FEMS Microbiology Letters 212 (2002) 209^216

FEMS Microbiology Letters 212 (2002) 209^216 www.fems-microbiology.org Construction of fusion vectors of corynebacteria: expression of glutathione-S...
Author: Darcy Pearson
4 downloads 1 Views 461KB Size
FEMS Microbiology Letters 212 (2002) 209^216

www.fems-microbiology.org

Construction of fusion vectors of corynebacteria: expression of glutathione-S-transferase fusion protein in Corynebacterium acetoacidophilum ATCC 21476 Preeti Srivastava, J.K. Deb



Department of Biochemical Engineering and Biotechnology, Indian Institute of Technology, Delhi, New Delhi 110 016, India Received 25 October 2001; received in revised form 29 April 2002; accepted 4 May 2002 First published online 29 May 2002

Abstract A series of fusion vectors containing glutathione-S-transferase (GST) were constructed by inserting GST fusion cassette of Escherichia coli vectors pGEX4T-1, -2 and -3 in corynebacterial vector pBK2. Efficient expression of GST driven by inducible tac promoter of E. coli was observed in Corynebacterium acetoacidophilum. Fusion of enhanced green fluorescent protein (EGFP) and streptokinase genes in this vector resulted in the synthesis of both the fusion proteins. The ability of this recombinant organism to produce several-fold more of the product in the extracellular medium than in the intracellular space would make this system quite attractive as far as the downstream processing of the product is concerned. 9 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Corynebacterium ; Glutathione-S-transferase fusion vector ; Enhanced green £uorescent protein; Streptokinase

1. Introduction Escherichia coli has been the workhorse of genetic engineers both for basic studies as well as for the expression of recombinant proteins. There are some limitations of E. coli as a recombinant host. Normally, the organism does not secrete protein. This is not advantageous from the downstream processing point of view. Also, it is not a food grade bacterium. In this respect, Bacillus subtilis could be useful but it has high proteolytic activity. Alternatively, Bacillus brevis has been used for this purpose [1]. Secretory nature and no detectable extracellular proteases have been an advantage of this organism. Generally, food grade organisms such as Lactococcus lactis [2,3] are useful for the production of commercially important proteins. Soil corynebacteria, such as Corynebacterium glutamicum, are another class of food grade bacteria that can serve as potential hosts for the production of recombinant proteins. During the past two decades, considerable progress has taken place in our understanding of the molecular

* Corresponding author. Tel. : +91 (11) 6591006; Fax : +91 (11) 6868521. E-mail address : [email protected] (J.K. Deb).

biology of coryneform bacteria through the development of e¡ective transformation systems and cloning vectors. Most of these vectors were constructed from endogenous plasmids [4,5]. These plasmids varied over a wide molecular range, from as low as 3 kb to as high as 55 kb. Exhaustive reviews on the molecular biology of these plasmids and cloning vectors constructed with them have appeared in recent years [6,7]. Considerable progress has been made in the development of C. glutamicum^E. coli shuttle vectors, corynebacterial expression vectors for cloning of amino acid biosynthesis pathway genes and promoter probe vectors [8,9]. Also, expression of heterologous proteins such as B. subtilis protease subtilisin has been achieved in corynebacteria using vector pEP2 [10]. The gene encoding ¢bronectin binding protein 85A of Mycobacterium tuberculosis has also been expressed in C. glutamicum using pBL1 replicon based vectors containing tac promoter of E. coli or endogenous cspB gene promoter [11]. It was reported earlier by Morinaga et al. [12] that common E. coli promoters such as lacUV5, tac and trp function very well in Brevibacterium lactofermentum. The former two promoters have been found to be 50% active in this organism compared to their activity in E. coli. Later, search for endogenous promoters has led to the cloning and identi¢cation of promoters in C. glutamicum

0378-1097 / 02 / $22.00 9 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII : S 0 3 7 8 - 1 0 9 7 ( 0 2 ) 0 0 7 4 0 - 1

