115 strain (Invitrogen) and the B. subtilis DB428 strain (provided by Arnold) were used as the final hosts. The pPIC9 and pHILS1 (Invitrogen) vectors were used for the extracellular expression in P. pastoris, while the shuttle pBE3 vector (provided by Arnold) was used for the extracellular expression in B. subtilis. Construction of expression plasmids and transformation of P. pastoris Two Alitrx expression constructs were created for the expression in P. pastoris. In both, the complete Alitrx coding region was amplified by PCR using the recombinant E. coli pTR99A vector as a template. The primers, Alit Rev AGAGAACTAGTGAATTCATG GCTATGATGACGTTG and Alit Forw AGAGAA CTAGTGAATTCTTACTGTAATACATCTGC were designed to contain the EcoRI restriction site at each end. The PCR product was purified by agarose gel extraction, cut with EcoRI, and inserted into EcoRI cut pPIC9 and pHILS1 plasmids, in frame with the Saccharomyces cerevisiae a-factor signal peptide and the Pichia pastoris PHO1 leader sequence, respectively.

These two recombinant vectors were subjected to DNA sequencing to confirm the nucleotide sequence and the correct frame with the peptide leader sequences. Manipulation of P. pastoris Procedures suggested by the providers of the P. pastoris expression system were used (Invitrogen). The expression plasmids were linearized with BglII and the DNAs were introduced into the P. pastoris strain GS115 by electroporation. The transformants were selected for ability to grow in the absence of histidine (hisþ ) and then were further screened for disruption of the AOX gene by selecting for slow growth in the presence of methanol (muts ). A total of 100 hisþ muts /mutþ transformants were tested for expression of Alitrx in 10 mL cultures, by checking the cell-free induction medium on 15% acrylammide Tris–glycine gels followed by Coomassie blue staining. The Alitrx transformed colonies that produced the highest levels of Alitrx were used for further analysis. Having identified the best protein-producing Alitrx transformants, named ppAlitrxR and ppAlitrxH8, and determined the optimal growth pH after induction and the time-course, intermediate-scale culture was carried out using an incubator Certomat (Braun). The selected clones were grown in YPD medium [1.34% yeast nitrogen base without amino acids (Bio 101), 2% dextrose] at 30 °C for 12 h. The cultures were then transferred to BMGY medium (0.1 M potassium phosphate buffer, pH 6.0 (Sigma), 1.34% yeast nitrogen base without amino acids, 1% glycerol, 400 lg/L D-biotin (Sigma), 1% yeast extract, and 2% Bacto-peptone) and continued growing at 30 °C with vigorous shaking, until the OD600 was approximately 1. Then, the cells were pelleted and, thereafter, resuspended on one-half volume of induction medium in BMMY (same as BMGY but 1% methanol in place of glycerol, plus 1% casamino acids). The induction was carried out at 29 °C, 240 rpm, in 3 L flasks for 5 days, after which time the expression level of the protein does not increase further. Construction of expression plasmids and transformation of B. subtilis The Alitrx coding region contained in the recombinant E. coli vector pTR99A was amplified by PCR using two primers designed to contain a SmaI re striction site at the 50 end (TCCCCCGGGATGGC TATGATGACGTTGA) and a BamHI restriction site at the 30 end (CGCGCGGATCCTTACTGTAATAC ATCTGC). This fragment was purified by agarose gel electrophoresis, cut with SmaI and BamHI restriction enzymes, and subcloned in frame with the B. subtilis subtilisin E signal peptide in the pBE3 vector [20] in-

