Cloning, Characterization and Expression in ... - Semantic Scholar
Journal of General Microbiology (1987), 133, 317-322.
Printed in Great Britain
317
Cloning, Characterization and Expression in Escherichia coli of a Leucine Biosynthetic Gene from Streptomyces rochei By J U N E H E R C O M B , ’ G E O R G T H I E R B A C H , 2 S I M O N B A U M B E R G 3 J . H. PARISH1* Department of Biochemistry, University of Leeds, b e d s LS2 9JT, UK Asta- Werke A G, Forschung Chemie, Organisch-Biotechnologie,Kantstrasse 2, 0-4802 Halle-Kiinsebeck, FRG 3Department of Genetics, University of Leeds, Leeds LS2 9JT, UK
AND
(Received 29 May 1986; revised 10 September 1986)
Leucine and histidine biosynthetic genes from Streptomyces rochei HPl that complemented auxotrophic mutations in S. lividans TK54 were cloned in pIJ61. DNA from one leucine recombinant plasmid was subcloned into pBR322. From the latter, a recombinant plasmid was obtained that complemented the leuA mutation in Escherichia coli CV5 12 but not other leucine markers in E. coli. Analysis of this and several subclones, including mutant plasmids constructed in uitro, established that the cloned S. rochei gene was expressed in E. cofi from the tetracycline promoter of pBR322 to produce a polypeptide of 67 kDa; the corresponding coding region was shown to be within a 1.7 kbp DNA fragment. Blot hybridization revealed corresponding homologous genes in several other streptomycetes.
INTRODUCTION
Several streptomyces genes have been cloned and expressed in streptomyces (Bibb et al., 1983). There are fewer examples of expression of cloned streptomyces genes in Escherichia coli: these include genes for antibiotic resistance (e.g. Vara et al., 1985; Gil et al., 1985; Schupp et al., 1983), a gene for an extracellular enzyme (Robbins et al., 1981) and a few examples of genes for biosynthetic enzymes. Of these, the cloned p-aminobenzoic acid synthetase gene appears to be involved in candicidin biosynthesis (Gil & Hopwood, 1983). There is one previously published example of gene for an enzyme from an amino acid biosynthetic pathway, argG, and this furnishes one of the very few known examples of expression in E. coli from a streptomyces promoter (Meade, 1985); more generally, it is to be anticipated that expression from such promoters will not be generally applicable to the characterization of biosynthetic genes from streptomycetes. Our studies on amino acid biosynthesis in streptomycetes arose from work on a newly isolated organism, strain HP1 (see Results) and attempts, so far unsuccessful, to clone its genes for cellulolytic enzymes. We were concerned to establish whether HP1 genes of any kind might be expressed in the cloning host, Streptomyces Iividans and in E. coli. Amino acid biosynthetic markers were chosen and, of the clones identified in S. lividans, leucine was selected for further study because there are relatively few ‘dedicated’ leucine biosynthetic enzymes that might be candidates for possible complementation experiments. Biosynthesis of leucine in most bacteria involves three steps (referred to collectively as the isopropylmalate pathway) unique to leucine biosynthesis. In enteric bacteria, the genes for the corresponding enzymes are clustered in a single operon. The products of the genes (in E. coli) are or-isopropylmalate synthase (EC 4.1.3.12) (IeuA), P-isopropylmalate dehydratase (EC 4.2.1.33) (IeuC, IeuD) and P-isopropylmalate dehydrogenase (EC 1.1.1.85) (leuB) (Calvo, 1983). 0001-3494 0 1987 SGM
318
J . HERCOMB A N D OTHERS
METHODS Bacterial strains and media. Streptomyces rochei HP1 was isolated from Jamaican laterite soil by the methods of Mullings & Parish (1984). Strains and plasmids used in the present work are listed in Table 1. For growth of streptomycetes, R2YE agar and MM agar are described by Hopwood et al. (1985). MM agar was supplemented (when necessary) with leucine (50 mg 1-l) and histidine (75 mg 1-l). YEME medium was as described by Hopwood et al. (1985) except that the glycine concentration was 0.1 % (w/v) for experiments in which protoplasts were to be isolated and was omitted in other cases. E. coli was grown using L broth and L agar (Maniatis et al., 1982) and M9 medium and M9 agar (Miller, 1972), supplemented when necessary with leucine and/or proline (40 mg 1-l). Antibiotics were added as filter-sterilized concentrates (in water, except for thiostrepton, which was dissolved in dimethyl sulphoxide) to the following final concentrations (mg 1-l) : thiostrepton, 50 : neomycin, 10: ampicillin, 50: tetracycline, 12.5. Bacterial growth. E. coli strains were incubated on plates at 37 "C and grown in shake culture at 37 "C. Streptomycetes were grown similarly at 30 "C. Enzymes and reagents. Molecular biological reagents were purchased from BRL, Boehringer and New England Biolabs. Other reagents and antibiotics were obtained from various laboratory suppliers; thiostrepton was a generous