Mol Biotechnol DOI 10.1007/s12033-014-9740-6
RESEARCH
High-Yield Expression in Escherichia coli, Purification and Application of Budding Yeast K2 Killer Protein Monika Podoliankait_e • Juliana Luksˇa • Gintautas Vysˇniauskis Jolanta Sereikait_e • Vytautas Melvydas • Saulius Serva • Elena Servien_e
•
Ó Springer Science+Business Media New York 2014
Abstract Saccharomyces cerevisiae K2 toxin is a highly active extracellular protein, important as a biocontrol agent for biotechnological applications in the wine industry. This protein is produced at negligible levels in yeast, making difficult to isolate it in amounts sufficient for investigation and generation of analysis tools. In this work, we demonstrate the use of a bacterial system for expression of the recombinant K2 protein, suitable for generation of antibodies specific for toxin of the yeast origin. Synthesis of the full-length S. cerevisiae K2 preprotoxin in Escherichia coli was found to be toxic to the host cell, resulting in diminished growth. Such effect was abolished by the introduction of the C-terminal truncation into K2 protein, directing it into non-toxic inclusion body fraction. The obtained protein is of limited solubility thus, facilitating the
Monika Podoliankait_e and Juliana Luksˇa have contributed equally to this study. M. Podoliankait_e J. Luksˇa G. Vysˇniauskis V. Melvydas E. Servien_e (&) Laboratory of Genetics, Institute of Botany, Nature Research Centre, Akademijos Str. 2, 08412 Vilnius, Lithuania e-mail: [email protected] G. Vysˇniauskis Department of Biopharmacy, State Research Institute for Innovative Medicine, Zygimantu 9, 01102 Vilnius, Lithuania J. Sereikait_e S. Serva E. Servien_e Department of Chemistry and Bioengineering, Faculty of Fundamental Sciences, Vilnius Gediminas Technical University, Sauletekio 11, 10223 Vilnius, Lithuania S. Serva Department of Biochemistry and Molecular Biology, Faculty of Natural Sciences, Vilnius University, Ciurlionio 21, 03101 Vilnius, Lithuania
purification by simple and efficient chromatography-free procedure. The protein aggregates were successfully refolded into a soluble form yielding sufficient amounts of a tag-less truncated K2 protein suitable for polyclonal antibody production. Antibodies were raised in rabbit and found to be specific for detection of both antigen and native S. cerevisiae K2 toxin. Keywords Saccharomyces cerevisiae K2 toxin Escherichia coli K2 recombinant protein Antibody
Introduction The toxins of the yeast origin confer a growth advantage to their hosts by increasing survival in clinical, environmental, and industrial ecosystems. Nowadays, particular interest in yeast killer strains or isolated toxins has surged. They find application in industry (food protection from spoiling, wine production), agriculture (phytopathogen control), or medicine (creation of a new generation vaccines and antibiotics, development of antifungal immunotherapy) [1–3]. Killer proteins and toxin-recognizing antibodies have been tested as a potential antiviral and anticancer agents [4]. Therefore, much attention is focused on creation of killer-specific monoclonal and polyclonal antibodies used for in-depth studies of killer systems and practical applications [5]. K2 killer protein of Saccharomyces cerevisiae is the most frequent toxin among yeast dominating in the vineyard-winery ecosystem, where up to half of yeast strains are hosts for killer systems, and some 80 % of them are of K2-type [3, 6]. Double-stranded RNA-derived K2 toxin is synthesized as a precursor of 362 amino acids, consisting of the N-terminal secretion signal, a, and b subunits
123
Mol Biotechnol
(172 and 140 amino acids, respectively). During the passage through the yeast secretory pathway, K2 protein is enzymatically processed into a mature cytotoxic a/b heterodimer [7, 8]. Functionality of protein is highly sensitive for sequence alterations resulting in activity being lost easily [7]. The protein is secreted; however, high biologic activity determines mass quantity of the killer protein remain negligible. The yield of yeast K2 toxin, isolated from 3 L of culture medium, is less than 1 lg [9]. This possesses a significant burden for the biochemical research requiring considerable amounts of protein, e.g., producing of the specific antibodies and developing of the immunodetection systems. Isolation of many bioactive peptides from natural sources is often expensive and time consuming. Therefore, the bacteria, in particular Escherichia coli, are employed as the hosts for large-scale production of proteins [10, 11]. However, the host stress response systems can be triggered upon the massive expression of the recombinant genes impairing plasmid replication and stability, cell growth and viability [12]. The produced foreign protein is a target for proteolytic degradation or insoluble aggregate formation. Aggregation can be minimized by precise control of the induction parameters, such as time and/or temperature, aimed to reduce the expression rate, adjusting the codon usage, protein engineering, or co-expression of the chaperones [12–15]. On the other hand, formation of the protein aggregates in inclusion bodies (IBs) might be beneficial, in order to simplify protein purification procedure, reducing number of purification steps and significantly improving cost-efficiency ratio [11, 13, 16]. During this work, the prokaryotic expression system was engaged to produce significant amounts of yeast K2 killer protein. Toxicity of the full-length foreign protein was neutralized by mutagenesis and directing it into the state of limited solubility. Expression conditions were optimized to allow refolding of tag-less target protein. Time and costefficient strategy for recombinant protein purification was elaborated, and genuine K2 toxin-specific antibodies were generated.
Materials and Methods Bacterial and Yeast Strains Escherichia coli strain DH5a (F-(/80dD(lacZ)M15) ? recA1 endA1 gyrA96 thi1 hsdR17 (rk mk ) supE44 relA1 deoR D (lacZYA-argF) U169) [17] was used as the host for recombinant plasmid construction and amplification. Protein expression was conducted in E. coli strain BL21 (DE3) (F- ompT hsdSB (rB mB ) gal dcm) [18].
123
Saccharomyces cerevisiae strains M437 (wt, HM/HM [kil–K2]) and a0 1 (MATa leu2-2 [Kil-0]) were used for K2 toxin isolation and for testing of the dsRNA-free strain [19]. Construction of Vectors for Expression in E. coli General procedures for the construction and analysis of recombinant DNAs were performed according to Green and Sambrook [20]. Restriction enzymes (SalI, Eco147I, XhoI), T4 DNA ligase, calf intestine alkaline phosphatase (CIAP), DNA size marker (GeneRulerTM DNA Ladder mix) were obtained from ThermoFisher Scientific (Vilnius, Lithuania) and used according to the manufacturer’s recommendations. The recombinant p28-kil2 plasmid (Fig. 1a) was obtained by the insertion of a K2 preprotoxin gene into the pET28 plasmid, downstream the phage T7 promoter. For this purpose, sequence encoding K2 toxin precursor was excised from the pYEX12 plasmid [8] using SalI and ligated into the vector pET28, digested by the same restriction enzyme and dephosphorylated by CIAP. For the construction of the p28-kilD plasmid (Fig. 1b), 30 -truncated K2 preprotoxin gene was excised from the pYEX12 using SalI and Eco147I and subsequently cloned into the pET28 plasmid, cut by XhoI, end blunted by Klenow polymerase, digested by SalI and dephosphorylated by CIAP. Bacterial Transformations Plasmid electroporation into E. coli DH5a strain was performed by a Micro Pulser (BioRad, Hercules, CA, USA) using a pre-set program (1,800 V, 4 ms). Plasmid DNA was isolated using alkaline lysis method [21] and performing column purification according to recommendations of ThermoFisher Scientific. Transformation of E. coli BL21 (DE3) strain was carried out using calcium chloride method [20]. Induction and Expression of Recombinant K2 Preprotoxin Escherichia coli strain BL21 (DE3) containing the recombinant p28-kil2 or p28-kilD plasmids was grown in liquid LB medium supplemented with kanamycin at 50 lg/mL at 37 °C until OD600 reached 0.6. Then, the expression of full-length and truncated K2 proteins (recK2 and recK2D, respectively) was induced by adding IPTG to the final concentrations of 0.1 or 0.4 mM and cultivation of cells for 2 or 18 h at either 18 or 30 °C on the rotary shaker. For the analysis of the recK2 expression level, 1 mL aliquots of culture were taken at different time points after
Mol Biotechnol
Fig. 1 Principal scheme of p28-kil2 (a) and p28-kilD (b) plasmids, highlighting deletion introduced into K2 protein (c). ori pMB1 replication origin, KnR gene for kanamycin resistance, f1 f1 bacteriophage replication origin, lacI operator region, T7 pr T7
bacteriophage promoter, T7 ter T7 bacteriophage terminator, Kil2 S. cerevisiae K2 preprotoxin gene, KilD truncated K2 preprotoxin gene, recK2 recombinant K2 protein, SalI restriction endonuclease site
induction with IPTG. Cells were harvested by centrifugation at 10,0009g for 10 min, pellets resuspended in 0.1 mL of 50 mM Tris–HCl (pH 8.0) buffer and mixed with the gel sample buffer. Samples were heated at 85 °C for 5 min and loaded on SDS-PAGE gel. For the examination of the soluble and inclusion body fractions, 50 mL of the BL21-[p28-kilD] culture was taken at 2 and 18 h after induction, and cells were separated by centrifugation at 4,0009g for 5 min. The cell pellets were suspended in 2 mL of 50 mM Tris–HCl (pH 8.0) buffer and sonicated in short pulses (10 s on/30 s off) for 10–30 min on ice using Sonics Vibracell VCX 750. The fraction of soluble cellular proteins was collected, and analyzed by SDS-PAGE. For the analysis of insoluble cellular proteins, the debris obtained upon cell disruption and centrifugation were suspended in 1 mL of 50 mM Tris–HCl (pH 8.0) buffer supplied with 3 M urea and 1 % Triton X-100 and incubated on ice for 30 min. Supernatant was collected and analyzed by SDS-PAGE according to Laemmli [22].
