- Email: [email protected]
Fate of transforming bacterial genome following incorporation into competent cells of Bacillus subtilis: a continuous length of incorporated DNA
JOURNAL OF BIOSCIENCE AND BIOENGINEERING Vol. 101, No. 3, 257–262. 2006 DOI: 10.1263/jbb.101.257
© 2006, The Society for Biotechnology, Japan
Fate of Transforming Bacterial Genome Following Incorporation into Competent Cells of Bacillus subtilis: a Continuous Length of Incorporated DNA Yukiko Saito,1 Hisataka Taguchi,1 and Takashi Akamatsu1* Department of Applied Microbial Technology, Faculty of Biotechnology and Life Science, Sojo University, Ikeda 4-22-1, Kumamoto 860-0082, Japan1 Received 2 November 2005/Accepted 27 December 2005
In contrast to the conventional transformation of Bacillus subtilis using purified DNA, those using DNA in lysed protoplasts have a high transformation efficiency and enable whole-genome transfer into competent B. subtilis [Akamatsu, T. and Taguchi, H., Biosci. Biotechnol. Biochem., 65, 823–829 (2001)]. Here, we examined the length of incorporated continuous DNA by analyzing the cotransfer ratio with selected and unselected markers, on the basis of a new experimental design. The cotransfer ratio of a selected marker with an unselected marker on the opposite side of the genetic map of the B. subtilis chromosome was about 5.6% and could be interpreted as congression (double transformation) ratio. In the wild-type strain, the cotransfer ratio of cysA (113 kb position on 4215 kb of B. subtilis chromosome) with metC (1384 kb) and leuB (2891 kb) was 0.77%, twice the value (5.6% × 5.6% = 0.31%) calculated from the congression ratio. Moreover, in a genetic background, the cotransfer ratios of metC with cysA and leuB, and metC with cysA and arg1 (3012 kb) were 2.7% and 7.2%, respectively. These results strongly suggest that the length of continuous DNA incorporated into B. subtilis is most probably greater than 1271 kb. When the DNA from the protoplast lysate was fragmented by mixing, the cotransfer ratios of arg1 with metC, and arg1 with metC and trpC (2374 kb) were 2.8% and 0.16%, respectively. A high cotransfer ratio (2.7–7.2%) could not, therefore, be obtained using the fragmented DNA. Based on these observations, we propose a working hypothesis on the mechanism of the transformation of competent B. subtilis by DNA in protoplast lysates (LP transformation). [Key words: Bacillus subtilis, competent, transformation, DNA uptake, continuous length]
Bacillus subtilis is a gram-positive bacterium capable of developing competence and being transformed with purified DNA (1–3). Double-stranded DNA is bound to a competent-cell surface (4). The limited cleavage of DNA by nuclease occurs, yielding fragments of 15–30 kb. About half of the DNA is degraded by a competence-specific nuclease and the remaining single-stranded DNA is then transported to the cytosol. The DNA taken up interacts with the recipient DNA to form a D-loop and recombines with a homologous region to yield a heteroduplex recombinant molecule. Finally, the expression of donor genetic information, the segregation of the heteroduplex, and replication occur resulting in the formation of a transformant clone (5, 6). A large number of proteins required for transformation have been identified in B. subtilis (6–12). All of them are regulated at the transcriptional level and depend on the comK transcriptional factor for their synthesis. ComG and ComC proteins form the DNA-binding apparatus, whereas ComE and ComF proteins form the DNA transport machinery. An attractive working hypothesis has been proposed for
such transformation process and for other Com proteins in B. subtilis (13). In contrast to conventional transformation methods using purified DNA, those using DNA in lysed protoplasts result in a high transformation efficiency (14–17). Transformants are obtained at a frequency of one transformant per ten genomes, 200–1000-fold higher than that of conventional methods. Since lysed-protoplast transformation (LP transformation) also depends on the comK transcriptional factor (1), both transformation mechanisms seem to be basically similar. Linkage analysis between two markers with a distance of 1673 kb leads to a constant cotransfer ratio (4–5%), even when the number of donor DNA molecules is apparently 60-fold lower than the number of competent cells (1, 17). Thus, the uptake of 4215 kb of whole-genome DNA into competent B. subtilis is one of the most significant features of LP transformation. From the linkage relationship between two markers, the length of continuous DNA taken up into B. subtilis could not be evaluated, because the cotransformation ratio sharply decreased with increasing marker distance of the chromosome up to 121 kb and showed a plateau from 121 kb to 1628 kb (4–5% cotransfer ratio) (1). This observation indi-
* Corresponding author. e-mail: [email protected] phone: +81-(0)96-326-3929 fax: +81-(0)96-323-1330 257
258
J. BIOSCI. BIOENG.,
SAITO ET AL.
cates two possibilities. First, the genome taken up is fragmented into many pieces (for example, about twenty pieces, 200-kb average fragment length) and, therefore, the cotransformation linkage between two markers separated by 100 kb can only be observed. Second, although the DNA taken up is very long (several fragmented pieces of genome DNA), the recombination of the DNA with a recipient chromosome occurs only within a short distance (100 kb). We developed a theoretical design evaluating the length of incorporated continuous DNA by comparison of two cotransfer ratios with selected and unselected markers (cotransformation and congression) to discriminate between the two possibilities. In this study, we describe the incorporation of a continuous length of DNA of more than 1271 kb that corresponds to about one-third of the whole-genome length, and propose a working hypothesis on the mechanism of the transformation of competent B. subtilis by DNA in protoplast lysates (LP transformation).
MATERIALS AND METHODS Bacterial strains The bacterial strains used in this study are listed in Table 1. B. subtilis AU1 (cysA metC hisH leuB arg1 ist1 ist2) is an ist1 ist2 mutant derivative with a high frequency of interspecific transformation by a long heterologous DNA fragment in protoplast lysates of B. amyloliquefaciens and a high frequency of cotransformation, as will be described elsewhere in detail. Media Luria–Bertani (LB) medium (19) was used for bacterial growth. The LB agar medium contained 15 g of agar in 1 l of LB medium. Spizizen minimal medium (20) (SM medium) was used as a basal medium for the preparation of competent cells. For the isolation of auxotrophs or transformants, low Spizizen minimal medium (LSM medium) (19) was used, supplemented with amino acids, when necessary. Preparation of B. subtilis protoplasts Protoplasts were prepared basically as described by Akamatsu and Sekiguchi (21, 22). B. subtilis 168S (trpC rpsL smo1) was cultivated overnight on LB agar medium at 37°C and suspended in LB medium. An aliquot of the bacterial suspension (100–1000 cells) was inoculated into 25 ml of LB medium and cultivated overnight at 37°C. The overnight culture was added to 25 ml of LB medium at an inoculation ratio of 1% and incubated with shaking for 3 h at 37°C. The culture was again added to 25 ml of LB medium at an inoculation ratio of 1% and incubated for 2–3 h. When the cell growth reached the mid-log to early stationary phase, cells from 20 ml of the culture were harvested by centrifugation (7200×g for 1–2 min) and suspended in 4 ml of SMM (0.5 M sucrose–0.02 M maleate buffer [pH 6.5]– 0.02 M MgCl2) containing lysozyme at a final concentration of 250 mg per liter. The total mixture was gently shaken for 15– 45 min at 40°C as a shallow layer (e.g., 4 ml of suspension in a 100-ml Erlenmeyer flask). Three milliliters of the suspension was TABLE 1. B. subtilis strains used in this study Strain
Genotype
AYG2 AU1 HH4 168S AC820
cysA14 metC3 trpC2 leuB8 arg1 aroG932 cysA14 metC3 hisH leuB8 arg1 ist1 ist2 cysA14 metC3 hisH leuB8 arg1 trpC2 rpsL smo1 hisH rpsL smo1
Reference, source or derivation 1 This study 1 17 18
transferred into a centrifuge tube (Nalgen Oak Ridge centrifuge tube 3118-0050; Nalge Nunc International, Rochester, New York, USA) containing 10 ml of SMM and centrifuged (4000×g for 5– 10 min). The pellet was suspended in SMM buffer. Protoplast density was measured from absorbance at 660 nm (A660). An A660 of 2.88 corresponded to 2.74 ×109 protoplasts per ml. After dilution with SMM buffer, an aliquot (0.1 ml) of protoplast suspension was used as the source of donor DNA. Preparation of competent cells Competent cells were prepared basically as described by Anagnostopoulos and Spizizen (23). B. subtilis AYG2 was cultivated overnight on LB agar medium at 37°C and suspended in LB medium. An aliquot of the bacterial suspension (100–1000 cells) was inoculated into 25 ml of LB medium and cultivated overnight at 37°C. The overnight culture was added to 25 ml of LB medium at an inoculation ratio of 1% and incubated with shaking for 3 h at 37°C. Cells from 20 ml of the culture were harvested by centrifugation (7200×g for 1–2 min) and suspended in an aliquot (5 ml) of medium I (Spizizen minimal medium supplemented with 200 mg/l casamino acids and, when necessary, 50 mg/l amino acids). Absorbance at 660 nm was measured and the culture was added to medium I in a 100-ml Erlenmeyer flask (initial A660 = 0.33, 10 ml total culture) and cultivated for 4 h at 37°C. Two milliliters of culture was added to 18 ml of medium II (Spizizen minimal medium supplemented with 100 mg/l casamino acids and, when necessary, 5 mg/l amino acids), and cultivated for 1.5 h at 37°C. One milliliter of the competent cell culture was used for transformation. Preparation of fragmented DNA solution An aliquot (0.1 ml) of a protoplast suspension was added to 1 ml of hypotonic medium II. DNA in the protoplast lysates was then fragmented by mixing (30 s, maximal power, twice) using a Vortex Genie mixer (Scientific Industries, New York, NY, USA) and used as donor DNA, when necessary. Transformation with DNA obtained by gentle lysis of protoplasts (LP transformation) An aliquot (0.1 ml) of a protoplast suspension was added to 1 ml of hypotonic competent-cell culture, and the mixture of DNA in protoplast lysates and competent cells was incubated at 37°C for 30 min. Then the cells were plated out onto minimal agar medium with appropriate nutrients. After singlecolony isolation, each colony, when necessary, was further transferred to diagnostic agar plates and unselected markers were identified.
