Microbial Cell Factories Recombinant Production of Streptococcus Equisimilis Streptokinase by Streptomyces Lividans

Background: Streptokinase (SK) is a potent plasminogen activator with widespread clinical use as a thrombolytic agent. It is naturally secreted by several strains of beta-haemolytic streptococci. The low yields obtained in SK production, lack of developed gene transfer methodology and the pathogenesis of its natural host have been the principal reasons to search for a recombinant source for this important therapeutic protein. We report here the expression and secretion of SK by the Gram-positive bacterium Streptomyces lividans. The structural gene encoding SK was fused to the Streptomyces venezuelae CBS762.70 subtilisin inhibitor (vsi) signal sequence or to the Streptomyces lividans xylanase C (xlnC) signal sequence. The native Vsi protein is translocated via the Sec pathway while the native XlnC protein uses the twin-arginine translocation (Tat) pathway.


Background
Streptokinases are proteins translocated to the growth medium by many strains of beta-haemolytic streptococci. Streptokinase is not an enzyme per se but rather a potent activator that interacts with plasminogen to form a stoichiometric 1:1 complex. This interaction results in the activation of plasminogen to plasmin, which is the active fibrinolytic component of the circulatory system [1]. SK was the first drug introduced as a therapy for acute myocardial infarction more than 40 years ago [2]. It is now the leading fibrinolytic agent in the treatment of thromboembolic conditions [3] and is included in the World Health Organization Model List of Essential Medicines.
The Streptococcus equisimilis H46A skc gene encoding streptokinase has been cloned and expressed in several heterologous hosts due to the pathogenicity of its natural host. Haemolytic streptococci secrete several toxins that complicate the downstream purification. Besides, genetic modification of the natural host is restricted as rather few genetic tools are available. As a result, the recombinant production of this protein in E. coli has been widely used, including the use of the protein SKC-2 naturally secreted by Streptococcus equisimilis ATCC 9542 [4,5]. High-level expression of skc in E. coli has been reported, but the formation of inclusion bodies consisting of highly aggregated SK molecules makes its recovery in an active form difficult [4,5]. High level of intracellular SK has also been obtained during continuous fermentation of recombinant Pichia pastoris but protein recovery requires cell lysis [6].
Since the recovery of extracellular proteins is generally easier than that of cytoplasmic proteins, the expression and subsequent secretion of SK have been studied in several heterologous hosts like Escherichia coli, Bacillus subtilis and Pichia pastoris [7][8][9]. In case of B. subtilis, the use of the sixextracellular-protease-deficient strain, WB600, greatly improved the yield of recombinant SK. The protein was also secreted into the culture medium by P. pastoris, but it was found to be heavily glycosylated. The biological activity of both secreted streptokinases was proved. A recent study using Schizosaccharomyces pombe as host, reported the expression of SK and its secretion into the periplasmic fraction without glycosylation and significant degradation or modification. However, conventional chromatographic approaches used before to purify SK from other hosts were inadequate because of cofractionation of a few proteins of similar size with SK through all the chromatographic steps [10].
As it is not possible to predict which host will be the best for the production of a protein, the aim of this work was to evaluate Streptomyces lividans as host for recombinant production of SK. S. lividans has been successfully used for the production of several proteins of bacterial and eukary-otic origin [11][12][13]. The advantages of the S. lividans host include its natural ability to secrete high levels of bioactive molecules into the extracellular medium, limited protease activity, its biological safety and well-established fermentation technology [14]. In the present study, the S. lividans system has been tested for the secretory production of the streptokinase from Streptococcus equisimilis group C by using the Sec and the recently described twin-arginine translocation (Tat) pathway in S. lividans [15]. The sequence encoding mature SK was fused to the Secdependent signal sequence of Streptomyces venezuelae CBS762.70 subtilisin inhibitor [16] and the twin-arginine signal sequence of S. lividans xylanase C [17], respectively. SK production in S. lividans was evaluated and purification of the secreted SK protein was carried out.

