SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References

The complete nucleotide sequence of the large linear plasmid pSLA2-L in Streptomyces rochei strain 7434AN4 has been determined. pSLA2-L was found to be 210 614 bp long with a GC content of 72.8% and carries 143 open reading frames. It is especially noteworthy that three-quarters of the pSLA2-L DNA is occupied by secondary metabolism-related genes, namely two type I polyketide synthase (PKS) gene clusters for lankacidin and lankamycin, a mithramycin synthase-like type II PKS gene cluster, a carotenoid biosynthetic gene cluster and many regulatory genes. In particular, the lankacidin PKS is unique, because it may be a mixture of modular- and iterative-type PKSs and carries a fusion protein of non-ribosomal peptide synthetase and PKS. It is also interesting that all the homologues of the afsA, arpA, adpA and strR genes in the A-factor regulatory cascade in Streptomyces griseus were found on pSLA2-L, and disruption of the afsA homologue caused non-production of both lankacidin and lankamycin. These results, together with the finding of three possible replication origins at 50–63 kb from the right end, suggest that the present form of pSLA2-L might have been generated by a series of insertions of the biosynthetic gene clusters into the left side of the original plasmid.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References

Streptomyces are filamentous soil bacteria that contain a linear chromosome and produce many useful secondary metabolites including antibiotics. The biosynthetic genes for antibiotics in Streptomyces form a gene cluster and are usually located on the chromosome. However, in three cases, a large linear plasmid is known to be involved in antibiotic production; SCP1 in Streptomyces coelicolor A3(2) for methylenomycin (Kirby et al., 1975; Kirby and Hopwood, 1977; Chater and Bruton, 1985; Kinashi et al., 1987; Kinashi and Shimaji-Murayama, 1991; Redenbach et al., 1998), pSLA2-L in Streptomyces rochei for lankacidin and lankamycin (Kinashi et al., 1994; 1998; Suwa et al., 2000) and pPZG103 in Streptomyces rimosus for oxytetracycline (Gravius et al., 1994; Pandza et al., 1998).

Streptomyces rochei strain 7434AN4 produces two structurally       unrelated       polyketide       antibiotics,       a       14-membered macrolide lankamycin (LM) (Keller-Schierlein and Roncari, 1964) and an unusual 17-membered macrolide lankacidin (LC) (Harada et al., 1969; Uramoto et al., 1969) (Fig. 1). The latter is used as a feed additive to protect and treat pork infection with Treponema hyodysenteriae. Strain 7434AN4 carries three linear plasmids, pSLA2-L (210 kb), M (100 kb) and S (17 kb) (Kinashi et al., 1994). Hybridization experiments using the typical polyketide synthase (PKS) gene probes, eryAI (Donadio and Katz, 1992) for erythromycin (EM) and actI (Fernandez-Moreno et al., 1992) for actinorhodin, identified their homologous regions on pSLA2-L (Kinashi et al., 1998). Sequencing and targeting experiments (Suwa et al., 2000) confirmed that two eryAI-homologous regions on PstI fragment A are parts of a large type I PKS gene cluster for LM (see Fig. 2 for the PstI restriction map). On the other hand, the actI-homologous region extending over PstI fragments H and I proved not to be involved in the production of LM or LC. Thus, we had not identified the location of the lankacidin synthase (lkc) gene cluster or the chemical structure of the polyketide metabolite, the synthesis of which is coded by the type II PKS gene cluster.

image

Figure 1. The chemical structures of three polyketide antibiotics, lankacidin C (LC), lankamycin (LM) and erythromycin (EM). R1, 4-acetyl-l-arcanose; R2, d-chalcose; R3, l-cladinose; R4, d-desosamine.

Download figure to PowerPoint

image

Figure 2. Location and functional classification of 143 ORFs deduced by the complete nucleotide sequence of pSLA2-L and targeting experiments. The ordered six cosmids and the three terminal plasmids, pE4, pHindE and pE3, are also shown together with the recognition sites for PstI. Each ORF is shown as a box above or under the DNA line based on their direction of translation (right or left). In addition to the five biosynthetic gene clusters, regulatory genes, resistance genes and replication and maintenance genes are shown by coloured boxes, while hypothetical genes are indicated by white boxes.

Download figure to PowerPoint

Streptomyces linear plasmids and chromosomes replicate bidirectionally from a centrally located replication origin towards both ends (Chang and Cohen, 1994; Musialoski et al., 1994). Replication in the leading strand is completed as far as the 3′ ends, but that in the lagging strand does not reach the 5′ ends. The 3′ protruding ends of the replication intermediate may form a foldback secondary structure, which was suggested to function in the fill-in DNA synthesis primed by terminal protein (Huang et al., 1998; Qin and Cohen, 1998). Recently, terminal protein genes have been cloned from S. rochei and Streptomyces lividans (Bao and Cohen, 2001; Yang et al., 2002).

To answer several important questions about pSLA2-L, such as where the lankacidin synthase (lkc) gene cluster is, what the functions of the type II PKS genes are, why pSLA2-L carries so many PKS genes, and how the genes necessary for plasmid replication and maintenance are organized, we have started the sequencing project of pSLA2-L. Here, we describe the complete nucleotide sequence of pSLA2-L and analysis of genes based on homology search and gene disruption, which revealed an unusually condensed gene organization for secondary metabolism.