FEMSLE 10528 26-6-02

210

P. Srivastava, J.K. Deb / FEMS Microbiology Letters 212 (2002) 209^216

[13]. The study also revealed normal consensus sequences at about 35 bp and 10 bp upstream of the transcription start site, similar to that observed in many other bacterial genes. These promoters were also found to be functional in E. coli. Earlier reports from this laboratory described the construction of several vectors, including corynebacteria^ E. coli shuttle vector, based on pBL1 replicon [14,15]. A detailed study on the stability of these vectors in Corynebacterium acetoacidophilum in continuous culture showed the vectors to be extremely stable. The study also led to the identi¢cation of minimum replicon. Since none of them were expression vectors, we were interested in exploiting similarities in promoter functions in E. coli and in corynebacteria for the construction of diverse expression vectors using strong tac promoter of E. coli. One of the useful strategies for the expression of heterologous protein is to construct a fusion vector that would result in the expression of foreign gene as a fusion product. This is favored under such circumstances where the expression of heterologous protein in the host is di⁄cult for various reasons. The fusion protein produced can sometimes be puri¢ed by a⁄nity chromatography. One of the genes extensively used for this purpose in E. coli is glutathioneS-transferase (GST). To explore if the construction of fusion vector based on GST would lead to successful expression of this enzyme in corynebacteria, we constructed vectors using tac promoter and GST gene in C. acetoacidophilum. Our ultimate objective was to clone, express and purify heterologous protein in corynebacteria. The present work describes the construction of GST fusion vectors, study of expression of GST in C. acetoacidophilum and of genes for green £uorescent protein and streptokinase as GST fusion proteins.

2. Materials and methods

2.2. Recombinant DNA techniques All DNA manipulations were carried out according to Sambrook et al. [16]. Restriction enzymes were obtained from Boehringer Mannheim, Germany and MBI Fermentas. Restriction digestion was carried out according to their protocols. Puri¢cation of DNA from agarose gels was carried out using Qiaquick gel extraction kit (Qiagen, Chatsworth, GA, USA). For blunt end ligation, the vector DNA was treated with calf intestinal phosphatase (CIP) and the dephosphorylated DNA fragments were puri¢ed by Qiaquick column (Qiagen). 2.3. Transformation of E. coli and C. acetoacidophilum Transformation of CaCl2 treated E. coli cells was carried out according to Sambrook et al. [16]. Preparation of protoplasts of C. acetoacidophilum and their transformation were carried out according to Mukherjee et al. [15]. 2.4. Plasmid stability studies Single colony of plasmid bearing cells was inoculated in BHI medium containing appropriate antibiotic and allowed to grow overnight at 30‡C. The overnight grown culture was inoculated in 100 ml of fresh BHI medium (0.5% inoculum) and was grown without selection pressure for another 24 h. This culture was used as inoculum for fresh 100 ml BHI medium and the above process repeated for 3 days. Every day, a sample of overnight grown culture was serially diluted and aliquots of appropriately diluted samples were plated on BHI agar plates. The colonies from these plates were replica plated on BHI agar plates with and without antibiotics. The fraction of plasmid bearing cells was calculated by dividing the number of colonies growing on antibiotic-containing plates by those growing on antibiotic-free plates.

2.1. Bacterial strains, plasmids and growth conditions 2.5. Preparation of cell extracts C. acetoacidophilum ATCC 21476 was the corynebacterial strain used in the present study. E. coli DH5K was used as the host for transformation by E. coli plasmids. E. coli plasmids used in the present study were pTrc99A, pEGFPC1, pUCsk and pGEX4T-1^3. For the construction of fusion vectors, the corynebacterial plasmid pBK2 containing pBL1 replicon was used. Both C. acetoacidophilum and E. coli were maintained in Luria agar (LA) and were routinely grown in Luria broth (LB). The transformants were grown in the medium supplemented with appropriate antibiotics at the following concentrations : ampicillin (40 Wg ml31 ) and kanamycin (40 Wg ml31 ). Brain-heart infusion (BHI) and hypertonic brain-heart infusion media [14] were used respectively for the growth and preparation of protoplasts of C. acetoacidophilum.

E. coli cells harboring pGEX plasmid were grown in LB containing ampicillin (40 Wg ml31 ) at 37‡C. C. acetoacidophilum harboring plasmids were grown in BHI medium containing kanamycin (40 Wg ml31 ) at 30‡C. The exponentially growing cells were subjected to IPTG (1 mM) induction and the cells were harvested after desired time by centrifugation and washed several times with 20 mM potassium phosphate bu¡er (pH 7.0). E. coli cells were broken by 30 s each of alternate sonication and chilling for a total duration of 10 min. The same procedure was followed for corynebacteria except that the duration in this case was 30 min. The sonicated cell suspension was centrifuged at 10 000 rpm for 15 min and the supernatant used for GST assay.