Construction of SK expression/secretion vectors
In order to establish the expression and secretion of SK from Streptococcus equisimilis ATCC9542 in S. lividans, the skc-2 gene was amplified by PCR using chromosomal DNA as template and finally cloned in appropriate vectors in Streptomyces. The constructed expression/secretion vectors pOVsiSK and pOXlnCSK encode the fusion proteins Vsi-SK and XlnC-SK, respectively (Table 1). Vsi-SK consists of the Sec-dependent Vsi signal peptide, the first two amino acid residues of mature Vsi followed by the mature SK, while XlnC-SK is composed of the Tat-dependent XlnC signal peptide, the first three amino acids of mature XlnC and mature SK. In this way, the signal peptidase cleavage sites of Vsi and XlnC are remained and a proper processing of the precursor proteins is achieved. Both fusion genes were placed under control of the vsi promoter, of which has been proved that it efficiently promotes heterologous gene expression [13,16].

Secretion of SK by S. lividans
S. lividans transformants carrying pOVsiSK and pOXl-nCSK, respectively, were grown at lab scale in rich BTSB medium and at several time intervals the presence of SK in the culture filtrate was assessed by Western blot analysis. In consequence of the poor SK yield using the XlnC signal peptide as mediator for translocation, only S. lividans [pOVsiSK] was tested under fermentation conditions.
Using small-scale fermentation conditions, the secretory SK production reached up to 15 mg SK/l, which corresponds to a fourfold improvement of secretory SK yield in S. lividans compared to lab-scale conditions.

Purification of recombinant SK secreted by S. lividans
Having defined the fermentation conditions for the secretory production of recombinant SK, the protein was purified from the extracellular culture fraction. The protein fraction obtained through ammonium sulfate precipitation (45% saturation) was dissolved in 20 mM Tris-HCl (pH 6.0), dialyzed against the same buffer and then applied on a DEAE-Sephacel column. The proteins were eluted in 20 mM Tris-HCl, 150 mM NaCl, pH 6.0. Samples from the various purification steps were analysed by SDS-PAGE followed by Coomassie staining (Fig. 3A) and immunodetection of the recombinant SK using a monoclonal anti-SK antibody (Fig. 3B). Samples from culture supernatant and 45% ammonium sulphate saturation fraction of S. lividans TK 24 [pOW15] were included as negative controls (Fig. 3, lanes 1 and 2). This experiment revealed that the purified proteins correspond to SK. Pooled elution fractions with a purity grade of 58% contained the 44-kDa degradation product which co-eluted with the 47-kDa mature SK.
In order to determine the specificity of signal peptidase processing and the nature of the approximately 44-kDa protein, an N-terminal sequence was carried out on both the 47-and 44-kDa proteins obtained from S. lividans TK 24 [pOVsiSK] culture supernatant. N-terminal residues of both purified proteins were those predicted from the sequence ( Table 2). The Vsi signal peptide was thus correctly cleaved off and the N-terminus of mature Vsi-SK released by S. lividans remained intact. This result also implicates that the processing of mature Vsi-SK secreted by Streptomyces occurred at its C-terminal end.
The specific activity of the partially purified proteins secreted by S. lividans [pOVsiSK] was amounted to 2661 IU/mg protein (Table 3). In consequence of incompatibility between the crude culture medium and the chromogenic substrate assay, we were not able to determine the initial specific activity of SK.