Results and discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References

Basic features of the pSLA2-L sequence

pSLA2-L was determined to be a 210 614 bp linear plasmid with a 1992 bp terminal inverted repeat (TIR) at both ends. The GC content of pSLA2-L is 72.8%, which is similar to the 72.1% overall GC of the S. coelicolor A3(2) chromosome (Bentley et al., 2002). pSLA2-L carries 143 open reading frames (ORFs) that were numbered from the left end as shown in Fig. 2. Most of them (111 ORFs, 78%) are predicted from database similarities, whereas the remaining 32 ORFs (22%) are hypothetical and mainly located near the right end of pSLA2-L. This is unusual for a plasmid, because predicted genes are few in plasmids without a pressure to keep indispensable genes.

The most interesting feature of pSLA2-L is that three-quarters of its DNA is occupied by biosynthetic gene clusters for secondary metabolites. Two large type I PKS gene clusters for lankacidin (lkc, ORF4–ORF18) and lankamycin (lkm, ORF24–ORF53) occupy the left half of the plasmid. After these clusters, another type I PKS gene cluster and a type II PKS gene cluster are located. Intriguingly, all these clusters are closely packed without a long intervening sequence. Three pairs of possible replication initiation proteins (ORF87 and ORF89; ORF93 and ORF94; ORF95 and ORF96) were found at 50–63 kb from the right end, and a set of two partition genes, parA and parB, are located at ORF98 and ORF97. A carotenoid biosynthetic gene cluster (ORF104–ORF110) is present at the right side of three replication origins. Regulatory genes for secondary metabolism are distributed over the entire region of the plasmid. A TTA codon was found in five ORFs (ORF89, ORF112, ORF131, ORF138 and ORF139), making their translation dependent on the leucyl-tRNA coded by bldA (Leskiw et al., 1991). All these ORFs encode a hypothetical protein except ORF89, which codes a replication initiation protein. Predicted functions of the genes grouped into the four main biosynthetic gene clusters are summarized in Table 1. The complete nucleotide sequence of pSLA2-L has been submitted to the DDBJ/EMBL/GenBank databases under accession number AB088224.

Table 1. . Organization of the four main biosynthetic gene clusters on pSLA2-L.
ORFPredicted functionORFPredicted function
  1. The order of domains in the PKS and NRPS genes is shown in parenthesis. KR°, non-functional KR domain.

Lankacidin synthase (lkc) gene cluster  40Glycosyltransferase
 4Pyrroloquinoline quinone biosynthesis protein B (PqqB) 41NDP-3-methyl-4-keto-2,6-dideoxyhexose
 5Pyrroloquinoline quinone biosynthesis protein E (PqqE) 4-ketoreductase
 6Pyrroloquinoline quinone biosynthesis protein D (PqqD) 42NDP-3-keto-6-deoxyhexose 3-ketoreductase
 7Pyrroloquinoline quinone biosynthesis protein C (PqqC) 43Glycosyltransferase
 8Pyrroloquinoline quinone biosynthesis protein A (PqqA) 44ABC transporter
 9ABC transporter 45NDP-hexose 3-O-methyltransferase
10ABC transporter 46NDP-4-keto-6-deoxyhexose 3,5-epimerase
11Isochorismatase 47Adenosylhomocysteinase
12Type I PKS (KS, ACP, TE) 485,10-Methylenetetrahydrofolate reductase
13Type I PKS (KR, ACP, KS, KR, ACP, KS) 495-Methyltetrahydrofolate-homocysteine
14Amino oxidase S-methyltransferase
15Acyltransferase 50Glucose kinase
16Type I PKS (KR, MT, ACP, ACP, KS) 51S-adenosylmethionine synthetase
17Dehydratase 52NDP-glucose synthase
18NRPS–PKS (condensation, adenylation, PCP; KS) 53NDP-4-keto-6-deoxyhexose 2,3-dehydratase
Lankamycin synthase (lkm) gene cluster  mtm-like type II PKS gene cluster 
24NDP-hexose 4,6-dehydratase 62Cyclase/aromatase
25Thioesterase 63Ketoreductase
26P450-like hydroxylase 64Oxygenase
27NDP-4-keto-2,6-dideoxyhexose 3-C-methyltransferase 65Acyl-CoA ligase
28NDP-hexose 3-O-methyltransferase 66Cyclase
29NDP-4-keto-2,6-dideoxyhexose 2,3-enoyl reductase 67Acyl carrier protein
30β-Glycosidase 68Chain length factor
31Glycosyltransferase 69β-Ketoacyl synthase
32NDP-4-keto-6-deoxyhexose 3,4-isomerase 70Thioesterase
33LkmAIII, modules 5 and 6 71SARP family regulatory protein
(KS, AT, KR, ACP; KS, AT, KR, ACP, TE)  
34LkmAII, modules 3 and 4Carotenoid synthesis (crt) gene cluster 
(KS, AT, KR°, ACP; KS, AT, DH, ER, KR, ACP)104Methyltransferase (CrtV)
35LkmAI, loading module, modules 1 and 2105Phytoene synthase (CrtB)
(AT, ACP; KS, AT, KR, ACP; KS, AT, KR, ACP)106Phytoene dehydrogenase (CrtI)
36O-acyltransferase107Geranylgeranyl pyrophosphate synthase (CrtE)
37P450-like hydroxylase108Lycopene cyclase (CrtY)
38NDP-6-deoxyhexose 3,4-dehydratase109Hypothetical
39NDP-4,6-dideoxyhexose 3,4-enoyl reductase110Dehydrogenase (CrtU)