gift from Mr S. J. Lucania of E. J. Squibb and Sons, New Brunswick, NJ, USA. Plasmid isolation,restriction, ligation, transformation,electrophoresis,blot hybridization and plasmid reconstruction. Plasmid pBR322 was isolated from E. coli JA221; pIJ61 was isolated from S . lividans 1326. General molecular biological methods and experiments specific to E. coli with the exceptions noted in the following paragraph were according to Maniatis et al. (1982); all manipulations involving streptomycetes were according to Hopwood er al. (1985). Transformants acquiring pIJ61 and its derivatives were detected by replica plating the pocks on to antibiotic-selective minimal or supplemented plates as described in the reference. Electroelution of restriction fragments from gels was by the direct method of Maniatis et al. (1982). Analysis of protein synthesis by using E. coli minicells was based on the methods of Hallewell & Sherratt (1976). Minicell strains were grown to stationary phase in 500 ml L broth and harvested by centrifugation at 4 "C for 10 min at 12000 r.p.m. The pellets were resuspended in 4 ml M9 medium and minicells were further purified by sedimentation through 40 ml linear gradients of 5-20% (w/v) sucrose in M9 medium, for 20 min at 3000 r.p.m. Minicells were recovered, washed, resuspended in M9 medium and purified by two further centrifugations through sucrose gradients. After washing, the minicells were resuspended in 0.5 ml M9 medium supplemented with glycerol (20%, v/v). Minicells were labelled in M9 medium supplemented with cycloserine (100 pg ml-l). Minicells from a portion (0.25 ml) of glycerol suspension were washed twice by centrifugation and resuspended in 0.1 ml medium. After 1 h at 37 "C, 25 pCi (925 kBq) [35S]methionine(0.5-1 Ci pmol-l) was added. After 30 min, the minicells were harvested by centrifugation, resuspended in SDS-running buffer (50 pl) and (after heating and removal of residues) the solution was analysed by SDS electrophoresis using 4C-labelled standard proteins as markers. RESULTS
Bacterium H P l Bacterium HP 1 was isolated as a cellulose-degrading streptomycete. A fuller description is being prepared for publication elsewhere. The organism has no unusual nutritional requirements, is aerobic and grows at temperatures between 20 and 45°C. Analysis of key taxonomic characters determined by Dr E. M. H. Wellington (Williams & Wellington, 1982; Williams et al., 1983) suggested that it is a representative of the species S. rochei. Isolation of plJ61 leucine and histidine recombinants A ligation mixture of a partial Sau3A digest of S . rochei HP1 DNA and a complete BamHI digest of pIJ61 DNA was used to transform S. liuidans TK54. The pocks were almost confluent on a total of 20 plates. We estimated that the total number of recombinants was approximately 20000. Of these, 27 were scored as apparent histidine recombinants and three as apparent leucine recombinants. Plasmids isolated from each of 16 His+ recombinants tested generated the His+ phenotype on transformation. Of the three Leu+ recombinants, one was rejected on the grounds that it was apparently a chromosomal revertant or recombinant (it proved to be neomycin resistant) and of the remaining two only one, pGT29, conferred the Leu+ phenotype on retransformation. Restriction enzyme analysis of pGT29 and the His+ recombinants revealed inserts of between 5 and 10 kbp (data not shown).
Cloning of a Streptomyces rochei leucine gene
319
Table 1. Bacterial strains and plasmids Relevant characters Streptomyces strains S . coelicolor M130 S . glaucescens NCIB 9619 S . griseus ATCC 12475 S . lividans TK54 S . lividans 1326 S . rochei HPl Streptoverticillium verticillium JC N 4924 E. coli strains CV5 12 CV520 CV526 DS4 10 HBlOl JA221 E. coli plasmids pBR322 pJHl pJH20 pJH 107 pJH108 pJHl1 pJHl11 pJH211 pJH311 pJH411 Streptomyces plasmids pIJ61 pGT29
leu2 his2
Source or reference D. A. Hopwood NCIB ATCC D. A. Hopwood D. A. Hopwood This work Riken, Japan
F+ leuA371 F+ leuCI71 F+ leuDlOl
Somers et al. (1973) Somers et al. (1973) Somers et al. (1973) Dougan & Sherratt (1977) Maniatis et al. (1982) Laboratory strain
amp amp amp amp amp amp amp amp amp amp
Maniatis et al. (1982) This work This work This work This work This work This work This work This work This work
minA minB rpsL leuB proA2 tet Leu+* Leu-* Leu+* Leu+* Leu+* Leu-* Leu-* Leu-* Leu-*
neo tsr ltz tsr ltz; pIJ61-HP1 recombinant: complements leu2 in S. liuidans
D. A. Hopwood This work