D.F. Anna et al. / Protein Expression and Purification 30 (2003) 179–184

stead of the subtilisin coding region previously cut off by using SmaI and BamHI restriction sites. The recombinant vector was checked by sequencing and was used to transform DB428 B. subtilis cells by the two-step protocol [21]. The transformants were selected on LB plates containing Kanamicin. Manipulation of B. subtilis The B. subtilis strain was routinely grown with vigorous agitation at 37 °C in synthetic medium [21] or in complex medium. For expression studies the sporulation medium 2 SG [22] was used. After having determined the optimal expression conditions in small scale, the growth was done at 37 °C for 24 h with vigorous agitation (260 rpm). To determine the expression of the protein, the cell-free induction medium was analyzed by SDS–PAGE electrophoresis followed by Coomassie blue staining. Activity assay and purification of recombinant Alitrx The turbidimetric assay of the reduced insulin precipitation was used to monitor the protein activity according to Holmgren [10,23]. In consequence of the reduction of the inter-chain disulfide bridges between chains A and B of insulin catalyzed by the Alitrx, the turbidity of the solution increases due to the precipitation of the free B chains. To the standard assay mixture containing 0.1 M sodium phosphate buffer (pH 7.0)/0.13 mM bovine insulin was added to 1 mM dithiothreitol (DTT) and the Alitrx in a final volume of 1 mL at 30 °C. One unit is defined as the amount of protein that produces an increase of 0.1 OD in 1 min at 650 nm. The recombinant protein present in the cell-free growth medium was purified nearby to homogeneity by two successive heat treatments (at 80 °C, respectively) for 20 min. Samples were extensively dialyzed against 20 mM Tris/HCl buffer (pH 8.0)/2 mM EDTA and loaded onto a Resource Q Column (Amersham) equilibrated in the same buffer. The column was eluted with a linear gradient of 0–0.2 M NaCl at a flow rate of 60 mL/h, with the active pools obtained with low ionic strength.

Results and discussion Expression in P. pastoris Two P. pastoris expression plasmids were constructed and used to transform the yeast. To explore which signal peptide normally used for P. pastoris secretion was best for our protein, two constructs were made by placing the

181

Alitrx coding sequence downstream of the P. pastoris alcohol oxidase promoter. The first construct used the P. pastoris PHO1 signal sequence, while the second used the S. cerevisiae a-factor leader sequence. The Alitrx gene was amplified by PCR and ligated into EcoRI cut vectors containing signal peptides, as described under Materials and methods, to yield plasmids ppHLSAlit and ppPICAlit. For both the constructs, approximately 100 Hisþ clones were tested for expression by direct checking of culture samples with Coomassie blue staining after electrophoretic separation on 15% SDS– PAGE gels. Ten clones secreting a protein of the expected molecular weight (approximately 10–12 kDa) (data not shown) were chosen for further analysis, as they appeared to secrete adequate levels of protein. The cell-free induction media, together with an equal amount of protein from cells transformed with a nonrecombinant expression vector (negative control), were assayed for insulin reductase activity. Comparable activities were found in each supernatant. Small-scale expression studies were performed in BMMY medium to further optimize the expression conditions with a strain transformed with each plasmid, with the aim of obtaining the highest expression levels (active protein per liter of broth) possible. GS115/ppAlitR (containing the ppPICAlit recombinant plasmid) and GS115/ppAlitH8 (containing the ppHLSAlit recombinant plasmid) clones were induced in large shake-flask cultures, as described in Materials and methods, at a dilution of one-half volume of induction medium in BMMY for a period of 5 days. During this period, increasing amounts of Alitrx accumulated. After this time, the level of secreted protein was approximately 0.9 g/L for GS115ppAlitH8 and 0.8 g/L for GS115/ppAlitR, respectively. The induction was sustained by adding methanol to a concentration of 1% every 24 h. The optimum pH was determined to be 6.0. In order to inhibit further the proteases of P. pastoris, casamino acids were also added. Under these conditions, other protein products were also detected in the supernatant, even though Alitrx was the predominant protein in the induction media (Fig. 1). The supernatant was processed by two successive thermal precipitations at 80 °C for 20 min. The protein purified after this precipitation was approximately 75% pure (Fig. 1) and had a high catalytic activity, as confirmed by the turbidimetric assay. However, a second prominent and sharp band was also present and was further investigated. To remove this second protein, the sample was subject to anionic-exchange chromatography, as summarized in Table 1. Our data indicated that, after this step, the two proteins were separated and that only the fraction containing the lower band showed insulin reductase activity, while the upper band was an unknown P. pastoris protein, as suggested by sequence determination and the literature data [24].