Lankacidin synthase (lkc) gene cluster (ORF4–ORF18)

Several type I PKS genes were found close to the left end of pSLA2-L, although the typical type I PKS probe, eryAI, did not hybridize to this region. To reveal the function of these genes, a targeting plasmid pC10PR was constructed from cosmid C10, which is located a little bit right of cosmid B10 (Fig. 2). In pC10PR, only the left and right ends of the cosmid insert were retained, and the central 36.8 kb region (nt 12 541–49 309) was replaced by a kanamycin resistance gene cassette. Using this plasmid, the pSLA2-L DNA extending over ORF12–ORF30 was deleted. Gene disruptants did not produce LC or LM, which indicated that the deleted region carried the lkc gene cluster as well as the left part of the lkm gene cluster. The details of gene disruption experiments in this study will be described elsewhere in due course, and only the main results of the bioassay are shown in Fig. 3.

image

Figure 3. Effects on antibiotic production of the disruption of genes around the lankacidin synthase (lkc) gene cluster and two regulatory genes.

Download figure to PowerPoint

The range of the lkc gene cluster was delimited as follows. As shown in Fig. 3, disruptants of ORF3 produced both LM and LC, whereas disruptants of ORF7 produced only LM. As ORF4–ORF8 forms a gene cluster for pyrroloquinoline quinone (PQQ) synthesis (Meulenberg et al., 1992), the left end of the lkc gene cluster was located at ORF4. In accordance with this speculation, addition of PQQ to the liquid culture of the ORF7 disruptant in a final concentration of 2 µg ml−1 restored synthesis of LC. The unusual 17-membered lactone ring of LC may be formed by nucleophilic attack of C-2 on C-18 in an imide bonding. As PQQ is a cofactor for oxidoreductases, it may function in an oxidation reaction to form an imide from an amide intermediate. As PQQ synthesis genes have been found only in Gram-negative bacteria, this is a first report from Gram-positive bacteria. Disruptants of ORF18 did not produce LC but produced LM, whereas disruptants of ORF19 produced both antibiotics (Fig. 3). This result located the right border of the lkc cluster at ORF18 and determined the size of the cluster to be 31 kb. The lkc-PKS genes are quite similar to the type I pks-PKS genes found in the Bacillus subtilis genome (Kunst et al., 1997), the functions of which have not been clarified.

The macrolide skeleton of LC is synthesized from a starter glycine molecule followed by eight acetate molecules (Uramoto et al., 1978). All the methyl groups at C-2, C-4, C-10 and C-16 are not derived from propionate but from methionine by C-methylation. This fact distinguishes the lkc-PKS from the usual modular-type PKSs. ORF18, the first PKS gene in the cluster, encodes a fusion protein of non-ribosomal peptide synthetase (NRPS) and PKS. NRPS–PKS fusion genes have been found in several biosynthetic gene clusters for secondary metabolites, for example antibiotic TA (Paitan et al., 1999), myxothiazol (Silakowski et al., 1999), microcystin (Nishizawa et al., 2000) and albicidin (Huang et al., 2001). The NRPS region in ORF18 has domains for condensation, adenylation and thiolation (PCP), whereas the PKS region possesses only a β-ketoacyl synthase (KS) domain. The adenylation domain contains amino acid residues DILQTLVEAK at the positions 235, 236, 239, 278, 299, 301, 322, 330, 331 and 517 (nt 33 131–33 979), which are important for recognizing amino acids (Stachelhaus et al., 1999). Antibiotic TA and saframycin Mx1 (Pospiech et al., 1995) also incorporate a glycine molecule, and their adenylation domains contain similar residues, DILQLRLVETK and DILQTLVETA respectively. These results suggest the signature sequence for glycine recognition, which was not described by Stachelhaus et al. (1999). Thus, ORF18 may be involved in the initial condensation reaction of glycine and acetate.

Although eight condensation reactions of acetate are necessary for the synthesis of the lankacidin skeleton, only five KS domains are located in the lkc gene cluster (Table 1; ORF18, ORF16, ORF13 × 2, ORF12). Additional type I PKS genes (ORF56–ORF61) were found in the centre of pSLA2-L, disruption of which gave no effect on LC synthesis (see below). So, the lkc-PKS may be a mixture of modular- and iterative-type PKSs and accomplishes eight condensation reactions using five KS domains. In this context, it is interesting that ORF16 carries two almost identical acyl carrier protein (ACP) domains in tandem alignment (Table 1). Tandem ACP domains were also found in XabB, an albicidin biosynthesis protein, in Xanthomonas albilineans, and are suggested to be involved in iterative condensation reactions (Huang et al., 2001). In addition, it is noteworthy that the order of domains in the lkc-PKS (KR-ACP-KS) is different from that in the typical type I PKSs (KS-AT-KR-ACP). The lkc-PKS contains none of the acyltransferase (AT), dehydratase (DH) and enoyl reductase (ER) domains but, instead, AT and DH genes are located separately at ORF15 and ORF17 respectively. The methyltransferase (MT) domain in ORF16 may function in C-methylation, and the thioesterase (TE) domain in ORF12 may be involved in lactone formation.