RESULTS Experimental design for evaluation of the length of continuous DNA incorporated into competent B. subtilis The transformation of B. subtilis by DNA involves the following steps: (i) the binding of incorporated DNA to the recipient chromosome, (ii) the recombination of the DNA, (iii) the expression of donor information, and (iv) the formation of a transformant clone (6). Although 4215 kb of the whole genome of donor B. subtilis can be incorporated into competent B. subtilis (1), the accurate length of fragments greater than 100 kb is not known. Continuous DNA length can be evaluated by analyzing the difference in efficiency between congression (double transformation) and cotransformation (Fig. 1). In the case in which A+ transformants of recipient cells with markers a, b, and c are examined for the unselected markers (b, c), let P (A, B) or P (A, C) be the probability that markers A and B, or A and C will be coinherited. Let P (A, B, C) be the probability that A, B and C will be coinherited. Three types of coinheritance of the three
VOL. 101, 2006
FATE OF TRANSFORMING BACTERIAL GENEOME
259
FIG. 1. Evaluation of the length of continuous donor DNA incorporated into competent Bacillus subtilis. Case 1: Let P (A, B), P (A, C), or P (A, B, C) be the probability that markers A and B, A and C, or A and B and C will be coinherited. P (A, B, C) will be calculated by P (A, B) ×P (A, C) and is nearly equal to P (A, C) ×P (A, C), providing that the A, B, and C markers are on separate fragments and the fragmentation of DNA (for example, into 20 pieces) results in a linkage between markers being linked by distances of 100 kb. Case 2: Donor DNA in a recipient cell is fragmented into several pieces only, and marker C is on a different DNA fragment from that carrying A and B markers. P (A, B, C) is then expected to be greater than P (A, C) ×P (A, C).
markers are possible: congression, cotransformation, or congression and cotransformation. When A+ B+ C+ transformants are obtained by congression (Fig. 1, case 1), P (A, B, C) can be calculated by P (A, B) ×P (A, C) and is almost equal to P (A, C) ×P (A, C). On the other hand, P (A, B, C) is expected to be higher than P (A, C) ×P (A, C), providing that A and B are on the same DNA fragment and C is on a different DNA fragment. The P (A, B) due to cotransformation is higher than that of P (A, C) due to congression, because congression requires the independent bindings of two DNA fragments and recombination with the recipient chromosome, whereas cotransformation requires the binding of one DNA fragment and recombination (Fig. 1, case 2). When genetic marker C is located on the opposite side of A and a higher efficiency of P (A, B, C) than that of P (A, C)×P (A, C) is obtained, this indicates that A and B are on the same DNA fragment. This interpretation results in the evaluation of the length of continuous DNA corresponding to the distance between A and B. The continuous length of DNA is greater than one million bases To determine congression ratio, pairs of genetic markers were chosen on the opposite side of the B. subtilis chromosome. metC is located at 118° of the chromosome (24). leuB, arg1, and aroG are at 247°, 257°, and 260°, respectively (Fig. 2). The Met+ transformants of competent B. subtilis AYG2 (cysA metC trpC leuB arg1 aroG) generated by using DNA in the protoplast lysate of B. subtilis 168S (trpC rpsL smo1) were selected and examined for Aro+, Arg+, and Leu+. The coinheritance ratios of the double markers were 5.7%, 5.7%, and 5.7%, respectively. These results strongly suggest that the coinherited ratio is a congression ratio. This was confirmed by determining the coinherited ratios of aroG to metC, arg1 to metC, and cysA (9.7°) to hisH (202°). The ratios were 5.7%, 5.7%, and 5.1%, respectively. All these observations lead to the conclusion that a value of 5.6% represents a congression ratio, corresponding to the P (A, C) in Fig. 1.
FIG. 2. Genetic map of Bacillus subtilis chromosome. The B. subtilis genome has 4,214,841 bases (24). One degree, therefore, corresponds to about 11.7 kb. The number in the circle indicates the map position (degrees) on the B. subtilis chromosome. The distance of the origin to cysA, rpsL, metC, hisH, trpC, leuB, arg1 or aroG is 113 kb, 130 kb, 1384 kb, 2370 kb, 2374 kb, 2891 kb, 3012 kb or 3045 kb, respectively.
Cys+ transformants of B. subtilis AYG2 generated by using DNA in the protoplast lysate of B. subtilis AC820 (hisH rpsL smo1) were selected and analyzed for the phenotype of unselected markers (Table 2). Cys+ Met+ Aro+ recombinants were obtained at a 1.0% ratio of Cys+ recombinants. If three markers are on separate DNA fragments, a ratio of Cys+ Met+ Aro+ to Cys+ of 0.31% (5.6% × 5.6%) is expected, which is clearly lower than the observed value of 1.0%. The observed value strongly suggests that the DNA fragment carries two of the three markers. Since the distances of cysA to metC, metC to leuB, and leuB to cysA are 1271 kb, 1507 kb and 1437 kb, the length of continuous DNA is most probably greater than 1271 kb. This conclusion was also confirmed from the 0.51–1.3% coinherited ratios of Cys+ with two of the three regions (metC, hisH, and leuC-arg1-aroG
260
J. BIOSCI. BIOENG.,
SAITO ET AL.
