APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Feb. 2000, p. 825–827 0099-2240/00/$04.00⫹0 Copyright © 2000, American Society for Microbiology. All Rights Reserved.
Vol. 66, No. 2
Cloning and Sequencing of an Alkaline Protease Gene from Bacillus lentus and Amplification of the Gene on the B. lentus Chromosome by an Improved Technique PER LINÅ JØRGENSEN,1 MARTIN TANGNEY,2 POUL ERIK PEDERSEN,1 SVEN HASTRUP,1 BØRGE DIDERICHSEN,1 AND STEEN T. JØRGENSEN1* Bacterial Gene Technology, Novo Nordisk A/S, 2880 Bagsværd, Copenhagen, Denmark,1 and Department of Biological Sciences, Heriot-Watt University, Riccarton, Edinburgh EH14 4AS, Scotland, United Kingdom2 Received 25 August 1999/Accepted 19 November 1999
A gene encoding an alkaline protease was cloned from an alkalophilic bacillus, and its nucleotide sequence was determined. The cloned gene was used to increase the copy number of the protease gene on the chromosome by an improved gene amplification technique. Members of the genus Bacillus are widely used in industry in the large-scale production of enzymes, such as proteases (4). Of particular industrial importance are proteases with activity at alkaline pH and high temperature. The best-known application for these alkaline proteases is their use in household detergents. Genes encoding alkaline protease have been cloned and sequenced from a variety of organisms, some of which are used commercially. To maximize industrial enzyme production, the host organism is generally manipulated to carry multiple copies of the gene. This can be accomplished by cloning the gene on a replicating plasmid, but preferentially this is achieved by amplification of the gene on the chromosome, as such amplification offers the more stable alternative. With an organism such as Bacillus subtilis, which can easily be transformed and has well-developed genetics, it is relatively straightforward to amplify genes on a chromosome; however, other Bacillus species frequently used in industry are more difficult to transform and less amenable to genetic manipulation. It is possible to use temperature-sensitive plasmids, such as pE194, as a means of achieving gene amplification on a chromosome; however, the incorporation of the plasmid origin of replication in the host chromosome is known to destabilize amplified structures (8, 13). There is therefore an interest in the development of improved amplification methods for industrial bacilli. The alkalophilic species Bacillus lentus is an important industrial organism which produces a commercially important alkaline protease. In this report we describe the cloning and sequencing of the gene encoding an alkaline protease from the B. lentus isolate NCIB 10309 and the generation of a strain which contains multiple copies of this gene on the chromosome. In the process of generating such a strain we have developed a general method for gene amplification in poorly transformable bacteria. Cloning of a B. lentus alkaline protease gene. Chromosomal DNA from B. lentus NCIB 10309 (6) was digested with the enzyme Sau3A, and fragments between 1.5 and 6.5 kb were purified and ligated into a general-purpose cloning vector, pSX50, developed at our laboratories (S. Hastrup, S. Branner, F. Norris, S. B. Petersen, L. Nørskov-Lauridsen, V. J. Jensen,
and D. Aaslyng, 13 July 1989, International patent publication number WO89/06279), which had been digested with the enzyme BamHI. Ligated DNA was transformed into the protease-deficient mutant B. subtilis DN497 (S. T. Jørgensen, P. L. Jørgensen, and B. Diderichsen, 27 June 1991, International patent publication number WO91/09129) and transformants were selected on Luria-Bertani agar plates supplemented with chloramphenicol, 1% skim milk, and 0.2% xylose. Proteaseproducing transformants, which arose at a frequency of 10⫺4, were identified by a clear halo around a colony. Two clones that carried the gene for subtilisin were isolated. The gene encodes a protein of 380 amino acids, consisting of a signal peptide, a propeptide, and a mature protein, in common with other alkaline proteases. The mature protein contains 269 amino acids and is currently marketed commercially under the trade name Savinase. A detailed description of its properties and its crystal structure have been presented elsewhere (2, 9, 11). The protein sequence has been deposited in the SwissProt database with the accession number P29600. A new strategy for integration and gene amplification. To increase production of the protease, it was desirable to amplify the gene on the B. lentus chromosome. Amplified sequences on the B. subtilis chromosome have been shown to be destabilized