Lankamycin synthase (lkm) gene cluster (ORF24–ORF53)

The lankamycin synthase (lkm) gene cluster extends over 68 kb from ORF24 to ORF53. Three large modular-type PKS genes, lkmAI (ORF35), lkmAII (ORF34) and lkmAIII (ORF33), are homologous to eryAI, eryAII and eryAIII, respectively, for 6-deoxyerythronolide B synthesis in Saccharopolyspora erythraea (Donadio and Katz, 1992). As in the case of eryAII, the ketoreductase (KR) domain in module 3 of lkmAII is non-functional, leaving the ketone group at C-9 unreduced, because the conserved motif GXGXXGXXXA, proposed to be the NADP(H) binding site (Scrutton et al., 1990), was replaced by AXSXXGXXXA (nt    71 624–71 653).    Many    genes    for    the    synthesis    of d-chalcose and 4-acetyl-l-arcanose and their attachments to the macrolide ring are located at both sides of lkmAI–AIII.

The macrolide ring structures of LM and erythromycin (EM) are quite similar including the stereochemistry of asymmetric carbons. The differences are the hydroxylation at the C-6, C-8 and C-12 positions, and the substitution at C-13 (Fig. 1). At the last position, LM contains a 1(s)-methyl-2(s)-hydroxypropyl group in place of the ethyl group in EM. Based on the modular organization of the lkm-PKS, we expected the presence of an extra module in lkmAI (ORF35) responsible for an additional C-2 unit. However, this was not the case. Therefore, the loading module of lkmAI has a different specificity; 2-methyl-3-hydroxybutyrate may be recognized as a starter, or 2-methylbutyrate may be incorporated first as in the case of avermectin synthesis (Chen et al., 1989) and then hydroxylated at C-15. Two P450-like hydroxylase genes (ORF26 and ORF37) in the lkm cluster may function in hydroxylation of C-8 or C-15.

mtm-like type II PKS gene cluster (ORF62–ORF71)

A type II PKS gene cluster, which is most similar to the mithramycin synthesis (mtm) gene cluster (Prado et al., 1999), is located from ORF62 to ORF71. An interesting feature of this cluster is the presence of a thioesterase (TE) gene (ORF70) upstream of the minimal PKS genes (ORF67–ORF69). No TE gene has been found in the type II PKS gene clusters, although it is common in the type I PKS gene clusters. Not surprisingly, the ORF70 product shows a high similarity to TE of the type I PKS for rifamycin synthesis in Amycolatopsis mediterranei (August et al., 1998). As ORF68, ORF69 and ORF70 overlap each other by 4 bp and have the same translation direction, they are likely to be translationally coupled (Zalkin and Ebbole, 1988). As reported previously, deletion of ORF66 (cyclase), ORF67 (ACP) and ORF68 (chain length factor) had no effect on the production of LM or LC (Suwa et al., 2000). Disruption of ORF69 (KS) and ORF70 (TE) gave no effect either. So, this gene cluster does not function or the type II polyketide metabolite could not be detected by the assay system we used.

Other biosynthetic gene clusters

At the left side of the mtm-like type II PKS cluster, several type I PKS genes (ORF56–ORF61) form a small cluster. ORF72 and ORF73 are also type I PKS genes. However, each PKS gene is too small to accommodate several functional domains for a large type I PKS, and the cluster is interrupted by insertion of the mtm-like type II PKS gene cluster. Moreover, disruption of ORF60, which only carries a KS domain, had no effect on antibiotic production. Therefore, these type I PKS genes may not function.

Homology search indicated that ORF104–ORF110 form a crt gene cluster for carotenoid biosynthesis. Seven crt genes (crtY, T, U, V, B, I, E  ) have been cloned and analysed from S. griseus (Schumann et al., 1996). Six of the seven crt genes except for crtT were identified on pSLA2-L, but their order (crtV, B, I, E, Y, U  ) is different from that in S. griseus.

Regulatory and resistance genes

During our effort to identify the lkc gene cluster preceding the sequencing project, we deleted a DNA region (nt 133 076–134 866) that contains ORF74, ORF75 and ORF76. Unexpectedly, deletion of this region caused non-production of both LC and LM. ORF74, ORF75 and ORF76 encode a γ-butyrolactone receptor-like protein (Horinouchi and Beppu, 1992), a SARP family regulatory protein (Wietzorrek and Bibb, 1997) and an NRPS-like protein respectively. Gene disruption of each of ORF74 and ORF75 revealed that only ORF75 is indispensable for the synthesis of both antibiotics. γ-Butyrolactone receptors contain a helix–turn–helix (HTH) DNA-binding motif and are grouped into the TetR family regulatory proteins (Hillen and Beren, 1994). SARP family regulatory proteins contain a helix–turn–helix–loop–helix DNA-binding motif and act as pathway-specific regulators in Streptomyces. In addition to ORF74 and ORF75, similar regulatory genes were found on pSLA2-L; tetR family repressor genes such as ORF79, ORF82, ORF92, ORF99 and ORF126, and SARP family regulatory genes such as ORF55 and ORF71. The conserved DNA-binding motifs of these regulatory proteins are compared with those of similar TetR and SARP regulatory proteins in Streptomyces(Fig. 4).

image

Figure 4. Amino acid alignments of the N-terminal regions of TetR family repressor proteins (A) and SARP family regulatory proteins (B). The α-helix regions in the DNA-binding motifs are underlined. ArpA in S. griseus (accession no. AB021882), BarA in S. virginiae (D32251), FarA in Streptomyces sp. (AB001683), SAR in S. ambofaciens (Y18862), TylS in S. fradiae (AF145049) and DnrI in S. peucetius (M80237).