TABLE 2. Linkage analysis of selected marker to unselected markers in Cys+ recombinantsa
TABLE 3. Cotransfer analysis in B. subtilis AU1 genetic backgrounda
Class % ratio Ab Ratio Bc + + – Cys Met His 0.51 1.6 0.51 1.6 Cys+ Met+ Leu+ 0.51 1.6 Cys+ Met+ Arg+ 1.0 3.3 Cys+ Met+ Aro+ 1.3 4.1 Cys+ His– Leu+ 0.77 2.5 Cys+ His– Arg+ 0.51 1.6 Cys+ His– Aro+ <0.26 <15 Cys+ Met+ His– Leu+ 0.26 15 Cys+ Met+ His– Arg+ <0.26 <15 Cys+ Met+ His– Aro+ a Cys+ recombinants with two or three other markers. b % ratio A shows the % ratio of Cys+ with two or three other markers to Cys+. c Ratio B shows the ratio of % ratio A to the double or triple congression ratio. One milliliter of recipient culture (about 1.0 ×108 cells/ml) was mixed with 0.1 ml of protoplast suspension (2.74 ×105 protoplasts/ml). Cys+ transformants of B. subtilis AYG2 (cys14 metC3 trpC2 leuB8 arg1 aroG932) were selected using DNA in protoplast lysates of B. subtilis AC820. The 489 Cys+ transformants were examined for the phenotype of the unselected markers. The coinheritance ratio of cysA (9.7°) to hisH (202°) is 5.1%.
Class % ratio Ab Ratio Bc + + + Met Cys His 2.3 7.2 5.3 17 Met+ Cys+ Trp– 2.7 8.5 Met+ Cys+ Leu+ 7.2 23 Met+ Cys+ Arg+ 1.1 3.6 Met+ Str-r His+ 5.7 18 Met+ Str-r Trp– 3.8 12 Met+ Str-r Leu+ 7.6 24 Met+ Str-r Arg+ 1.1 3.6 Met+ His+ Leu+ 1.9 6.0 Met+ His+ Arg+ 0.76 2.4 Met+ Trp– Leu+ 6.8 22 Met+ Trp– Arg+ 0.38 22 Met+ Cys+ His+ Leu+ 0.76 43 Met+ Cys+ His+ Arg+ 0.38 22 Met+ Cys+ Trp– Leu+ 4.2 237 Met+ Cys+ Trp– Arg+ 0.38 22 Met+ Str-r His+ Leu+ 0.76 43 Met+ Str-r His+ Arg+ 0.38 22 Met+ Str-r Trp– Leu+ 5.7 324 Met+ Str-r Trp– Arg+ a Met+ recombinants with two or three other markers. b % ratio A shows the % ratio of Met+ with two or three other markers to Met+. c Ratio B shows the ratio of %ratio A to the double or triple congression ratio. 264 Met+ transformants were selected and analyzed. One milliliter of recipient culture (about 1.0 ×108 cells/ml) was mixed with 0.1 ml of protoplast suspension (2.74 ×104 cells/ml).
regions), which were also higher than 0.31% (Table 2). Evaluation of the length of continuous DNA in another genetic background of B. subtilis Continuous DNA length was examined in another genetic background of B. subtilis. A high cotransfer ratio in the genetic background of B. subtilis AU1 (cysA metC hisH leuB arg1 ist1 ist2) has been observed, compared with that of the wild-type strain (unpublished results). Met+ transformants of B. subtilis AU1 generated by using DNA in the protoplast lysate of B. subtilis 168S (trpC rpsL smo1) were, therefore, selected and analyzed for the phenotype of unselected markers (Table 3). Met+ Cys+ Leu+ and Met+ Cys+ Arg+ recombinants were obtained with 2.7% and 7.2% the efficiencies of Met+ recombinants, respectively. Met+ Str-r Leu+ and Met+ Str-r Arg+ recombinants were also obtained with 3.8% and 7.6% the efficiencies of Met+ recombinants. These efficiencies are approximately 10- to 20-fold higher than 0.31%, which was calculated using the congression ratio of 5.6%. To determine whether significant congression was observed in B. subtilis AU1, the fragmented DNA prepared as described in Materials and Methods was used for transformation of B. subtilis HH4 and B. subtilis AU1. As shown in Table 4, the ratios of Arg+ Met+ to Arg+ and Arg+ His+ Met+ to Arg+ were 2.8% and 2.5%, and 0.16% and 0.19% in the wild-type and AU1, respectively. These results indicate that the congression ratio of B. subtilis HH4 is quite similar to that of B. subtilis AU1, and that the high cotransfer ratio between the separated markers in B. subtilis AU1 is not most probably derived from the enhancement of double transformation. These observations strongly suggest that the continuous DNA length is also greater than 1271 kb. When the other coinheritance ratios of Met+ with two of the three regions (cysA-rpsL, hisH-trpC, and leuB-arg1 regions) were examined, ratios of 0.76–6.8% were also obtained (Table 3). Moreover, the coinheritance ratios of Met+ with the three regions (cysA-rpsL, hisH-trpC, and leuB-arg1 regions) were
from 0.38% to 5.7% (Table 3), which were 20- to 300-fold higher than the ratio of 0.018% calculated using the congression ratio of 5.6%. All these observations indicate that the length of continuous DNA is very large (greater than 1000 kb). DISCUSSION The transformation of competent B. subtilis with DNA in protoplast lysates of B. subtilis has three characteristic features. The first feature is the high efficiency of transformation approaching unity (17). The frequency is one transformant per ten molecules of genomic DNA and is 200–1000fold higher than that of conventional transformation using purified DNA. The second feature of the LP transformation is the large amount of DNA incorporated, corresponding to 4215 kb of the whole-genome DNA (1). The third is the large length of continuous DNA incorporated (greater than 1271 kb) that was observed in this study (Tables 2 and 3). The coinheritance ratio of cysA (9.7°) to hisH (202°) was 5.1% (Table 2), most probably corresponding to a congression ratio. The ratio B of P (cysA, metC, aroG) to the double congression ratio was 3.3 in B. subtilis AYG2 (Table 2). The 3.3-fold difference is significant and leads to the conclusion that the region from cysA to metC or aroG is on a continuous DNA fragment. Therefore, the length of continuous DNA incorporated is greater than 1271 kb, which corresponds to about one-third of the whole-genome length. The 3.3-fold difference is significant but low for the following reasons: (i) the strain is recombination-proficient, (ii) two independent recombinations occur in a long DNA fragment with the
VOL. 101, 2006
FATE OF TRANSFORMING BACTERIAL GENEOME
261
TABLE 4. Congression frequency of competent B. subtilis HH4 and B. subtilis AU1 by fragmented DNA % congression Arg+ Met+ Arg+ His+ Met+ Arg+ His+ 6 HH4 (cysA metC hisH leuB arg1) 168S (rpsL trpC smo1) 2.74 ×10 2.4 2.8 0.16 2.5 2.5 0.19 AU1 (cysA metC hisH leuB arg1 ist1 ist2) 168S (rpsL trpC smo1) 2.74 ×106 8 7 One milliliter of recipient culture (about 1.0 ×10 cells/ml) was mixed with 0.1 ml of protoplast suspension (2.74 ×10 protoplasts/ml). The donor protoplast suspension was diluted by hypotonic medium (medium II). Then, DNA in the protoplast lysate was fragmented with mixing (30 s, twice) on a Vortex Genie mixer (Scientific Industries). Arg+ transformants of B. subtilis HH4 (cys14 metC3 hisH leuB8 arg1) were selected using DNA in protoplast lysates of B. subtilis 168S (rpsL trpC2 smo1). The linkages of other markers to arg1 were also examined. Recipient (genotype)
Donor (genotype)
cysA and metC, or cysA and aroG regions, and (iii) a recombination occurs in a different DNA fragment containing the aroG or metC marker (Fig. 1). B. subtilis AYG2 is not a specific strain but a standard one, and is often used as a recipient strain for transformation and PBS1 phage transduction (19, 25). A similar ratio of P (cysA, metC, aroG) to P (cysA hisH) x P (cysA hisH) was also obtained in the LP transformation of B. subtilis HH4 with the DNA of B. subtilis 168S protoplasts. Thus, the ratio was generally observed in the wild-type genetic background of B. subtilis 168 derivatives. The length of continuous DNA incorporated was also determined in B. subtilis AU1. This strain is an isogenic derivative of B. subtilis AYG2, which was obtained from mutagenesis with N-methyl-N′-nitro-N-nitrosoguanidine (NTG) by two-step mutagenesis and sequential breeding. The ratio B of P (metC, cysA, leuB), P (metC, rpsL, leuB), or (metC, cysA, arg1) to the double congression ratio was 8.5, 12, or 23, corresponding to values 5–14 fold higher than those of the wild-type strain (Tables 2 and 3). The transformation efficiencies and congression ratios of both strains by fragmented DNA are similar (Table 4). DNA longer than fragmented DNA is, thus, necessary for the 5–14-fold difference. One probability is that both homologous recombination and recombination-dependent DNA replication proposed by Kogoma (26) occur in the mutant genetic background. In the case of the repair of DNA double-strand breaks, long linear DNA fragments with DNA lesions (nicks or gaps) form a Holliday structure with a recipient chromosome leading to replication using the recipient chromosome as a template. Therefore, a significantly long linkage greater than 1300 kb is observed in the mutant genetic background. Inducible stable DNA replication has been reported to be a unique type of recombination-dependent DNA replication (26), and several sites serve as origins of replication. When a whole genome containing such sites is incorporated into a competent cell, recombination-dependent replication may occur between a long continuous donor DNA fragment and the host chromosomal DNA, or between a donor DNA fragment and the host chromosomal DNA with a double-strand break. Although a long linkage greater than 1300 kb is obtained, the LP transformation of B. subtilis is different from protoplast fusion for the following reasons (21, 22); The protoplast fusion of B. subtilis requires the use of a fusion-inducing reagent such as polyethylene glycol. Transformants in the B. amyloliquefaciens genetic background cannot be obtained by LP transformation, (unpublished results), whereas such strains are frequently obtained by protoplast fusion (27). Finally, LP transformation is a DNase I-sensitive proc-
Number of protoplasts
ess (unpublished result). We propose a working hypothesis on the mechanism of the LP transformation of competent B. subtilis. Donor chromosomal DNA is covered by various DNA-binding proteins. The proteins probably interact with DNA receptor proteins and lead to high-frequency transformation. When the DNA is taken up into competent B. subtilis, the DNA is a long continuous double-stranded DNA fragment (greater than 1300 kb) with DNA lesions (nicks and/or gaps). The DNA lesions result in proficient recombination and an abrupt decrease in the linkage between two markers. Purified double-stranded DNA without DNA-binding proteins is attacked by various nucleases (4, 6) and is converted to almost single-stranded DNA. Several experiments based on the proposed model are currently underway.