* These are derivatives of pGT29-pBR322 recombinants (see Results); the Leu phenotype refers to the ability/inability of the plasmid to complement the leuA mutation in E. coli CV512. Construction, analysis and phenotypic characterization of pBR322-pGT29 recombinants A partial Sau3A digest of pGT29 DNA was ligated with a complete BamHI digest of pBR322 DNA and was used to transform E. coli strains HBlOl (IeuB) and CV512 (IeuA). Leu+ colonies were detected only from the transformation of strain CV512. Plasmid pJHl was isolated from one of these recombinants. Retransformation experiments with this confirmed that it was able to complement the leuA mutation in strain CV512 but it was unable to complement IeuB, leuCor leuD mutations in the strains tested (Table 1). Sub-clones and deletion mutants derived from pJHl were constructed and tested for the capacity to complement the leuA mutation (Fig. 1). The physical maps of these plasmids and of pGT29 were obtained from single and double restriction digests. Details of the pGT29 map were confirmed by Southern blot hybridization of several single digests of the plasmid using labelled pJHl DNA as a probe. Sequences derived from vectors (pBR322 and pIJ61) were identified from the published restriction maps for these plasmids (Maniatis et al., 1982; Hopwood et al., 1985). From an alignment of thee maps (Fig. I), we conclude that deletions of pBR322 sequences occurred in rho in the experiments that generated pJHlO7 and pJH108. The ClaI site of pBR322, which lies within the tet promoter, was removed (by end filling) from pJHll to generate pJHl11. Five plasmids with the same properties as pJHl11 were isolated independently. The absence of the Cla site was confirmed by restriction analysis and the transformants were Leu-, thus establishing that the streptomyces gene was expressed from the tet promoter in the Leu+ pJH plasmids. The validity of the method was confirmed by analysis of a tetracycline-sensitive pBR322 derivative constructed by the same method, and the inability of
320
J . HERCOMB AND OTHERS
ti
(i
I
I
V V I
1
P I
B G
G B P B V P I
I
I
I l l
BVP
V
PECHP
G BP
V
PECHP
G B
V
PECHP
G B P
V
P ECHP
G BP B
I
V
B
P
V H E
H
I
I
I
I
I
1
G
B
V
B
P
V
I
I
H
1 pGT29 (+)
V
i PJHl(+) P
V
3 ' pJH20(-) PG BPBV
2 pJH107(+) PBV
.--
V
111
P ECHP
I
pJHlO8(+)
II
G B P BV
pJHll (+)
V
P ECHP
V
P ECHP
G B
BV
I pJH211 (-) ' -
--------
P BV
pJH411 (-)
I kbp
H
Fig. 1. Restriction endonuclease maps of recombinant plasmids encoding the S . rochei Leu gene and of certain subclones. The double arrow under the map of pJHl1 represents the proposed limits of the Leu gene; sites i, ii and iii are referred to in the text. The leucine phenotype conferred by each plasmid is shown by or - in parentheses against the plasmid name. I pIJ61 , DNA;=, pBR322 DNA; -, cloned HP1 DNA; ----, deleted regions. Restriction sites are denoted: B, BurnHI; C, ClaI; E, EcoRI; G, BgnI; H, HindIII; P, PstI; V, PvuII. The origins of plasmids pGT29 and pJHll are described in the text. pJH20 was obtained by religation of a BarnHI digest of pJH1; pJH107 and pJHlO8 were obtained by cloning a partial Sau3A digest of pJHl into pBR322 (deletions of pBR322 sequences were generated in these constructions). pJHl1 was generated by religation of a PvuII digest of pJHl. The remaining three plasmids were all derived from pJHl1 by deletions generated by BurnHI (pJH21 l), BglII-Hind111 (pJH311) and partial digestion with Psi1 (pJH411).
+
p J H l l l to synthesize the leucine biosynthetic enzyme was confirmed by studies on the translation products from the plasmid (see below). Protein synthesis due to expression of plasmid genes was measured in E . coliminicells (Fig. 2). In addition to the pBR322 proteins, pJHl and pJHl1 direct the synthesis of a polypeptide (p67) of 67 kDa. This is presumptively the product of the leucine biosynthetic gene. The protein was greatly reduced in amount in the mutant, pJHl11, in which the tet promoter was inactivated. Protein p67 was replaced by a truncated protein of 55 kDa in mutant pJH211 and by two proteins of approximately 46 kDa in pJH311. No protein corresponding to p67 was observed with pJH411. The coding sequence corresponding to a polypeptide of the size of p67 is approximately 1.7 kbp; the corresponding values for the two truncated proteins are 1.4 kbp and 1.1 kbp. From these data and the maps and phenotypes of the plasmids of Table 1, we deduced the limits within which the gene must lie (see Discussion). IdentiJicationof sequences homologous to the cloned leucine gene in other streptomycetes The internal PstI-Psi1 fragment from pJHll that contains the leucine gene was labelled in vitro and used as a hybridization probe for genomic DNA from several streptomycetes digested with PstI. As a control, genomic DNA from S. rochei HPl was also digested with BamHI to confirm the size of the homologous region from the source organism. Homologous fragments were discovered in all the strains tested, i.e. the seven StreptomyceslStreptoverticillium strains listed in Table 1. Moreover the PstI fragments were all similar in size except in the cases of S.
Cloning of a Streptomyces rochei leucine gene
32 1
Fig. 2. SDS-PAGE analysis of 35S-labelledproteins synthesized in minicells containing plasmids. The results in the two photographs were obtained in separate experiments and are therefore calibrated separately.
