Abstract
Previously, we showed that a 145-kb DNA region, including the entire kanamycin biosynthetic gene cluster (with two kanamycin resistance genes), was tandemly amplified up to 36-fold in an industrial strain of Streptomyces kanamyceticus. Strain improvement had included the use of increased kanamycin resistance as an initial potential indicator of higher kanamycin productivity. We were able to recapitulate the DNA amplification by cultivating S. kanamyceticus under selection for kanamycin resistance. To identify the genes required for amplification, various chromosome deletions were constructed, and the DNA amplification was shown to depend on orf1082 (zouA), present in a putative mobile genetic element. ZouA consists of 1,481 amino acids and is homologous to the products of traA-like genes of some conjugative plasmids. These genes encode relaxases that initiate DNA transfer during conjugation by single-strand nicking at oriT. As in the original high-producing strain, DNA amplification occurred between 16-nucleotide (nt) sites (RsA and RsB) containing 14 identical nucleotides. Interestingly, RsA lies just 80 bp upstream of the initiation codon of zouA and is partially contained in an inverted repeat structure similar to those found in plasmid oriT sequences, suggesting that it might function in a manner similar to that of oriT. We therefore propose that DNA amplification in S. kanamyceticus is initiated by relaxase-mediated recombination between oriT-related sequences.
INTRODUCTION
DNA amplification is a common phenomenon in many organisms, from bacteria to higher eukaryotes. When subjected to environmental stress, organisms may adapt by increasing the copy numbers of pertinent genes; examples include genes for resistance to antibiotics, anticancer drugs, and heavy metals such as copper and mercury (1, 20, 23). Moreover, additional copies of genes produced by DNA amplification are likely to play an important role in evolution (2), while gene amplification can also result in cancer (21).
Microorganisms produce an extensive variety of natural products (often referred to as secondary metabolites) with a wide range of biological activities that have important applications in medicine (e.g., as antibiotics, immunosuppressants, and anticancer agents) and in agriculture (e.g., as veterinary products and herbicides). The industrial-scale production of these compounds by fermentation usually requires the empirical isolation of high-producing mutants. The nature of such mutants is generally not known, but in some cases increased production is accompanied by the amplification of the corresponding biosynthetic gene clusters (6, 19, 25), suggesting that amplification may be one factor contributing to improved production.
Kanamycin is an aminoglycoside antibiotic made by Streptomyces kanamyceticus (24) and is one of the most commercially successful anti-infectives. Although the clinical use of kanamycin has been limited to the treatment of tuberculosis by the emergence of antibiotic resistance, the semisynthetic kanamycin derivatives amikacin, dibekacin, and arbekacin, which are effective against kanamycin-resistant pathogens, are used widely (13). In our previous work (25), we showed that a 145-kb DNA region (AUD [amplified unit of DNA]), including the entire kanamycin biosynthetic gene cluster and associated kanamycin resistance genes, was tandemly amplified in an industrial production strain (Fig. 1). Kanamycin production by this strain had been improved by iterative mutagenesis and screening, using the level of kanamycin resistance of the producing organism as a potential indicator of improved productivity. The AUD was located between the recombination sites RsA and RsB, which appear to consist of 16-nucleotide (nt) sequences with 14 identical nucleotides. Analysis of the new junction created by recombination between RsA and RsB revealed that recombination had occurred in the conserved hexanucleotide sequence 5′-TGGTCC-3′. The deletion of a 106.6-kb DNA region between sites RsC and RsD (incomplete direct repeats of 666 bp with 645 bp of identical sequence) accompanied the amplification. Interestingly, CRISPRs (clustered regularly interspaced short palindromic repeats) (11, 17), their associated Cas protein (CRISPR-associated protein) genes (11), and several genes homologous to those found in insertion sequence (IS) elements, conjugative plasmids, and bacteriophages are present in the region between RsC and RsD. Together with a GC content lower than those of flanking regions, this strongly suggests that the DNA between RsC and RsD is, or was, derived from a mobile genetic element. RsA, essential for DNA amplification, lies between RsC and RsD and was thus presumably not present in the ancestral kanamycin-producing strain.
Fig. 1.
Schematic representation of the S. kanamyceticus chromosome structures. The thick line indicates the AUD between RsA and RsB. S, SspI recognition site. (A) Region of the wild-type chromosome containing the recombination sites, which are indicated by RsA, RsB, RsC, and RsD. The positions of the CRISPR sequences (CRISPR 1 to CRISPR 4) and of the kanamycin biosynthetic gene cluster (Km cluster) are indicated. The position and direction of PCR primers used in this study are shown. The locations of the cosmid inserts are shown above. (B) Chromosome structure with duplicated AUD. (C) Chromosome structure with highly amplified AUD. n, multiple copies of the AUD.
There are two mechanisms for DNA amplification in bacteria (1). One is gene duplication and amplification. Duplications can arise by RecA-dependent unequal homologous recombination between tandem repeats present on the same or different chromosomes or by RecA-independent random end joining. Once duplication has occurred, higher-level amplification can occur by the RecA-dependent recombination of the longer perfect tandem repeats generated by the initial duplication. The other mechanism is rolling-circle replication, which involves the entrapment of a replication fork in a rolling circle through recombination (e.g., see reference 25) and can generate high-level amplification in a single round of DNA replication. Either mechanism might explain the DNA amplification found in the improved kanamycin-producing strain.
Here we describe a gene required for the amplification of a large DNA segment, including the entire kanamycin biosynthetic gene cluster, in S. kanamyceticus. We demonstrate that DNA amplification can be recapitulated by cultivating S. kanamyceticus under selective pressure for kanamycin resistance. Furthermore, we discovered that a closely linked gene encoding a putative relaxase resembling conjugal transfer proteins of some plasmids plays an essential role in DNA amplification.
MATERIALS AND METHODS
Strains and media.