REFERENCES 1. Akamatsu, T. and Taguchi, H.: Incorporation of the whole chromosomal DNA in protoplast lysates into competent cells of Bacillus subtilis. Biosci. Biotechnol. Biochem., 65, 823– 829 (2001). 2. Priest, F. G.: Systematics and ecology of Bacillus subtilis, p. 3–16. In Sonnenshein, A. L., Hoch, J. A., and Losick, R. (ed.), Bacillus subtilis and other gram-positive bacteria. American Society for Microbiology, Washington, D.C. (1993). 3. Wilson, G. A. and Young, F. E.: Intergenotic transformation of the Bacillus subtilis genospecies. J. Bacteriol., 111, 705– 716 (1972). 4. Dubnau, D.: Genetic transformation in Bacillus subtilis, p. 147–178. In Dubnau, D. (ed.), Molecular biology. Academic Press, New York (1982). 5. Dubnau, D.: Genetic transformation in Bacillus subtilis: a review with emphasis on the recombination mechanism, p. 14– 27. In Schlessinger, D. (ed.), Microbiology. American Society for Microbiology, Washington, D.C. (1976). 6. Dubnau, D.: Binding and transport of transforming DNA by Bacillus subtilis: the role of type-IV pilin-like proteins-a review. Gene, 192, 191–198 (1997). 7. Hahn, J., Albano, M., and Dubnau, D.: Isolation and characteization of competence mutants in Bacillus subtilis. J. Bacteriol., 169, 3104–3109 (1987). 8. Hahn, J., Inamine, G., Kozlov, Y., and Dubnau, D.: Characterization of comE, a late competence operon of Bacillus subtilis required for the binding and uptake of transforming DNA. Mol. Microbiol., 10, 99–111 (1993). 9. Hahn, J., Kong, L., and Dubnau, D.: The regulation of competence transcription factor synthesis constitutes a critical control point in the regulation of competence in Bacillus subtilis. J. Bacteriol., 176, 5753–5761 (1994). 10. Van Sinderen, D., ten Berg, A., Hayema, B. J., Hamoen, L., and Venema, G.: Molecular cloning and sequence of comK, a gene required for genetic competence in Bacillus subtilis. Mol. Microbiol., 11, 695–703 (1994).
262
J. BIOSCI. BIOENG.,
SAITO ET AL.
11. Van Sinderen, D. and Venema, G.: comK acts as an autoregulatory control switch in the signal transduction route to competence in Bacillus subtilis. J. Bacteriol., 176, 5762–5770 (1994). 12. Van Sinderen, D., Luttinger, A., Kong, L., Dubnau, D., Venema, G., and Hamoen, L.: comK encodes the competence transcription factor (CTF), the key regulatory protein for competence developement in Bacillus subtilis. Mol. Microbiol., 15, 455–462 (1995). 13. Chen, I. and Dubnau, D.: DNA uptake during bacterial transformation. Nat. Rev. Microbiol., 2, 241–249 (2004). 14. Kelly, M. S. and Pritchard, R. H.: Unstable linkage between genetic markers in transformation. J. Bacteriol., 89, 1314–1321 (1965). 15. Bettinger, G. E. and Young, F. E.: Transformation of Bacillus subtilis using gently lysed L-forms: a new mapping technique. Biochem. Biophys. Res. Commun., 55, 1105–1111 (1973). 16. Bettinger, G. E. and Young, F. E.: Transformation of Bacillus subtilis: transforming ability of deoxyribonucleic acid in lysates of L-forms or protoplasts. J. Bacteriol., 122, 987–993 (1975). 17. Akamatsu, T. and Sekiguchi, J.: Characterization of chromosome and plasmid transformation in Bacillus subtilis using gently lysed protoplasts. Arch. Microbiol., 146, 353–357 (1987). 18. Akamatsu, T., Moriyama, H., and Kaneda, T.: Genetic mapping of bfmA mutation causing fatty acid deficiency in Bacillus subtilis. Can. J. Microbiol., 39, 629–633 (1993). 19. Akamatsu, T. and Taguchi, H.: Interspecific transformation of Bacillus subtilis competent cells by chromosomal DNA in
20. 21.
22. 23. 24.
25.
26. 27.
lysates of protoplasts of Bacillus amyloliquefaciens. Biosci. Biotechnol. Biochem., 64, 275–279 (2000). Spizizen, J.: Transformation of biochemically deficient strains of Bacillus subtilis by deoxyribonucleate. Proc. Natl. Acad. Sci. USA, 44, 1072–1078 (1958). Akamatsu, T. and Sekiguchi, J.: An improved method of protoplast regeneration for Bacillus species and its application to protoplast fusion and transformation. Agric. Biol. Chem., 48, 651–655 (1984). Akamatsu, T. and Sekiguchi, J.: Genetic mapping by means of protoplast fusion in Bacillus subtilis. Mol. Gen. Genet., 208, 254–262 (1987). Anagnostopoulos, C. and Spizizen, J.: Requirements for transformation in Bacillus subtilis. J. Bacteriol., 81, 741–746 (1961). Kunst, F., Ogasawara, N., Moszer, I., Albertini, A. M., Alloni, G., Azevedo, V., Bertero, M. G., Bessieres, P., Bolotin, A., Borchert, S., and other 141 authors: The complete genome sequence of the gram-positive bacterium Bacillus subtilis. Nature, 390, 249–256 (1997). Akamatsu, T. and Taguchi, H.: A simple and rapid method for intra- and interspecific transformation of Bacillus subtilis on solid media by DNA in protoplast lysates. Biosci. Biotechnol. Biochem., 65, 446–448 (2001). Kogoma, T.: Stable DNA replication: interplay between DNA replication, homologous recombination, and transcription. Microbiol. Mol. Biol. Rev., 61, 212–238 (1997) Akamatsu, T. and Sekiguchi, J.: Selection methods in bacilli for recombinants and transformants of intra- and interspecific fused protoplasts. Arch. Microbiol., 134, 303–308 (1987).