remove contaminants, the pellet fraction was further washed with 1 mL of 50 mM Tris–HCl (pH 8.0) containing 0.5 M NaCl and 0.2 % Triton X-100. Lysate was centrifugated at 16,0009g for 10 min, and the supernatant was carefully removed. IBs were dissolved by adding 1 mL 50 mM Tris– HCl (pH 8.0), containing 3 M urea and 1 % Triton-X100, carefully resuspending, and incubating on ice for 60 min. Then, the suspension was centrifugated at 16,0009g for 15 min, and supernatant was subjected to two-step dialysis (2 h each) against 1.5 M urea (first step) and 50 mM Tris– HCl (pH 8.0) buffer only (second step). The purity and quantity of target protein were evaluated by SDS-PAGE, gel visualized by Coomassie Brilliant Blue R250 staining.
Purification of the Recombinant K2 Protein BL21 cells containing the p28-kilD plasmid were grown overnight at 37 °C in LB medium containing kanamycin (50 lg/mL) and reinoculated into 100 mL of fresh LB medium. As the culture reached OD600 = 0.6, production of recombinant protein was induced by adding IPTG (0.1 mM final concentration) and grown for additional 2 h at 30 °C. The bacteria were harvested by centrifugation at 4,0009g for 5 min, and washed by 50 mM Tris–HCl (pH 8.0) buffer. 100 mg of wet cells was resuspended in 2 mL of the same buffer, and sonicated in short pulses for 30 min on ice using Sonics Vibracell VCX 750. The suspension of disrupted cells was centrifuged at 1,0009g for 10 min for separation of unbroken cells, and then at 16,0009g for 15 min at 4 °C—for isolation of insoluble IBs. In order to
Preparation and Concentration of the Yeast K2 Toxin The K2 toxin-producing S. cerevisiae strain M437 was grown in synthetic medium (2 % dextrose; 6 mM K2HPO4; 8 mM MgSO4; 8 mM (NH4)2SO4; adjusted to pH 4.0 with the 75 mM phosphate–citrate buffer, containing 5 % glycerol) for 4 days at 18 °C. Yeast cells were separated by centrifugation at 3,0009g for 10 min; supernatant was filtered through a 0.22 lm sterile polyvinylidene fluoride membrane and concentrated 1,000-fold by ultrafiltration through Amicon PM-10 membrane. Concentrated supernatant was subjected to SDS-PAGE and used for evaluation of the antibody specificity. Removal of dsRNA from the Yeast The K2 killer yeast strain M437 was spread onto YPD-agar plates (1 % yeast extract, 2 % peptone, 2 % dextrose, and 2 % agar) at a density of 5 9 103 cells/plate and incubated at 37 °C for 4 days. The surviving colonies were replica plated on MB agar plates (0.5 % yeast extract, 0.5 % peptone, and 2 % dextrose) overlaid with sensitive to K2 toxin S. cerevisiae strain a0 1 and grown for 2 days at 25 °C.
123
Mol Biotechnol
Colonies lacking killer phenotype were selected and used for negative control preparation for testing of antibody. Production of the Polyclonal Antiserum Antibodies were raised against recombinant K2 protein in rabbits. A sample of the recombinant K2 protein (200 lg) was resolved on 10 % SDS-polyacrylamide gel. The gel fragment with the band of interest was homogenized in 1 mL PBS buffer, mixed with complete Freund’s adjuvant (1:1, v/v) and injected into the rabbit. Three boost injections with equivalent amounts of protein in Freund’s incomplete adjuvant were performed. Seven days after the last immunization, the rabbit was bled. All procedures involving experimental rabbits were performed under controlled laboratory conditions in strict accordance with the Lithuanian and European legislation (Law on the Care, Keeping and Use of Animals of the Republic of Lithuania, No. 0209). The blood was allowed to coagulate at 37 °C temperature for 30 min, and then kept at 4 °C overnight. The antiserum separated from the blood clot was centrifuged at 3,0009g for 10 min. Immunoglobulin G-enriched fraction was prepared by the precipitation with ammonium sulfate and stored at -20 °C in 20–50 lL aliquots. Such antiserum was further used for the immunological studies. Indirect ELISA Indirect ELISA was used to estimate the titre of polyclonal antibodies against native and recombinant K2 proteins. 50 lL solution of antigen (4 lg/mL) in immobilization buffer (PBS, pH 7.2) was loaded into each well of the ELISA plate, and incubated overnight at 37 °C. The plate was blocked with 100 lL blocking buffer (PBS, 3 % BSA, 0.05 % Tween 20) for 1 h at room temperature (RT), and incubated with 100 lL of polyclonal antiserum at various dilutions ranging from 1:100 to 1:128,000 in PBS buffer for 1 h at RT. Following additional washing, the plate was incubated with horseradish peroxidase-conjugated antirabbit IgG secondary antibody (GE Healthcare, Freiburg, Germany) for 1 h at RT. The enzymatic reaction was detected with OPD substrate (Sigma-Aldrich, St. Louis, MO, USA). The reaction was stopped by adding 50 lL 2 M H2SO4 and the absorption at 490 nm was recorded. Western Blot Analysis For analysis of recombinant and native K2 toxin, purified recK2 protein or 1,000-fold concentrated cell-free culture medium of M437 strain were resolved on 12 % SDSpolyacrylamide gel in duplicate. One gel was stained with Coomassie Brilliant Blue R250. The other gel was transferred onto PVDF membrane in transblot buffer (25 mM
123
Fig. 2 SDS-PAGE analysis of recK2 and recK2D expression level in strains bearing p28-kil2 and p28-kilD constructs. M protein marker (PageRuler Unstained Protein Ladder, ThermoFisher Scientific); 0 h uninduced cells, 2 and 18 h induction time at 18 °C with 0.1 mM IPTG; recK2/recK2D recombinant K2 protein and truncated version, accordingly
Tris, 192 mM glycine, 20 % methanol, and 0.1 % SDS), blocked for 1 h at RT with 5 % of milk powder in TTBS, containing 0.05 % Tween 20, and washed three times with TTBS. Western blot analysis was carried out using primary polyclonal antibody developed against recK2 protein (dilution in TTBS was determined experimentally) and two versions of secondary antibodies: an alkaline phosphataseconjugated or horseradish peroxidase (HRP)-conjugated goat anti-rabbit-IgG antiserum (Merck Millipore, Darmstadt, Germany). Colorimetric signal detection of the immunoprecipitate was performed using, respectively, a NBT/BCIP or TMB solutions (ThermoFisher Scientific) according to manufacturer’s recommendations.
Results Construction of the Bacterial Strains, Producing the K2 Protein In this study, we performed the construction of a p28-kil2 bacterial plasmid, bearing a gene which encodes the fulllength S. cerevisiae K2 under the T7 promoter (Fig. 1a). Following several unsuccessful chemical transformation attempts into E. coli DH5a cells, the electroporation was applied, resulting in small number of transformants. The purification of plasmid DNA and restriction analysis were carried out. The selected recombinant plasmids with correct location of K2 preprotoxin gene in respect to T7 promoter was retransformed into E. coli BL21 (DE3) strain. A comparison of the protein expression level in induced and non-induced cells harboring the recombinant K2 preprotoxin (recK2), respectively, is presented in Fig. 2. RecK2 migrates in SDS-PAGE gel as a *45 kDa protein. It was found that full-length killer protein was
Mol Biotechnol
Both full-length and a truncated version of the killer protein produced at basal level under non-inducing conditions did not affect the growth of the host. The growth dynamics of both BL21-[p28-kil2] and BL21-[p28-kilD] strains were comparable (Fig. 3a). However, when IPTG was added, increase in cell density of BL21-[p28-kil2] strain was reduced more than twofold (cells grew from OD600 = 0.6 to OD600 = 0.9 only) (Fig. 3b) comparing to that of truncated protein-producing cells, which grew similarly to mock cells (bearing only the pET28 vector) from OD600 = 0.6 to OD600 = 1.3 (Fig. 3b). Even the modest synthesis level of full-length recombinant K2 protein (following the induction by IPTG) diminishes the growth of bacteria. Notably, the truncated K2 protein, synthesized even at higher levels, did not affect the growth of the E. coli host, as represented in Fig. 3b.