182

D.F. Anna et al. / Protein Expression and Purification 30 (2003) 179–184

Fig. 1. SDS–PAGE stained with Coomassie blue of the protein samples expressed in Pichia pastoris. Lanes 1 and 6: Fermentas Molecular Weight Marker (kDa). Lane 2 shows the cell free induction medium obtained from GS115 transformed with the non-recombinant plasmid pPIC9. Lanes 3–5 show the samples from the clone GS115/ppAlitR, while lanes 7–9 show the samples from the clone GS115ppAlitH8. Lanes 3 and 7: cell-free induction medium; lanes 4 and 8: samples after a two-step thermal precipitation at 80 °C; lanes 5 and 9: active pools obtained from an anionic-exchange chromatography.

Fig. 2. SDS–PAGE analysis of the protein samples expressed in Bacillus subtilis. Different samples obtained during the purification procedure of the recombinant Alitrx were analyzed by SDS–PAGE. Lane 1: cell-free sporulation medium; lane 2: sample after a one-step thermal precipitation at 80 °C; lane 3: active pools obtained from an anionicexchange chromatography; lane 4: Fermentas Molecular Weight Marker (kDa).

Expression in B. subtilis For expression in B. subtilis, we placed the Alitrx coding region under control of the B. subtilis AprE (subtilisin E) gene promoter, which is highly expressed during sporulation, and in frame with the peptide secretion signal of this gene [20]. We selected positive clones by means of Kanamicin resistance and after restriction and PCR analysis to confirm the presence of the recombinant plasmid, we carried out shake-flask culture expression with three of these clones to determine the best parameters for expression. Cell-free sporulation medium samples were examined by 15% SDS–PAGE followed by Coomassie blue staining for the presence of Alitrx. As already described for P. pastoris, we also evaluated catalytic activity of Alitrx. As there were no differences between these clones, we performed all further expression optimization experiments on one of these, namely bsAlitrx1. Expression was induced by growing the strain with vigorous agitation inoculated to an initial OD600 of 0.15 for 24 h. Under these conditions, only a few contaminating products were detected, and Alitrx was the predominant protein in the sporulation medium (Fig. 2). We monitored protein production at 1-h intervals for 26 h of growth by measuring insulin-reduction and by SDS–PAGE. The optimal time of induction was determined to be 24 h;

after this period, the quantity of our protein decreased, suggesting that, although the strain used was a ‘‘protease -deficient’’ strain, proteases were still produced. The supernatant from 100 mL of culture medium was collected and used for purification, as described for P. pastoris. The results of the purification procedure are summarized in Table 2. Unlike P. pastoris, the supernatant was subjected to only one round of thermal precipitation at 80 °C for twenty min, which was sufficient to eliminate most contaminating proteins present in the medium (Fig. 2; Table 2). Also in this case, the supernatant was subjected to an anion-exchange chromatography step, as described in Materials and methods and as shown in Fig. 2, in order to further remove contaminating proteins. Comparison of the recombinant thioredoxins The specific activity of each protein was characterized by turbidimetric assay and no differences were observed between Alitrx expressed in P. pastoris, B. subtilis, and E. coli. Fig. 3 shows the progress of DTT-dependent reduction of bovine insulin disulfides in the presence of increasing quantities of pure recombinant Alitrx. The activity curve obtained for 10 lM of protein proved

Table 1 Purification of the Alitrx from Pichia pastoris Purification step

Total volume (mL)

Activity (U/mL)

Protein (mg/mL)

Specific activity (U/mg)

Total units (U)

Purification factor

Yield (%)

Cell-free supernatant Heat-treated supernatant Anion-exchange chromatographya

300 250 0.5

1.14 6.88 6.77

1.60 1.18 0.88

0.7 5.8 7.9

342 1720 1692

8.3 11.3

100 503 495

a

Only 0.5 mL over 250 mL was passed through the resin and the purified protein was concentrated to 0.5 mL.