FEMSLE 10528 26-6-02

P. Srivastava, J.K. Deb / FEMS Microbiology Letters 212 (2002) 209^216

211

Fig. 1. Construction of corynebacterial GST fusion vector pBKGEXm2. Gray bar, the larger fragment of pBK2; black bar, the smaller fragment of pGEXm2.

2.6. Assay of GST activity

2.7. Screening of clones for streptokinase

The activity of the enzyme in the cell extract was determined by the method utilizing glutathione and 1-chloro2,4-dinitrobenzene (CDNB) as substrates for GST. The yellow product S-(2,4-dinitrophenyl)glutathione formed was detected by measurement of absorbance at 340 nm [17].

The transformants were transferred from hypertonic BHI plates to LA plates and incubated overnight at 30‡C. The plates were overlaid with molten agar containing defatted milk and plasminogen, incubated at 30‡C and positive clones were identi¢ed by the zone of clearance around the colonies.

FEMSLE 10528 26-6-02

212

P. Srivastava, J.K. Deb / FEMS Microbiology Letters 212 (2002) 209^216

Table 1 Intracellular and extracellular activities of GST Sample

Intracellular GST activity U ml

E. coli (DH5K) pGEX4T-2 C. acetoacidophilum pBKGEXm2

100 89.5

31

Extracellular GST activity

Total U

U ml

Total U

100 89.5

2.0 20.8

41.6 416.6

2.8. Streptokinase assay Streptokinase activity was determined by a method similar to that described by Castellino et al. [18]. Chromozym PL (Boehringer Mannheim) was used as the substrate.

31

Total activity (U)

Extracellular/intracellular

141.6 506.1

0.4 4.6

Brie£y, the cell extract containing streptokinase was mixed with plasminogen and pre-incubated at 37‡C for 5 min. The substrate was then added, and the release of 4-nitroaniline was monitored by measuring absorbance at 405 nm.

Fig. 2. Construction of corynebacterial GST^GFP fusion vector pBKGEXm2EGFP.

FEMSLE 10528 26-6-02

P. Srivastava, J.K. Deb / FEMS Microbiology Letters 212 (2002) 209^216

213

2.9. Sodium dodecyl sulfate^polyacrylamide gel electrophoresis (SDS^PAGE) Proteins were fractionated by electrophoresis in a 12% SDS^polyacrylamide gel according to the method of Laemmli [19].

3. Results and discussion 3.1. Construction of GST fusion vectors of corynebacteria For the construction of a GST fusion vector, the plasmid pGEX4T-2 was initially modi¢ed to generate pGEXm2. For this purpose, pGEX4T-2 was digested by SspI and the smaller fragment containing the GST cassette was puri¢ed by gel elution. Similarly, pTrc99A was digested with SspI to generate two fragments and the larger fragment carrying the lacIq gene, the pBR322 origin of replication and the ampr gene was gel eluted and dephosphorylated by CIP. The two gel eluted fragments were ligated to obtain the plasmid pGEXm2. This plasmid has a single SphI site which is derived from the pTrc99A fragment. The above manipulation made it possible to get the lacIq regulator gene and the GST cassette in a single fragment when pGEXm2 was digested with SphI and ScaI. Since pBK2 has multiple closely spaced SphI sites in the non-essential region of the plasmid, the above fragment will have a compatible cohesive end for insertion in pBK2. Initially, pBK2 was digested with MluI and blunt ended with Klenow DNA polymerase. Subsequent digestion by SphI generated two fragments. The larger fragment containing the pBL1 origin of replication was ligated to the SphI^ScaI fragment containing the lacIq and GST cassette of pGEXm2 to obtain pBKGEXm2 (Fig. 1). The same strategy was followed to construct pBKGEXm1 and pBKGEXm3 ¢rst by constructing pGEXm1 and pGEXm3 respectively. 3.2. Expression of GST To check if Ptac-driven GST can be expressed in C. acetoacidophilum, recombinants were obtained by transformation of the protoplasts with pBKGEXm2 and isolation of kanamycin-resistant colonies. Stability studies showed that the plasmid was 100% stable up to 60 generations (data not shown). For extraction of enzymes from cells, the recombinant cells were grown in hypertonic BHI medium in the presence of 40 Wg ml31 kanamycin. When the cells reached an OD600nm of 0.4, induction by IPTG was initiated. The cells were harvested at di¡erent times after the induction and enzyme assays carried out on the cell-free extracts as described in Section 2. The optimum time for harvesting of cells after induction was found to be 3 h. Cell extract was also prepared from E. coli DH5K bearing pGEX4T-2 of approximately the same cell density