S. kanamyceticus JCM4775 (wild type) and its derivative 12-6 (kanamycin-overproducing strain) (25) were obtained from the Meiji Seika Kaisha culture collection. S. kanamyceticus strains were cultivated in S liquid medium consisting of 3% corn steep liquor, 0.25% dry yeast, and 0.15% CaCl2 (pH 7.5). Escherichia coli strains DH5α [F− φ80d lacZΔM15 Δ(lacZYA-argF)U169 deoR recA endA1 hsdR17(rK− mK+) phoA supE44 λ− thi-1 gyrA96 relA1], JM109 [recA1 Δ(lac-proAB) endA1 gyrA96 thi hsdR17 supE44 relA1/F′ traD36 proAB lacIqZΔM15], and BW25113 [Δ(araD-araB)567 ΔlacZ4787(::rrnB-4) lacIp-4000(lacIq)λ− rpoS369(Am) rph-1 Δ(rhaD-rhaB)568 hsdR514] harboring pIJ790 (10) were used for plasmid constructions. E. coli ET12567 [F− dam-13::Tn9 dcm-6 hsdM hsdR zjj-202::Tn10 recF143 galK2 galT22 ara-14 lacY1 xyl-5 leuB6 thi-1 tonA31 rpsL136 hisG4 tsx-78 mtl-1 glnV44] harboring pUZ8002 was used for conjugations. E. coli strains were cultivated in LB medium consisting of 1% tryptone, 0.5% yeast extract, and 0.5% NaCl supplemented with suitable antibiotics.
Generation of strain RsAcos3.
A cosmid, AB201, in which the apramycin resistance gene [aac(3)-IV] and an EcoRV restriction site were inserted at the right end of the insert present in cosmid 4-5 (which included nt 1 to 33392 of GenBank accession no. AB254080 [25]), was constructed as follows. A 1.3-kb HindIII-EcoRI fragment containing the apramycin resistance gene and oriT was prepared from plasmid pIJ773 (10) and used as a template to generate a 1.4-kb DNA fragment by PCR amplification using primers RsA1U (5′-CACGGCACGGAATACCACTGCGTGCCCGTCGACGACGGTATTCCGGGGATCCGTCGACC-3′) and RsA1L (5′-CCAGGTCGGGAAGGGTGCTCTCCGCGCGAGCGGAGGTGATATCTTGATTTGAGAGGACCAAGGTATCGTCACTTCTGTAGGCTGGAGCTGCTTC-3′) (underlining and boldface type indicate the EcoRV site and 16 nt of RsA, respectively). E. coli BW25113/pIJ790 containing cosmid 4-5 was transformed with the 1.4-kb DNA fragment using E. coli Red/ET-mediated recombination (5), yielding cosmid AB201 (including nt 1 to 32510 of GenBank accession no. AB254080, the RsA site, the apramycin resistance gene, and an EcoRV site).
Cosmid AB202, in which the streptomycin resistance gene and a BsrG1 restriction site were generated at the left end of the insert of cosmid 5-13 (including nt 114645 to 146821 of GenBank accession no. AB254080), was constructed as follows. The 1.8-kb HindIII-EcoRI fragment containing the streptomycin resistance gene was prepared from plasmid pIJ778 (10) and used as a template to generate a 1.9-kb DNA fragment by PCR amplification using primers RsA2U (5′-CTCGCGCGGGAGCACCCCAGGCTGCCTGCAGAAAACTGTACATTCCGGGGATCCGTCGACC-3′ [the underlined sequence is a BsrG1 site]) and RsA2L (5′-AGTTCGCATCGCCCATCTAAGGAACTGGTGGGCCTTAGCTGTAGGCTGGAGCTGCTTC-3′). E. coli BW25113/pIJ790 containing cosmid 5-13 was transformed with the above-described 1.9-kb DNA fragment, and E. coli Red/ET-mediated recombination was used to generate cosmid AB202.
Cosmid AB201 was digested with BsrG1 (nt 16650 of GenBank accession no. AB254080) and EcoRV, and a ca. 16-kb BsrG1-EcoRV fragment was prepared. This fragment was inserted between the BsrG1 and EcoRV (nt 128993 of GenBank accession no. AB254080) sites of cosmid AB202, yielding cosmid 203-7. Cosmid 203-7 was introduced into E. coli ET12567/pUZ8002 (12), and the resulting strain was used for conjugation with S. kanamyceticus JCM4775. Apramycin-resistant S. kanamyceticus exconjugants were screened for double-crossover recombination by PCR, resulting in strain RsAcos3.
Generation of strain AB305cure.
A 3.37-kb DNA fragment (nt 47230 to 50602 of GenBank accession no. AB254080) was obtained by PCR using cosmid 2-1 (including nt 29373 to 75609 of GenBank accession no. AB254080) as a template and primers AfrU (5′-GGAGAAGCATGCGAGGACAAGTCGCGGCTTGAAC-3′ [the underlined sequence is an SphI site]) and AfrLRV (5′-CAGGCGGATCCCTGCGATATCCGTAGCGCGCATAAACGAAGAA-3′ [the underlined sequences are BamHI and EcoRV sites]). The PCR fragment was digested with BamHI and SphI and inserted into BamHI-plus-SphI-cleaved pUC118, yielding pAB301. A 3.98-kb DNA fragment (nt 87961 to 91943 of GenBank accession no. AB254080) was obtained by PCR using cosmid 1-3 (including nt 65356 to 99182 of GenBank accession no. AB254080) as a template and primers BfrU (5′-GCAGATGGATCCAGAGTCTAGATTCAGCTCGTTGATCACCATGTC-3′ [the underlined sequences are BamHI and XbaI sites]) and BfrL (5′-CAGGCGAATTCCGCGTGGAATCGCTCCGCATCTT-3′ [the underlined sequence is an EcoRI site]). The PCR fragment was digested with BamHI and EcoRI and inserted into BamHI-plus-EcoRI-cleaved pUC118, yielding pAB302. The BamHI-EcoRI fragment from pAB302 was inserted into BamHI-plus-EcoRI-cleaved pAB301, yielding pAB303. A 1-kb BclI fragment containing the thiostrepton resistance gene, tsr, from pIJ702 (12) was cloned into the BamHI site of pUC118, yielding pUC118tsr. pUC118tsr was digested with XbaI and SmaI, and the resulting XbaI-SmaI fragment containing tsr was inserted between the EcoRV and XbaI sites of pAB303, yielding pAB304. pAB304 was digested with SphI and EcoRI, and the resulting SphI-EcoRI fragment was inserted into SphI-plus-EcoRI-cleaved pSET152 (3), yielding pAB305.