by the presence of an active plasmid origin of replication in the integrated DNA (8, 13). We previously developed a strategy which avoids the presence of a functional plasmid replication system on the chromosome (7, 10). The amplification method is dependant upon the construction of a pUB110-derived plasmid incorporating two critically located plus origins of replication (⫹ori). Such plasmids have the capacity to form two separate progeny vectors, which we term the replicative and nonreplicative vectors, as described in the work of Tangney et al. (10). Each of the progeny vectors necessarily contains its own ⫹ori, but by design only one of these progeny vectors (the replicative vector) encodes the trans-acting replication protein. As a consequence, the nonreplicative vector can be maintained only in a culture where either the replicative vector is also present (to supply the essential replication protein in trans) or it has integrated into the host chromosome and is replicated along with the chromosomal DNA. Therefore, by screening for loss of the replicative vector it is possible to identify clones where the nonreplicative vector DNA has integrated into the chromosome. In practice, the parental plasmid is designed so that the nonreplicative vector contains the DNA sequence with
* Corresponding author. Mailing address: Bacterial Gene Technology, Novo Nordisk A/S, 2880 Bagsværd, Copenhagen, Denmark. Phone: 45 4442 2809. Fax: 45 4442 7303. E-mail:
[email protected]. 825
826
JØRGENSEN ET AL.
FIG. 1. Construction of the integrating plasmid pPL2002. Abbreviations are as follows: bla, the -lactamase gene; cat, the chloramphenicol acetyltransferase gene; ⫹ori-pE194, the ⫹ori of pE194; repF, the gene encoding the replication protein of pE194; ⌬repF, a deleted repF gene; erm, the erythromycin resistance gene; and S-309, the subtilisin gene from B. lentus NCIB 10309. Arrows indicate the directions of transcription.
APPL. ENVIRON. MICROBIOL.
homology to the host chromosome that is the target for integration into the chromosome, by means of homologous recombination. As this vector cannot replicate autonomously, any subsequently derived amplified structures will not contain an active origin of plasmid replication. While this method is useful, it may be further optimized. It involves the construction of large complex plasmids that may be difficult to introduce into poorly transformable species, such as most of the industrially significant Bacillus species (including B. lentus). Furthermore, the technique necessitates screening for the spontaneous loss of a plasmid moiety from a cell in which the plasmid can naturally replicate. To circumvent these problems, we considered the use of an alternative plasmid, pE194 (Fig. 1). Like pUB110, this plasmid replicates by the rolling-circle method of replication (3), which involves a replication protein that functions in trans at an origin of replication (⫹ori). Plasmid pE194 is temperature sensitive for replication; therefore, growth at elevated temperatures (e.g., 45°C) allows for the simple identification of cells which have lost the replicating plasmid. This property confers a crucial advantage over those of the pUB110-based amplification system. We envisaged that a plasmid construct which itself cannot replicate in Bacillus but contains the ⫹ori should be maintained at the permissive temperature when it is transformed into a recipient cell which already contains resident pE194, providing the necessary trans-acting replication protein, as depicted in Fig. 2A. If this was the case, then a construct which contains the ⫹ori from pE194, as well as a resistance marker and a region of homology to the chromosome (as depicted in Fig. 2A), should be readily propagated in the presence of pE194 at the permissive temperature but not at the nonpermissive temperature. Growth at the elevated temperature in the presence of the appropriate antibiotic should therefore facilitate the simple isolation of clones wherein the described construct has integrated into the chromosome by means of homologous recombination between the vector and the chromosome, as depicted in Fig. 2B. Such a chromosome would contain an amplification unit, thereby facilitating gene amplification.
FIG. 2. (A) Depiction of plasmids pPL2002 and pE194 when replicating in a Bacillus genetic background. The black rectangle represents a functional RepF protein (the product of the intact repF gene), which can act at both ⫹ori pE194 sequences as indicated by the dashed lines. (B) Integration of pPL2002 into the B. lentus chromosome via the gene for subtilisin. Abbreviations are given in the legend to Fig. 1.