Fig. 3. SDS^PAGE of induction of EGFP by recombinant C. acetoacidophilum: Lane 1, uninduced ; lane 2, induced ; lane M, molecular mass markers. GST^EGFP fusion protein band is indicated by an arrow.

as that of C. acetoacidophilum cells. This served as positive control. Extracts of E. coli DH5K and C. acetoacidophilum harboring pBK2 were used as negative controls. Although the intracellular GST activities in the two strains were comparable, the total GST activity of C. acetoacidophilum recombinant was 3.5 times more than that of E. coli (Table 1) in spite of the fact that the copy number of E. coli plasmid pGEX4T-2 was higher than that of C. acetoacidophilum plasmid pBKGEXm2 (data not shown). This is quite remarkable, since it was earlier found that tac promoter-driven expression of CAT activities was the same in both B. lactofermentum and E. coli [12]. Moreover, it was observed that though the intracellular GST activity was comparable in the two strains, the extracellular activities di¡ered considerably. The bulk of the activity was intracellular in E. coli, but the intracellular activity in C. acetoacidophilum was 20% of its extracellular activity. These results show that the enzyme is mainly secreted by the latter strain, which is quite attractive from the downstream processing point of view. It was also observed that sometimes heterologous genes need modi¢cation of their codons to suit those of the host for optimum expression [20]. Our results suggest that the native codons of GST are recognized e⁄ciently by C. acetoacidophilum. The better expression of GST in C. acetoacidophilum may be due to more e⁄cient transcription from Ptac or due to e⁄cient translation of GST codons or both. 3.3. Expression of EGFP fusion protein in C. acetoacidophilum Since Ptac-driven GST is e⁄ciently expressed in C. acetoacidophilum, it would also be of interest to study the expression of heterologous protein as fusion partner of GST in this host. The green £uorescent protein EGFP

FEMSLE 10528 26-6-02

214

P. Srivastava, J.K. Deb / FEMS Microbiology Letters 212 (2002) 209^216

Fig. 4. Construction of corynebacterial GST^streptokinase fusion vector pBKGEXm2sk. Striped bar, the HincII fragment containing streptokinase gene (sk).

was chosen for this purpose due to its diverse applications. The fusion vector was constructed by digesting plasmid pEGFPC1 sequentially with MluI and NheI. The smaller fragment was gel eluted and the ends were ¢lled with Klenow DNA polymerase. This fragment was inserted at the SmaI site of pBKGEXm2. This would allow 50% of the population of the insert to be joined in the right orientation and in the right frame. The resulting vector pBKGEXm2EGFP (Fig. 2) was used to transform C. acetoacidophilum and the transformants screened by £uores-

cence at an excitation wavelength of 480 nm. The positive clones expressing a functional GFP were identi¢ed by green £uorescence. The fusion protein appeared as an intense band in polyacrylamide gel of the IPTG-induced sample (Fig. 3). There are several versions of GFP available. Genes like GFPuv contain codons which are preferred in E. coli, while others such as EGFP, used in the present study, contain silent base mutations that correspond to human codon usage preferences. Leblon and coworkers [21] reported expression of GFPuv in C. glutami-

FEMSLE 10528 26-6-02

P. Srivastava, J.K. Deb / FEMS Microbiology Letters 212 (2002) 209^216

215

Table 2 Intracellular and extracellular activities of streptokinase Sample

Intracellular U ml

C. acetoacidophilum pBKGEXm2sk

36

31

Extracellular Total U

U ml

1800

560

cum. Sun et al. [22] reported that EGFP is expressed e⁄ciently in Streptomyces coelicolor due to the codon usage of jelly¢sh, the natural source of this protein, corresponds to that of many GC-rich Streptomyces genes. The present work showed that such codons are also expressed well in corynebacteria although the GC content of C. acetoacidophilum is not as high as that of S. coelicolor. The results are important because they indicate that it is possible to express heterologous eukaryotic genes e⁄ciently in corynebacteria as far as this protein is concerned.