pAB305 was introduced into E. coli ET12567/pUZ8002, and the resulting strain was used for conjugation with S. kanamyceticus JCM4775. The resulting apramycin-resistant S. kanamyceticus exconjugants were screened by PCR to yield single-crossover strain AB305. AB305 was grown in liquid medium and plated for single colonies. Forty-eight apramycin-sensitive colonies were obtained from 5,400 colonies, and seven colonies were also thiostrepton resistant. The replacement of nt 50603 to 87960 of GenBank accession no. AB254080 by tsr was confirmed by PCR, and the resulting strain was named AB305cure.
Generation of strain M27.
A 1.4-kb DNA fragment containing the apramycin resistance gene and oriT was amplified by PCR using the 1.3-kb HindIII-EcoRI fragment from pIJ773 as a template and primers 97682U (5′-TCTTCTGTCGTCTCATCCATCGTGCTGGCCTTCGATGACATTCCGGGGATCCGTCGACC-3′) and 120181L (5′-GGGAAAGTACGGGAAAAGATCTCGGTTACTCGCGATCCATGTAGGCTGGATCTGCTTC-3′). E. coli BW25113/pIJ790 containing cosmid 3-7 (including nt 90866 to 124998 of GenBank accession no. AB254080) was transformed with the 1.4-kb DNA fragment using E. coli Red/ET-mediated recombination, yielding cosmid 3-7::AB402. Cosmid 3-7::AB402 was introduced into E. coli ET12567/pUZ8002, and the resulting E. coli strain was used for conjugation with S. kanamyceticus strain AB305cure. PCR was used to screen the resulting apramycin-resistant S. kanamyceticus exconjugant colonies for double-crossover recombination, resulting in strain M27.
Generation of strain M29.
A cosmid library of M27 genomic DNA was constructed in SuperCos 1 (Stratagene). Cosmid 1-10 was selected from colonies resistant to both ampicillin and apramycin and confirmed by nucleotide sequence analysis of both ends of the cloned insert to contain the NdeI site at nt 29213 of GenBank accession no. AB254080 and the AflII site at nt 139611 of GenBank accession no. AB254080. To facilitate replacement of the NdeI-AflII fragment of cosmid 203-7 with the NdeI-AflII fragment of cosmid 1-10, the apramycin resistance gene of cosmid 203-7 was replaced with a streptomycin resistance gene. A 1.9-kb DNA fragment containing the streptomycin resistance gene was amplified by PCR using the 1.8-kb HindIII-EcoRI fragment from pIJ778 as a template and primers RsA1Ussp (5′-CACGGCACGGAATACCACTGCGTGCCCGTCGACGACAATATTCCGGGGATCCGTCGACC-3′) and RsA1LRV (5′-CAGACTCTGAGTGATATCTTGATTTGAGAGGACCAAGGTTGTAGGCTGGAGCTGCTTC-3′). E. coli BW25113/pIJ790 containing cosmid 203-7 was transformed with the 1.9-kb DNA fragment using E. coli Red/ET-mediated recombination, yielding cosmid 203-7::str. Cosmid 1-10 was digested with NdeI, AflII, and DraI and ligated with cosmid 203-7::str that had been digested with NdeI and AflII. The ligation product was packaged into λ phage heads in vitro and mixed with E. coli XL1-Blue, and transductants were isolated on LB agar medium containing ampicillin and apramycin. Cosmid AB501 was selected, and the replacement of the NdeI-AflII fragment was confirmed by PCR. However, restriction enzyme analysis of cosmid AB501 indicated the occurrence of an unexpected deletion during plasmid construction. Nucleotide sequence analysis revealed that the unexpected deletion extended from nt 120621 to nt 139619 of GenBank accession no. AB254080 and that cosmid AB501 consisted of nt 16650 to 29218, 87961 to 97640, 120062 to 120620, and 139620 to 146821 of GenBank accession no. AB254080, with tsr between nt 16650 to 29218 and nt 87961 to 97640 and the apramycin resistance gene between nt 87961 to 97640 and nt 120062 to 120620. Cosmid AB501 was introduced into E. coli ET12567/pUZ8002, and the resulting E. coli strain was used for conjugation with S. kanamyceticus JCM4775. One hundred apramycin-resistant S. kanamyceticus exconjugants were examined for susceptibility to apramycin and thiostrepton, and two colonies showed the required thiostrepton resistance. PCR analysis of genomic DNA confirmed that both strains had undergone double-crossover recombination between nt 16650 to 29218 and nt 139620 to 146821. One of the strains was named M29.
Generation of strain AB113-2.
A 1.4-kb DNA fragment containing the apramycin resistance gene was amplified by PCR using the 1.3-kb HindIII-EcoRI fragment from pIJ773 as a template and primers M13U (5′-GGAGCACTTGCCGGTCTGGCCCAGAACGCGGACGCCGTCATTCCGGGGATCCGTCGACC-3′) and M13L (5′-AGAGCAGTCAGGCTGGCAACCGCACATCCACGCGATCGTTGTAGGCTGGAGCTGCTTC-3′). E. coli BW25113/pIJ790 containing cosmid 5-13 was transformed with the 1.4-kb DNA fragment using E. coli Red/ET-mediated recombination, yielding cosmid 5-13::AB113. Cosmid 5-13::AB113 was introduced into E. coli ET12567/pUZ8002, and the resulting strain was used for conjugation with S. kanamyceticus JCM4775. PCR was used to screen the resulting apramycin-resistant S. kanamyceticus exconjugants for double-crossover recombination, and strain AB113-2 was isolated.