Synthesis of the recK2D Protein
Fig. 3 The effect of full size and truncated K2 protein on the growth of E. coli cells. a Non-inducing conditions, b induction by 0.1 mM IPTG
produced in E. coli at low to negligible level either after short (2 h) or overnight (18 h) induction (Fig. 2). The amount of the target protein did not increase after 2 days induction period (data not shown). Given the difficult introduction into E. coli cells of a plasmid with full-length preprotoxin gene, and the low expression level of the target protein, we modified the K2 preprotoxin gene. Previously, we found that the removal of C-terminus of K2 protein completely abolishes the killing activity of toxin in yeast S. cerevisiae [23]. Thus, we introduced a deletion into the coding sequence of K2 preprotoxin gene resulting in a new plasmid p28-kilD (Fig. 1b) featuring a 16 aa C-terminal truncation of the recombinant killer protein (Fig. 1c). The expression level of such protein in both induced and non-induced cells harboring the p28-kilD plasmid is presented in Fig. 2. The distinct band corresponding to truncated K2 preprotoxin (recK2D) is clearly visible after both short (2 h) and overnight (18 h) induction periods. RecK2D protein was expressed up to 33 % of total protein content after overnight induction at 18 °C. Effect of Recombinant K2 Protein on Growth of Bacteria The growth of bacteria producing yeast killer preprotoxin was evaluated under non-inducing and inducing conditions.
As described above, we found that the maximum level of synthesis of yeast killer preprotoxin is achieved in bacterial cells containing the p28-kilD plasmid. Therefore, an appropriate E. coli strain was cultivated under different conditions by altering the temperature, the amount of IPTG and the induction time. Level of the synthesized product was evaluated by SDS-PAGE. We found that under the cultivation temperature of 18 °C and short induction time (2 h), small amount of soluble recK2D was detected (Fig. 4a, I). Virtually, no protein of interest was observed in insoluble fraction (Fig. 4b, I). Increase of the induction time up to 18 h resulted in 10–12 times larger accumulation of the insoluble aggregates (Fig. 4b, II), still leaving *30 % of protein in the soluble form (Fig. 4a, II). It was observed that the major part of recK2D protein was found in insoluble form, when cultivation temperature was elevated up to 30 °C (Fig. 4b, III, IV). The SDS-PAGE analysis shows further redistribution of recK2D toward the fraction of insoluble cellular proteins when cultures grown at 30 °C were subjected to induction for 2 and 18 h, respectively. We found that the overnight cultivation of E. coli at 30 °C results in an accumulation of recK2D completely in the insoluble aggregates (Fig. 4b, IV), as soluble protein was not found in the cytoplasmic fraction (Fig. 4a, IV). Induction for 2 h allowed to obtain about 30 % less of target protein in comparison to overnight cultivation at 30 °C, albeit it had a higher solubility (Fig. 4 a, b, III, IV). The shift of cultivation temperature to 18 °C followed by 18 h induction did not increase the solubility of recK2D protein (Fig. 4a, b, II). Our data also show that the amount of IPTG (0.1 and 0.4 mM) used did not alter neither expression level nor solubility of recK2D protein (Fig. 4). Based on our experiments, optimal conditions for
123
Mol Biotechnol
Fig. 4 SDS-PAGE analysis of the recK2D synthesis. a Cytoplasmic fraction, b insoluble fraction. M protein marker (PageRuler Unstained Protein Ladder, ThermoFisher Scientific). Induction conditions I 2 h, 18 °C; II 18 h, 18 °C; III 2 h, 30 °C; IV 18 h, 30 °C. Filled circle 0.1 mM IPTG, open circle 0.4 mM IPTG
recombinant K2 preprotoxin production were determined as follows: induction by 0.1 mM IPTG for 2 h at 30 °C. Purification of Recombinant K2D Protein and Production of the Polyclonal Antiserum During this study, the protocol for production of truncated recombinant K2 protein was developed (Fig. 5a). Samples, representing each purification step, were collected and tested by SDS-PAGE (Fig. 5b). For this purpose, bacteria, cultivated for 2 h at 30 °C in presence of 0.1 mM IPTG, were collected and disrupted by ultrasonication. The soluble cytoplasmic protein fraction (CF) and insoluble fraction of IBs were separated by centrifugation. For the purification of recK2D preprotoxin, we have chosen the IBs containing [90 % of the target protein, as observed after ultrasonication (Fig. 5b). This fraction was subjected to several washing procedures by using (in separate assays) either 0.5 M NaCl (Fig. 5b, 1), 0.2 % Triton X-100 (Fig. 5b, 3), or both 0.5 M NaCl and 0.2 % Triton X-100 (Fig. 5b, 2). After the washing step, the aggregated IBs were dissolved in 3 M urea and 1 % Triton X-100 and analyzed by SDS-PAGE. The least amount of impurities was detected in the sample where IBs were washed by the solution of both NaCl and Triton X-100 (Fig. 5b, IBw 2). Dissolved IBs were subsequently subjected to dialysis to remove urea while keeping the purified protein in a soluble form. Several assays were carried out: a single 48 h
123
dialysis or a two-step (2 h each) dialysis. The former dialysis approach promoted some protein degradation, while the two-step dialysis was more favorable, and as a consequence a purified soluble protein was obtained. RecK2D protein was approx. 90 % pure as judged by densitometric analysis of the bands presented in Fig. 5b, far right lane. The purification procedure resulted in the final yield of 1 mg purified recK2D protein per 100 mg of wet biomass of bacteria. The resulted recombinant K2 protein was applied for rabbit immunization and polyclonal antibody production (described in details in ‘‘Materials and Methods’’ section). In this work, about 80 mL of antiserum raised against recK2D protein was obtained. The titre and sensitivity of antibody were determined using indirect ELISA. It was shown that the antiserum is specific to recK2D protein expressed in bacteria. The antibody titre was found to be 1:16,000 (Fig. 6a). The antiserum was tested for the reactivity toward the genuine mature K2 toxin produced in yeast cells, and the titre of 1:300 was established (Fig. 6b). The results obtained from ELISA analysis show that the polyclonal antibodies in serum were able to react with both recombinant antigen and native K2 killer protein from yeast. The specificity of generated antibodies was addressed by Western blot analysis (Fig. 7). The purified recK2D protein and 1,000-fold concentrated yeast extracellular fraction (containing K2 toxin) were resolved by SDS-PAGE and then probed by antiserum raised against recK2D expressed in E. coli. Dilutions of antiserum were optimized and most appropriate established: 1:10,000 for recK2D protein or 1:300 for yeast K2 toxin detection. Western blot analysis confirms the results obtained by ELISA by displaying the positive signal for the purified recK2D (Fig. 7a) and the native K2 toxin (Fig. 7b). Western blot analysis of yeast K2 protein was performed using two different secondary antibody conjugates (based on AP or HRP), and in both cases K2 toxin migrates on SDS-PAGE corresponding to approx. 20 kDa (Fig. 7b, II, III). The antibody did not react with 1,000-fold concentrated medium of dsRNA lacking cell-free culture (Fig. 7b, I), confirming specificity toward K2 toxin.