Download figure to PowerPoint

In S. griseus, the A-factor regulatory cascade positively controls both streptomycin production and spore formation (Horinouchi and Beppu, 1992; Ohnishi et al., 1999). Namely, afsA is involved in the synthesis of A-factor, which binds to ArpA, a γ-butyrolactone receptor-type repressor, and releases it from the promoter of adpA. The AdpA protein, a global transcriptional regulator, binds to the promoter of strR, which codes a positive regulator for streptomycin biosynthesis (Beyer et al., 1996). All the homologues in this cascade were found on pSLA2-L; ORF85, ORF74, ORF116 and ORF3 are similar to afsA, arpA, adpA and strR respectively. This regulatory cascade may actually function in S. rochei, because disruption of the afsA-like ORF85 caused non-production of both LC and LM (Fig. 3). However, disruption of the strR-like ORF3 did not affect the production of LC or LM.

Another transcriptional regulatory gene is located at ORF101. Five antibiotic resistance genes were also found; ABC transporter genes at ORF9, ORF10 and ORF44, and efflux transporter genes at ORF20 and ORF91.

Replication and maintenance genes

We have recently cloned and sequenced a replication origin and the telomeres of pSLA2-L (Hiratsu et al., 2000). The telomere sequence is homologous to typical ones of Streptomyces chromosomes and linear plasmids (Huang et al., 1998). On the other hand, the replication origin (ori1), comprising repL1 (ORF95) and repL2 (ORF96), shows no similarity to the replication initiation genes of Streptomyces linear plasmids, pSLA2 (= pSLA2-S), pSCl1 (Chang et al., 1996) and SCP1 (Redenbach et al., 1999). The replication origin of SCP1 is composed of a pair of genes, a primase/helicase-like gene and a gene coding a protein with 130 amino acids. The complete nucleotide sequence of pSLA2-L identified two pairs of such genes close to ori1. ORF89 and ORF94 are similar to the primase/helicase-like gene, and ORF87 and ORF93 to the 130-amino-acid protein gene. The DNA fragments that contained each of these pairs, ORF87–ORF89 (nt 147 764–153 994) and ORF93–ORF94 (nt 154 338–159 297) were able to replicate in circular form in S. lividans (K. Hiratsu and H. Kinashi, unpublished) and were named ori3 and ori2 respectively (Fig. 2). It is not known which of the three replication origins actually functions in S. rochei. In addition, ORF102 is identical to the terminal protein gene (tpgR1) cloned from S. rochei (Bao and Cohen, 2001). ORF97 and ORF98 code for the parB and parA genes and may function in an efficient partition of the pSLA2-L DNA during cell division.

Possible generation mechanism of pSLA2-L

The complete nucleotide sequence of pSLA2-L has revealed many characteristic features of this large linear plasmid: (i) it contains three possible replication origins; (ii) a PQQ biosynthetic gene cluster has been found in Gram-positive bacteria; (iii) the lkc-PKS carries a fusion gene of NRPS and PKS and may be a mixture of the modular- and iterative-type PKSs; (iv) the loading module of the lkm-PKS may incorporate 2-methyl-3-hydroxybutyrate or 2-methylbutyrate as a starter; (v) the mtm-like type II PKS gene cluster contains a type I TE gene; (vi) pSLA2-L contains all the homologues in the A-factor regulatory cascade in S. griseus; and (vii) the biosynthetic gene clusters and genes related to secondary metabolism are present in an unusually condensed organization. The last is closely related to the long-standing question how the antibiotic biosynthetic genes have been horizontally transferred in prokaryotes and, in some cases, beyond the barrier between prokaryotes and eukaryotes.

Three replication origins are located at 50–63 kb from the right end, and most of the genes for secondary metabolism are located on their left side. Helicase-like ORFs were found at both inside ends of TIR-L (ORF1) and TIR-R (ORF143). ORF143 at the right end shows homology to the SCP1 helicase (SCP1.136; accession no. AL590463) over the entire region, whereas the C-terminal half of ORF1 at the left end is deleted just at the inside end of TIR-L. These results suggest that the biosynthetic gene clusters have been inserted into the left side of the original plasmid to generate the present form of pSLA2-L. A series of insertions might have truncated ORF1 at the left end and also have generated several small hypothetical genes, such as ORF54 and ORF111, at the border of the gene clusters. However, any traces of transposition events, such as direct repeats and transposase genes, have not been detected there.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References

Bacterial strains, cosmids and plasmids

A cosmid library was constructed for total DNA of S. rochei strain 51252 (Kinashi et al., 1994), a mutant that carries only one plasmid pSLA2-L, by using the vector Supercos1 and Escherichia coli Sure2, and then probed with the pSLA2-L DNA to obtain a sublibrary (Hiratsu et al., 2000). Forty cosmids in the pSLA2-L sublibrary were ordered by rounds of cross-hybridization using RNA end probes and, finally, six cosmids (B10, A1, 14F1, A8, C7 and A7) were aligned on a linear map as shown in Fig. 2. As the left and right end regions had previously been cloned in three plasmids, pE4, pHindE and pE3 (Kinashi et al., 1998), we had in hand DNA fragments covering the entire region of pSLA2-L for sequencing. The sequences of the left and right extreme ends have been reported recently (Hiratsu et al., 2000). E. coli XL1-Blue and pUC19 were used for cloning, nested deletion and sequencing.