D.F. Anna et al. / Protein Expression and Purification 30 (2003) 179–184

183

Table 2 Purification of the Alitrx from Bacillus subtilis Purification step

Total volume (mL)

Activity (U/mL)

Protein (mg/mL)

Specific activity (U/mg)

Total units (U)

Cell-free supernatant Heat-treated supernatant Anion-exchange chromatographya

100 90 0.5

1.05 3.75 6.2

1.241 0.989 0.612

0.85 3.75 5.62

105 338 558

a

Purification factor

Yield (%)

4.4 6.5

100 322 531

Only 0.5 mL over 250 mL was passed through the resin and the purified protein was concentrated to 0.5 mL.

The excellent yield and simple purification make the P. pastoris and B. subtilis expression systems convenient for Alitrx production and commercial exploitation.

Acknowledgments

Fig. 3. Insulin reductase activity of pure recombinant Alitrx expressed in Pichia pastoris. The DTT-dependent reduction of bovine insulin disulfides was done at 30 °C in the presence of increasing concentrations of pure recombinant thioredoxin, as described in Materials and methods. Spontaneous precipitation reaction in the absence of recombinant Alitrx (s) or in the presence of different concentrations of pure Alitrx: (d) 1 lM; (I) 4 lM; and (N) 8 lM. The progress curve obtained for (j) 10 lM protein was identical with that obtained for 8 lM, suggesting that the assay was saturated. The same results were obtained for Alitrx expressed from Bacillus subtilis.

identical to that for 8 lM, indicating that the assay was at Alitrx saturation. As shown in Tables 1 and 2, the activity of thioredoxin in supernatants was underestimated since, after purification, an approximately 500% gain of activity units was observed.

Conclusion A simple production and purification scheme has been developed to obtain nearly pure Alitrx in large quantities using either of two expression systems secreting heterologous proteins, P. pastoris and B. subtilis. In either system, thioredoxin displayed characteristics and catalytic activity similar to the protein previously expressed in E. coli. The level of expression obtained using these secreted systems was approximately 70-fold higher than that observed with the best E. coli expression strain. Moreover, the maximum level of expression achieved with the constructs (approximately 0.9 and 0.5 g/L of the P. pastoris and B. subtilis cell-free growth medium, respectively, using shake-flask culture conditions) is expected to increase significantly in high-density fermenter cultures.

This research was supported by a collaborative grant from the Campania Region to Prof. Mose Rossi. We thank D. Cavaliere, G. Imperato, and C. Sole for their technical assistance. We are grateful to Arnold for kindly giving us B. subtilis vector and strain.