VOL. 66, 2000
GENE AMPLIFICATION IN AN ALKALOPHILIC BACILLUS
Amplification of the alkaline protease gene on the B. lentus chromosome. We tested the efficacy of the new approach by amplifying the alkaline protease gene of B. lentus. Plasmid pPL2002 was constructed as shown in Fig. 1. An MboI fragment of pE194, from positions 1 to 1585 (6), containing the ⫹ori of pE194 and a segment of the gene encoding the replication protein RepF was cloned, in Escherichia coli, into pDN3000 (5) at the unique BglII site, resulting in plasmid pPL1957. Plasmid pSX120 contains a chloramphenicol resistance gene and the alkaline protease gene from B. lentus on a 3.3-kb EcoRI/BamHI fragment. This fragment was cloned into pPL1957 to generate plasmid pPL2002 (Fig. 1). This plasmid therefore contains the ampicillin resistance marker and functions for replication in E. coli; it also contains the cloned alkaline protease gene, a cat gene (which is functional in a Bacillus genetic background), the ⫹ori from pE194, and a truncated repF gene. Plasmid pPL2002 was constructed such that it does not produce an active RepF protein and consequently cannot replicate autonomously in a Bacillus genetic background. However, it does possess a ⫹ori sequence from pE194 and therefore should be maintained in a cell where a functional RepF protein is provided in trans. Accordingly, B. lentus was transformed by protoplast transformation (1) with pE194, and erythromycinresistant transformants were regenerated at the replication permissive temperature of 30°C. One such transformant was designated PL2156. This strain was subsequently transformed with plasmid pPL2002 by selecting for erythromycin- and chloramphenicol-resistant transformants at 30°C. One such transformant was designated PL2157. Strain PL2157 was grown overnight at 30°C. Dilutions of this culture were subsequently plated out onto chloramphenicol plates, and the plates were incubated at 45°C. Chloramphenicol-resistant colonies arising at this temperature can be a result of homologous recombination between the plasmid-borne and chromosomal copies of the alkaline protease gene, as depicted in Fig. 2B. One such colony was designated PL2158. There was no evidence of pE194 in this strain. Southern hybridization analysis confirmed the integration of pPL2002 and revealed that spontaneous amplification (to approximately four copies), had occurred (data not shown). Amplified strains also had elevated protease levels. The stability of the chromosomally integrated copies of pPL2002 in strain PL2158 was assessed in the absence of antibiotic selection in large-scale industrial fermentations (1,500 liters). Following fermentation (8 to 10 generations), culture samples were plated out for single colonies on nonselective plates and subsequently replica plated onto plates containing chloramphenicol. Of 100 colonies, 98 were found to be still resistant to chloramphenicol. Southern hybridization analysis of 20 of these, selected at random, revealed the presence of pPL2002, apparently at the same copy number in all of the colonies examined (not shown). Conclusions. We report the cloning, sequencing, and specific amplification of a protease gene on the chromosome of an alkalophilic bacillus. The strategy employed to generate the amplified strain is derived from a method which we previously developed using the plasmid pUB110 that precludes the presence of an active plasmid origin of replication on the amplified chromosome. In the new approach we adapted the earlier method for use with the temperature-sensitive plasmid pE194. The new strategy simplifies the plasmid constructions but perhaps even more significantly allows a direct screen for clones in
827