Generation of strain AB1-3⑧.
A 1.4-kb DNA fragment containing the apramycin resistance gene was amplified by PCR using the 1.3-kb HindIII-EcoRI fragment from pIJ773 as a template and primers M8U (5′-TCAAGACCTCCGATACGGGCTTCTGTGCCGTTCAGTCGAATTCCGGGGATCCGTCGACC-3′) and M8L (5′-CAACGCCGTCGACCTCTACGGCGAGGACACGGTGGAGAATGTAGGCTGGAGCTGCTTC-3′). E. coli BW25113/pIJ790 containing cosmid 1-3 was transformed with the 1.4-kb DNA fragment using E. coli Red/ET-mediated recombination, yielding cosmid 1-3::AB108. Cosmid 1-3::AB108 was introduced into E. coli ET12567/pUZ8002, and the resulting strain was used for conjugation with S. kanamyceticus JCM4775. PCR screening of the resulting apramycin-resistant S. kanamyceticus exconjugants for double-crossover recombination resulted in strain AB1-3⑧, with nt 90730 to 93870 internal to orf1082 (nt 90183 to 94628 of GenBank accession no. AB254080) deleted.
Construction of pKM2003.
Two oligonucleotides, Sma-Stu-1 (5′-GGGAGGCCTA-3′) and Sma-Stu-2 (5′-AGCTTAGGCCTCCC-3′), were annealed and inserted into the HindIII-SmaI sites of pUC119, yielding pUC119-Stu. The 6.6-kb SmaI fragment (nt 88479 to 95063 of GenBank accession no. AB254080) from cosmid 1-3 was inserted into the SmaI site of pUC119-Stu, and the orientation of the insert was determined by KpnI digestion, yielding pKM2001, in which the KpnI site (nt 94889 of GenBank accession no. AB254080) was located close to the HindIII site of pUC119-Stu. The 4.1-kb StuI fragment (nt 135493 to 139615 of GenBank accession no. AB254080) from cosmid 5-13 was inserted into the StuI site of pKM2001, yielding pKM2002, in which the orientation of the StuI fragment was the same as that of the SmaI fragment. The 10.7-kb HindIII-EcoRI fragment of pKM2002 was ligated with the 2.8-kb HindIII-EcoRI fragment of pSET153 (constructed from pSET152 by replacing the SphI site with a HindIII site by linker ligation), yielding pKM2003.
RESULTS
Specific DNA amplification occurs reiteratively in S. kanamyceticus.
We first investigated whether the DNA amplification previously observed for S. kanamyceticus, which included the entire kanamycin biosynthetic gene cluster, could be recapitulated. S. kanamyceticus JCM4775 was grown in S liquid medium containing 250 μg/ml kanamycin for 48 h at 28°C, and part of the culture was then transferred into S liquid medium containing 2,000 μg/ml kanamycin and further cultivated for 48 h at 28°C. Genomic DNA samples were prepared from the mycelium obtained from each culture, and PCR, using primers KM-16′ and KM-17′ (see Table 1 for all diagnostic PCR primers), was performed to detect the new junction generated by recombination between RsA and RsB. The appearance of this new junction fragment would reflect amplification (minimally duplication) of the DNA segment between RsA and RsB. Genomic DNA from S. kanamyceticus 12-6, in which DNA amplification was first shown to occur (25), was used as a positive control. Although no amplified band was observed when genomic DNA prepared from the mycelium cultivated without kanamycin was used as a template, a 1.2-kb band was amplified in the genomic DNA samples prepared from the mycelium cultivated with either 250 μg/ml or 2,000 μg/ml of kanamycin, with the latter yielding a band of higher intensity (Fig. 2 A). Nucleotide sequence analyses of these PCR fragments confirmed that recombination had occurred in the hexanucleotide sequence 5′-TGGTCC-3′, which is conserved in RsA and RsB, as found previously for the genomic DNA of S. kanamyceticus 12-6 (25). This experiment was repeated, and the 1.2-kb band was always and only detected when genomic DNAs from the mycelium cultivated with kanamycin were used as templates. Thus, we were able to demonstrate that the DNA amplification discovered in an industrial strain of the kanamycin producer S. kanamyceticus could indeed be recapitulated under controlled laboratory conditions by applying selection for kanamycin resistance.
Table 1.
Diagnostic PCR primers used in this study
| Primer | Sequence | Direction | Location (nt)a |
|---|---|---|---|
| KM-16′ | CCGGCACTTCCGCTCCAA | Forward | 6139–6156 |
| KM-17′ | GCGGGTTCGCCAACTCCA | Reverse | 95793–95810 |
| KM-18′ | CTCGACAAGGTCTGCAAGCC | Forward | 5275–5294 |
| KM-25 | CCGCTCTCATTCGGTCAG | Forward | 94644–94661 |
| KM-36 | ACAGGAACAGCGAGGGATAG | Reverse | 136450–136469 |
| KM-37 | TCTGCTCACCTCTGCGTCAG | Forward | 28231–28250 |
| KM-201 | CCATCCCGTCGAAGAGCC | Reverse | 136081–136098 |
| KM-202 | CCCCTGACTTTCGTCGAG | Reverse | 139639–139656 |
| KM-203 | TCTGCGTCAGCCACGGAC | Forward | 28241–28258 |
| M19′L | ATCTTGATTTGAGAGGACCA | Reverse | 94708–94727 |
Fig. 2.
Detection of the new recombination junction by PCR. Lane 1, 200-bp DNA ladder (the brightest band is 1 kb) used as molecular size markers; lanes 2 to 5, PCR products derived from genomic DNA templates from S. kanamyceticus JCM4775 cultivated without kanamycin (lane 2) and with 250 μg/ml (lane 3) or 2,000 μg/ml (lane 4) of kanamycin and from S. kanamyceticus 12-6 (lane 5). (A) The RsA/RsB junction (1.2-kb amplified product) was detected using primers KM-16′ and KM-17′. (B) The natural RsA site (no recombination, yielding a 1.2-kb amplified product) was detected using primers KM-25 and KM-17′. (C) The RsC/RsD junction (926-bp amplified product) was detected by using primers KM-36 and KM-37.
