Previous Article | Next Article ![]()
Applied and Environmental Microbiology, April 2004, p. 2508-2513, Vol. 70, No. 4
0099-2240/04/$08.00+0 DOI: 10.1128/AEM.70.4.2508-2513.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Division of Bioscience and Biotechnology, Department of Ecological Engineering, Toyohashi University of Technology, Toyohashi, Aichi 441-8580,1 Mitsubishi Kagaku Institute of Life Sciences, Machida-shi, Tokyo 194-8511, Japan2
Received 20 October 2003/ Accepted 17 December 2003
|
|
|---|
|
|
|---|
In contrast to these studies of the integration process, investigation of the reverse process has been limited to two experimental approaches (9, 19). One approach includes dissection of part of the B. subtilis genome to physically separate a 300-kb genomic segment that becomes an independent replicon or a 300-kb subgenome (9, 10). The other approach involves the copy and transfer of the cloned heterologous DNAs to an independent replicon by the Bacillus recombinational transfer mechanism (BReT) (19). The mechanism of BReT-mediated transfer is indicated in Fig. 1. It requires the two sequences called landing pad sequences (LPSs), where homologous recombination occurs during B. subtilis transformation. Intervening DNA between the two LPSs is copied and transferred to the BReT plasmid. Initially, the BReT system was developed to recover the heterologous DNAs cloned into the two halves of pBR322 sequences of the B. subtilis genome vector (19). The pBReT plasmid pGETS103 has the two halves of pBR322 sequences that are used as the two LPSs to designate the cloned DNA (19). It was demonstrated that pGETS103 stably maintains DNA segments of up to 100 kb due to its
-type replicative form for B. subtilis (2, 19).
![]() View larger version (35K): [in a new window] |
FIG. 1. Transfer of genomic segments into plasmids by recombinational transfer. Target segments of the genome are shadowed. The pBReT vector comprised pGETS113 (bold lines) and LPSs (parts of the open arrow). Recombinational transfer proceeds by homologous recombination between LPSs of the genome and the pBReT vector. The pBReT plasmid replicates as a single copy in the recombinant. The two genome loci are enlarged. A dark rhomb indicates the location of menA. Construction of the pBReT vector for the pps operon is described in Table 1. pBReT vectors for the other genome locus are represented by that for the 24.6-kb segment. The pBR322 sequence of pGETS1203 drawn by two lines is eliminated upon linearization.
|
A region carrying the genes with known function was first examined. B. subtilis Marburg strain produces a fungicidal lipopeptide, plipastatin, the peptide moiety of which is synthesized in a nonribosomal manner by the five peptide synthetases encoded by ppsA, ppsB, ppsC, ppsD, and ppsE (18). The five genes residing in the 38.2-kb-long pps operon (18) are ppsA (7.7 kb), ppsB (7.7 kb), ppsC (7.7 kb), ppsD (10.8 kb), and ppsE (3.9 kb). Cloning of this large DNA segment is not reported. Sequence information was downloaded from http://www.Pasteur.fr/Bio/SubtiList.html (13). Two LPSs, one of 1.9 kb (designated [LPS-P]) and one of 2.0 kb (designated [LPS-S]), were prepared by PCR-mediated amplification using primer sets listed in Table 1. Insertion of [LPS-P] and [LPS-S] into pGETS113 in the correct order and orientation was done as indicated in Fig. 1 and Table 1. The pBReT vector, pGETS113-ppsAE, was obtained.