© 2006, The Society for Biotechnology, Japan
Fate of Transforming Bacterial Genome Following Incorporation into Competent Cells of Bacillus subtilis: a Continuous Length of Incorporated DNA Yukiko Saito,1 Hisataka Taguchi,1 and Takashi Akamatsu1* Department of Applied Microbial Technology, Faculty of Biotechnology and Life Science, Sojo University, Ikeda 4-22-1, Kumamoto 860-0082, Japan1 Received 2 November 2005/Accepted 27 December 2005
In contrast to the conventional transformation of Bacillus subtilis using purified DNA, those using DNA in lysed protoplasts have a high transformation efficiency and enable whole-genome transfer into competent B. subtilis [Akamatsu, T. and Taguchi, H., Biosci. Biotechnol. Biochem., 65, 823–829 (2001)]. Here, we examined the length of incorporated continuous DNA by analyzing the cotransfer ratio with selected and unselected markers, on the basis of a new experimental design. The cotransfer ratio of a selected marker with an unselected marker on the opposite side of the genetic map of the B. subtilis chromosome was about 5.6% and could be interpreted as congression (double transformation) ratio. In the wild-type strain, the cotransfer ratio of cysA (113 kb position on 4215 kb of B. subtilis chromosome) with metC (1384 kb) and leuB (2891 kb) was 0.77%, twice the value (5.6% × 5.6% = 0.31%) calculated from the congression ratio. Moreover, in a genetic background, the cotransfer ratios of metC with cysA and leuB, and metC with cysA and arg1 (3012 kb) were 2.7% and 7.2%, respectively. These results strongly suggest that the length of continuous DNA incorporated into B. subtilis is most probably greater than 1271 kb. When the DNA from the protoplast lysate was fragmented by mixing, the cotransfer ratios of arg1 with metC, and arg1 with metC and trpC (2374 kb) were 2.8% and 0.16%, respectively. A high cotransfer ratio (2.7–7.2%) could not, therefore, be obtained using the fragmented DNA. Based on these observations, we propose a working hypothesis on the mechanism of the transformation of competent B. subtilis by DNA in protoplast lysates (LP transformation). [Key words: Bacillus subtilis, competent, transformation, DNA uptake, continuous length]
Bacillus subtilis is a gram-positive bacterium capable of developing competence and being transformed with purified DNA (1–3). Double-stranded DNA is bound to a competent-cell surface (4). The limited cleavage of DNA by nuclease occurs, yielding fragments of 15–30 kb. About half of the DNA is degraded by a competence-specific nuclease and the remaining single-stranded DNA is then transported to the cytosol. The DNA taken up interacts with the recipient DNA to form a D-loop and recombines with a homologous region to yield a heteroduplex recombinant molecule. Finally, the expression of donor genetic information, the segregation of the heteroduplex, and replication occur resulting in the formation of a transformant clone (5, 6). A large number of proteins required for transformation have been identified in B. subtilis (6–12). All of them are regulated at the transcriptional level and depend on the comK transcriptional factor for their synthesis. ComG and ComC proteins form the DNA-binding apparatus, whereas ComE and ComF proteins form the DNA transport machinery. An attractive working hypothesis has been proposed for
such transformation process and for other Com proteins in B. subtilis (13). In contrast to conventional transformation methods using purified DNA, those using DNA in lysed protoplasts result in a high transformation efficiency (14–17). Transformants are obtained at a frequency of one transformant per ten genomes, 200–1000-fold higher than that of conventional methods. Since lysed-protoplast transformation (LP transformation) also depends on the comK transcriptional factor (1), both transformation mechanisms seem to be basically similar. Linkage analysis between two markers with a distance of 1673 kb leads to a constant cotransfer ratio (4–5%), even when the number of donor DNA molecules is apparently 60-fold lower than the number of competent cells (1, 17). Thus, the uptake of 4215 kb of whole-genome DNA into competent B. subtilis is one of the most significant features of LP transformation. From the linkage relationship between two markers, the length of continuous DNA taken up into B. subtilis could not be evaluated, because the cotransformation ratio sharply decreased with increasing marker distance of the chromosome up to 121 kb and showed a plateau from 121 kb to 1628 kb (4–5% cotransfer ratio) (1). This observation indi-
* Corresponding author. e-mail: [email protected] phone: +81-(0)96-326-3929 fax: +81-(0)96-323-1330 257
258
J. BIOSCI. BIOENG.,
SAITO ET AL.
cates two possibilities. First, the genome taken up is fragmented into many pieces (for example, about twenty pieces, 200-kb average fragment length) and, therefore, the cotransformation linkage between two markers separated by 100 kb can only be observed. Second, although the DNA taken up is very long (several fragmented pieces of genome DNA), the recombination of the DNA with a recipient chromosome occurs only within a short distance (100 kb). We developed a theoretical design evaluating the length of incorporated continuous DNA by comparison of two cotransfer ratios with selected and unselected markers (cotransformation and congression) to discriminate between the two possibilities. In this study, we describe the incorporation of a continuous length of DNA of more than 1271 kb that corresponds to about one-third of the whole-genome length, and propose a working hypothesis on the mechanism of the transformation of competent B. subtilis by DNA in protoplast lysates (LP transformation).
MATERIALS AND METHODS Bacterial strains The bacterial strains used in this study are listed in Table 1. B. subtilis AU1 (cysA metC hisH leuB arg1 ist1 ist2) is an ist1 ist2 mutant derivative with a high frequency of interspecific transformation by a long heterologous DNA fragment in protoplast lysates of B. amyloliquefaciens and a high frequency of cotransformation, as will be described elsewhere in detail. Media Luria–Bertani (LB) medium (19) was used for bacterial growth. The LB agar medium contained 15 g of agar in 1 l of LB medium. Spizizen minimal medium (20) (SM medium) was used as a basal medium for the preparation of competent cells. For the isolation of auxotrophs or transformants, low Spizizen minimal medium (LSM medium) (19) was used, supplemented with amino acids, when necessary. Preparation of B. subtilis protoplasts Protoplasts were prepared basically as described by Akamatsu and Sekiguchi (21, 22). B. subtilis 168S (trpC rpsL smo1) was cultivated overnight on LB agar medium at 37°C and suspended in LB medium. An aliquot of the bacterial suspension (100–1000 cells) was inoculated into 25 ml of LB medium and cultivated overnight at 37°C. The overnight culture was added to 25 ml of LB medium at an inoculation ratio of 1% and incubated with shaking for 3 h at 37°C. The culture was again added to 25 ml of LB medium at an inoculation ratio of 1% and incubated for 2–3 h. When the cell growth reached the mid-log to early stationary phase, cells from 20 ml of the culture were harvested by centrifugation (7200×g for 1–2 min) and suspended in 4 ml of SMM (0.5 M sucrose–0.02 M maleate buffer [pH 6.5]– 0.02 M MgCl2) containing lysozyme at a final concentration of 250 mg per liter. The total mixture was gently shaken for 15– 45 min at 40°C as a shallow layer (e.g., 4 ml of suspension in a 100-ml Erlenmeyer flask). Three milliliters of the suspension was TABLE 1. B. subtilis strains used in this study Strain
Genotype
AYG2 AU1 HH4 168S AC820
cysA14 metC3 trpC2 leuB8 arg1 aroG932 cysA14 metC3 hisH leuB8 arg1 ist1 ist2 cysA14 metC3 hisH leuB8 arg1 trpC2 rpsL smo1 hisH rpsL smo1
Reference, source or derivation 1 This study 1 17 18
transferred into a centrifuge tube (Nalgen Oak Ridge centrifuge tube 3118-0050; Nalge Nunc International, Rochester, New York, USA) containing 10 ml of SMM and centrifuged (4000×g for 5– 10 min). The pellet was suspended in SMM buffer. Protoplast density was measured from absorbance at 660 nm (A660). An A660 of 2.88 corresponded to 2.74 ×109 protoplasts per ml. After dilution with SMM buffer, an aliquot (0.1 ml) of protoplast suspension was used as the source of donor DNA. Preparation of competent cells Competent cells were prepared basically as described by Anagnostopoulos and Spizizen (23). B. subtilis AYG2 was cultivated overnight on LB agar medium at 37°C and suspended in LB medium. An aliquot of the bacterial suspension (100–1000 cells) was inoculated into 25 ml of LB medium and cultivated overnight at 37°C. The overnight culture was added to 25 ml of LB medium at an inoculation ratio of 1% and incubated with shaking for 3 h at 37°C. Cells from 20 ml of the culture were harvested by centrifugation (7200×g for 1–2 min) and suspended in an aliquot (5 ml) of medium I (Spizizen minimal medium supplemented with 200 mg/l casamino acids and, when necessary, 50 mg/l amino acids). Absorbance at 660 nm was measured and the culture was added to medium I in a 100-ml Erlenmeyer flask (initial A660 = 0.33, 10 ml total culture) and cultivated for 4 h at 37°C. Two milliliters of culture was added to 18 ml of medium II (Spizizen minimal medium supplemented with 100 mg/l casamino acids and, when necessary, 5 mg/l amino acids), and cultivated for 1.5 h at 37°C. One milliliter of the competent cell culture was used for transformation. Preparation of fragmented DNA solution An aliquot (0.1 ml) of a protoplast suspension was added to 1 ml of hypotonic medium II. DNA in the protoplast lysates was then fragmented by mixing (30 s, maximal power, twice) using a Vortex Genie mixer (Scientific Industries, New York, NY, USA) and used as donor DNA, when necessary. Transformation with DNA obtained by gentle lysis of protoplasts (LP transformation) An aliquot (0.1 ml) of a protoplast suspension was added to 1 ml of hypotonic competent-cell culture, and the mixture of DNA in protoplast lysates and competent cells was incubated at 37°C for 30 min. Then the cells were plated out onto minimal agar medium with appropriate nutrients. After singlecolony isolation, each colony, when necessary, was further transferred to diagnostic agar plates and unselected markers were identified.