coelicolor and Stu. uerticillium, in which the complementary fragments were larger. The size of the corresponding fragment in S . liuidans could not be deduced because the DNA was not completely digested. There is thus qualitative evidence that the sequence is not unique to S . rochei HPl . DISCUSSION
The complementation of the leu2 mutation by a gene that also complements leuA in E . coli shows that the leu2 gene of S . liuidans is a structural gene for a-isopropylmalate synthase. It is, therefore, the first ‘dedicated’ gene of the leucine biosynthetic pathway in the organism. As pJHl does not complement E. coli leuB, leuC or leuD mutations, it is probable that the other streptomyces Leu genes are not present in the clone and, indeed, two of the Leu genes in S. coelicolor are separated by approximately half the chromosome (Hopwood et al., 1985). Comparison of the maps and phenotypes of plasmids pJH1, pJH20 and pJHl1 establishes that PstI site i (Fig. 1) lies within the region essential for the Leu+ phenotype and is, therefore, within or immediately proximal to the coding sequence, and the properties of plasmids pJHl1, pJH107 and pJH108 establish that the region distal to the BamHI site ii is not required. The absence of a polypeptide product from pJH411 suggests the start of the 1.7 kbp coding region is distal to the PstI site iii (Fig. 1). We propose that the truncated 1.4 kbp coding region in pJH211 defines the start of the coding region as being between the sites iii and ii. The polypeptide produced by pJH311 can only be accounted for by assuming that it arises adventitiously, for example by the accidental generation of a ribosome-binding site. The estimated molecular mass of the streptomyces isopropylmalate synthase subunit, 67 kDa, is similar to that for the corresponding yeast enzyme (65-67 kDa; Chang et al., 1984) but is substantially larger than that for the salmonella protein (50 kDa; Calvo, 1983). In the light of the homology between the cloned gene and unique sequences from other streptomycetes, we
322
J . HERCOMB A N D O T H E R S
conclude that this size is likely to be true of the enzymes from the streptomycete group in general. Despite the difference in size from the enzyme of enteric bacteria, the complementation data show that the streptomyces enzyme functions in E. coli. The method we have described should be applicable to other amino acid biosynthetic genes from streptomycetes. We are grateful to Dr E. M. H. Wellington for the taxonomic assignment of strain HP1, to Dr M. J. Bibb and Professor D. A. Hopwood for help and advice and to Mr K. Ainley for excellent technical assistance with the studies that led to the isolation of pGT29. The work was supported, in part, by an SERC Research Grant to J. H. P. and S. B., and J. H. is an SERC Research Student. REFERENCES
BIBB,M. J., CHATER, K. F. & HOPWOOD, D. A. (1983). Developments in Streptomyces cloning. In Experimental Manipulation of Gene Expression, pp. 54-80. Edited by M. Inouye. London: Academic Press. CALVO, J. M. (1983). Leucine biosynthesis in prokaryotes. In Amino Acids - Biosynthesis and Genetic Regulation, pp. 267-285. Edited by K. M. Herrmann & R. L. Somerville. Reading, Mass.: AddisonWesley. CHANG,L. L.-F., CUNNINGHAM, T. A., GATZEK, P. R., CHEN,W.-J. &KOHLHAW, G. B. (1984). Cloning and characterization of yeast leu4, one of two genes responsible for a-isopropylmalate synthesis. Genetics 108, 91-106. DOUGAN, G. & SHERRATT, D. J. (1977). The transposon T n l as a probe for studying ColE1 structure and function. Molecular and General Genetics 151, 151160. GIL, J. A. & HOPWOOD,D. A. (1983). Cloning and expression of a p-aminobenzoic acid synthetase gene of the candicidin-producing Streptomyces griseus. Gene 25, 119-132. GIL, J. A., KIESER,H. M. & HOPWOOD, D. A. (1985). Cloning of a chloramphenicol acetyltransferase gene of Streptomyces acrimycini and its expression in Streptomyces and Escherichia coli. Gene 38, 1-8. HALLEWELL, R. A. & SHERRATT, D. J. (1976). Isolation and characterisation of ColE2 plasmid mutants unable to kill colicin-sensitive cells. Molecular and General Genetics 146, 239-245. HOPWOOD, D. A., BIBB,M. J., CHATER, K. F., KIESER, T., BRUTON,C. J., KIESER,H. M., LYDIATE, D. J., SMITH,C. P., WARD,J. M. & SCHREMPF, H. (1985). Genetic Manipulation of Streptomyces - a Luboratory Manual. Norwich: John Innes Foundation. E. F. & SAMBROOK, J. (1982). MANIATIS, T., FRITSCH, Molecular Cloning: a Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. MEADE,H. (1985). Cloning of argG from Streptomyces:
loss of gene in Arg- mutants of S. cattleya. BiolTechnology 3, 917-918. MILLER,J. H. (editor) (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. MULLINGS,R. & PARISH,J. H. (1984). Mesophilic aerobic Gram negative cellulose degrading bacteria from aquatic habitats and soils. Journal of Applied Bacteriology 57, 455-468. ROBBINS, P. W., WIRTH,D. F. & HERING,C. (1981). Expression of the Streptomyces enzyme endoglycosidase H in Escherichia coli. Journal of Biological Chemistry 256, 10640-1 0644. SCHUPP, T., TOUPET,c . , STALHAMMER-CARLEMALM, M. & MEYER,J. (1983). Expression of a neomycin phosphotransferase gene from Streptomyces fradiae in Escherichia coli after interplasmidic recombination. Molecular and General Genetics 189, 27-33. SOMERS, J. M., AMZALLAG, A. & MIDDLETON, K. B. (1973). Genetic fine structure of the leucine operon of E. coli K12. Journal of Bacteriology 113, 12681272. VARA, J., MALPARTIDA, F., HOPWOOD,D. A. & JIMENEZ, A. (1985). Cloning and expression of a puromycin N-acetyltransferase gene from Strepromyces alboniger in Streptomyces lividans and Escherichia coli. Gene 33, 197-206. S. T. & WELLINGTON, E. M. H. (1982). WILLIAMS, Actinomycetes. In Methods of Soil Analysis, Part 2. Chemical and Microbiological Properties, pp. 969987. Edited by A. L. Page, R. H. Miller & D. R. Keeney. Madison, Wis. : ASA-SSSA Agronomy Monograph no. 9 (2nd edn). WILLIAMS,S. T., GOODFELLOW, M., WELLINGTON, J. C., ALDERSON, G., SNEATH, E. M. H., VICKERS, P. H. A., SACKIN,M. J. & MORTIMER, A. M. (1983). A probability matrix for identification of some streptomycetes. Journal of General Microbiology 129, 1815-1 830.