Discussion In this study, we present the cloning, expression, and purification scheme of yeast K2 killer protein in E. coli. Base level synthesis of the full size preprotoxin under noninducing conditions is tolerated by bacteria resulting in negligible (if any) impact on the growth. However, this is in contrary to the conditions of induced expression, leading to more than twofold decrease in proliferation of bacteria
Mol Biotechnol
Fig. 5 Purification scheme for recombinant K2D preprotoxin (a) and SDS-PAGE analysis of the recK2D protein-expressing BL-21 cells (b). M Protein marker (PageRuler Unstained Protein Ladder, ThermoFisher Scientific); CrEx sonicated bacterial cells, CF cytoplasmic protein fraction, IB fraction of inclusion bodies, wash samples of IB
Fig. 6 ELISA of the antiserum against recK2D protein (a) and native K2 toxin (b). a The titre was measured, while the dilutions of sera were at 1:2,000, 1:4,000, 1:8,000, 1:16,000, 1:32,000, 1:64,000, and 1:128,000, when recK2D was used as immunogen; b dilutions of sera were 1:100, 1:200, 1:400, 1:800, 1:1,600, and 1:3,200, when native K2 toxin was used as immunogen
even if the modest expression level is achieved. C-terminal deletion of K2 protein not only completely abrogates the killing activity in yeast [23] but also ceases the negative
washing including 0.5 M NaCl (1), 0.5 M NaCl and 0.2 % Triton X-100 (2), 0.2 % Triton X-100 (3); IBw dissolved inclusion bodies after washing, respective to wash solutions; Dialysis samples after either 24 h, 48 h or (2 ? 2 h)—two-step dialysis for 2 h each
impact on bacterial cell proliferation even at relatively high expression levels. Since the adverse effect on bacterial growth is typical for full-length preprotoxin only, one can envision a structural role of the C-terminal fragment (Fig. 1c), which aggravates the formation of insoluble aggregates. RecK2D, devoid of this fragment, is inactivated by directing it into non-toxic IBs, relieving bacterial cells of stress induced by this harmful foreign protein. The expression of truncated killer preprotoxin was optimized by direction into easily resoluble IBs, allowing a simple and efficient separation from the host proteins. Usually, the formation of IBs is deleterious for biotechnology. To recover the biologically active protein, IBs have to be isolated and solubilized, and the target protein has to be refolded to its native state. The renaturation of target protein is rather complicated process and often leads to aggregation and misfolding [24, 25]. However, the separation of the soluble target protein from cytoplasmic host proteins is also a relatively difficult task as this compartment comprises the vast majority of the total cellular proteins. Despite the fact that IBs are highly pure protein deposits, and over-expressed recombinant protein may represent up to 95 % of the total IB content [26], the final yields of bioactive protein are often low [27]. However, in our case the formation of IBs was essential for high-level expression and efficient purification. Keeping the expression conditions of recK2D at a level sequestering the protein between insoluble and cytoplasmic fractions resulted in easily soluble IBs, as demonstrated by their solubilization in the buffer containing 3 M urea and subsequent refolding by two-step dialysis. The resulted protein is stable for several months. Two-step dialysis was found to be reasonable for efficient removal of urea. Graduate
123
Mol Biotechnol
The polyclonal antibodies raised against recK2D in rabbit were found to possess the required specificity not only toward truncated recK2 but also the native K2 killer toxin from yeast. This is an important property of such antibodies given the structurally different genuine K2 protein formed by processing of preprotoxin in its natural host. In summary, the expression of S. cerevisiae K2 preprotoxin in E. coli leads to diminished growth of bacteria and limited target protein production. C-terminal truncated protein was accumulated in the cytoplasm as non-toxic IBs. Expression of recK2D in insoluble yet easily refoldable form allowed a simple and efficient chromatography-free purification of the obtained aggregates. The resulted protein was used for rabbit immunization and production of the polyclonal serum. Indirect ELISA and Western blot analysis confirmed the specificity of raised polyclonal antibodies toward both K2 preprotoxin expressed in bacteria and the native K2 toxin. Overall, the described workflow might serve as an example of the use of bacterial expression systems to prepare antigen for raise of antibodies against cytotoxic eukaryotic proteins. Fig. 7 Detection of recK2D protein (a) and K2 toxin of yeast origin (b) by SDS-PAGE and Western blot analysis. M protein marker, S purified recK2D protein, I no dsRNA cell-free yeast M437 culture probed with anti-rabbit IgG-AP conjugate, II, III concentrated extracellular fraction of active K2 toxin-producing yeast strain M437 probed with anti-rabbit IgG-AP and HRP conjugates, respectively. An asterisk marks position of either recK2D (a) or native K2 (b) protein
decrease of urea ensured solubilization of the target protein and preservation from proteolysis, observed during prolonged single-step dialysis. Overall, our approach presents a simple and efficient chromatography-free purification of a soluble tag-less protein. K2 killer protein from yeast S. cerevisiae possesses high biological activity even at the negligible concentrations [9]. Immunization of one rabbit using the native protein would require cultivating of 2,400 L of K2 toxin-producing S. cerevisiae culture media, necessary to obtain 800 lg of protein. Besides economic cost and inadequate culture conditions between fermenter- and flask-based biomass cultivation, concentration of such quantities of media to volume of a few millilitres, a standard approach of preparation of native K2 toxin, possesses a significant resource burden for antigen supply. The simplicity and low cost of bacterial cultivation are a compelling advantage over any other expression systems, and therefore E. coli is the most preferable choice [11]. A sufficient amount of recK2D protein used for immunization was easily prepared from only 100 mL of E. coli culture in a simple and cheap approach.
123
Acknowledgments This work was supported by Grants of Research Council of Lithuania (LMT) to ES (Nos. MIP-061/2011, MIP-42/ 2013). We would like to thank Dr. J. Urbonavicˇius for helpful comments on the manuscript.
References 1. Goretti, M., Turchetti, B., Buratta, M., Branda, E., Corazzi, L., Vaughan-Martini, A., et al. (2009). In vitro antimycotic activity of a Williopsis saturnus killer protein against food spoilage yeasts. International Journal of Food Microbiology, 131, 178–182. 2. Magliani, W., Conti, S., Giovati, L., Zanello, P. P., Sperinde`, M., Ciociola, T., et al. (2012). Antibody peptide based antifungal immunotherapy. Frontiers in Microbiology, 3, 1–9. 3. Maqueda, M., Zamora, E., Alvarez, M., & Ramirez, M. (2012). Characterization, ecological distribution and population dynamics of Saccharomyces sensu stricto killer yeasts in the spontaneous grape must fermentations of southwestern Spain. Applied and Environmental Microbiology, 78, 735–743. 4. Magliani, W., Conti, S., Cunha, R. L., Travassos, L. R., & Polonelli, L. (2009). Antibodies as crypts of antiinfective and antitumor peptides. Current Medicinal Chemistry, 16, 2305–2323. 5. Breinig, F., Sendzik, T., Eisfeld, K., & Schmitt, M. (2006). Dissecting toxin immunity in virus-infected killer yeast uncover intrinsic strategy of self-protection. Proceedings of the National Academy of Sciences of the United States of America, 103, 3810–3815. 6. Meinhardt, F., & Klassen, R. (2009). Yeast killer toxins: Fundamentals and applications, in physiology and genetics. In J. Anke & D. Weber (Eds.), The mycota (pp. 109–130). Berlin: Springer. 7. Dignard, D., Whiteway, M., Germain, D., Tessier, D., & Thomas, D. Y. (1991). Expression in yeast of a cDNA copy of the K2
Mol Biotechnol
8.
9.
10.
11.
12.
13.
14.
15.
16.
killer toxin gene. Molecular and General Genetics, 227, 127–136. Mesˇkauskas, A., & Cˇitavicˇius, D. (1992). The K2-type killer toxin- and immunity-encoding region from Saccharomyces cerevisiae: Structure and expression in yeast. Gene, 111, 135–139. Lebionka, A., Servien_e, E., & Melvydas, M. (2002). Isolation and purification of yeast Saccharomyces cerevisiae K2 killer toxin. Biologija, 4, 7–9. Peternel, S., & Kornel, R. (2010). Isolation of biologically active nanomaterial (inclusion bodies) from bacterial cells. Microbial Cell Factories, 9, 1–16. Pandey, N., Sachan, A., Chen, Q., Ruebling-Jass, K., Bhalla, R., Panguluri, K., et al. (2013). Screening and identification of genetic loci involved in producing more/denser inclusion bodies in Escherichia coli. Microbial Cell Factories, 12, 43. Villaverde, A., & Carrio, M. (2003). Protein aggregation in recombinant bacteria: Biological role of inclusion bodies. Biotechnology Letters, 25, 1385–1395. Zerbs, S., Frank, A. M., & Collart, F. R. (2009). Bacterial systems for production of heterologous proteins. Methods in Enzymology, 463, 149–168. Narayanan, N., Khan, M., & Chou, P. (2011). Enhancing functional expression of heterologous Burkholderia lipase in Escherichia coli. Molecular Biotechnology, 47, 130–143. Gauci, C., Jenkins, D., & Lightowlers, M. W. (2011). Strategies for optimal expression of vaccine antigens from taeniid cestode parasites in Escherichia coli. Molecular Biotechnology, 48, 277–289. Kamionka, M. (2011). Engineering of therapeutic proteins production in Escherichia coli. Current Pharmaceutical Biotechnology, 12, 268–274.
17. Woodcock, D. M., Crowther, P. J., Doherty, J., Jefferson, S., Decruz, E., Noyer-Weidner, M., et al. (1989). Quantitative evaluation of Escherichia coli host strains for tolerance to cytosine methylation in plasmid and phage recombinants. Nucleic Acids Research, 17, 3469–3478. 18. Studier, F. W., & Moffatt, B. A. (1986). Use of bacteriophage T7 RNA polymerase to direct selective high-level expression of cloned genes. Journal of Molecular Biology, 189, 113–130. 19. Gulbiniene, G., Kondratiene, L., Jokantaite, T., Serviene, E., Melvydas, V., & Petkuniene, G. (2004). Occurence of killer yeast strains in fruit and berry wine yeast populations. Food Technology and Biotechnology, 42, 159–163. 20. Green, M., & Sambrook, J. (2012). Molecular cloning. New York: Cold Spring Harbor Laboratory Press. 21. Birnboim, H., & Doly, J. (1979). A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Research, 7, 1513–1523. 22. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680–685. 23. Gulbinien_e, G. (2002). Characterization of the functional organization of yeast K2 killer preprotoxin gene. Biologija, 3, 37–40. 24. Panda, A. K. (2003). Bioprocessing of therapeutic proteins from the inclusion bodies of Escherichia coli. Advances in Biochemical Engineering/Biotechnology, 85, 43–93. 25. Burgess, R. R. (2009). Refolding solubilized inclusion body proteins. Methods in Enzymology, 463, 259–282. 26. Ventura, S., & Villaverde, A. (2006). Protein quality in bacterial inclusion bodies. Trends in Biotechnology, 24, 179–185. 27. Lilie, H., Schwarz, E., & Rudolph, R. (1998). Advances in refolding of proteins produced in E. coli. Current Opinion in Biotechnology, 9, 497–501.