Nucleotide sequencing and analysis

We used a combination of two strategies for sequencing: (i) shotgun clones of the ordered cosmids partially digested with Sau3AI; and (ii) nested exonuclease-III deletion clones of BamHI, EcoRI and KpnI fragments. Nucleotide sequences were determined by the dideoxy termination method using a model 4200 sequencer (LI-COR) and a Thermo Sequenase cycle sequencing kit (Amersham Pharmacia Biotech) or an ABI-373S sequencer (PE Biosystems) and DYEnamic Terminator cycle sequencing premix kit (Amersham Pharmacia Biotech). Sequence data were assembled by sequencher 3.1.1 (Gene Codes) and analysed by genetyx-mac 10.1 (Software Development, Tokyo) and the frameplot program (Ishikawa and Hotta, 1999; http:/http:www.nih.go.jpjuncgi-binframeplot.pl). Gene prediction was carried out based on homology search using the blastp program (Altschul et al., 1997) and the unique codon usage in Streptomyces (Bibb et al., 1984). Where possible, we chose an initiation codon (ATG, GTG, CTG or TTG) that is preceded by an upstream ribosome binding site. If this could not be identified, we chose the most upstream initiation codon.

DNA manipulation and gene disruption

Streptomyces total DNA was prepared by a neutral method as described previously (Suwa et al., 2000). The targeting vectors were constructed using a shuttle vector pRES18 (Ishikawa et al., 1996) and a kanamycin resistance gene cassette from pUC4-KIXX (Barany, 1985). They were propagated once in E. coli ET12567 (dam dcm hsdM; MacNeil et al., 1992) to overcome the restriction and modification barrier in Streptomyces, and introduced into S. rochei strain 51252 as described previously (Suwa et al., 2000). One randomly selected transformant, which was resistant to both kanamycin and thiostrepton, was continuously cultured in liquid YEME medium (Kieser et al., 2000) containing 10 µg ml−1 kanamycin to induce a second crossover for gene replacement. Protoplasts were prepared and regenerated to obtain single colonies.

TLC bioautography

Streptomyces rochei strains were reciprocally cultured in 100 ml of tryptic soy broth (TSB, Difco) medium in a 500 ml Sakaguchi flask at 28°C. The broth filtrates were extracted with ethyl acetate, applied to a silica gel thin-layer chromatography (TLC) plate (Kieselgel 60 F254; Merck) and developed with a mixture of chloroform and methanol (10:1). The TLC plate was placed in contact with a bioassay plate for 20 min and then incubated at 28°C overnight. The bioassay plate was composed of two layers. The bottom layer contained TSB medium−1% agar, while the top layer contained TSB medium−0.7% agar supplemented with 2% of the overnight culture in TSB medium of the indicator organism, Micrococcus luteus.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References