which integration has occurred, by virtue of the temperaturesensitive nature of the plasmid. Strains in which integration has occurred are readily isolated by growth at the nonpermissive temperature in the presence of a selective antibiotic. Such strains can subsequently be used to derive clones where further gene amplification has occurred. While in the example presented here we detected spontaneous amplification of the alkaline protease gene on the B. lentus chromosome, it is also possible to screen for clones with multiple copies by successive subculturing of amplified strains in increasing concentrations of the relevant antibiotic, as described for other methods (10, 12). The new strategy for gene amplification in poorly transformable bacteria represents a significant advance on current technology and should be applicable to any organism in which pE194 can replicate. The method has already been successfully exploited in our laboratories at Novo Nordisk (Copenhagen, Denmark) for a number of Bacillus species (unpublished data). Furthermore, that we could readily adapt the technique from pUB110 to pE194 demonstrates the general nature of the technique and suggests that it should be adaptable not only for other rolling-circle plasmids but indeed for any plasmid where there is a requirement for a trans-acting replication factor. Apart from industrial applications of the technique, it may also be of use in the functional analyses of the many organisms whose genomes are currently being sequenced. The figures are reproduced by permission of Novo Nordisk A/S. REFERENCES 1. Akamatzu, T., and J. Sekiguchi. 1984. An improved method of protoplast regeneration for Bacillus species and its application to protoplast fusion and transformation. Agric. Biol. Chem. 48:651–655. 2. Betzel, C., S. Klupsch, S. Branner, and K. S. Wilson. 1996. Crystal structures of the alkaline proteases Savinase and Esperase from Bacillus lentus, p. 49–61. In R. Bott and C. Betzel (ed.), Subtilisin enzymes: practical protein engineering. Plenum Press, New York, N.Y. 3. Bron, S. 1990. Plasmids, p. 75–175. In C. R. Harwood and S. M. Cutting (ed.), Molecular biological methods for Bacillus. J. Wiley & Sons, Ltd., London, United Kingdom. 4. de Boer, A. S., F. G. Priest, and B. Diderichsen. 1994. On the industrial use of Bacillus licheniformis—a review. Appl. Microbiol. Biotechnol. 40:595–598. 5. Diderichsen, B., U. Wedsted, L. Hedegaard, B. R. Jensen, and C. Sjøholm. 1990. Cloning of aldB, which encodes ␣-acetolactate decarboxylase, an exoenzyme from Bacillus brevis. J. Bacteriol. 172:4315–4321. 6. Horinouchi, S., and B. Weisblum. 1982. Nucleotide-sequence and functional map of pE194 a plasmid that specifies inducible resistance to macrolide lincosamide and streptogramin type-b antibiotics. J. Bacteriol. 150:804–814. 7. Jørgensen, S. T., M. Tangney, P. L. Jørgensen, and B. Diderichsen. 1998. Integration and amplification of a cyclodextrin glycosyltransferase gene from Thermoanaerobacter sp. ATCC 53627 on the Bacillus subtilis chromosome. Biotechnol. Tech. 12:15–19. 8. Noirot, P., M. A. Petit, and S. D. Ehrlich. 1987. Plasmid replication stimulates DNA recombination in Bacillus subtilis. J. Mol. Biol. 196:39–48. 9. Outtrup, H., and C. O. L. Boyce. 1990. Microbial proteinases and biotechnology, p. 227–254. In W. M. Fogarty and C. T. Kelly (ed.), Microbial enzymes and biotechnology, 2nd ed. Elsevier Applied Science, London, United Kingdom. 10. Tangney, M., P. L. Jørgensen, B. Diderichsen, and S. T. Jørgensen. 1995. A new method for integration and stable DNA amplification in poorly transformable bacilli. FEMS Microbiol. Lett. 125:107–114. 11. von der Osten, C., S. Branner, S. Hastrup, L. Hedegaard, M. D. Rasmussen, H. Bisgaard-Frantzen, S. Carlsen, and J. M. Mikkelsen. 1993. Protein engineering of subtilisins to improve stability in detergent formulations. J. Biotechnol. 28:55–68. 12. Young, M., and D. Hranueli. 1988. Chromosomal gene amplification in gram-positive bacteria, p. 157–200. In J. A. Thomson (ed.), Recombinant DNA and bacterial fermentation. CRC Press, Inc., Boca Raton, Fla. 13. Young, M., and S. D. Ehrlich. 1989. Stability of reiterated sequences in the Bacillus subtilis chromosome. J. Bacteriol. 171:2653–2656.