PCR using KM-25 and KM-17′ as primers can detect the natural RsA site (which would remain if recombination and amplification did not occur); however, these primers failed to amplify the natural RsA when using genomic DNA from S. kanamyceticus 12-6 as a template (Fig. 2B). Thus, S. kanamyceticus 12-6 contained only chromosomes in which the DNA segment between RsA and RsB was amplified. Conversely, since the amplified natural RsA band was clearly detected when genomic DNAs from S. kanamyceticus JCM4775 cultivated with kanamycin were used, these strains harbor a mixture of chromosomes, some with and others without amplification between RsA and RsB.
In S. kanamyceticus 12-6, the deletion of the DNA segment between RsC and RsD accompanied the amplification of the DNA region between RsA and RsB. This deletion can be detected by the PCR amplification of a 926-bp fragment using primers KM-36 and KM-37. This 926-bp fragment was not amplified when genomic DNA from the mycelium of S. kanamyceticus JCM4775 grown without kanamycin was used as a template but appeared when genomic DNA from the mycelium of the same strain cultivated with either 250 μg/ml or 2,000 μg/ml of kanamycin was used; the intensity of the band from the genomic DNA derived from the culture grown with 2,000 μg/ml of kanamycin was again higher than that from the sample cultivated with 250 μg/ml of kanamycin (Fig. 2C). These results indicated that DNA amplification mediated by recombination between RsA and RsB was reproducibly linked to the deletion of the DNA segment between RsC and RsD and that the degree of amplification increased in parallel with the concentration of kanamycin present in the culture medium.
Identification of the gene required for DNA amplification.
We next tried to identify genes required for amplification. RsA, essential for amplification, lies between RsC and RsD. Given the lower GC content of the RsC-RsD region (68.4 mol%) than that of the flanking sequences (left side, 71.8 mol%; right side, 72.5 mol%), and the presence of homologues of genes found in IS elements, conjugative plasmids, and bacteriophages (i.e., transposase, type IV secretory pathway protein, and integrase genes) within it, it seems likely that the DNA between RsC and RsD is, or was, derived from a mobile genetic element. CRISPRs and CRISPR-associated protein genes (cas) are also present in this region (25) and could potentially be involved in DNA amplification. To test this, strain RsAcos3 was generated (Fig. 3), in which most (nt 32511 to 128992 of GenBank accession no. AB254080) of the DNA present between RsC and RsD (nt 28935 to 135581 of GenBank accession no. AB254080) was deleted but in which a 34-bp DNA segment (nt 94693 to 94726 of GenBank accession no. AB254080) containing RsA was retained. RsAcos3 was cultivated in liquid medium containing kanamycin, genomic DNA was isolated, and PCR amplification was attempted using primers KM-18′ (located 0.9-kb upstream of RsB) and M19′L. Recombination between RsA and RsB would result in the amplification of a 937-bp band. However, no such band was observed even when genomic DNA prepared from the mycelium grown with 2,000 μg/ml of kanamycin as a template was used. This suggested that the gene(s) required for DNA amplification was present in the deleted segment (nt 32511 to 128992 of GenBank accession no. AB254080).
Fig. 3.
Schematic representation of the chromosome structure of constructed deletion mutants and their ability to form the RsA/RsB junction. Dotted lines indicate deleted DNA regions. +, formation of an RsA/RsB junction; −, no formation of an RsA/RsB junction.
S. kanamyceticus strains deleted for different segments of the RsC-RsD region were next constructed (Fig. 3): AB305cure (deletion of nt 50603 to 87960 of GenBank accession no. AB254080), M27 (deletion of nt 50603 to 87960 and 97641 to 120061 of GenBank accession no. AB254080), M29 (deletion of nt 29219 to 87960, 97641 to 120061, and 120621 to 139619 of GenBank accession no. AB254080), and AB113-2 (deletion of nt 118626 to 130558 of GenBank accession no. AB254080). All of these strains were able to amplify the DNA region between RsA and RsB, indicating that the 9.68-kb DNA region (nt 87961 to 97640 of GenBank accession no. AB254080) containing RsA also carried a gene or genes required for DNA amplification.
Of eight open reading frames (orf1079 to orf1086) present in this DNA region, only the product of orf1082 showed homology to proteins known to interact directly with DNA. Thus, an orf1082 gene disruptant (strain AB1-3⑧) was constructed and cultivated with kanamycin. PCR analysis using genomic DNA of AB1-3⑧ as a template, and KM-16′ and KM-17′ as primers, failed to amplify the 1.2-kb band indicative of amplification, indicating that orf1082 is essential for the amplification of the DNA segment between RsA and RsB.
orf1082 encodes a 1,481-amino-acid protein with 39% and 42% sequence identities to TraA-like proteins encoded by conjugative Streptomyces plasmids pFP11 and pFP1, respectively (27). Moreover, a conserved-domain search (15, 16) revealed TrwC superfamily (cI08490) and Tra_Ti (TIGR02768) domains, both implicated in conjugation, located between amino acid residues 350 to 580 and 680 to 1100 of Orf1082, respectively. TraA-like proteins encode relaxases that initiate DNA transfer during conjugation by single-strand nicking at oriT (4, 18). While there is no proven role for TraA proteins in conjugation in Streptomyces thus far (27), there is one in conjugation in the related actinomycete Rhodococcus erythropolis AN12 (26), and it is conceivable that Orf1082 might also function as a relaxase. Interestingly, the conserved hexanucleotide sequence of RsA lies just 80 bp upstream of the initiation codon of orf1082, and an inverted repeat sequence similar to those found in other oriT sites encompasses part of RsA (Fig. 4). Thus, RsA may function in a fashion analogous to that of oriT. Since orf1082 is essential for the DNA amplification observed for S. kanamyceticus, we named it “zouA” (derived from zouhuku, which means amplification in Japanese).