We thank A. Akita and T. Kurokawa for partial sequencing of pSLA2-L, and K. Sakaguchi for his encouragement from the start of this project. We also thank C. W. Chen for helpful suggestions on the manuscript.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References
  • Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., Miller, W., and Lipman, D.J. (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs, Nucleic Acids Res 25: 33893402.
  • August, P.R., Tang, L., Yoon, Y., Ning, J., Muller, S., Yu, R., et al. (1998) Biosynthesis of the ansamycin antibiotic rifamycin: deductions from the molecular analysis of the rif biosynthetic gene cluster of Amycolatopsis mediterranei S699. Chem Biol 5: 6979.
  • Bao, K., and Cohen, S.N. (2001) Terminal proteins essential for the replication of linear plasmids and chromosomes in Streptomyces. Genes Dev 15: 15181527.
  • Barany, F. (1985) Single-stranded hexameric linkers: a system for in-phase insertion mutagenesis and protein engineering. Gene 37: 111123.
  • Bentley, S.D., Chater, K.F., Cerdeno-Tarraga, A.-M., Challis, G.L., Thomson, N.R., James, K.D., et al. (2002) Complete genome sequence of the model actinomycete Streptomyces coelicolor A3(2). Nature 417: 141147.
  • Beyer, S., Distler, J., and Piepersberg, W. (1996) The str gene cluster for the biosynthesis of 5′-hydroxystreptomycin in Streptomyces glaucescens GLA.0 (ETH 22794): new operons and evidence for pathway-specific regulation by StrR. Mol Gen Genet 250: 775784.
  • Bibb, M.J., Findlay, P.R., and Johnson, M.W. (1984) The relationship between base composition and codon usage in bacterial genes and its use for the simple and reliable identification of protein-coding sequences. Gene 30: 157166.
  • Chang, P.C., and Cohen, S.N. (1994) Bidirectional replication from an internal origin in a linear Streptomyces plasmid. Science 265: 952954.
  • Chang, P.C., Kim, E.S., and Cohen, S.N. (1996) Streptomyces linear plasmids that contain a phage-like, centrally located, replication origin. Mol Microbiol 22: 789800.
  • Chater, K.F., and Bruton, C.J. (1985) Resistance, regulatory and production genes for the antibiotic methylenomycin are clustered. EMBO J 4: 18931897.
  • Chen, T.S., Inamine, E.S., Hensens, O.D., Zink, D., and Ostlind, D.A. (1989) Directed biosynthesis of avermectins. Arch Biochem Biophys 269: 544547.
  • Donadio, S., and Katz, L. (1992) Organization of the enzymatic domains in the multifunctional polyketide synthase involved in erythromycin formation in Saccharopolyspora erythraea. Gene 111: 5160.
  • Fernandez-Moreno, M.A., Martinez, E., Boto, L., Hopwood, D.A., and Malpartida, F. (1992) Nucleotide sequence and deduced functions of a set of cotranscribed genes of Streptomyces coelicolor A3(2) including the polyketide synthase for the antibiotic actinorhodin. J Biol Chem 267: 1927819290.
  • Gravius, B., Glocker, D., Pigac, J., Pandza, K., Hranueli, D., and Cullum, J. (1994) The 387 kb linear plasmid pPZG101 of Streptomyces rimosus and its interactions with the chromosome. Microbiology 140: 22712277.
  • Harada, S., Higashide, E., Fugono, T., and Kishi, T. (1969) Isolation and structures of T-2636 antibiotics. J Antibiot 27: 22392244.
  • Hillen, W., and Beren, C. (1994) Mechanism underlying expression of Tn10 encoded tetracycline resistance. Annu Rev Microbiol 48: 345369.
  • Hiratsu, K., Mochizuki, S., and Kinashi, H. (2000) Cloning and analysis of the replication origin and the telomeres of the linear plasmid pSLA2-L in Streptomyces rochei. Mol Gen Genet 263: 10151021.
  • Horinouchi, S., and Beppu, T. (1992) Autoregulatory factors and communication in actinomycetes. Annu Rev Microbiol 46: 377398.
  • Huang, C.-H., Lin, Y.-S., Yang, Y.-L., Huang, S.-W., and Chen, C.W. (1998) The telomeres of Streptomyces chromosomes contain conserved palindromic sequences with potential to form complex secondary structures. Mol Microbiol 28: 905916.
  • Huang, G., Zhang, L., and Birch, R.G. (2001) A multifunctional polyketide-peptide synthetase essential for albicidin biosynthesis in Xanthomonas albilineans. Microbiology 147: 631642.
  • Ishikawa, J., and Hotta, K. (1999) FramePlot: a new implementation of the frame analysis for predicting protein-coding regions in bacterial DNA with a high G+C content. FEMS Microbiol Lett 174: 251253.
  • Ishikawa, J., Niino, Y., and Hotta, K. (1996) Construction of pRES18 and pRES19, Streptomyces-Escherichia coli shuttle vectors carrying multiple cloning sites. FEMS Microbiol Lett 145: 113116.
  • Keller-Schierlein, W., and Roncari, G. (1964) Stoffwechselprodukte von Actinomyceten. 46 Mitteilung: Die Konstitution des Lankamycins. Helv Chim Acta 47: 78103.
  • Kieser, T., Bibb, M.J., Buttner, M.J., Chater, K.F., and Hopwood, D.A. (2000) Practical Streptomyces Genetics. Norwich: The John Innes Foundation.
  • Kinashi, H., and Shimaji-Murayama, M. (1991) Physical characterization of SCP1, a giant linear plasmid from Streptomyces coelicolor. J Bacteriol 173: 15231529.
  • Kinashi, H., Shimaji, M., and Sakai, A. (1987) Giant linear plasmids in Streptomyces which code for antibiotic biosynthesis genes. Nature 328: 454456.
  • Kinashi, H., Mori, E., Hatani, A., and Nimi, O. (1994) Isolation and characterization of large linear plasmids from lankacidin-producing Streptomyces species. J Antibiot 47: 14471455.
  • Kinashi, H., Fujii, S., Hatani, A., Kurokawa, T., and Shinkawa, H. (1998) Physical mapping of the linear plasmid pSLA2-L and localization of the eryAI and actI homologs. Biosci Biotech Biochem 62: 18921897.