|
View this table: [in a new window] |
TABLE 1. Plasmids for recombinational transfer
|
(mrr-hsdRMS-mcrBC)
80lacZ
M15
lacX74 recA1 deoR araD139
(ara-leu)7697 galU galK rpsL(Strr) endA1 nupG] supplied by Invitrogen Inc. (Carlsbad, Calif.). Subcloning to construct pBReT vectors was done by using E. coli JA221 (F hsdR hsdM+ trp leu lacY recA1), and JM109 [F endA1 recA1 gyrA96 thi hsdR17(rK mK+) relA1 supE44
(lac-proAB)(F' traD36 proAB lacIq lacZ
M15)] was routinely used as a host for molecular cloning. Luria-Bertani broth was used for all bacteria at 37°C. Tetracycline (10 µg/ml) and ampicillin (50 µg/ml) were added for JA221 or JM109 selection. Manipulation of the plasmid in E. coli was done according to the established method (16). A large amount of plasmid was purified with a CsCl gradient formed by ultracentrifugation in the presence of ethidium bromide (16). Restriction enzymes were obtained from Toyobo (Tokyo, Japan). Blunting of DNA fragments and ligation were done using the TaKaRa blunting kit and TaKaRa DNA ligation kit (Takara Shuzo, Kyoto, Japan). B. subtilis 168 trpC2 (1A1) was from the Bacillus Genetic Stock Center (Columbus, Ohio). The competent 1A1 cell was prepared by the two-step culture method developed by Anagnostopoulos and Spizizen (1) with modifications as described previously (8). [LPS-P] and [LPS-S] of pGETS113-ppsAE were physically separated by complete NotI digestion prior to B. subtilis transformation. Only plasmid that copied the intervening 38.2-kb DNA by the BReT mechanism was established as a circular plasmid and rendered B. subtilis resistant to chloramphenicol (Fig. 1). One microgram of pGETS113-ppsAE yielded 143 chloramphenicol-resistant colonies selected at the 5-µg/ml concentration. Plasmids prepared from four colonies by the minipreparation method were identical in their SfiI fragments, and a representative plasmid was referred to as pGETS113-ppsABCDE. The structure of the 38.2-kb DNA segment plus [LPS-P] and [LPS-S] in pGETS113-ppsABCDE was checked using 11 restriction enzymes, BglII, EcoRI, EcoRV, HindIII, KpnI, SacI, SalI, SmaI, PstI, PvuII, and NotI (data not shown). The number and sizes of the fragments yielded by restriction enzyme digestion were all identical to those predicted using nucleotides of the pps locus (1,959,404 to 1,997,591) (13). Examples of HindIII and BglII digests are presented in Fig. 2A. To examine the gene function of the copied segment, the pGETS113-ppsABCDE plasmid was used to transform BUSY8519, a mutant that lacked the pps operon. Plipastatin production was fully restored in the chloramphenicol-resistant transformant as shown in Fig. 2B. These results proved that the five pps genes of pGETS113-ppsABCDE are functional and that no structural alteration was introduced during the BReT process, regardless of the high level of repetition in sequences in the pps operon (18) such as the 2.8-kb segments of ppsA and ppsC (97% identity in nucleotide sequences), the 1.8-kb segments (97% identity) of ppsD and ppsA, and the 1.7-kb segments (97% identity) of ppsC and ppsD.
![]() ![]() View larger version (107K): [in a new window] |
FIG. 2. (A) pBReT plasmids containing the pps operon. pGETS113-ppsAE (lane 1) and pGETS113-ppsABCDE prepared from B. subtilis (lane 2) and from E. coli JM109 (lane 3) were digested by the indicated restriction enzyme. Lambda/HindIII marker fragments were run in lane M with their sizes indicated on the left. Fragments shown by open arrowheads came from the internal region of the pBReT vector pGETS113-ppsAE. The number and sizes of the other fragments were identical to those predicted from a sequence data file constructed with nucleotide sequences of pGETS113 and of the pps locus (1,959,404 to 1,997,591) (13). (B) Complementation of plipastatin production by the pps operon of the pBReT plasmid. B. subtilis strain BEST8628 (plipastatin producer) and BUSY8519 (plipastatin nonproducer) are our laboratory stock. Samples were prepared according to the method described in reference 18 from the culture in 40 ml of ACS medium in 200-ml Erlenmeyer flasks at 30°C and 120 strokes/min. Three-day cultures were used for BEST8628 and BUSY8519, and four-day cultures were used for plasmid transformants. The indicated high-performance liquid chromatography (HPLC) peaks represent plipastatin stereoisomers within lipid and peptide moieties (18) and surfactins. Conditions for reverse-phase HPLC were the same as those described in reference 18.