Printed in Great Britain
317
Cloning, Characterization and Expression in Escherichia coli of a Leucine Biosynthetic Gene from Streptomyces rochei By J U N E H E R C O M B , ’ G E O R G T H I E R B A C H , 2 S I M O N B A U M B E R G 3 J . H. PARISH1* Department of Biochemistry, University of Leeds, b e d s LS2 9JT, UK Asta- Werke A G, Forschung Chemie, Organisch-Biotechnologie,Kantstrasse 2, 0-4802 Halle-Kiinsebeck, FRG 3Department of Genetics, University of Leeds, Leeds LS2 9JT, UK
AND
(Received 29 May 1986; revised 10 September 1986)
Leucine and histidine biosynthetic genes from Streptomyces rochei HPl that complemented auxotrophic mutations in S. lividans TK54 were cloned in pIJ61. DNA from one leucine recombinant plasmid was subcloned into pBR322. From the latter, a recombinant plasmid was obtained that complemented the leuA mutation in Escherichia coli CV5 12 but not other leucine markers in E. coli. Analysis of this and several subclones, including mutant plasmids constructed in uitro, established that the cloned S. rochei gene was expressed in E. cofi from the tetracycline promoter of pBR322 to produce a polypeptide of 67 kDa; the corresponding coding region was shown to be within a 1.7 kbp DNA fragment. Blot hybridization revealed corresponding homologous genes in several other streptomycetes.
INTRODUCTION
Several streptomyces genes have been cloned and expressed in streptomyces (Bibb et al., 1983). There are fewer examples of expression of cloned streptomyces genes in Escherichia coli: these include genes for antibiotic resistance (e.g. Vara et al., 1985; Gil et al., 1985; Schupp et al., 1983), a gene for an extracellular enzyme (Robbins et al., 1981) and a few examples of genes for biosynthetic enzymes. Of these, the cloned p-aminobenzoic acid synthetase gene appears to be involved in candicidin biosynthesis (Gil & Hopwood, 1983). There is one previously published example of gene for an enzyme from an amino acid biosynthetic pathway, argG, and this furnishes one of the very few known examples of expression in E. coli from a streptomyces promoter (Meade, 1985); more generally, it is to be anticipated that expression from such promoters will not be generally applicable to the characterization of biosynthetic genes from streptomycetes. Our studies on amino acid biosynthesis in streptomycetes arose from work on a newly isolated organism, strain HP1 (see Results) and attempts, so far unsuccessful, to clone its genes for cellulolytic enzymes. We were concerned to establish whether HP1 genes of any kind might be expressed in the cloning host, Streptomyces Iividans and in E. coli. Amino acid biosynthetic markers were chosen and, of the clones identified in S. lividans, leucine was selected for further study because there are relatively few ‘dedicated’ leucine biosynthetic enzymes that might be candidates for possible complementation experiments. Biosynthesis of leucine in most bacteria involves three steps (referred to collectively as the isopropylmalate pathway) unique to leucine biosynthesis. In enteric bacteria, the genes for the corresponding enzymes are clustered in a single operon. The products of the genes (in E. coli) are or-isopropylmalate synthase (EC 4.1.3.12) (IeuA), P-isopropylmalate dehydratase (EC 4.2.1.33) (IeuC, IeuD) and P-isopropylmalate dehydrogenase (EC 1.1.1.85) (leuB) (Calvo, 1983). 0001-3494 0 1987 SGM
318
J . HERCOMB A N D OTHERS
METHODS Bacterial strains and media. Streptomyces rochei HP1 was isolated from Jamaican laterite soil by the methods of Mullings & Parish (1984). Strains and plasmids used in the present work are listed in Table 1. For growth of streptomycetes, R2YE agar and MM agar are described by Hopwood et al. (1985). MM agar was supplemented (when necessary) with leucine (50 mg 1-l) and histidine (75 mg 1-l). YEME medium was as described by Hopwood et al. (1985) except that the glycine concentration was 0.1 % (w/v) for experiments in which protoplasts were to be isolated and was omitted in other cases. E. coli was grown using L broth and L agar (Maniatis et al., 1982) and M9 medium and M9 agar (Miller, 1972), supplemented when necessary with leucine and/or proline (40 mg 1-l). Antibiotics were added as filter-sterilized concentrates (in water, except for thiostrepton, which was dissolved in dimethyl sulphoxide) to the following final concentrations (mg 1-l) : thiostrepton, 50 : neomycin, 10: ampicillin, 50: tetracycline, 12.5. Bacterial growth. E. coli strains were incubated on plates at 37 "C and grown in shake culture at 37 "C. Streptomycetes were grown similarly at 30 "C. Enzymes and reagents. Molecular biological reagents were purchased from BRL, Boehringer and New England Biolabs. Other reagents and antibiotics were obtained from various laboratory suppliers; thiostrepton was a generous gift from Mr S. J. Lucania of E. J. Squibb and Sons, New Brunswick, NJ, USA. Plasmid