RESULTS Experimental design for evaluation of the length of continuous DNA incorporated into competent B. subtilis The transformation of B. subtilis by DNA involves the following steps: (i) the binding of incorporated DNA to the recipient chromosome, (ii) the recombination of the DNA, (iii) the expression of donor information, and (iv) the formation of a transformant clone (6). Although 4215 kb of the whole genome of donor B. subtilis can be incorporated into competent B. subtilis (1), the accurate length of fragments greater than 100 kb is not known. Continuous DNA length can be evaluated by analyzing the difference in efficiency between congression (double transformation) and cotransformation (Fig. 1). In the case in which A+ transformants of recipient cells with markers a, b, and c are examined for the unselected markers (b, c), let P (A, B) or P (A, C) be the probability that markers A and B, or A and C will be coinherited. Let P (A, B, C) be the probability that A, B and C will be coinherited. Three types of coinheritance of the three
VOL. 101, 2006
FATE OF TRANSFORMING BACTERIAL GENEOME
259
FIG. 1. Evaluation of the length of continuous donor DNA incorporated into competent Bacillus subtilis. Case 1: Let P (A, B), P (A, C), or P (A, B, C) be the probability that markers A and B, A and C, or A and B and C will be coinherited. P (A, B, C) will be calculated by P (A, B) ×P (A, C) and is nearly equal to P (A, C) ×P (A, C), providing that the A, B, and C markers are on separate fragments and the fragmentation of DNA (for example, into 20 pieces) results in a linkage between markers being linked by distances of 100 kb. Case 2: Donor DNA in a recipient cell is fragmented into several pieces only, and marker C is on a different DNA fragment from that carrying A and B markers. P (A, B, C) is then expected to be greater than P (A, C) ×P (A, C).
markers are possible: congression, cotransformation, or congression and cotransformation. When A+ B+ C+ transformants are obtained by congression (Fig. 1, case 1), P (A, B, C) can be calculated by P (A, B) ×P (A, C) and is almost equal to P (A, C) ×P (A, C). On the other hand, P (A, B, C) is expected to be higher than P (A, C) ×P (A, C), providing that A and B are on the same DNA fragment and C is on a different DNA fragment. The P (A, B) due to cotransformation is higher than that of P (A, C) due to congression, because congression requires the independent bindings of two DNA fragments and recombination with the recipient chromosome, whereas cotransformation requires the binding of one DNA fragment and recombination (Fig. 1, case 2). When genetic marker C is located on the opposite side of A and a higher efficiency of P (A, B, C) than that of P (A, C)×P (A, C) is obtained, this indicates that A and B are on the same DNA fragment. This interpretation results in the evaluation of the length of continuous DNA corresponding to the distance between A and B. The continuous length of DNA is greater than one million bases To determine congression ratio, pairs of genetic markers were chosen on the opposite side of the B. subtilis chromosome. metC is located at 118° of the chromosome (24). leuB, arg1, and aroG are at 247°, 257°, and 260°, respectively (Fig. 2). The Met+ transformants of competent B. subtilis AYG2 (cysA metC trpC leuB arg1 aroG) generated by using DNA in the protoplast lysate of B. subtilis 168S (trpC rpsL smo1) were selected and examined for Aro+, Arg+, and Leu+. The coinheritance ratios of the double markers were 5.7%, 5.7%, and 5.7%, respectively. These results strongly suggest that the coinherited ratio is a congression ratio. This was confirmed by determining the coinherited ratios of aroG to metC, arg1 to metC, and cysA (9.7°) to hisH (202°). The ratios were 5.7%, 5.7%, and 5.1%, respectively. All these observations lead to the conclusion that a value of 5.6% represents a congression ratio, corresponding to the P (A, C) in Fig. 1.
FIG. 2. Genetic map of Bacillus subtilis chromosome. The B. subtilis genome has 4,214,841 bases (24). One degree, therefore, corresponds to about 11.7 kb. The number in the circle indicates the map position (degrees) on the B. subtilis chromosome. The distance of the origin to cysA, rpsL, metC, hisH, trpC, leuB, arg1 or aroG is 113 kb, 130 kb, 1384 kb, 2370 kb, 2374 kb, 2891 kb, 3012 kb or 3045 kb, respectively.
Cys+ transformants of B. subtilis AYG2 generated by using DNA in the protoplast lysate of B. subtilis AC820 (hisH rpsL smo1) were selected and analyzed for the phenotype of unselected markers (Table 2). Cys+ Met+ Aro+ recombinants were obtained at a 1.0% ratio of Cys+ recombinants. If three markers are on separate DNA fragments, a ratio of Cys+ Met+ Aro+ to Cys+ of 0.31% (5.6% × 5.6%) is expected, which is clearly lower than the observed value of 1.0%. The observed value strongly suggests that the DNA fragment carries two of the three markers. Since the distances of cysA to metC, metC to leuB, and leuB to cysA are 1271 kb, 1507 kb and 1437 kb, the length of continuous DNA is most probably greater than 1271 kb. This conclusion was also confirmed from the 0.51–1.3% coinherited ratios of Cys+ with two of the three regions (metC, hisH, and leuC-arg1-aroG
260
J. BIOSCI. BIOENG.,
SAITO ET AL.
TABLE 2. Linkage analysis of selected marker to unselected markers in Cys+ recombinantsa
TABLE 3. Cotransfer analysis in B. subtilis AU1 genetic backgrounda
Class % ratio Ab Ratio Bc + + – Cys Met His 0.51 1.6 0.51 1.6 Cys+ Met+ Leu+ 0.51 1.6 Cys+ Met+ Arg+ 1.0 3.3 Cys+ Met+ Aro+ 1.3 4.1 Cys+ His– Leu+ 0.77 2.5 Cys+ His– Arg+ 0.51 1.6 Cys+ His– Aro+ <0.26 <15 Cys+ Met+ His– Leu+ 0.26 15 Cys+ Met+ His– Arg+ <0.26 <15 Cys+ Met+ His– Aro+ a Cys+ recombinants with two or three other markers. b % ratio A shows the % ratio of Cys+ with two or three other markers to Cys+. c Ratio B shows the ratio of % ratio A to the double or triple congression ratio. One milliliter of recipient culture (about 1.0 ×108 cells/ml) was mixed with 0.1 ml of protoplast suspension (2.74 ×105 protoplasts/ml). Cys+ transformants of B. subtilis AYG2 (cys14 metC3 trpC2 leuB8 arg1 aroG932) were selected using DNA in protoplast lysates of B. subtilis AC820. The 489 Cys+ transformants were examined for the phenotype of the unselected markers. The coinheritance ratio of cysA (9.7°) to hisH (202°) is 5.1%.