31

Total activity (U)

Extracellular/intracellular

29 800

15.6

Total U 28 000

phy on glutathione^Sepharose column. The advantage of the system thus developed is that the presence of the bulk of the activity in the medium would make the downstream processing of this protein cheaper since the recovery of protein by an energy-intensive cell disruption process can be avoided.

Acknowledgements This study was supported by a ¢nancial grant (No. SP/ SO/D-68/96) from the Department of Science and Technology, Government of India to J.K.D.

3.4. Expression of streptokinase fusion protein in C. acetoacidophilum Streptokinase is a thrombolytic drug extensively used as e¡ective therapy for improving survival and preserving left ventricular function [23]. High-level expression of streptokinase has been achieved in various laboratories using E. coli as host [24]. At the industrial level also recombinant E. coli is used to produce streptokinase. Also, there are several reports of recombinant streptokinase produced in B. subtilis [25]. It was demonstrated by Klessen et al. [26] that the HincII fragment of streptokinase is functionally active. Since GST was mostly secreted into the medium by C. acetoacidophilum, it would be of interest to express streptokinase as a GST fusion product to check if the fusion protein is also e⁄ciently secreted. We used the HincII fragment of the streptokinase gene for the construction of streptokinase fusion vector. The fragment was retrieved from streptokinase gene cloned in an E. coli plasmid pUCsk, which contains a 2.5 kb PstI fragment bearing the streptokinase gene cloned at the single PstI site of pUC19. The plasmid was digested with BanI which forms several small fragments and one large fragment. The latter was puri¢ed, digested with HincII and the fragment carrying the streptokinase gene was cloned at the SmaI site of plasmid pBKGEXm2. The resulting vector pBKGEXm2sk (Fig. 4) was introduced into C. acetoacidophilum cells by protoplast transformation and clones were selected for kanamycin resistance. The positive clones were identi¢ed by the assay based on clear zones formed around the colonies on plasminogen^milk agar plates. The determination of streptokinase activities in the intracellular extract of the cells and in the culture medium showed that more than 90% of the activity was secreted out into the medium (Table 2). The ratio of extracellular to intracellular streptokinase activities was higher than that observed with GST. The GST^streptokinase fusion protein thus synthesized is amenable to puri¢cation by a⁄nity chromatogra-

References [1] Ichikawa, Y., Yamagata, H., Tochikubo, K. and Udaka, S. (1993) Very e⁄cient extracellular production of cholera toxin B subunit using Bacillus brevis. FEMS Microbiol. Lett. 111, 219^224. [2] Wells, J.M., Wilson, P.W., Norton, P.M., Gasson, M.J. and Le Page, R.W.F. (1993) Lactococcus lactis : High level expression of tetanus toxin fragment C and protection against challenge. Mol. Microbiol. 8, 1155^1162. [3] Steidler, L., Wells, J.M., Raemaekers, A., Vandekerckhove, J., Fiers, W. and Remaut, E. (1995) Secretion of biologically active murine interleukin-2 by Lactococcus lactis subspecies lactis. Appl. Environ. Microbiol. 61, 1627^1629. [4] Cadenas, R.F., Fernandez-Gonzalez, C., Martin, J.F. and Gil, J.A. (1996) Construction of new cloning vectors for Brevibacterium lactofermentum. FEMS Microbiol. Lett. 137, 63^68. [5] Yoshihama, M., Higashiro, K., Rao, E.A., Akedo, M., Shanabruch, W.G., Follettie, M.T., Walker, G.C. and Sinskey, A.J. (1985) Cloning vector systems for Corynebacterium glutamicum. J. Bacteriol. 162, 591^597. [6] Martin, J.F., Mateos, L.M. and Gil, J.A. (1987) Cloning systems of amino acid producing corynebacteria. Bio/Technology 5, 137^146. [7] Deb, J.K. and Nath, N. (1999) Plasmids of corynebacteria. FEMS Microbiol. Lett. 175, 11^20. [8] Eikmanns, B.J., Kleinertz, E., Liebl, W. and Sahm, H. (1991) A family of Corynebacterium glutamicum^Escherichia coli shuttle vectors for cloning, controlled gene expression, and promoter probing. Gene 102, 93^98. [9] Eikmanns, B.J. (1992) Identi¢cation, sequence analysis, and expression of a Corynebacterium glutamicum gene cluster encoding the three glycolytic enzymes glyceraldehyde-3-phosphate dehydrogenase, 3-phosphoglycerate kinase, and triosephosphate isomerase. J. Bacteriol. 174, 6076^6086. [10] Billman-Jacobe, H., Wang, L., Kortt, A., Stewert, D. and Radford, A. (1995) Expression and secretion of heterologous proteases by Corynebacterium glutamicum. Appl. Environ. Microbiol. 61, 1610^1613. [11] Salim, K., Haedens, V., Content, J., Leblon, G. and Huygen, K. (1997) Heterologous expression of the Mycobacterium tuberculosis gene encoding antigen 85A in Corynebacterium glutamicum. Appl. Environ. Microbiol. 63, 4392^4400.