isolation,restriction, ligation, transformation,electrophoresis,blot hybridization and plasmid reconstruction. Plasmid pBR322 was isolated from E. coli JA221; pIJ61 was isolated from S . lividans 1326. General molecular biological methods and experiments specific to E. coli with the exceptions noted in the following paragraph were according to Maniatis et al. (1982); all manipulations involving streptomycetes were according to Hopwood er al. (1985). Transformants acquiring pIJ61 and its derivatives were detected by replica plating the pocks on to antibiotic-selective minimal or supplemented plates as described in the reference. Electroelution of restriction fragments from gels was by the direct method of Maniatis et al. (1982). Analysis of protein synthesis by using E. coli minicells was based on the methods of Hallewell & Sherratt (1976). Minicell strains were grown to stationary phase in 500 ml L broth and harvested by centrifugation at 4 "C for 10 min at 12000 r.p.m. The pellets were resuspended in 4 ml M9 medium and minicells were further purified by sedimentation through 40 ml linear gradients of 5-20% (w/v) sucrose in M9 medium, for 20 min at 3000 r.p.m. Minicells were recovered, washed, resuspended in M9 medium and purified by two further centrifugations through sucrose gradients. After washing, the minicells were resuspended in 0.5 ml M9 medium supplemented with glycerol (20%, v/v). Minicells were labelled in M9 medium supplemented with cycloserine (100 pg ml-l). Minicells from a portion (0.25 ml) of glycerol suspension were washed twice by centrifugation and resuspended in 0.1 ml medium. After 1 h at 37 "C, 25 pCi (925 kBq) [35S]methionine(0.5-1 Ci pmol-l) was added. After 30 min, the minicells were harvested by centrifugation, resuspended in SDS-running buffer (50 pl) and (after heating and removal of residues) the solution was analysed by SDS electrophoresis using 4C-labelled standard proteins as markers. RESULTS
Bacterium H P l Bacterium HP 1 was isolated as a cellulose-degrading streptomycete. A fuller description is being prepared for publication elsewhere. The organism has no unusual nutritional requirements, is aerobic and grows at temperatures between 20 and 45°C. Analysis of key taxonomic characters determined by Dr E. M. H. Wellington (Williams & Wellington, 1982; Williams et al., 1983) suggested that it is a representative of the species S. rochei. Isolation of plJ61 leucine and histidine recombinants A ligation mixture of a partial Sau3A digest of S . rochei HP1 DNA and a complete BamHI digest of pIJ61 DNA was used to transform S. liuidans TK54. The pocks were almost confluent on a total of 20 plates. We estimated that the total number of recombinants was approximately 20000. Of these, 27 were scored as apparent histidine recombinants and three as apparent leucine recombinants. Plasmids isolated from each of 16 His+ recombinants tested generated the His+ phenotype on transformation. Of the three Leu+ recombinants, one was rejected on the grounds that it was apparently a chromosomal revertant or recombinant (it proved to be neomycin resistant) and of the remaining two only one, pGT29, conferred the Leu+ phenotype on retransformation. Restriction enzyme analysis of pGT29 and the His+ recombinants revealed inserts of between 5 and 10 kbp (data not shown).
Cloning of a Streptomyces rochei leucine gene
319
Table 1. Bacterial strains and plasmids Relevant characters Streptomyces strains S . coelicolor M130 S . glaucescens NCIB 9619 S . griseus ATCC 12475 S . lividans TK54 S . lividans 1326 S . rochei HPl Streptoverticillium verticillium JC N 4924 E. coli strains CV5 12 CV520 CV526 DS4 10 HBlOl JA221 E. coli plasmids pBR322 pJHl pJH20 pJH 107 pJH108 pJHl1 pJHl11 pJH211 pJH311 pJH411 Streptomyces plasmids pIJ61 pGT29
leu2 his2
Source or reference D. A. Hopwood NCIB ATCC D. A. Hopwood D. A. Hopwood This work Riken, Japan
F+ leuA371 F+ leuCI71 F+ leuDlOl
Somers et al. (1973) Somers et al. (1973) Somers et al. (1973) Dougan & Sherratt (1977) Maniatis et al. (1982) Laboratory strain
amp amp amp amp amp amp amp amp amp amp
Maniatis et al. (1982) This work This work This work This work This work This work This work This work This work
minA minB rpsL leuB proA2 tet Leu+* Leu-* Leu+* Leu+* Leu+* Leu-* Leu-* Leu-* Leu-*
neo tsr ltz tsr ltz; pIJ61-HP1 recombinant: complements leu2 in S. liuidans
D. A. Hopwood This work