Fig. 4.
Nucleotide sequences around RsA and RsB. Underlining indicates the 16-nt sequences of RsA and RsB. The conserved hexanucleotides are shown in boldface type. Arrows indicate inverted repeats. Dotted lines in arrows indicate nonidentical nucleotides in the inverted repeats.
Reconstitution of the DNA amplification in the RsC-RsD-deleted strain.
In our previous study, strain 12-6-4, which carries only one copy of the kanamycin biosynthetic gene cluster, was isolated through homologous recombination of the amplified DNA segment from a high-kanamycin-producing strain, strain 12-6 (25). Strain 12-6-4 had lost the ability to amplify DNA since it lacks zouA and RsA (contained in the deleted 106.6-kb DNA region between RsC and RsD). We tried to restore the ability of strain 12-6-4 to amplify DNA by introducing a plasmid containing zouA and RsA.
pKM2003, with the 6.6-kb SmaI fragment (nt 88479 to 95063 of GenBank accession no. AB254080) containing zouA and RsA and the 4.1-kb StuI fragment (nt 135493 to 139615 of GenBank accession no. AB254080) containing RsD and its downstream region, was introduced into strain 12-6-4 by conjugation. A resulting transconjugant (12-6-4/pKM2003) was confirmed by PCR to have arisen through the chromosomal integration of pKM2003 by single crossover in the StuI fragment (Fig. 5 A). Strain 12-6-4/pKM2003 was cultivated with and without kanamycin, and genomic DNA was prepared from the resulting mycelium. To detect recombination between RsA and RsB, PCR analysis was carried out using primers KM-16′ and KM-201; a 1,020-bp amplification product would be indicative of recombination between RsA and RsB. When genomic DNA from strains 12-6 and 12-6-4/pKM2003 grown in the absence of kanamycin was used as templates, no 1,020-bp band was observed. However, the 1,020-bp amplified product was readily detected when genomic DNA from 12-6-4/pKM2003 grown with kanamycin was used as a template (Fig. 5B). Nucleotide sequence analysis of this band indicated that recombination had occurred in the hexanucleotide sequence (5′-TGGTCC-3′) conserved between RsA and RsB.
Fig. 5.
Schematic representation of chromosome structure after the introduction of pKM2003 by homologous recombination (A) and detection of the new recombination junction by PCR (B, C, and D). (A) The thin line shows the DNA derived from the pSET152 vector. The position and direction of primers used for PCR analyses are shown. (B) PCR detection of the RsA/RsB junction (1,020-bp amplified band) using primers KM-16′ and KM-201. Lane 1, 200-bp DNA ladder (the brightest band is 1 kb) used as molecular size markers; lanes 2 to 5, PCR analysis using genomic DNA templates from S. kanamyceticus 12-6 (lane 2), S. kanamyceticus 12-6-4/pKM2003 cultivated without (lane 3) or with (lane 4) kanamycin, and S. kanamyceticus 12-6-4/pKM2003(Ams) (lane 5). (C) PCR detection of the natural RsA site (no recombination, resulting in a 1,031-bp amplified band) was detected using primers KM-25 and KM-201. Lane 1, 200-bp DNA ladder (the brightest band is 1 kb) used as molecular size markers; lanes 2 to 5, PCR analysis using genomic DNA templates from S. kanamyceticus 12-6 (lane 2), S. kanamyceticus 12-6-4/pKM2003 cultivated without (lane 3) or with (lane 4) kanamycin, and S. kanamyceticus 12-6-4/pKM2003(Ams) (lane 5). (D) Deletion of the vector DNA by homologous recombination between the 4.1-kb StuI fragments (resulting in a 4.1-kb amplified band) was detected using primers KM-202 and KM-203. Lane 1, λ-HindIII was used as a molecular size marker; lanes 2 to 5, PCR analysis using genomic DNA templates from S. kanamyceticus 12-6 (lane 2), S. kanamyceticus 12-6-4/pKM2003 cultivated without (lane 3) or with (lane 4) kanamycin, and S. kanamyceticus 12-6-4/pKM2003(Ams) (lane 5).
PCR analysis using primers KM-25 and KM-201 would give a 1,031-bp amplified band if RsA from pKM2003 did not recombine with RsB or any other site. No 1,031-bp band was observed when genomic DNA from strain 12-6 was used as a template as a negative control, but it was observed clearly with genomic DNA from 12-6-4/pKM2003 regardless of whether or not it was grown in the presence of kanamycin (Fig. 5C). This indicated that 12-6-4/pKM2003 cultivated with kanamycin harbors a mixture of chromosomes, as shown previously for JCM4775 when grown with the same antibiotic. It was suggested above that in JCM4775, DNA amplification accompanied the deletion of the DNA segment between RsC and RsD. If this was the case for 12-6-4/pKM2003, the DNA segment between the 4.1-kb StuI fragments should be deleted with the DNA amplification, and a chromosome including the amplified DNA should lose the apramycin resistance gene derived from pKM2003. Therefore, we postulated that a strain containing only chromosomes with the amplified DNA could be obtained by screening for apramycin-sensitive (Ams) colonies derived from strain 12-6-4/pKM2003 cultivated with kanamycin. A liquid culture of 12-6-4/pKM2003 grown with kanamycin was plated onto agar medium, and several hundred single colonies were screened for susceptibility to apramycin, resulting in the isolation of strain 12-6-4/pKM2003(Ams). Genomic DNA was isolated from 12-6-4/pKM2003(Ams) and analyzed by PCR. An amplified band of the expected size was observed when primers KM-16′ and KM-201 were used (Fig. 5B) but not when primers KM-25 and KM-201 were used (Fig. 5C), indicating that 12-6-4/pKM2003(Ams) contained only chromosomes with the amplified DNA. In another PCR experiment, KM-203, located upstream of RsC, and KM-202, located downstream of the 4.1-kb StuI fragment, were used as primers. A 4.1-kb amplified band was observed when genomic DNA from strains 12-6 and 12-6-4/pKM2003(Ams) was used as a template but not when DNA from 12-6-4/pKM2003 cultivated with kanamycin was used (Fig. 5D), suggesting that the ratio of the number of chromosomes with the amplified DNA to the number of those lacking it was very low. Genomic DNA from 12-6-4/pKM2003(Ams) was digested with SspI, which does not cut the amplified unit of DNA, and separated by pulsed-field gel electrophoresis. Southern blot analysis revealed a ladder of bands (data not shown), indicating that 12-6-4/pKM2003(Ams) contained a mixture of chromosomes in which the copy number of the amplified unit of DNA varied, as found previously for strain 12-6 (25).