123
RESEARCH
High-Yield Expression in Escherichia coli, Purification and Application of Budding Yeast K2 Killer Protein Monika Podoliankait_e • Juliana Luksˇa • Gintautas Vysˇniauskis Jolanta Sereikait_e • Vytautas Melvydas • Saulius Serva • Elena Servien_e
•
Ó Springer Science+Business Media New York 2014
Abstract Saccharomyces cerevisiae K2 toxin is a highly active extracellular protein, important as a biocontrol agent for biotechnological applications in the wine industry. This protein is produced at negligible levels in yeast, making difficult to isolate it in amounts sufficient for investigation and generation of analysis tools. In this work, we demonstrate the use of a bacterial system for expression of the recombinant K2 protein, suitable for generation of antibodies specific for toxin of the yeast origin. Synthesis of the full-length S. cerevisiae K2 preprotoxin in Escherichia coli was found to be toxic to the host cell, resulting in diminished growth. Such effect was abolished by the introduction of the C-terminal truncation into K2 protein, directing it into non-toxic inclusion body fraction. The obtained protein is of limited solubility thus, facilitating the
Monika Podoliankait_e and Juliana Luksˇa have contributed equally to this study. M. Podoliankait_e J. Luksˇa G. Vysˇniauskis V. Melvydas E. Servien_e (&) Laboratory of Genetics, Institute of Botany, Nature Research Centre, Akademijos Str. 2, 08412 Vilnius, Lithuania e-mail: [email protected] G. Vysˇniauskis Department of Biopharmacy, State Research Institute for Innovative Medicine, Zygimantu 9, 01102 Vilnius, Lithuania J. Sereikait_e S. Serva E. Servien_e Department of Chemistry and Bioengineering, Faculty of Fundamental Sciences, Vilnius Gediminas Technical University, Sauletekio 11, 10223 Vilnius, Lithuania S. Serva Department of Biochemistry and Molecular Biology, Faculty of Natural Sciences, Vilnius University, Ciurlionio 21, 03101 Vilnius, Lithuania
purification by simple and efficient chromatography-free procedure. The protein aggregates were successfully refolded into a soluble form yielding sufficient amounts of a tag-less truncated K2 protein suitable for polyclonal antibody production. Antibodies were raised in rabbit and found to be specific for detection of both antigen and native S. cerevisiae K2 toxin. Keywords Saccharomyces cerevisiae K2 toxin Escherichia coli K2 recombinant protein Antibody
Introduction The toxins of the yeast origin confer a growth advantage to their hosts by increasing survival in clinical, environmental, and industrial ecosystems. Nowadays, particular interest in yeast killer strains or isolated toxins has surged. They find application in industry (food protection from spoiling, wine production), agriculture (phytopathogen control), or medicine (creation of a new generation vaccines and antibiotics, development of antifungal immunotherapy) [1–3]. Killer proteins and toxin-recognizing antibodies have been tested as a potential antiviral and anticancer agents [4]. Therefore, much attention is focused on creation of killer-specific monoclonal and polyclonal antibodies used for in-depth studies of killer systems and practical applications [5]. K2 killer protein of Saccharomyces cerevisiae is the most frequent toxin among yeast dominating in the vineyard-winery ecosystem, where up to half of yeast strains are hosts for killer systems, and some 80 % of them are of K2-type [3, 6]. Double-stranded RNA-derived K2 toxin is synthesized as a precursor of 362 amino acids, consisting of the N-terminal secretion signal, a, and b subunits
123
Mol Biotechnol
(172 and 140 amino acids, respectively). During the passage through the yeast secretory pathway, K2 protein is enzymatically processed into a mature cytotoxic a/b heterodimer [7, 8]. Functionality of protein is highly sensitive for sequence alterations resulting in activity being lost easily [7]. The protein is secreted; however, high biologic activity determines mass quantity of the killer protein remain negligible. The yield of yeast K2 toxin, isolated from 3 L of culture medium, is less than 1 lg [9]. This possesses a significant burden for the biochemical research requiring considerable amounts of protein, e.g., producing of the specific antibodies and developing of the immunodetection systems. Isolation of many bioactive peptides from natural sources is often expensive and time consuming. Therefore, the bacteria, in particular Escherichia coli, are employed as the hosts for large-scale production of proteins [10, 11]. However, the host stress response systems can be triggered upon the massive expression of the recombinant genes impairing plasmid replication and stability, cell growth and viability [12]. The produced foreign protein is a target for proteolytic degradation or insoluble aggregate formation. Aggregation can be minimized by precise control of the induction parameters, such as time and/or temperature, aimed to reduce the expression rate, adjusting the codon usage, protein engineering, or co-expression of the chaperones [12–15]. On the other hand, formation of the protein aggregates in inclusion bodies (IBs) might be beneficial, in order to simplify protein purification procedure, reducing number of purification steps and significantly improving cost-efficiency ratio [11, 13, 16]. During this work, the prokaryotic expression system was engaged to produce significant amounts of yeast K2 killer protein. Toxicity of the full-length foreign protein was neutralized by mutagenesis and directing it into the state of limited solubility. Expression conditions were optimized to allow refolding of tag-less target protein. Time and costefficient strategy for recombinant protein purification was elaborated, and genuine K2 toxin-specific antibodies were generated.
Materials and Methods Bacterial and Yeast Strains Escherichia coli strain DH5a (F-(/80dD(lacZ)M15) ? recA1 endA1 gyrA96 thi1 hsdR17 (rk mk ) supE44 relA1 deoR D (lacZYA-argF) U169) [17] was used as the host for recombinant plasmid construction and amplification. Protein expression was conducted in E. coli strain BL21 (DE3) (F- ompT hsdSB (rB mB ) gal dcm) [18].
123
Saccharomyces cerevisiae strains M437 (wt, HM/HM [kil–K2]) and a0 1 (MATa leu2-2 [Kil-0]) were used for K2 toxin isolation and for testing of the dsRNA-free strain [19]. Construction of Vectors for Expression in E. coli General procedures for the construction and analysis of recombinant DNAs were performed according to Green and Sambrook [20]. Restriction enzymes (SalI, Eco147I, XhoI), T4 DNA ligase, calf intestine alkaline phosphatase (CIAP), DNA size marker (GeneRulerTM DNA Ladder mix) were obtained from ThermoFisher Scientific (Vilnius, Lithuania) and used according to the manufacturer’s recommendations. The recombinant p28-kil2 plasmid (Fig. 1a) was obtained by the insertion of a K2 preprotoxin gene into the pET28 plasmid, downstream the phage T7 promoter. For this purpose, sequence encoding K2 toxin precursor was excised from the pYEX12 plasmid [8] using SalI and ligated into the vector pET28, digested by the same restriction enzyme and dephosphorylated by CIAP. For the construction of the p28-kilD plasmid (Fig. 1b), 30 -truncated K2 preprotoxin gene was excised from the pYEX12 using SalI and Eco147I and subsequently cloned into the pET28 plasmid, cut by XhoI, end blunted by Klenow polymerase, digested by SalI and dephosphorylated by CIAP. Bacterial Transformations Plasmid electroporation into E. coli DH5a strain was performed by a Micro Pulser (BioRad, Hercules, CA, USA) using a pre-set program (1,800 V, 4 ms). Plasmid DNA was isolated using alkaline lysis method [21] and performing column purification according to recommendations of ThermoFisher Scientific. Transformation of E. coli BL21 (DE3) strain was carried out using calcium chloride method [20]. Induction and Expression of Recombinant K2 Preprotoxin Escherichia coli strain BL21 (DE3) containing the recombinant p28-kil2 or p28-kilD plasmids was grown in liquid LB medium supplemented with kanamycin at 50 lg/mL at 37 °C until OD600 reached 0.6. Then, the expression of full-length and truncated K2 proteins (recK2 and recK2D, respectively) was induced by adding IPTG to the final concentrations of 0.1 or 0.4 mM and cultivation of cells for 2 or 18 h at either 18 or 30 °C on the rotary shaker. For the analysis of the recK2 expression level, 1 mL aliquots of culture were taken at different time points after
Mol Biotechnol
Fig. 1 Principal scheme of p28-kil2 (a) and p28-kilD (b) plasmids, highlighting deletion introduced into K2 protein (c). ori pMB1 replication origin, KnR gene for kanamycin resistance, f1 f1 bacteriophage replication origin, lacI operator region, T7 pr T7
bacteriophage promoter, T7 ter T7 bacteriophage terminator, Kil2 S. cerevisiae K2 preprotoxin gene, KilD truncated K2 preprotoxin gene, recK2 recombinant K2 protein, SalI restriction endonuclease site
induction with IPTG. Cells were harvested by centrifugation at 10,0009g for 10 min, pellets resuspended in 0.1 mL of 50 mM Tris–HCl (pH 8.0) buffer and mixed with the gel sample buffer. Samples were heated at 85 °C for 5 min and loaded on SDS-PAGE gel. For the examination of the soluble and inclusion body fractions, 50 mL of the BL21-[p28-kilD] culture was taken at 2 and 18 h after induction, and cells were separated by centrifugation at 4,0009g for 5 min. The cell pellets were suspended in 2 mL of 50 mM Tris–HCl (pH 8.0) buffer and sonicated in short pulses (10 s on/30 s off) for 10–30 min on ice using Sonics Vibracell VCX 750. The fraction of soluble cellular proteins was collected, and analyzed by SDS-PAGE. For the analysis of insoluble cellular proteins, the debris obtained upon cell disruption and centrifugation were suspended in 1 mL of 50 mM Tris–HCl (pH 8.0) buffer supplied with 3 M urea and 1 % Triton X-100 and incubated on ice for 30 min. Supernatant was collected and analyzed by SDS-PAGE according to Laemmli [22].