  • Kirby, R., and Hopwood, D.A. (1977) Genetic determination of methylenomycin synthesis by the SCP1 plasmid of Streptomyces coelicolor A3(2). J Gen Microbiol 98: 239252.
  • Kirby, R., Wright, L.F., and Hopwood, D.A. (1975) Plasmid-determined antibiotic synthesis and resistance in Streptomyces coelicolor. Nature 254: 265267.
  • Kunst, F., Ogasawara, N., Moszer, I., Albertini, A.M., Alloni, G., Azevedo, V., et al. (1997) The complete genome sequence of the gram-positive bacterium Bacillus subtilis. Nature 390: 249256.
  • Leskiw, B.K., Lawlor, E.J., Fernandez-Abalos, J.M., and Chater, K.F. (1991) TTA codons in some genes prevent their expression in a class of developmental, antibiotic-negative, Streptomyces mutants. Proc Natl Acad Sci USA 88: 24612465.
  • MacNeil, D.J., Gewain, K.M., Ruby, C.L., Dezeny, G., Gibbons, P.H., and MacNeil, T. (1992) Analysis of Streptomyces avermitilis genes required for avermectin biosynthesis utilizing a novel integration vector. Gene 111: 6168.
  • Meulenberg, J.J.M., Sellink, E., Riegman, N.H., and Postma, P.W. (1992) Nucleotide sequence and structure of the Klebsiella pneumoniae pqq operon. Mol Gen Genet 232: 284294.
  • Musialoski, M.S., Flett, F., Scott, G.B., Hobes, G., Smith, C.P., and Oliver, S.G. (1994) Functional evidence that the principal DNA replication origin of the Streptomyces coelicolor chromosome is close to the dnaA-gyrB region. J Bacteriol 176: 51235125.
  • Nishizawa, T., Ueda, A., Asayama, M., Fujii, K., Harada, K., Ochi, K., and Shirai, M. (2000) Polyketide synthase gene coupled to the peptide synthetase module involved in the biosynthesis of the cyclic heptapeptide microcystin. J Biochem 127: 779789.
  • Ohnishi, Y., Kameyama, S., Onaka, H., and Horinouchi, S. (1999) The A-factor regulatory cascade leading to streptomycin biosynthesis in Streptomyces griseus: identification of a target gene of the A-factor receptor. Mol Microbiol 34: 102111.
  • Paitan, Y., Alon, G., Orr, E., Ron, E.Z., and Rosenberg, E. (1999) The first gene in the biosynthesis of the polyketide antibiotic TA of Myxococcus xanthus codes for a unique PKS module coupled to a peptide synthetase. J Mol Biol 286: 465474.
  • Pandza, S., Biukovic, G., Paravic, A., Dadbin, A., Cullum, J., and Hranueli, D. (1998) Recombination between the linear plasmid pPZG101 and the linear chromosome of Streptomyces rimosus can lead to exchange of ends. Mol Microbiol 28: 11651176.
  • Pospiech, A., Cluzel, B., Bietenhader, J., and Schupp, T. (1995) A new Myxococcus xanthus gene cluster for the biosynthesis of the antibiotic saframycin Mx1 encoding a peptide synthetase. Microbiology 141: 17931803.
  • Prado, L., Lombo, F., Brana, A.F., Mendez, C., Rohr, J., and Salas, J.A. (1999) Analysis of two chromosomal regions adjacent to genes for a type II polyketide synthase involved in the biosynthesis of the antitumor polyketide mithramycin in Streptomyces argillaceus. Mol Gen Genet 261: 216225.
  • Qin, Z., and Cohen, S.N. (1998) Replication at the telomeres of the Streptomyces linear plasmid pSLA2. Mol Microbiol 28: 893903.
  • Redenbach, M., Ikeda, K., Yamasaki, M., and Kinashi, H. (1998) Cloning and physical mapping of the EcoRI fragments of the giant linear plasmid SCP1. J Bacteriol 180: 27962799.
  • Redenbach, M., Bibb, M., Gust, B., Seitz, B., and Spychaj, A. (1999) The linear plasmid SCP1 of Streptomyces coelicolor A3(2) possesses a centrally located replication origin and shows significant homology to the transposon Tn4811. Plasmid 42: 174185.
  • Schumann, G., Nurnberger, H., Sandmann, G., and Krugel, H. (1996) Activation and analysis of cryptic crt genes for carotenoid biosynthesis from Streptomyces griseus. Mol Gen Genet 252: 658666.
  • Scrutton, N.S., Berry, A., and Perham, R.N. (1990) Redesign of the coenzyme specificity of a dehydrogenase by protein engineering. Nature 343: 3843.
  • Silakowski, B., Schairer, H.U., Ehret, H., Kunze, B., Weinig, S., Nordsiek, G., et al. (1999) New lessons for combinatorial biosynthesis from Myxobacteria. The myxothiazol biosynthetic gene cluster of Stigmatella aurantiaca DW4/3-1. J Biol Chem 274: 3739137399.
  • Stachelhaus, T., Mootz, H.D., and Marahiel, M.A. (1999) The specificity-conferring code of adenylation domains in nonribosomal peptide synthetases. Chem Biol 6: 493505.
  • Suwa, M., Sugino, H., Mori, E., Sasaoka, A., Fujii, S., Shinkawa, H., et al. (2000) Identification of two polyketide synthase gene clusters on the linear plasmid pSLA2-L in Streptomyces rochei. Gene 246: 123131.
  • Uramoto, M., Otake, N., Ogawa, Y., and Yonehara, H. (1969) The structures of bundlin A (lankacidin) and bundlin B. Tetrahedron Lett 27: 22492254.
  • Uramoto, M., Otake, N., Cary, L., and Tanabe, M. (1978) Biosynthetic studies with carbon-13. Lankacidin group of antibiotics. J Am Chem Soc 100: 36163617.
  • Wietzorrek, A., and Bibb, M. (1997) A novel family of proteins that regulates antibiotic production in streptomycetes appears to contain an OmpR-like DNA-binding fold. Mol Microbiol 25: 11811184.
  • Yang, C.C., Huang, C.H., Li, C.Y., Tsay, Y.G., Lee, S.C., and Chen, C.W. (2002) The terminal proteins of linear Streptomyces chromosomes and plasmids: a novel class of replication priming proteins. Mol Microbiol 43: 297305.
  • Zalkin, H., and Ebbole, D.J. (1988) Organization and regulation of genes encoding biosynthetic enzymes in Bacillus subtilis. J Biol Chem 263: 15951598.