DISCUSSION
In this study, we showed that the amplification of the DNA segment between RsA and RsB found previously for an industrial kanamycin-producing strain of S. kanamyceticus (strain 12-6) could be recapitulated by cultivating S. kanamyceticus JCM4775 (wild type) in the presence of kanamycin. Moreover, we demonstrated that orf1082 was essential for DNA amplification and named it zouA.
zouA lies in the DNA region between RsC and RsD that appears to be, or to have arisen from, a mobile genetic element. ZouA contains TrwC relaxase and TraA_Ti domains, suggesting that it might function similarly to relaxases involved in the conjugation of some plasmids. In addition, RsA was present immediately upstream of the initiation codon of orf1082 and partly contained in an inverted repeat similar to those found for oriT sequences (18). Thus, RsA is likely to be a cognate target for the ZouA relaxase.
A relaxase is a single-strand DNA transesterase (4). In some conjugative plasmids, such relaxases, together with other proteins, assemble at a site, oriT, to form a DNA-protein complex called the relaxosome. The relaxase cleaves one of the oriT DNA strands by reversible transesterification to generate a protein-DNA covalent intermediate. A tyrosine in relaxase joins to the DNA by a phosphotyrosyl linkage. The plasmid DNA is presumably recircularized after transfer following the transesterification of the protein-DNA covalent intermediate by the 3′ OH at the end of the transferred strand. If the reaction of the intermediate occurs with the 3′ OH at the end of another DNA molecule, recombination between the two DNA molecules will occur. For the relaxases known to date, there have been reports describing oriT-specific recombination by TraI from the E. coli F plasmid (22), TrwC from IncW plasmid R388 (14), TraX from plasmid pAD1 of Enterococcus faecalis (8), and NikB from plasmid R64 (9). In addition, there is an example of DNA amplification in which both the relaxase and oriT are involved (7). pAMα1 of E. faecalis contains two oriT types, oriTB and oriTE; the 4.1-kb DNA segment between these two sites, which includes a tetracycline resistance gene, is tandemly duplicated by a mechanism that depends on a relaxase encoded on the same plasmid. After duplication, the copy number of the segment between oriTB and oriTE can be increased in a RecA-dependent manner by increasing the concentration of tetracycline in the culture medium.
ZouA is presumed to nick one strand of duplex DNA at RsA, forming a protein-DNA complex intermediate by covalently binding to the 5′ end of the cleaved strand via a phosphotyrosyl linkage, and to generate the new recombination junction RsA/RsB by transferring it to the strand in the nicked RsB site through transesterification. Although the tyrosine residue covalently bound to the DNA strand was identified for the TrwC and TraI relaxases, we do not know which tyrosine residue in ZouA might be involved in covalent binding to the DNA strand. RsB lies in orf25, outside the kanamycin biosynthetic gene cluster (25). orf25 encodes a protein belonging to the TetR family of transcriptional regulators, and highly homologous proteins occur in many Streptomyces spp. Thus, it seems unlikely that RsB is a natural, cognate oriT for ZouA. It is perhaps more likely that RsB (Fig. 4) is an oriT-like sequence recognized fortuitously, and perhaps inefficiently, by ZouA but detectable by selecting for kanamycin resistance.
The relaxosome usually consists of a relaxase and other proteins. It remains to be determined whether ZouA exerts transesterification activity in a strand- and site-specific manner alone or with other proteins. When strain 12-6-4 was used as a host, the 6.6-kb SmaI fragment (nt 88479 to 95063 of GenBank accession no. AB254080) containing zouA and RsA was sufficient for DNA amplification. There are only two complete genes, zouA and orf1081, in the SmaI fragment. orf1081 encodes a 201-amino-acid protein homologous to metal-dependent phosphohydrolases, but there is no precedent for the involvement of such enzymes in the relaxosome. However, host proteins from distantly located S. kanamyceticus genes might also be involved in the formation of the relaxosome.
The reproducible amplification of the DNA segment between RsA and RsB in S. kanamyceticus JCM4775 was followed by the deletion of the DNA segment between RsC and RsD. When strain 12-6-4 was used to generate the AUD, the DNA segment derived from the integration of pKM2003 into the chromosome was also deleted by homologous recombination between the direct repeats of the 4.1-kb StuI fragment, followed by recombination between RsA and RsB. These results are consistent with a rolling-circle model in which the entrapment of a replication fork in a rolling circle resulting from recombination between RsA and RsB would result in the amplification of the DNA segment between RsA and RsB. This appears more likely as a mechanism of DNA amplification by ZouA rather than a model based on unequal crossing over, in which misaligned pairing and recombination between RsA and RsB on different chromosomes or sister chromatids would generate a duplication of the DNA segment between RsA and RsB; it seems more likely that homologous recombination between RsC and RsD, or between the two tandem 4.1-kb StuI fragments, was used for the resolution of the rolling-circle structure.
ACKNOWLEDGMENTS
We thank Toyomi Sato for helpful discussion and encouragement in executing this research and David Hopwood and Keith Chater for comments on the manuscript.
M.B. acknowledges financial support from the United Kingdom Biotechnology and Biological Sciences Research Council.
Footnotes
Published ahead of print on 4 February 2011.