remove contaminants, the pellet fraction was further washed with 1 mL of 50 mM Tris–HCl (pH 8.0) containing 0.5 M NaCl and 0.2 % Triton X-100. Lysate was centrifugated at 16,0009g for 10 min, and the supernatant was carefully removed. IBs were dissolved by adding 1 mL 50 mM Tris– HCl (pH 8.0), containing 3 M urea and 1 % Triton-X100, carefully resuspending, and incubating on ice for 60 min. Then, the suspension was centrifugated at 16,0009g for 15 min, and supernatant was subjected to two-step dialysis (2 h each) against 1.5 M urea (first step) and 50 mM Tris– HCl (pH 8.0) buffer only (second step). The purity and quantity of target protein were evaluated by SDS-PAGE, gel visualized by Coomassie Brilliant Blue R250 staining.
Purification of the Recombinant K2 Protein BL21 cells containing the p28-kilD plasmid were grown overnight at 37 °C in LB medium containing kanamycin (50 lg/mL) and reinoculated into 100 mL of fresh LB medium. As the culture reached OD600 = 0.6, production of recombinant protein was induced by adding IPTG (0.1 mM final concentration) and grown for additional 2 h at 30 °C. The bacteria were harvested by centrifugation at 4,0009g for 5 min, and washed by 50 mM Tris–HCl (pH 8.0) buffer. 100 mg of wet cells was resuspended in 2 mL of the same buffer, and sonicated in short pulses for 30 min on ice using Sonics Vibracell VCX 750. The suspension of disrupted cells was centrifuged at 1,0009g for 10 min for separation of unbroken cells, and then at 16,0009g for 15 min at 4 °C—for isolation of insoluble IBs. In order to
Preparation and Concentration of the Yeast K2 Toxin The K2 toxin-producing S. cerevisiae strain M437 was grown in synthetic medium (2 % dextrose; 6 mM K2HPO4; 8 mM MgSO4; 8 mM (NH4)2SO4; adjusted to pH 4.0 with the 75 mM phosphate–citrate buffer, containing 5 % glycerol) for 4 days at 18 °C. Yeast cells were separated by centrifugation at 3,0009g for 10 min; supernatant was filtered through a 0.22 lm sterile polyvinylidene fluoride membrane and concentrated 1,000-fold by ultrafiltration through Amicon PM-10 membrane. Concentrated supernatant was subjected to SDS-PAGE and used for evaluation of the antibody specificity. Removal of dsRNA from the Yeast The K2 killer yeast strain M437 was spread onto YPD-agar plates (1 % yeast extract, 2 % peptone, 2 % dextrose, and 2 % agar) at a density of 5 9 103 cells/plate and incubated at 37 °C for 4 days. The surviving colonies were replica plated on MB agar plates (0.5 % yeast extract, 0.5 % peptone, and 2 % dextrose) overlaid with sensitive to K2 toxin S. cerevisiae strain a0 1 and grown for 2 days at 25 °C.
123
Mol Biotechnol
Colonies lacking killer phenotype were selected and used for negative control preparation for testing of antibody. Production of the Polyclonal Antiserum Antibodies were raised against recombinant K2 protein in rabbits. A sample of the recombinant K2 protein (200 lg) was resolved on 10 % SDS-polyacrylamide gel. The gel fragment with the band of interest was homogenized in 1 mL PBS buffer, mixed with complete Freund’s adjuvant (1:1, v/v) and injected into the rabbit. Three boost injections with equivalent amounts of protein in Freund’s incomplete adjuvant were performed. Seven days after the last immunization, the rabbit was bled. All procedures involving experimental rabbits were performed under controlled laboratory conditions in strict accordance with the Lithuanian and European legislation (Law on the Care, Keeping and Use of Animals of the Republic of Lithuania, No. 0209). The blood was allowed to coagulate at 37 °C temperature for 30 min, and then kept at 4 °C overnight. The antiserum separated from the blood clot was centrifuged at 3,0009g for 10 min. Immunoglobulin G-enriched fraction was prepared by the precipitation with ammonium sulfate and stored at -20 °C in 20–50 lL aliquots. Such antiserum was further used for the immunological studies. Indirect ELISA Indirect ELISA was used to estimate the titre of polyclonal antibodies against native and recombinant K2 proteins. 50 lL solution of antigen (4 lg/mL) in immobilization buffer (PBS, pH 7.2) was loaded into each well of the ELISA plate, and incubated overnight at 37 °C. The plate was blocked with 100 lL blocking buffer (PBS, 3 % BSA, 0.05 % Tween 20) for 1 h at room temperature (RT), and incubated with 100 lL of polyclonal antiserum at various dilutions ranging from 1:100 to 1:128,000 in PBS buffer for 1 h at RT. Following additional washing, the plate was incubated with horseradish peroxidase-conjugated antirabbit IgG secondary antibody (GE Healthcare, Freiburg, Germany) for 1 h at RT. The enzymatic reaction was detected with OPD substrate (Sigma-Aldrich, St. Louis, MO, USA). The reaction was stopped by adding 50 lL 2 M H2SO4 and the absorption at 490 nm was recorded. Western Blot Analysis For analysis of recombinant and native K2 toxin, purified recK2 protein or 1,000-fold concentrated cell-free culture medium of M437 strain were resolved on 12 % SDSpolyacrylamide gel in duplicate. One gel was stained with Coomassie Brilliant Blue R250. The other gel was transferred onto PVDF membrane in transblot buffer (25 mM
123
Fig. 2 SDS-PAGE analysis of recK2 and recK2D expression level in strains bearing p28-kil2 and p28-kilD constructs. M protein marker (PageRuler Unstained Protein Ladder, ThermoFisher Scientific); 0 h uninduced cells, 2 and 18 h induction time at 18 °C with 0.1 mM IPTG; recK2/recK2D recombinant K2 protein and truncated version, accordingly
Tris, 192 mM glycine, 20 % methanol, and 0.1 % SDS), blocked for 1 h at RT with 5 % of milk powder in TTBS, containing 0.05 % Tween 20, and washed three times with TTBS. Western blot analysis was carried out using primary polyclonal antibody developed against recK2 protein (dilution in TTBS was determined experimentally) and two versions of secondary antibodies: an alkaline phosphataseconjugated or horseradish peroxidase (HRP)-conjugated goat anti-rabbit-IgG antiserum (Merck Millipore, Darmstadt, Germany). Colorimetric signal detection of the immunoprecipitate was performed using, respectively, a NBT/BCIP or TMB solutions (ThermoFisher Scientific) according to manufacturer’s recommendations.
Results Construction of the Bacterial Strains, Producing the K2 Protein In this study, we performed the construction of a p28-kil2 bacterial plasmid, bearing a gene which encodes the fulllength S. cerevisiae K2 under the T7 promoter (Fig. 1a). Following several unsuccessful chemical transformation attempts into E. coli DH5a cells, the electroporation was applied, resulting in small number of transformants. The purification of plasmid DNA and restriction analysis were carried out. The selected recombinant plasmids with correct location of K2 preprotoxin gene in respect to T7 promoter was retransformed into E. coli BL21 (DE3) strain. A comparison of the protein expression level in induced and non-induced cells harboring the recombinant K2 preprotoxin (recK2), respectively, is presented in Fig. 2. RecK2 migrates in SDS-PAGE gel as a *45 kDa protein. It was found that full-length killer protein was
Mol Biotechnol
Both full-length and a truncated version of the killer protein produced at basal level under non-inducing conditions did not affect the growth of the host. The growth dynamics of both BL21-[p28-kil2] and BL21-[p28-kilD] strains were comparable (Fig. 3a). However, when IPTG was added, increase in cell density of BL21-[p28-kil2] strain was reduced more than twofold (cells grew from OD600 = 0.6 to OD600 = 0.9 only) (Fig. 3b) comparing to that of truncated protein-producing cells, which grew similarly to mock cells (bearing only the pET28 vector) from OD600 = 0.6 to OD600 = 1.3 (Fig. 3b). Even the modest synthesis level of full-length recombinant K2 protein (following the induction by IPTG) diminishes the growth of bacteria. Notably, the truncated K2 protein, synthesized even at higher levels, did not affect the growth of the E. coli host, as represented in Fig. 3b.