Discussion
In the present study, it was shown that SK from Streptococcus equisimilis ATCC9542 could be efficiently secreted in a bio-active form via the Sec pathway in Streptomyces lividans. Sec-routed secretion was obtained by using the regulatory signal sequences of S. venezuelae CBS762.70 subtilisin inhibitor gene. The Tat translocation route was also tested for the secretion of SK in S. lividans by means of a fusion of SK to the Tat-dependent signal peptide of S. lividans xylanase C. Yield was about 30 times higher when the SK was fused to the Sec-dependent Vsi signal peptide compared to the fusion with the Tat-dependent XlnC signal peptide. Although the use of the Tat pathway in most cases does not result in higher production yield compared to Sec-mediated secretion (e.g. Schaerlaekens et al. 2004), some proteins need to be secreted via the Tat pathway to obtain their bio-active conformation. This is the case for the homologous protein xylanase C [18], but also for the heterologous enhanced green fluorescent protein (EGFP) [19].
The maximum level of SK secreted by S. lividans was 15 mg/l of culture, but as a result of incompatibility between the crude culture medium and the chromogenic substrate assay, we were not able to determine the initial activity of  [9,10]. However, it is possible to establish a relative comparison with the total yield (24.5 mg/l) and the initial specific activity of SK secreted by S. pombe (1581 IU/mg protein) [10].
SK has a tendency to degrade very easily [20,21]. Several hosts, including the native host Streptococcus equisimilis, produce at least two major forms of SK [7,8,22]: the intact mature SK with a molecular mass of 47 kDa and a 44-kDa degradation product. This degradation product lacks 31 or 32 C-terminal residues whereas it retains the plasminogen activation capability [23]. Furthermore, C-terminal deletion mutants of SK lacking 40 [24] or 41 amino acids [25] exhibited normal plasminogen activator function. In addition to the 47-and 44-kDa bands, a 32-kDa degradation product was detected by Western blot. Since SK proteins which lack 18 or more N-terminal or 51 or more Cterminal amino acid residues are unlikely to be effective thrombolytic agents [24], the 32-kDa SK-related protein missing about 135 aa residues was not further investigated.
It was demonstrated that the post-translational modification at the C-terminus of native SK was caused by chymotrypsin-like activity [23]. Similar degradation of recombinant SK has been also reported to occur in heterologous hosts such as Streptococcus sanguis [23] and E. coli [26]. Chymotrypsin-like activity and several genes encoding chymotrypsin-like serine proteases have been reported in S. lividans 66 [27,28]. Since the first amino acid residues of the S. lividans-produced Vsi-SK were identical to those of the expected N-terminal sequence, the recombinant protein was proteolytically degraded at its C-terminal end.
In case of recombinant SK production in Lactococcus lactis, the protease susceptibility and hence the productivity of SK was dependent on the pH of the culture and the initial phosphate concentration of the medium. Suppression of the acid tolerance response, by which protease expression is induced, enhanced the SK yield 2.5 fold [29]. Results of a differential scanning calorimetry study on E. coli-derived recombinant S. equisimilis SK firmly indicated that at neutral and basic pH, the recombinant SK from Streptococcus equisimilis group C (ATCC 9542) has four domains, whereas gentle changes in the experimental conditions, such as mild acidification or increase in the NaCl concentration, decreased this number [30]. Consequently, pH and ionic strength of the production medium define the conformational status of SK and are thus important factors determining the protease susceptibility of the recombinant protein.
The specific activity of the partially purified SK (58% purity) secreted by S. lividans [pOVsiSK] was determined at 2661 IU/mg protein. We believe that further up-scaling of the fermentation process and optimisation of production medium and purification protocol, will surely improve yield of recombinant bio-active SK in S. lividans.

Conclusion
Heterologous expression of Streptococcus equisimilis ATCC9542 skc-2 in Streptomyces lividans was successfully achieved. SK can be translocated via both the Sec and the Tat pathway in S. lividans, but yield was about 30 times higher when the SK was fused to the Sec-dependent Vsi signal peptide compared to the fusion with the Tatdependent signal peptide of S. lividans xylanase C. Smallscale fermentation led to a fourfold improvement of secretory SK yield in S. lividans compared to lab-scale conditions. The plasminogen activity of the partially purified SK (58% purity) secreted by S. lividans [pOVsiSK] was determined at 2661 IU/mg protein. Once more, Streptomyces lividans was shown to be a valuable host for the production of a world-wide important, biopharmaceutical product in a bio-active form.