References [1] A. Holmgren, Thioredoxin. Annu. Rev. Biochem. 54 (1985) 237– 271. [2] A. Holmgren, M. Bjornstedt, Thioredoxin and thioredoxin reductase, Methods Enzymol. 25 (1995) 199–208. [3] H. Masutani, J. Yodoi, Thioredoxin, Overview Methods Enzymol. 347 (2002) 279–286. [4] A. Spector, G.Z. Yan, R.R. Huang, M.J. McDermott, P.R. Gascoyne, V. Pigiet, The effect of H2 O2 upon thioredoxinenriched lens epithelial cells, J. Biol. Chem. 263 (1988) 4984–4990. [5] M.R. Fernando, H. Nanri, S. Yoshitake, K. Nagata-Kuno, S. Minakami, Thioredoxin regenerates proteins inactivated by oxidative stress in endothelial cells, Eur. J. Biochem. 209 (1992) 917– 922. [6] V.P. Pigiet, B.J. Schuster, Thioredoxin-catalyzed refolding of disulfide-containing proteins, Proc. Natl. Acad. Sci. USA 83 (1986) 7643–7647. [7] B.B. Buchanan, P. Schurmann, P. Decottignies, R.M. Lozano, Thioredoxin: a multifunctional regulatory protein with a bright future in technology and medicine, Arch. Biochem. Biophys. 314 (1994) 257–260. [8] G. del Val, B.C. Yee, R.M. Lozano, B.B. Buchanan, R.W. Ermel, Y.M. Lee, O.L. Frick, Thioredoxin treatment increases digestibility and lowers allergenicity of milk, J. Allergy Clin. Immunol. 4 (1999) 690–697. [9] T. Okamoto, K. Asamitsu, T. Tetsuka, Thioredoxin and mechanism of inflammatory response, Methods Enzymol. 347 (2002) 349–360. [10] S. Bartolucci, A. Guagliardi, E. Pedone, D. De Pascale, R. Cannio, L. Camardella, M. Rossi, G. Nicastro, C. de Chiara, P. Facci, G. Mascetti, C. Nicolini, Thioredoxin from Bacillus acidocaldarius: characterization, high-level expression in Escherichia coli and molecular modelling, Biochem. J. 328 (1997) 277– 285. [11] E.M Pedone, S. Bartolucci, M. Rossi, M. Saviano, Computational analysis of the thermal stability in thioredoxins: a molecular dynamics approach, Biomol. Struct. Dyn. J. 16 (1998) 437–445.

184

D.F. Anna et al. / Protein Expression and Purification 30 (2003) 179–184

[12] E. Pedone, R. Cannio, M. Saviano, M. Rossi, S. Bartolucci, Prediction and experimental testing of Bacillus acidocaldarius thioredoxin stability, Biochem. J. 339 (1999) 309–317. [13] J.M. Cregg, T.S. Vedvick, W.C. Raschke, Recent advances in the expression of foreign genes in Pichia pastoris, Biotechnology 11 (1993) 905–910. [14] Y. Laroche, V. Storme, J. De Mentter, J. Messens, M. Lauwereys, High-level secretion and very efficient isotopic labeling of tick anticoagulant peptide (TAP) expressed in the methylotrophic yeast Pichia pastoris, Biotechnology 12 (1994) 1119–1124. [15] M. Romanos, Advances in the use of Pichia pastoris for high-level gene expression, Curr. Opin. Biotechnol. 6 (1995) 527–533. [16] K.H. Lam, K.C. Chow, W.K. Wong, Construction of an efficient Bacillus subtilis system for extracellular production of heterologous proteins, J. Biotechnol. 63 (1998) 167–177. [17] J. Sambrook, E.F. Fritsch, T. Maniatis, Molecular Cloning: A Laboratory Manual, second ed., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1989. [18] P.K. Smith, R.I. Khrohn, G.T. Hermanson, A.K. Mallia, F.H. Gartner, M.D. Provenzano, E.K. Fujimoto, N.M. Goeke, B.J.

[19]

[20]

[21] [22]

[23]

[24]

Olson, D.C. Klenk, Measurement of protein using bicinchoninic acid, Anal. Biochem. 150 (1985) 76–85. U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature (London) 227 (1970) 680–685. H. Zhao, F.H. Arnold, Functional and non-functional mutations distinguished by random recombination of homologous genes, Proc. Natl. Acad. Sci. USA 94 (1997) 7997– 8000. C.R. Harwood, S.M. Cutting, Molecular biological methods for Bacillus, John Willey & Sons, Chichester, UK, 1990. T.J. Leighton, R.H. Doi, The stability of messenger ribonucleic acid during sporulation in Bacillus subtilis, J. Biol. Chem. 246 (1971) 3189–3195. A. Holmgren, Thioredoxin catalyzes the reduction of insulin disulfides by dithiothreitol and dihydrolipoamide, J. Biol. Chem. 254 (1979) 9627–9632. N. Roy, S. Padmanabhan, M. Smith, L. Shi, M. Navre, G. Das, Expression of human gelatinase B in Pichia pastoris, Protein Expr. Purif. 16 (1999) 324–330.