Synthesis of the recK2D Protein
Fig. 3 The effect of full size and truncated K2 protein on the growth of E. coli cells. a Non-inducing conditions, b induction by 0.1 mM IPTG
produced in E. coli at low to negligible level either after short (2 h) or overnight (18 h) induction (Fig. 2). The amount of the target protein did not increase after 2 days induction period (data not shown). Given the difficult introduction into E. coli cells of a plasmid with full-length preprotoxin gene, and the low expression level of the target protein, we modified the K2 preprotoxin gene. Previously, we found that the removal of C-terminus of K2 protein completely abolishes the killing activity of toxin in yeast S. cerevisiae [23]. Thus, we introduced a deletion into the coding sequence of K2 preprotoxin gene resulting in a new plasmid p28-kilD (Fig. 1b) featuring a 16 aa C-terminal truncation of the recombinant killer protein (Fig. 1c). The expression level of such protein in both induced and non-induced cells harboring the p28-kilD plasmid is presented in Fig. 2. The distinct band corresponding to truncated K2 preprotoxin (recK2D) is clearly visible after both short (2 h) and overnight (18 h) induction periods. RecK2D protein was expressed up to 33 % of total protein content after overnight induction at 18 °C. Effect of Recombinant K2 Protein on Growth of Bacteria The growth of bacteria producing yeast killer preprotoxin was evaluated under non-inducing and inducing conditions.
As described above, we found that the maximum level of synthesis of yeast killer preprotoxin is achieved in bacterial cells containing the p28-kilD plasmid. Therefore, an appropriate E. coli strain was cultivated under different conditions by altering the temperature, the amount of IPTG and the induction time. Level of the synthesized product was evaluated by SDS-PAGE. We found that under the cultivation temperature of 18 °C and short induction time (2 h), small amount of soluble recK2D was detected (Fig. 4a, I). Virtually, no protein of interest was observed in insoluble fraction (Fig. 4b, I). Increase of the induction time up to 18 h resulted in 10–12 times larger accumulation of the insoluble aggregates (Fig. 4b, II), still leaving *30 % of protein in the soluble form (Fig. 4a, II). It was observed that the major part of recK2D protein was found in insoluble form, when cultivation temperature was elevated up to 30 °C (Fig. 4b, III, IV). The SDS-PAGE analysis shows further redistribution of recK2D toward the fraction of insoluble cellular proteins when cultures grown at 30 °C were subjected to induction for 2 and 18 h, respectively. We found that the overnight cultivation of E. coli at 30 °C results in an accumulation of recK2D completely in the insoluble aggregates (Fig. 4b, IV), as soluble protein was not found in the cytoplasmic fraction (Fig. 4a, IV). Induction for 2 h allowed to obtain about 30 % less of target protein in comparison to overnight cultivation at 30 °C, albeit it had a higher solubility (Fig. 4 a, b, III, IV). The shift of cultivation temperature to 18 °C followed by 18 h induction did not increase the solubility of recK2D protein (Fig. 4a, b, II). Our data also show that the amount of IPTG (0.1 and 0.4 mM) used did not alter neither expression level nor solubility of recK2D protein (Fig. 4). Based on our experiments, optimal conditions for
123
Mol Biotechnol
Fig. 4 SDS-PAGE analysis of the recK2D synthesis. a Cytoplasmic fraction, b insoluble fraction. M protein marker (PageRuler Unstained Protein Ladder, ThermoFisher Scientific). Induction conditions I 2 h, 18 °C; II 18 h, 18 °C; III 2 h, 30 °C; IV 18 h, 30 °C. Filled circle 0.1 mM IPTG, open circle 0.4 mM IPTG
recombinant K2 preprotoxin production were determined as follows: induction by 0.1 mM IPTG for 2 h at 30 °C. Purification of Recombinant K2D Protein and Production of the Polyclonal Antiserum During this study, the protocol for production of truncated recombinant K2 protein was developed (Fig. 5a). Samples, representing each purification step, were collected and tested by SDS-PAGE (Fig. 5b). For this purpose, bacteria, cultivated for 2 h at 30 °C in presence of 0.1 mM IPTG, were collected and disrupted by ultrasonication. The soluble cytoplasmic protein fraction (CF) and insoluble fraction of IBs were separated by centrifugation. For the purification of recK2D preprotoxin, we have chosen the IBs containing [90 % of the target protein, as observed after ultrasonication (Fig. 5b). This fraction was subjected to several washing procedures by using (in separate assays) either 0.5 M NaCl (Fig. 5b, 1), 0.2 % Triton X-100 (Fig. 5b, 3), or both 0.5 M NaCl and 0.2 % Triton X-100 (Fig. 5b, 2). After the washing step, the aggregated IBs were dissolved in 3 M urea and 1 % Triton X-100 and analyzed by SDS-PAGE. The least amount of impurities was detected in the sample where IBs were washed by the solution of both NaCl and Triton X-100 (Fig. 5b, IBw 2). Dissolved IBs were subsequently subjected to dialysis to remove urea while keeping the purified protein in a soluble form. Several assays were carried out: a single 48 h
123
dialysis or a two-step (2 h each) dialysis. The former dialysis approach promoted some protein degradation, while the two-step dialysis was more favorable, and as a consequence a purified soluble protein was obtained. RecK2D protein was approx. 90 % pure as judged by densitometric analysis of the bands presented in Fig. 5b, far right lane. The purification procedure resulted in the final yield of 1 mg purified recK2D protein per 100 mg of wet biomass of bacteria. The resulted recombinant K2 protein was applied for rabbit immunization and polyclonal antibody production (described in details in ‘‘Materials and Methods’’ section). In this work, about 80 mL of antiserum raised against recK2D protein was obtained. The titre and sensitivity of antibody were determined using indirect ELISA. It was shown that the antiserum is specific to recK2D protein expressed in bacteria. The antibody titre was found to be 1:16,000 (Fig. 6a). The antiserum was tested for the reactivity toward the genuine mature K2 toxin produced in yeast cells, and the titre of 1:300 was established (Fig. 6b). The results obtained from ELISA analysis show that the polyclonal antibodies in serum were able to react with both recombinant antigen and native K2 killer protein from yeast. The specificity of generated antibodies was addressed by Western blot analysis (Fig. 7). The purified recK2D protein and 1,000-fold concentrated yeast extracellular fraction (containing K2 toxin) were resolved by SDS-PAGE and then probed by antiserum raised against recK2D expressed in E. coli. Dilutions of antiserum were optimized and most appropriate established: 1:10,000 for recK2D protein or 1:300 for yeast K2 toxin detection. Western blot analysis confirms the results obtained by ELISA by displaying the positive signal for the purified recK2D (Fig. 7a) and the native K2 toxin (Fig. 7b). Western blot analysis of yeast K2 protein was performed using two different secondary antibody conjugates (based on AP or HRP), and in both cases K2 toxin migrates on SDS-PAGE corresponding to approx. 20 kDa (Fig. 7b, II, III). The antibody did not react with 1,000-fold concentrated medium of dsRNA lacking cell-free culture (Fig. 7b, I), confirming specificity toward K2 toxin.