Class % ratio Ab Ratio Bc + + + Met Cys His 2.3 7.2 5.3 17 Met+ Cys+ Trp– 2.7 8.5 Met+ Cys+ Leu+ 7.2 23 Met+ Cys+ Arg+ 1.1 3.6 Met+ Str-r His+ 5.7 18 Met+ Str-r Trp– 3.8 12 Met+ Str-r Leu+ 7.6 24 Met+ Str-r Arg+ 1.1 3.6 Met+ His+ Leu+ 1.9 6.0 Met+ His+ Arg+ 0.76 2.4 Met+ Trp– Leu+ 6.8 22 Met+ Trp– Arg+ 0.38 22 Met+ Cys+ His+ Leu+ 0.76 43 Met+ Cys+ His+ Arg+ 0.38 22 Met+ Cys+ Trp– Leu+ 4.2 237 Met+ Cys+ Trp– Arg+ 0.38 22 Met+ Str-r His+ Leu+ 0.76 43 Met+ Str-r His+ Arg+ 0.38 22 Met+ Str-r Trp– Leu+ 5.7 324 Met+ Str-r Trp– Arg+ a Met+ recombinants with two or three other markers. b % ratio A shows the % ratio of Met+ with two or three other markers to Met+. c Ratio B shows the ratio of %ratio A to the double or triple congression ratio. 264 Met+ transformants were selected and analyzed. One milliliter of recipient culture (about 1.0 ×108 cells/ml) was mixed with 0.1 ml of protoplast suspension (2.74 ×104 cells/ml).
regions), which were also higher than 0.31% (Table 2). Evaluation of the length of continuous DNA in another genetic background of B. subtilis Continuous DNA length was examined in another genetic background of B. subtilis. A high cotransfer ratio in the genetic background of B. subtilis AU1 (cysA metC hisH leuB arg1 ist1 ist2) has been observed, compared with that of the wild-type strain (unpublished results). Met+ transformants of B. subtilis AU1 generated by using DNA in the protoplast lysate of B. subtilis 168S (trpC rpsL smo1) were, therefore, selected and analyzed for the phenotype of unselected markers (Table 3). Met+ Cys+ Leu+ and Met+ Cys+ Arg+ recombinants were obtained with 2.7% and 7.2% the efficiencies of Met+ recombinants, respectively. Met+ Str-r Leu+ and Met+ Str-r Arg+ recombinants were also obtained with 3.8% and 7.6% the efficiencies of Met+ recombinants. These efficiencies are approximately 10- to 20-fold higher than 0.31%, which was calculated using the congression ratio of 5.6%. To determine whether significant congression was observed in B. subtilis AU1, the fragmented DNA prepared as described in Materials and Methods was used for transformation of B. subtilis HH4 and B. subtilis AU1. As shown in Table 4, the ratios of Arg+ Met+ to Arg+ and Arg+ His+ Met+ to Arg+ were 2.8% and 2.5%, and 0.16% and 0.19% in the wild-type and AU1, respectively. These results indicate that the congression ratio of B. subtilis HH4 is quite similar to that of B. subtilis AU1, and that the high cotransfer ratio between the separated markers in B. subtilis AU1 is not most probably derived from the enhancement of double transformation. These observations strongly suggest that the continuous DNA length is also greater than 1271 kb. When the other coinheritance ratios of Met+ with two of the three regions (cysA-rpsL, hisH-trpC, and leuB-arg1 regions) were examined, ratios of 0.76–6.8% were also obtained (Table 3). Moreover, the coinheritance ratios of Met+ with the three regions (cysA-rpsL, hisH-trpC, and leuB-arg1 regions) were
from 0.38% to 5.7% (Table 3), which were 20- to 300-fold higher than the ratio of 0.018% calculated using the congression ratio of 5.6%. All these observations indicate that the length of continuous DNA is very large (greater than 1000 kb). DISCUSSION The transformation of competent B. subtilis with DNA in protoplast lysates of B. subtilis has three characteristic features. The first feature is the high efficiency of transformation approaching unity (17). The frequency is one transformant per ten molecules of genomic DNA and is 200–1000fold higher than that of conventional transformation using purified DNA. The second feature of the LP transformation is the large amount of DNA incorporated, corresponding to 4215 kb of the whole-genome DNA (1). The third is the large length of continuous DNA incorporated (greater than 1271 kb) that was observed in this study (Tables 2 and 3). The coinheritance ratio of cysA (9.7°) to hisH (202°) was 5.1% (Table 2), most probably corresponding to a congression ratio. The ratio B of P (cysA, metC, aroG) to the double congression ratio was 3.3 in B. subtilis AYG2 (Table 2). The 3.3-fold difference is significant and leads to the conclusion that the region from cysA to metC or aroG is on a continuous DNA fragment. Therefore, the length of continuous DNA incorporated is greater than 1271 kb, which corresponds to about one-third of the whole-genome length. The 3.3-fold difference is significant but low for the following reasons: (i) the strain is recombination-proficient, (ii) two independent recombinations occur in a long DNA fragment with the
VOL. 101, 2006
FATE OF TRANSFORMING BACTERIAL GENEOME
261
TABLE 4. Congression frequency of competent B. subtilis HH4 and B. subtilis AU1 by fragmented DNA % congression Arg+ Met+ Arg+ His+ Met+ Arg+ His+ 6 HH4 (cysA metC hisH leuB arg1) 168S (rpsL trpC smo1) 2.74 ×10 2.4 2.8 0.16 2.5 2.5 0.19 AU1 (cysA metC hisH leuB arg1 ist1 ist2) 168S (rpsL trpC smo1) 2.74 ×106 8 7 One milliliter of recipient culture (about 1.0 ×10 cells/ml) was mixed with 0.1 ml of protoplast suspension (2.74 ×10 protoplasts/ml). The donor protoplast suspension was diluted by hypotonic medium (medium II). Then, DNA in the protoplast lysate was fragmented with mixing (30 s, twice) on a Vortex Genie mixer (Scientific Industries). Arg+ transformants of B. subtilis HH4 (cys14 metC3 hisH leuB8 arg1) were selected using DNA in protoplast lysates of B. subtilis 168S (rpsL trpC2 smo1). The linkages of other markers to arg1 were also examined. Recipient (genotype)
Donor (genotype)
cysA and metC, or cysA and aroG regions, and (iii) a recombination occurs in a different DNA fragment containing the aroG or metC marker (Fig. 1). B. subtilis AYG2 is not a specific strain but a standard one, and is often used as a recipient strain for transformation and PBS1 phage transduction (19, 25). A similar ratio of P (cysA, metC, aroG) to P (cysA hisH) x P (cysA hisH) was also obtained in the LP transformation of B. subtilis HH4 with the DNA of B. subtilis 168S protoplasts. Thus, the ratio was generally observed in the wild-type genetic background of B. subtilis 168 derivatives. The length of continuous DNA incorporated was also determined in B. subtilis AU1. This strain is an isogenic derivative of B. subtilis AYG2, which was obtained from mutagenesis with N-methyl-N′-nitro-N-nitrosoguanidine (NTG) by two-step mutagenesis and sequential breeding. The ratio B of P (metC, cysA, leuB), P (metC, rpsL, leuB), or (metC, cysA, arg1) to the double congression ratio was 8.5, 12, or 23, corresponding to values 5–14 fold higher than those of the wild-type strain (Tables 2 and 3). The transformation efficiencies and congression ratios of both strains by fragmented DNA are similar (Table 4). DNA longer than fragmented DNA is, thus, necessary for the 5–14-fold difference. One probability is that both homologous recombination and recombination-dependent DNA replication proposed by Kogoma (26) occur in the mutant genetic background. In the case of the repair of DNA double-strand breaks, long linear DNA fragments with DNA lesions (nicks or gaps) form a Holliday structure with a recipient chromosome leading to replication using the recipient chromosome as a template. Therefore, a significantly long linkage greater than 1300 kb is observed in the mutant genetic background. Inducible stable DNA replication has been reported to be a unique type of recombination-dependent DNA replication (26), and several sites serve as origins of replication. When a whole genome containing such sites is incorporated into a competent cell, recombination-dependent replication may occur between a long continuous donor DNA fragment and the host chromosomal DNA, or between a donor DNA fragment and the host chromosomal DNA with a double-strand break. Although a long linkage greater than 1300 kb is obtained, the LP transformation of B. subtilis is different from protoplast fusion for the following reasons (21, 22); The protoplast fusion of B. subtilis requires the use of a fusion-inducing reagent such as polyethylene glycol. Transformants in the B. amyloliquefaciens genetic background cannot be obtained by LP transformation, (unpublished results), whereas such strains are frequently obtained by protoplast fusion (27). Finally, LP transformation is a DNase I-sensitive proc-
Number of protoplasts
ess (unpublished result). We propose a working hypothesis on the mechanism of the LP transformation of competent B. subtilis. Donor chromosomal DNA is covered by various DNA-binding proteins. The proteins probably interact with DNA receptor proteins and lead to high-frequency transformation. When the DNA is taken up into competent B. subtilis, the DNA is a long continuous double-stranded DNA fragment (greater than 1300 kb) with DNA lesions (nicks and/or gaps). The DNA lesions result in proficient recombination and an abrupt decrease in the linkage between two markers. Purified double-stranded DNA without DNA-binding proteins is attacked by various nucleases (4, 6) and is converted to almost single-stranded DNA. Several experiments based on the proposed model are currently underway.