* These are derivatives of pGT29-pBR322 recombinants (see Results); the Leu phenotype refers to the ability/inability of the plasmid to complement the leuA mutation in E. coli CV512. Construction, analysis and phenotypic characterization of pBR322-pGT29 recombinants A partial Sau3A digest of pGT29 DNA was ligated with a complete BamHI digest of pBR322 DNA and was used to transform E. coli strains HBlOl (IeuB) and CV512 (IeuA). Leu+ colonies were detected only from the transformation of strain CV512. Plasmid pJHl was isolated from one of these recombinants. Retransformation experiments with this confirmed that it was able to complement the leuA mutation in strain CV512 but it was unable to complement IeuB, leuCor leuD mutations in the strains tested (Table 1). Sub-clones and deletion mutants derived from pJHl were constructed and tested for the capacity to complement the leuA mutation (Fig. 1). The physical maps of these plasmids and of pGT29 were obtained from single and double restriction digests. Details of the pGT29 map were confirmed by Southern blot hybridization of several single digests of the plasmid using labelled pJHl DNA as a probe. Sequences derived from vectors (pBR322 and pIJ61) were identified from the published restriction maps for these plasmids (Maniatis et al., 1982; Hopwood et al., 1985). From an alignment of thee maps (Fig. I), we conclude that deletions of pBR322 sequences occurred in rho in the experiments that generated pJHlO7 and pJH108. The ClaI site of pBR322, which lies within the tet promoter, was removed (by end filling) from pJHll to generate pJHl11. Five plasmids with the same properties as pJHl11 were isolated independently. The absence of the Cla site was confirmed by restriction analysis and the transformants were Leu-, thus establishing that the streptomyces gene was expressed from the tet promoter in the Leu+ pJH plasmids. The validity of the method was confirmed by analysis of a tetracycline-sensitive pBR322 derivative constructed by the same method, and the inability of
320
J . HERCOMB AND OTHERS
ti
(i
I
I
V V I
1
P I
B G
G B P B V P I
I
I
I l l
BVP
V
PECHP
G BP
V
PECHP
G B
V
PECHP
G B P
V
P ECHP
G BP B
I
V
B
P
V H E
H
I
I
I
I
I
1
G
B
V
B
P
V
I
I
H
1 pGT29 (+)
V
i PJHl(+) P
V
3 ' pJH20(-) PG BPBV
2 pJH107(+) PBV
.--
V
111
P ECHP
I
pJHlO8(+)
II
G B P BV
pJHll (+)
V
P ECHP
V
P ECHP
G B
BV
I pJH211 (-) ' -
--------
P BV
pJH411 (-)
I kbp
H
Fig. 1. Restriction endonuclease maps of recombinant plasmids encoding the S . rochei Leu gene and of certain subclones. The double arrow under the map of pJHl1 represents the proposed limits of the Leu gene; sites i, ii and iii are referred to in the text. The leucine phenotype conferred by each plasmid is shown by or - in parentheses against the plasmid name. I pIJ61 , DNA;=, pBR322 DNA; -, cloned HP1 DNA; ----, deleted regions. Restriction sites are denoted: B, BurnHI; C, ClaI; E, EcoRI; G, BgnI; H, HindIII; P, PstI; V, PvuII. The origins of plasmids pGT29 and pJHll are described in the text. pJH20 was obtained by religation of a BarnHI digest of pJH1; pJH107 and pJHlO8 were obtained by cloning a partial Sau3A digest of pJHl into pBR322 (deletions of pBR322 sequences were generated in these constructions). pJHl1 was generated by religation of a PvuII digest of pJHl. The remaining three plasmids were all derived from pJHl1 by deletions generated by BurnHI (pJH21 l), BglII-Hind111 (pJH311) and partial digestion with Psi1 (pJH411).
+
p J H l l l to synthesize the leucine biosynthetic enzyme was confirmed by studies on the translation products from the plasmid (see below). Protein synthesis due to expression of plasmid genes was measured in E . coliminicells (Fig. 2). In addition to the pBR322 proteins, pJHl and pJHl1 direct the synthesis of a polypeptide (p67) of 67 kDa. This is presumptively the product of the leucine biosynthetic gene. The protein was greatly reduced in amount in the mutant, pJHl11, in which the tet promoter was inactivated. Protein p67 was replaced by a truncated protein of 55 kDa in mutant pJH211 and by two proteins of approximately 46 kDa in pJH311. No protein corresponding to p67 was observed with pJH411. The coding sequence corresponding to a polypeptide of the size of p67 is approximately 1.7 kbp; the corresponding values for the two truncated proteins are 1.4 kbp and 1.1 kbp. From these data and the maps and phenotypes of the plasmids of Table 1, we deduced the limits within which the gene must lie (see Discussion). IdentiJicationof sequences homologous to the cloned leucine gene in other streptomycetes The internal PstI-Psi1 fragment from pJHll that contains the leucine gene was labelled in vitro and used as a hybridization probe for genomic DNA from several streptomycetes digested with PstI. As a control, genomic DNA from S. rochei HPl was also digested with BamHI to confirm the size of the homologous region from the source organism. Homologous fragments were discovered in all the strains tested, i.e. the seven StreptomyceslStreptoverticillium strains listed in Table 1. Moreover the PstI fragments were all similar in size except in the cases of S.
Cloning of a Streptomyces rochei leucine gene
32 1
Fig. 2. SDS-PAGE analysis of 35S-labelledproteins synthesized in minicells containing plasmids. The results in the two photographs were obtained in separate experiments and are therefore calibrated separately.