Bacterial strains and growth conditions
E. coli TG1 was used as host for cloning purposes. Culture conditions for E. coli were as described by Sambrook et al. [31]. Streptococcus equisimilis ATCC9542 cells were grown as described by Estrada et al. [4]. Streptomyces lividans TK24 was selected as host for heterologous protein production. Protoplast formation and subsequent transfor- mation of S. lividans were carried out as described by Kieser et al. [32]. Regeneration of S. lividans protoplasts and selection of transformants was carried out on MRYE medium [33]. When appropriate, thiostrepton (50 µg/ml in solid medium or 10 µg/ml in liquid medium) was added. Spore suspensions of S. lividans TK24 and derivatives were stored at -70°C in 20% (v/v) glycerol. Primary cultures of S. lividans strains were routinely cultured for 72 h (28°C, 240 rpm) in BTSB, which is a modified version of the medium described by Dyson and Schrempf [34]: 10% sucrose, 1% yeast extract, 1% glucose, 0.5% NaCl, 0.5% soya flour, 1.7% tryptone, 0.25% K 2 HPO 4 , pH 7.2. For monitoring recombinant protein expression and secretion, 1-ml primary cultures were inoculated to 0.5-L shake flask containing 0.1 L of BTSB medium and grown for 40 h at 28°C and 300 rpm. For production of SK, the recombinant strain was cultured for 48 h (28°C, 350 rpm) in a 2.5-L MBR reactor containing 1.5 L BTSB medium. The pH was controlled at 7.0 by the addition of 5 N NaOH.

Plasmid construction and recombinant DNA technology
DNA manipulations were carried out following standard procedures [31,32]. Restriction endonucleases and DNAmodifying enzymes were from Invitrogen and Roche Diagnostics. Upon isolation of S. equisimilis ATCC9542 chromosomal DNA as described by Estrada et al. [4], the skc-2 gene was PCR-amplified from the chromosome under standard conditions using the oligonucleotides SK-F1 and SK-R1 (see Table 4) and Taq polymerase. To allow in-frame fusion of the skc-2 gene downstream the signal sequences, the oligonucleotide SK-F1 was designed with an EcoRV site (GATATC) in its 5'end. Consequently, a silent mutation was introduced in the first codon of the mature part of the skc-2 gene: ATT was replaced by ATC, both encoding isoleucine.
The 1245-bp PCR fragment was ligated into pGEM ® -T Easy (Promega) and the resulting plasmid was denominated pGEM-SK. The DNA sequence was verified using the Thermo Sequenase Primer Cycle Sequencing Kit with 7deaza-dGTP on an ALFexpress apparatus (Amersham Biosciences, Rainham, UK). Subsequently, the 1.3-kb EcoRV/ EcoRI fragment of pGEM-SK was cloned into pBS-CBSS [16] successively treated with DraII, Klenow polymerase and EcoRI. The unique DraII site in pBS-CBSS is located two codons downstream the signal peptidase recognition site. skc-2 was also cloned into pBSVXM, a derivative of plasmid pBSVX [17] missing an EcoRI site. To remove the EcoRI site located upstream the vsi promoter in pBSVX, a site-directed mutagenesis was carried out by means of PCR using Pfu polymerase and the mutagenic oligonucleotides PBSXylE-F and -R (Table 4), which contain the desired mutation. As such, a unique EcoRI site located downstream the S. lividans xlnC signal sequence was available. In order to insert the skc-2 gene fused to the third codon of mature xlnC, the vector pBSVXM was digested with NsiI, treated with T4 DNA polymerase removing the 3'-protruding ends and finally treated with EcoRI. DNA sequence analyses of the newly constructed fusion genes confirmed their correctness.
Finally, both expression/secretion cassettes were isolated as BamHI/EcoRI-fragments and ligated in BamHI/EcoRIdigested pOW15. The vector with the vsi signal sequence was designated pOVsiSK and the vector with the xlnC signal sequence was denominated pOXlnCSK. Plasmids used in this study are listed in Table 1.