REFERENCES 1. Akamatsu, T. and Taguchi, H.: Incorporation of the whole chromosomal DNA in protoplast lysates into competent cells of Bacillus subtilis. Biosci. Biotechnol. Biochem., 65, 823– 829 (2001). 2. Priest, F. G.: Systematics and ecology of Bacillus subtilis, p. 3–16. In Sonnenshein, A. L., Hoch, J. A., and Losick, R. (ed.), Bacillus subtilis and other gram-positive bacteria. American Society for Microbiology, Washington, D.C. (1993). 3. Wilson, G. A. and Young, F. E.: Intergenotic transformation of the Bacillus subtilis genospecies. J. Bacteriol., 111, 705– 716 (1972). 4. Dubnau, D.: Genetic transformation in Bacillus subtilis, p. 147–178. In Dubnau, D. (ed.), Molecular biology. Academic Press, New York (1982). 5. Dubnau, D.: Genetic transformation in Bacillus subtilis: a review with emphasis on the recombination mechanism, p. 14– 27. In Schlessinger, D. (ed.), Microbiology. American Society for Microbiology, Washington, D.C. (1976). 6. Dubnau, D.: Binding and transport of transforming DNA by Bacillus subtilis: the role of type-IV pilin-like proteins-a review. Gene, 192, 191–198 (1997). 7. Hahn, J., Albano, M., and Dubnau, D.: Isolation and characteization of competence mutants in Bacillus subtilis. J. Bacteriol., 169, 3104–3109 (1987). 8. Hahn, J., Inamine, G., Kozlov, Y., and Dubnau, D.: Characterization of comE, a late competence operon of Bacillus subtilis required for the binding and uptake of transforming DNA. Mol. Microbiol., 10, 99–111 (1993). 9. Hahn, J., Kong, L., and Dubnau, D.: The regulation of competence transcription factor synthesis constitutes a critical control point in the regulation of competence in Bacillus subtilis. J. Bacteriol., 176, 5753–5761 (1994). 10. Van Sinderen, D., ten Berg, A., Hayema, B. J., Hamoen, L., and Venema, G.: Molecular cloning and sequence of comK, a gene required for genetic competence in Bacillus subtilis. Mol. Microbiol., 11, 695–703 (1994).
262
J. BIOSCI. BIOENG.,
SAITO ET AL.
11. Van Sinderen, D. and Venema, G.: comK acts as an autoregulatory control switch in the signal transduction route to competence in Bacillus subtilis. J. Bacteriol., 176, 5762–5770 (1994). 12. Van Sinderen, D., Luttinger, A., Kong, L., Dubnau, D., Venema, G., and Hamoen, L.: comK encodes the competence transcription factor (CTF), the key regulatory protein for competence developement in Bacillus subtilis. Mol. Microbiol., 15, 455–462 (1995). 13. Chen, I. and Dubnau, D.: DNA uptake during bacterial transformation. Nat. Rev. Microbiol., 2, 241–249 (2004). 14. Kelly, M. S. and Pritchard, R. H.: Unstable linkage between genetic markers in transformation. J. Bacteriol., 89, 1314–1321 (1965). 15. Bettinger, G. E. and Young, F. E.: Transformation of Bacillus subtilis using gently lysed L-forms: a new mapping technique. Biochem. Biophys. Res. Commun., 55, 1105–1111 (1973). 16. Bettinger, G. E. and Young, F. E.: Transformation of Bacillus subtilis: transforming ability of deoxyribonucleic acid in lysates of L-forms or protoplasts. J. Bacteriol., 122, 987–993 (1975). 17. Akamatsu, T. and Sekiguchi, J.: Characterization of chromosome and plasmid transformation in Bacillus subtilis using gently lysed protoplasts. Arch. Microbiol., 146, 353–357 (1987). 18. Akamatsu, T., Moriyama, H., and Kaneda, T.: Genetic mapping of bfmA mutation causing fatty acid deficiency in Bacillus subtilis. Can. J. Microbiol., 39, 629–633 (1993). 19. Akamatsu, T. and Taguchi, H.: Interspecific transformation of Bacillus subtilis competent cells by chromosomal DNA in
20. 21.
22. 23. 24.
25.
26. 27.
lysates of protoplasts of Bacillus amyloliquefaciens. Biosci. Biotechnol. Biochem., 64, 275–279 (2000). Spizizen, J.: Transformation of biochemically deficient strains of Bacillus subtilis by deoxyribonucleate. Proc. Natl. Acad. Sci. USA, 44, 1072–1078 (1958). Akamatsu, T. and Sekiguchi, J.: An improved method of protoplast regeneration for Bacillus species and its application to protoplast fusion and transformation. Agric. Biol. Chem., 48, 651–655 (1984). Akamatsu, T. and Sekiguchi, J.: Genetic mapping by means of protoplast fusion in Bacillus subtilis. Mol. Gen. Genet., 208, 254–262 (1987). Anagnostopoulos, C. and Spizizen, J.: Requirements for transformation in Bacillus subtilis. J. Bacteriol., 81, 741–746 (1961). Kunst, F., Ogasawara, N., Moszer, I., Albertini, A. M., Alloni, G., Azevedo, V., Bertero, M. G., Bessieres, P., Bolotin, A., Borchert, S., and other 141 authors: The complete genome sequence of the gram-positive bacterium Bacillus subtilis. Nature, 390, 249–256 (1997). Akamatsu, T. and Taguchi, H.: A simple and rapid method for intra- and interspecific transformation of Bacillus subtilis on solid media by DNA in protoplast lysates. Biosci. Biotechnol. Biochem., 65, 446–448 (2001). Kogoma, T.: Stable DNA replication: interplay between DNA replication, homologous recombination, and transcription. Microbiol. Mol. Biol. Rev., 61, 212–238 (1997) Akamatsu, T. and Sekiguchi, J.: Selection methods in bacilli for recombinants and transformants of intra- and interspecific fused protoplasts. Arch. Microbiol., 134, 303–308 (1987).