FEMSLE 10528 26-6-02

216

P. Srivastava, J.K. Deb / FEMS Microbiology Letters 212 (2002) 209^216

[12] Morinaga, Y., Tsuchiya, M., Miwa, K. and Sano, K. (1987) Expression of E. coli promoters in Brevibacterium lactofermentum using the shuttle vector pEB003. J. Biotechnol. 5, 305^312. [13] Patek, M., Eikmanns, B.J., Patek, J. and Sahm, J. (1996) Promoters of Corynebacterium glutamicum : cloning, molecular analysis and search for a consensus motif. Microbiology 142, 1297^1309. [14] Mukherjee, K.J., Deb, J.K. and Ramachandran, K.B. (1990) Construction of vector of Brevibacterium lactofermentum and study of its stability in continuous culture. J. Biotechnol. 16, 109^122. [15] Mukherjee, K.J., Deb, J.K. and Ramachandran, K.B. (1996) Studies on maintenance of vector pBK2 in Corynebacterium acetoacidophilum by coculture experiments in continuous culture. Enzyme Microb. Tech. 18, 555^560. [16] Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. [17] Meister, A. (1985) Glutamate, glutamine, glutathione and related compounds. Methods Enzymol. 113, 501^501. [18] Castellino, F.J., Sodetz, J.M., Brockway, W.J. and Siefring Jr., G.E. (1976) Streptokinase. Methods Enzymol. 45, 244^257. [19] Laemmli, U.K. (1990) Cleavage of structural proteins during the assembly of the head of the bacteriophage T4. Nature 227, 680^ 685.

[20] Jakoby, M., Carole-Estelle, N.-N. and Burkovski, A. (1999) Construction and application of new Corynebacterium glutamicum vectors. Biotechnol. Tech. 4, 437^441. [21] Ben-Samoun, K., Leblon, G. and Reyes, O. (1999) Positively regulated expression of the Escherichia coli araBAD promoter in Corynebacterium glutamicum. FEMS Microbiol. Lett. 174, 125^130. [22] Sun, J., Kelemen, G.H., Fernandez-Abalos, J.M. and Bibb, M.J. (1999) Green £uorescent protein as a reporter for spatial and temporal gene expression in Streptomyces coelicolor A3(2). Microbiology 145, 2221^2227. [23] White, H.D., Rivers, J.T. and Maslowsky, A.H. (1989) E¡ect of intravenous streptokinase as compared with that of tissue plasminogen activator on left ventricular function after ¢rst myocardial infarction. N. Engl. J. Med. 320, 817^821. [24] Estrada, M.P., Herna¤ndez, L. and Pe¤rez, A. (1992) High level expression of streptokinase in Escherichia coli. Bio/Technology 10, 1138^ 1142. [25] Wong, S.-L., Ye, R. and Nathoo, S. (1994) Engineering and production of streptokinase in a Bacillus subtilis expression^secretion system. Appl. Environ. Microbiol. 60, 517^523. [26] Klessen, C., Schmidt, H., Ferreti, J. and Malke, H. (1988) Tripartite streptokinase gene fusion vectors for Gram-positive and Gram-negative prokaryotes. Mol. Gen. Genet. 212, 295^300.

FEMSLE 10528 26-6-02