Detection of SK
The detection of SK in culture supernatants and cell lysates of S. lividans transformed with pOVsiSK or pOXlnCSK was performed using Western Blot and immunodetection. Gel electrophoresis of proteins was carried out on 10% SDSpolyacrylamide gels [35]. Separated proteins were visualized by Coomassie brilliant blue staining or transferred to a Hybond™-C extra membrane (GE Healthcare) by using a semidry transfer cell (Biometra) according to the manufacturer's recommendations. SK was detected using a mouse anti-SK monoclonal antibody (produced by Center for Genetic Engineering and Biotechnology, Sancti Spiritus, Cuba). HRP-conjugated goat anti-mouse antibody (Promega) was used as secondary antibody. Immunoreactive bands were visualized by brief exposure to 3,3diaminobenzidine or 4-chloro naftol (Sigma). Cell lysates were obtained according to Pimienta et al. [36]. The protein content of culture supernatants, cell lysates and purified fractions was determined using the Bradford method [37].

Name Sequence (5'-3' direction) Restriction sites (in italic)
The mutagenic base is in bold face.
The molecular size and the purification degree of recombinant SK protein were estimated from densitometric scanning of Coomassie brilliant blue-stained gels using a GENE GENIUS gel documentation system and GeneTools software (Syngene).
SK present in culture supernatants or anion exchange chromatography eluates was quantified by means of a general sandwich ELISA protocol (Abrahantes et al. unpublished results). The SK standard was kindly supplied by the Development Division, Center for Genetic Engineering and Biotechnology, Havana, Cuba. The coefficient of variation of the ELISA tests was less than 10%.
SK activity was monitored spectrophotometrically at 405 nm in a coupled SK-plasminogen assay employing the chromogenic substrate S-2251 (Kabi, Sweden) according to Hernández et al. [38]. The specific activity (IU/mg) was calculated by dividing the SK activity (IU/ml) with protein concentration (mg/ml).

Protein purification and chromatography
For purification of SK from recombinant S. lividans cultures, strains were grown for 2 days in 200 ml BTSB medium. Then, cultures were centrifuged and the mycelium was resuspended in 0.1 L of water. This suspension was transferred to 1.5 L BTSB in the MBR reactor and bacterial growth was continued for 2 days. Culture supernatant proteins were precipitated by addition of (NH 4 ) 2 SO 4 (45% saturation, 4°C) and collected by centrifugation (Hettich Universal 32R centrifuge, Sorvall, 1620A rotor, 4°C, 20 min, 8000 × g). The protein pellets were left overnight at 4°C in 0.1 L of 20 mM Tris-HCl buffer, pH 6.0. Then, the protein solution was dialyzed against 20 mM Tris-HCl buffer (pH 6.0) at 4°C for 20 h and was finally applied on a DEAE Sephacel column equilibrated with 20 mM Tris-HCl buffer, pH 6.0. The column was extensively washed with the mentioned buffer followed by 1 column volume of 20 mM Tris-HCl, 20 mM NaCl, pH 6.0. SK protein elution from the DEAE Sephacel column was carried out with 3 column volumes of 20 mM Tris-HCl, 150 mM NaCl, pH 6.0 at a flow rate of 0.5 ml/min. One ml fractions were collected. Fractions containing SK with a similar degree of purity, determined by means of SDS-PAGE followed by Coomassie staining, were pooled.

N-terminal amino acid sequence analysis
The purified SK was subjected to SDS-PAGE and blotted on a Hybond-P membrane (GE healthcare) as described by Ausubel et al. [39]. After Coomassie staining, the relevant protein bands were excised and subjected to sequencing. The N-terminal amino acid sequence of recombinant SK was determined by Edman degradation using an automatic 477A-1201 protein sequencing system (Applied Biosystems).