REFERENCES
- 1. Andersson D. I., Hughes D. 2009. Gene amplification and adaptive evolution in bacteria. Annu. Rev. Genet. 43:167–195 [DOI] [PubMed] [Google Scholar]
- 2. Bergthorsson U., Andersson D. I., Roth J. R. 2007. Ohno's dilemma: evolution of new genes under continuous selection. Proc. Natl. Acad. Sci. U. S. A. 104:17004–17009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Bierman M., et al. 1992. Plasmid cloning vectors for the conjugal transfer of DNA from Escherichia coli to Streptomyces spp. Gene 116:43–49 [DOI] [PubMed] [Google Scholar]
- 4. Byrd D. R., Matson S. W. 1997. Nicking by transesterification: the reaction catalysed by a relaxase. Mol. Microbiol. 25:1011–1022 [DOI] [PubMed] [Google Scholar]
- 5. Datsenko K. A., Wanner B. L. 2000. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci. U. S. A. 97:6640–6645 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Fierro F., et al. 1995. The penicillin gene cluster is amplified in tandem repeats linked by conserved hexanucleotide sequences. Proc. Natl. Acad. Sci. U. S. A. 92:6200–6204 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Francia M. V., Clewell D. B. 2002. Amplification of the tetracycline resistance determinant of pAMα1 in Enterococcus faecalis requires a site-specific recombination event involving relaxase. J. Bacteriol. 184:5187–5193 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Francia M. V., Clewell D. B. 2002. Transfer origins in the conjugative Enterococcus faecalis plasmids pAD1 and pAM373: identification of the pAD1 nic site, a specific relaxase and a possible TraG-like protein. Mol. Microbiol. 45:375–395 [DOI] [PubMed] [Google Scholar]
- 9. Furuya N., Komano T. 2003. NikAB- or NikB-dependent intracellular recombination between tandemly repeated oriT sequences of plasmid R64 in plasmid or single-stranded phage vectors. J. Bacteriol. 185:3871–3877 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Gust B., Challis G. L., Fowler K., Kieser T., Chater K. F. 2003. PCR-targeted Streptomyces gene replacement identifies a protein domain needed for biosynthesis of the sesquiterpene soil odor geosmin. Proc. Natl. Acad. Sci. U. S. A. 100:1541–1546 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Jansen R., van Embden J. D. A., Gaastra W., Schouls L. M. 2002. Identification of genes that are associated with DNA repeats in prokaryotes. Mol. Microbiol. 43:1565–1575 [DOI] [PubMed] [Google Scholar]
- 12. Kieser T., Bibb M. J., Buttner M. J., Chater K. F., Hopwood A. D. 2000. Practical Streptomyces genetics. John Innes Foundation, Norwich, United Kingdom [Google Scholar]
- 13. Kondo S., Hotta K. 1999. Semisynthetic aminoglycoside antibiotics: development and enzymatic modifications. J. Infect. Chemother. 5:1–9 [DOI] [PubMed] [Google Scholar]
- 14. Llosa M., Bolland S., Grandoso G., de la Cruz F. 1994. Conjugation-independent, site-specific recombination at oriT of the IncW plasmid R388 mediated by TrwC. J. Bacteriol. 176:3210–3217 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Marchler-Bauer A., Stephen H. B. 2004. CD-Search: protein domain annotations on the fly. Nucleic Acids Res. 32:W327–W331 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Marchler-Bauer A., et al. 2008. CDD: specific functional annotation with the Conserved Domain Database. Nucleic Acids Res. 37:D205–D210 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Mojica F. J. M., Díez-Villaseñor C., Soria E., Juez G. 2000. Biological significance of a family of regularly spaced repeats in the genomes of archaea, bacteria and mitochondria. Mol. Microbiol. 36:244–246 [DOI] [PubMed] [Google Scholar]
- 18. Parker C., Becker E., Zhang X., Jandle S., Meyer R. 2005. Elements in the co-evolution of relaxases and their origins of transfer. Plasmid 53:113–118 [DOI] [PubMed] [Google Scholar]
- 19. Peschke U., Schmidt H., Zhang H.-Z., Piepersberg W. 1995. Molecular characterization of the lincomycin-production gene cluster of Streptomyces lincolnensis 78-11. Mol. Microbiol. 16:1137–1156 [DOI] [PubMed] [Google Scholar]
- 20. Sandegren L., Andersson D. I. 2009. Bacterial gene amplification: implications for the evolution of antibiotic resistance. Nat. Rev. Microbiol. 7:578–588 [DOI] [PubMed] [Google Scholar]
- 21. Schwab M. 1998. Amplification of oncogenes in human cancer cells. Bioessays 20:473–479 [DOI] [PubMed] [Google Scholar]
- 22. Sherman J. A., Matson S. W. 1994. Escherichia coli DNA helicase I catalyzes a sequence-specific cleavage/ligation reaction at the F plasmid origin of transfer. J. Biol. Chem. 269:26220–26226 [PubMed] [Google Scholar]
- 23. Stark G. R., Wahl G. M. 1984. Gene amplification. Annu. Rev. Biochem. 53:447–491 [DOI] [PubMed] [Google Scholar]
- 24. Umezawa H., et al. 1957. Production and isolation of a new antibiotic, kanamycin. J. Antibiot. 10:181–188 [PubMed] [Google Scholar]
- 25. Yanai K., Murakami T., Bibb M. 2006. Amplification of the entire kanamycin biosynthetic gene cluster during empirical strain improvement of Streptomyces kanamyceticus. Proc. Natl. Acad. Sci. U. S. A. 103:9661–9666 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Yang J. C., et al. 2007. TraA is required for megaplasmid conjugation in Rhodococcus erythropolis AN12. Plasmid 57:55–70 [DOI] [PubMed] [Google Scholar]
- 27. Zhang R., Zeng A., Fang P., Qin Z. 2008. Characterization of replication and conjugation of Streptomyces circular plasmids pFP1 and pFP11 and their ability to propagate in linear mode with artificially attached telomeres. Appl. Environ. Microbiol. 74:3368–3376 [DOI] [PMC free article] [PubMed] [Google Scholar]