coelicolor and Stu. uerticillium, in which the complementary fragments were larger. The size of the corresponding fragment in S . liuidans could not be deduced because the DNA was not completely digested. There is thus qualitative evidence that the sequence is not unique to S . rochei HPl . DISCUSSION
The complementation of the leu2 mutation by a gene that also complements leuA in E . coli shows that the leu2 gene of S . liuidans is a structural gene for a-isopropylmalate synthase. It is, therefore, the first ‘dedicated’ gene of the leucine biosynthetic pathway in the organism. As pJHl does not complement E. coli leuB, leuC or leuD mutations, it is probable that the other streptomyces Leu genes are not present in the clone and, indeed, two of the Leu genes in S. coelicolor are separated by approximately half the chromosome (Hopwood et al., 1985). Comparison of the maps and phenotypes of plasmids pJH1, pJH20 and pJHl1 establishes that PstI site i (Fig. 1) lies within the region essential for the Leu+ phenotype and is, therefore, within or immediately proximal to the coding sequence, and the properties of plasmids pJHl1, pJH107 and pJH108 establish that the region distal to the BamHI site ii is not required. The absence of a polypeptide product from pJH411 suggests the start of the 1.7 kbp coding region is distal to the PstI site iii (Fig. 1). We propose that the truncated 1.4 kbp coding region in pJH211 defines the start of the coding region as being between the sites iii and ii. The polypeptide produced by pJH311 can only be accounted for by assuming that it arises adventitiously, for example by the accidental generation of a ribosome-binding site. The estimated molecular mass of the streptomyces isopropylmalate synthase subunit, 67 kDa, is similar to that for the corresponding yeast enzyme (65-67 kDa; Chang et al., 1984) but is substantially larger than that for the salmonella protein (50 kDa; Calvo, 1983). In the light of the homology between the cloned gene and unique sequences from other streptomycetes, we
322
J . HERCOMB A N D O T H E R S
conclude that this size is likely to be true of the enzymes from the streptomycete group in general. Despite the difference in size from the enzyme of enteric bacteria, the complementation data show that the streptomyces enzyme functions in E. coli. The method we have described should be applicable to other amino acid biosynthetic genes from streptomycetes. We are grateful to Dr E. M. H. Wellington for the taxonomic assignment of strain HP1, to Dr M. J. Bibb and Professor D. A. Hopwood for help and advice and to Mr K. Ainley for excellent technical assistance with the studies that led to the isolation of pGT29. The work was supported, in part, by an SERC Research Grant to J. H. P. and S. B., and J. H. is an SERC Research Student. REFERENCES
BIBB,M. J., CHATER, K. F. & HOPWOOD, D. A. (1983). Developments in Streptomyces cloning. In Experimental Manipulation of Gene Expression, pp. 54-80. Edited by M. Inouye. London: Academic Press. CALVO, J. M. (1983). Leucine biosynthesis in prokaryotes. In Amino Acids - Biosynthesis and Genetic Regulation, pp. 267-285. Edited by K. M. Herrmann & R. L. Somerville. Reading, Mass.: AddisonWesley. CHANG,L. L.-F., CUNNINGHAM, T. A., GATZEK, P. R., CHEN,W.-J. &KOHLHAW, G. B. (1984). Cloning and characterization of yeast leu4, one of two genes responsible for a-isopropylmalate synthesis. Genetics 108, 91-106. DOUGAN, G. & SHERRATT, D. J. (1977). The transposon T n l as a probe for studying ColE1 structure and function. Molecular and General Genetics 151, 151160. GIL, J. A. & HOPWOOD,D. A. (1983). Cloning and expression of a p-aminobenzoic acid synthetase gene of the candicidin-producing Streptomyces griseus. Gene 25, 119-132. GIL, J. A., KIESER,H. M. & HOPWOOD, D. A. (1985). Cloning of a chloramphenicol acetyltransferase gene of Streptomyces acrimycini and its expression in Streptomyces and Escherichia coli. Gene 38, 1-8. HALLEWELL, R. A. & SHERRATT, D. J. (1976). Isolation and characterisation of ColE2 plasmid mutants unable to kill colicin-sensitive cells. Molecular and General Genetics 146, 239-245. HOPWOOD, D. A., BIBB,M. J., CHATER, K. F., KIESER, T., BRUTON,C. J., KIESER,H. M., LYDIATE, D. J., SMITH,C. P., WARD,J. M. & SCHREMPF, H. (1985). Genetic Manipulation of Streptomyces - a Luboratory Manual. Norwich: John Innes Foundation. E. F. & SAMBROOK, J. (1982). MANIATIS, T., FRITSCH, Molecular Cloning: a Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. MEADE,H. (1985). Cloning of argG from Streptomyces:
loss of gene in Arg- mutants of S. cattleya. BiolTechnology 3, 917-918. MILLER,J. H. (editor) (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. MULLINGS,R. & PARISH,J. H. (1984). Mesophilic aerobic Gram negative cellulose degrading bacteria from aquatic habitats and soils. Journal of Applied Bacteriology 57, 455-468. ROBBINS, P. W., WIRTH,D. F. & HERING,C. (1981). Expression of the Streptomyces enzyme endoglycosidase H in Escherichia coli. Journal of Biological Chemistry 256, 10640-1 0644. SCHUPP, T., TOUPET,c . , STALHAMMER-CARLEMALM, M. & MEYER,J. (1983). Expression of a neomycin phosphotransferase gene from Streptomyces fradiae in Escherichia coli after interplasmidic recombination. Molecular and General Genetics 189, 27-33. SOMERS, J. M., AMZALLAG, A. & MIDDLETON, K. B. (1973). Genetic fine structure of the leucine operon of E. coli K12. Journal of Bacteriology 113, 12681272. VARA, J., MALPARTIDA, F., HOPWOOD,D. A. & JIMENEZ, A. (1985). Cloning and expression of a puromycin N-acetyltransferase gene from Strepromyces alboniger in Streptomyces lividans and Escherichia coli. Gene 33, 197-206. S. T. & WELLINGTON, E. M. H. (1982). WILLIAMS, Actinomycetes. In Methods of Soil Analysis, Part 2. Chemical and Microbiological Properties, pp. 969987. Edited by A. L. Page, R. H. Miller & D. R. Keeney. Madison, Wis. : ASA-SSSA Agronomy Monograph no. 9 (2nd edn). WILLIAMS,S. T., GOODFELLOW, M., WELLINGTON, J. C., ALDERSON, G., SNEATH, E. M. H., VICKERS, P. H. A., SACKIN,M. J. & MORTIMER, A. M. (1983). A probability matrix for identification of some streptomycetes. Journal of General Microbiology 129, 1815-1 830.