Discussion In this study, we present the cloning, expression, and purification scheme of yeast K2 killer protein in E. coli. Base level synthesis of the full size preprotoxin under noninducing conditions is tolerated by bacteria resulting in negligible (if any) impact on the growth. However, this is in contrary to the conditions of induced expression, leading to more than twofold decrease in proliferation of bacteria
Mol Biotechnol
Fig. 5 Purification scheme for recombinant K2D preprotoxin (a) and SDS-PAGE analysis of the recK2D protein-expressing BL-21 cells (b). M Protein marker (PageRuler Unstained Protein Ladder, ThermoFisher Scientific); CrEx sonicated bacterial cells, CF cytoplasmic protein fraction, IB fraction of inclusion bodies, wash samples of IB
Fig. 6 ELISA of the antiserum against recK2D protein (a) and native K2 toxin (b). a The titre was measured, while the dilutions of sera were at 1:2,000, 1:4,000, 1:8,000, 1:16,000, 1:32,000, 1:64,000, and 1:128,000, when recK2D was used as immunogen; b dilutions of sera were 1:100, 1:200, 1:400, 1:800, 1:1,600, and 1:3,200, when native K2 toxin was used as immunogen
even if the modest expression level is achieved. C-terminal deletion of K2 protein not only completely abrogates the killing activity in yeast [23] but also ceases the negative
washing including 0.5 M NaCl (1), 0.5 M NaCl and 0.2 % Triton X-100 (2), 0.2 % Triton X-100 (3); IBw dissolved inclusion bodies after washing, respective to wash solutions; Dialysis samples after either 24 h, 48 h or (2 ? 2 h)—two-step dialysis for 2 h each
impact on bacterial cell proliferation even at relatively high expression levels. Since the adverse effect on bacterial growth is typical for full-length preprotoxin only, one can envision a structural role of the C-terminal fragment (Fig. 1c), which aggravates the formation of insoluble aggregates. RecK2D, devoid of this fragment, is inactivated by directing it into non-toxic IBs, relieving bacterial cells of stress induced by this harmful foreign protein. The expression of truncated killer preprotoxin was optimized by direction into easily resoluble IBs, allowing a simple and efficient separation from the host proteins. Usually, the formation of IBs is deleterious for biotechnology. To recover the biologically active protein, IBs have to be isolated and solubilized, and the target protein has to be refolded to its native state. The renaturation of target protein is rather complicated process and often leads to aggregation and misfolding [24, 25]. However, the separation of the soluble target protein from cytoplasmic host proteins is also a relatively difficult task as this compartment comprises the vast majority of the total cellular proteins. Despite the fact that IBs are highly pure protein deposits, and over-expressed recombinant protein may represent up to 95 % of the total IB content [26], the final yields of bioactive protein are often low [27]. However, in our case the formation of IBs was essential for high-level expression and efficient purification. Keeping the expression conditions of recK2D at a level sequestering the protein between insoluble and cytoplasmic fractions resulted in easily soluble IBs, as demonstrated by their solubilization in the buffer containing 3 M urea and subsequent refolding by two-step dialysis. The resulted protein is stable for several months. Two-step dialysis was found to be reasonable for efficient removal of urea. Graduate
123
Mol Biotechnol
The polyclonal antibodies raised against recK2D in rabbit were found to possess the required specificity not only toward truncated recK2 but also the native K2 killer toxin from yeast. This is an important property of such antibodies given the structurally different genuine K2 protein formed by processing of preprotoxin in its natural host. In summary, the expression of S. cerevisiae K2 preprotoxin in E. coli leads to diminished growth of bacteria and limited target protein production. C-terminal truncated protein was accumulated in the cytoplasm as non-toxic IBs. Expression of recK2D in insoluble yet easily refoldable form allowed a simple and efficient chromatography-free purification of the obtained aggregates. The resulted protein was used for rabbit immunization and production of the polyclonal serum. Indirect ELISA and Western blot analysis confirmed the specificity of raised polyclonal antibodies toward both K2 preprotoxin expressed in bacteria and the native K2 toxin. Overall, the described workflow might serve as an example of the use of bacterial expression systems to prepare antigen for raise of antibodies against cytotoxic eukaryotic proteins. Fig. 7 Detection of recK2D protein (a) and K2 toxin of yeast origin (b) by SDS-PAGE and Western blot analysis. M protein marker, S purified recK2D protein, I no dsRNA cell-free yeast M437 culture probed with anti-rabbit IgG-AP conjugate, II, III concentrated extracellular fraction of active K2 toxin-producing yeast strain M437 probed with anti-rabbit IgG-AP and HRP conjugates, respectively. An asterisk marks position of either recK2D (a) or native K2 (b) protein
decrease of urea ensured solubilization of the target protein and preservation from proteolysis, observed during prolonged single-step dialysis. Overall, our approach presents a simple and efficient chromatography-free purification of a soluble tag-less protein. K2 killer protein from yeast S. cerevisiae possesses high biological activity even at the negligible concentrations [9]. Immunization of one rabbit using the native protein would require cultivating of 2,400 L of K2 toxin-producing S. cerevisiae culture media, necessary to obtain 800 lg of protein. Besides economic cost and inadequate culture conditions between fermenter- and flask-based biomass cultivation, concentration of such quantities of media to volume of a few millilitres, a standard approach of preparation of native K2 toxin, possesses a significant resource burden for antigen supply. The simplicity and low cost of bacterial cultivation are a compelling advantage over any other expression systems, and therefore E. coli is the most preferable choice [11]. A sufficient amount of recK2D protein used for immunization was easily prepared from only 100 mL of E. coli culture in a simple and cheap approach.
123
Acknowledgments This work was supported by Grants of Research Council of Lithuania (LMT) to ES (Nos. MIP-061/2011, MIP-42/ 2013). We would like to thank Dr. J. Urbonavicˇius for helpful comments on the manuscript.
References 1. Goretti, M., Turchetti, B., Buratta, M., Branda, E., Corazzi, L., Vaughan-Martini, A., et al. (2009). In vitro antimycotic activity of a Williopsis saturnus killer protein against food spoilage yeasts. International Journal of Food Microbiology, 131, 178–182. 2. Magliani, W., Conti, S., Giovati, L., Zanello, P. P., Sperinde`, M., Ciociola, T., et al. (2012). Antibody peptide based antifungal immunotherapy. Frontiers in Microbiology, 3, 1–9. 3. Maqueda, M., Zamora, E., Alvarez, M., & Ramirez, M. (2012). Characterization, ecological distribution and population dynamics of Saccharomyces sensu stricto killer yeasts in the spontaneous grape must fermentations of southwestern Spain. Applied and Environmental Microbiology, 78, 735–743. 4. Magliani, W., Conti, S., Cunha, R. L., Travassos, L. R., & Polonelli, L. (2009). Antibodies as crypts of antiinfective and antitumor peptides. Current Medicinal Chemistry, 16, 2305–2323. 5. Breinig, F., Sendzik, T., Eisfeld, K., & Schmitt, M. (2006). Dissecting toxin immunity in virus-infected killer yeast uncover intrinsic strategy of self-protection. Proceedings of the National Academy of Sciences of the United States of America, 103, 3810–3815. 6. Meinhardt, F., & Klassen, R. (2009). Yeast killer toxins: Fundamentals and applications, in physiology and genetics. In J. Anke & D. Weber (Eds.), The mycota (pp. 109–130). Berlin: Springer. 7. Dignard, D., Whiteway, M., Germain, D., Tessier, D., & Thomas, D. Y. (1991). Expression in yeast of a cDNA copy of the K2
Mol Biotechnol
8.
9.
10.
11.
12.
13.
14.
15.
16.
killer toxin gene. Molecular and General Genetics, 227, 127–136. Mesˇkauskas, A., & Cˇitavicˇius, D. (1992). The K2-type killer toxin- and immunity-encoding region from Saccharomyces cerevisiae: Structure and expression in yeast. Gene, 111, 135–139. Lebionka, A., Servien_e, E., & Melvydas, M. (2002). Isolation and purification of yeast Saccharomyces cerevisiae K2 killer toxin. Biologija, 4, 7–9. Peternel, S., & Kornel, R. (2010). Isolation of biologically active nanomaterial (inclusion bodies) from bacterial cells. Microbial Cell Factories, 9, 1–16. Pandey, N., Sachan, A., Chen, Q., Ruebling-Jass, K., Bhalla, R., Panguluri, K., et al. (2013). Screening and identification of genetic loci involved in producing more/denser inclusion bodies in Escherichia coli. Microbial Cell Factories, 12, 43. Villaverde, A., & Carrio, M. (2003). Protein aggregation in recombinant bacteria: Biological role of inclusion bodies. Biotechnology Letters, 25, 1385–1395. Zerbs, S., Frank, A. M., & Collart, F. R. (2009). Bacterial systems for production of heterologous proteins. Methods in Enzymology, 463, 149–168. Narayanan, N., Khan, M., & Chou, P. (2011). Enhancing functional expression of heterologous Burkholderia lipase in Escherichia coli. Molecular Biotechnology, 47, 130–143. Gauci, C., Jenkins, D., & Lightowlers, M. W. (2011). Strategies for optimal expression of vaccine antigens from taeniid cestode parasites in Escherichia coli. Molecular Biotechnology, 48, 277–289. Kamionka, M. (2011). Engineering of therapeutic proteins production in Escherichia coli. Current Pharmaceutical Biotechnology, 12, 268–274.
17. Woodcock, D. M., Crowther, P. J., Doherty, J., Jefferson, S., Decruz, E., Noyer-Weidner, M., et al. (1989). Quantitative evaluation of Escherichia coli host strains for tolerance to cytosine methylation in plasmid and phage recombinants. Nucleic Acids Research, 17, 3469–3478. 18. Studier, F. W., & Moffatt, B. A. (1986). Use of bacteriophage T7 RNA polymerase to direct selective high-level expression of cloned genes. Journal of Molecular Biology, 189, 113–130. 19. Gulbiniene, G., Kondratiene, L., Jokantaite, T., Serviene, E., Melvydas, V., & Petkuniene, G. (2004). Occurence of killer yeast strains in fruit and berry wine yeast populations. Food Technology and Biotechnology, 42, 159–163. 20. Green, M., & Sambrook, J. (2012). Molecular cloning. New York: Cold Spring Harbor Laboratory Press. 21. Birnboim, H., & Doly, J. (1979). A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Research, 7, 1513–1523. 22. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680–685. 23. Gulbinien_e, G. (2002). Characterization of the functional organization of yeast K2 killer preprotoxin gene. Biologija, 3, 37–40. 24. Panda, A. K. (2003). Bioprocessing of therapeutic proteins from the inclusion bodies of Escherichia coli. Advances in Biochemical Engineering/Biotechnology, 85, 43–93. 25. Burgess, R. R. (2009). Refolding solubilized inclusion body proteins. Methods in Enzymology, 463, 259–282. 26. Ventura, S., & Villaverde, A. (2006). Protein quality in bacterial inclusion bodies. Trends in Biotechnology, 24, 179–185. 27. Lilie, H., Schwarz, E., & Rudolph, R. (1998). Advances in refolding of proteins produced in E. coli. Current Opinion in Biotechnology, 9, 497–501.
123