|
To prove applicability to other internal DNAs, positional cloning of a 100-kb segment from another genome locus, 3,932,990 to 4,033,322 (13), was examined. First, cloning of the 24.6-kb region including an essential gene, menA (3,949,750 to 3,950,682), encoding 14-dihydroxy-2-naphthoate octaprenyltransferase (12) was attempted using the pBReT vector pGETS1203. pGETS1203 was constructed by insertion of [LPS-A] and [LPS-B(24.6)] into pGETS113 as indicated in Fig. 1 and Table 1. Transformation of 1A1 by pGETS1203 (1 µg) linearized with NotI yielded 10 chloramphenicol-resistant colonies. The plasmids isolated from these 10 recombinants had identical structures with respect to digestion by EcoRI, BamHI, and BglII. The cloned 24.6-kb segment of the representative pGETS1208 was confirmed using restriction endonucleases EcoRV, KpnI, SacI, SalI, SfiI, SmaI, PstI, PvuII, and NotI (data not shown except for PvuII in Fig. 3A). To examine positional cloning of a larger portion of this region, four pBReT vectors that designate 46.6, 58.8, 88.7, and 100.3 kb of this genome locus were separately constructed as described in Table 1. Transformation experiments with strain 1A1 were done in triplicate using 1 µg of pBReT vectors linearized with NotI. The average numbers of chloramphenicol-resistant transformants were as follows: 11 ± 0.6 with pGETS1203, 28 ± 3.5 with pGETS1204, 5 ± 1.2 with pGETS1205, 8 ± 2.1 with pGETS1207, and 11 ± 1.2 with pGETS1202. All the plasmids isolated from each pBReT transformant had identical structures (data not shown). Insert structures of the representative pBReT plasmids pGETS1209 (46.6 kb), pGETS1210 (58.8 kb), pGETS1212 (88.7 kb), and pGETS1201 (100.3 kb) were confirmed using several restriction endonucleases (data not shown except for PvuII in Fig. 3A). Figure 3B indicates that the 100.3-kb segment cloned in pGETS1201 detected no alteration of Southern signals of genome digests by PstI, PvuII, EcoRI, BglII, and SacI. Transfer of these positionally cloned segments to E. coli remained to be examined because of the lack of a pBR322 sequence in these pBReT plasmids. It was established by adopting a low-copy-number pBReT vector that the BReT mechanism also functioned for positional cloning of the internal DNA of the B. subtilis genome. The nearly equal numbers of chloramphenicol-resistant transformants indicate that the BReT proceeds at equal rates regardless of the DNA length to be copied. As discussed previously, the rate-determining step for the BReT is not the uptake-replication process but the process of formation of simultaneous homologous pairs (19). This conclusion is consistent with the similar degrees of efficiency of the cloning of the internal 100.3-kb in this study and the previous cloning of the 90.6-kb heterologous DNA originating from the Cyanobacterium genome into the B. subtilis genome (19). The successful positional cloning of internal DNA facilitates precise cloning of DNA fragments such as lysogenic phages that could not be obtained with other cloning systems.
![]() ![]() View larger version (144K): [in a new window] |
FIG. 3. (A) Positionally cloned inserts from pBReT plasmids. The indicated pBReT plasmids (cloned sizes [in kilobases] are in parentheses) were digested by PvuII and separated by contour-clamped homogeneous electric field (CHEF) gel electrophoresis. Running conditions for CHEF electrophoresis are given at the bottom. Lambda/HindIII marker fragments were run in lane M with their sizes indicated on the right. Left panel, gel stained with ethidium bromide. Right panel, Southern bands obtained using a mixture of probes made by pGETS1201 (100.3 kb) and lambda DNA digested with HindIII. The use of mixed probes established in reference 7 made lambda markers visible for direct comparison. The lambda/HindIII marker bands are seen in lane M. The PvuII fragments of the smaller pBReT plasmids are all hybridized by the longest insert (100.3 kb) of pGETS1201. Open arrowheads indicate two fragments containing the pGETS113 part divided by the internal PvuII site. (B) Identification of the positionally cloned region. DNA of 1A1 digested by the indicated restriction enzymes was separated by CHEF electrophoresis. Running conditions for CHEF are given at the bottom. Lambda/HindIII marker fragments were run in lane M with their sizes indicated on the right. Southern bands using the same mixed probe pGETS1201 (100.3 kb) and lambda DNA are consistent with those deduced from the sequence data (13).
|
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»