Differential and Dynamic Localization of Topoisomerases in Bacillus subtilis
http://www.100md.com
《细菌学杂志》
Institut fur Mikrobiologie, Stefan Meier Str. 19, Albert-Ludwigs Universitt Freiburg, 79104 Freiburg, Germany
ABSTRACT
Visualization of topoisomerases in live Bacillus subtilis cells showed that Topo I, Topo IV, and DNA gyrase differentially localize on the nucleoids but are absent at cytosolic spaces surrounding the nucleoids, suggesting that these topoisomerases interact with many regions of the chromosome. While both subunits of Topo IV were uniformly distributed throughout the nucleoids, Topo I and gyrase formed discrete accumulations, or foci, on the nucleoids in a large fraction of the cells, which showed highly dynamic movements. Three-dimensional time lapse microscopy showed that gyrase foci accumulate and dissipate within a 1-min time scale, revealing dynamic assembly and disassembly of subcelluar topoisomerase centers. Gyrase centers frequently colocalized with the central DNA replication machinery, suggesting a major role for gyrase at the replication fork, while Topo I foci were frequently close to or colocalized with the structural maintenance of chromosomes (SMC) chromosome segregation complex. The findings suggest that different areas of supercoiling exist on the B. subtilis nucleoids, which are highly dynamic, with a high degree of positive supercoiling attracting gyrase to the replication machinery and areas of negative supercoiling at the bipolar SMC condensation centers recruiting Topo I.
INTRODUCTION
Chromosomes must be compacted to fit into the nucleus or the bacterial cell, and DNA topology must be regulated to allow transcription and replication to take place. DNA compaction is regulated by histones or histone-like proteins, which wind DNA around themselves; by topoisomerases, which affect supercoiling directly through changes in the linking number (the number of times DNA strands pass around each other); and by SMC proteins. Homeostasis of global supercoiling of the chromosome (which is negatively supercoiled in mesophilic bacteria) is regulated by the antagonistic action of topoisomerase (Topo) II (DNA gyrase), which actively introduces negative supercoils, and by Topo I, which partially relaxes negative supercoils (7). There are two additional topoisomerases, whose functions in global supercoiling are less clear. Topo III is not essential (in contrast to the other three enzymes) and appears to be involved in DNA recombination (33), while Topo IV is required for the separation of interlinked sister chromosomes (15), which arise during replication and DNA recombination and would resist complete segregation of sister chromosomes. Topo I (encoded by the topA gene) and III belong to the type I family of topoisomerases. They are composed of a single polypeptide and change supercoiling through breakage, strand passage, and resealing of a single DNA strand. Conversely, type II topoisomerases (gyrase and Topo IV) break both strands of the DNA duplex and move an unbroken DNA strand through this gap before resealing it and are composed of two different polypeptides (encoded by gyrA and gyrB for gyrase and by parC and parE for Topo IV). In vitro, gyrase can relax positive supercoils and actively introduce negative supercoils and can decatenate interlinked DNA molecules (28). Topo IV preferentially relaxes positive supercoils and has a much higher decatenation activity than gyrase. Topo I and III can relax negative supercoils, and Topo III can also decatenate circular DNA molecules (4). However, their detailed contributions to chromosomal processes in vivo are still unclear.
SMC proteins are key players in various chromosome dynamics in most organisms. Their functions include cohesion of sister chromosomes, DNA double-strand break repair, chromosome compaction, and maintenance of chromosome structure (12). SMC proteins invariably act within protein complexes, containing an SMC homodimer (in prokaryotic cells) or a heterodimer, and additionally, non-SMC proteins. In prokaryotes, an SMC homodimer forms a complex with ScpA, belonging to the kleisin protein family, and with ScpB, both of which are widely conserved in bacteria and archaea (22, 27). In vitro, condensin can introduce positive writhe into DNA through an as yet ill-defined mechanism, that is, it bends the DNA into a right-handed superhelix (16). This tension can be relieved by Topo I, the net result being that negative supercoiling is introduced into DNA. In vivo, condensin and the prokaryotic SMC complex are thought to mediate chromosome compaction through the introduction of negative supercoiling into DNA. This is based on two important observations, (i) that reduction of Topo I activity is a suppressor of a mukB deletion in E. coli (the organism's equivalent of smc) (26), and (ii) that smc mutant cells are hypersensitive to inhibitors of DNA gyrase (20). The bacterial SMC complex localizes in two discrete centers, one within each cell half (21, 22), while DNA is replicated by centrally located replication machinery (18). DNA replication depends on removal of positive supercoils that arise ahead of the replication fork because of opening of the DNA duplex. These supercoils could be removed by gyrase and Topo IV, based on their biochemical properties. Alternatively, or additionally, the supercoils could be converted into precatenanes, when duplicated sister strands wind around each other behind the replication fork and become intertwined (7). In vitro, such entanglements can be removed by gyrase, Topo III, or Topo IV, but it is unclear which enzyme(s) mediates these functions in vivo.
In Escherichia coli, Topo IV has an unusual pattern of localization: one of its subunits (ParC) is associated with the replication machinery that is localized close to the cell center, whereas the other subunit, ParE, was found in DNA-free spaces, that is, mostly close to the cell poles or in the middle of cells containing two separated nucleoids (6). Topo IV activity was found to be repressed until late in the cell cycle, when DNA replication is almost or fully complete, suggesting that ParC is sequestered by the replication machinery and released to form a functional complex with ParE during the final steps of chromosome segregation, when DNA decatenation is most needed (6). Conversely, in Bacillus subtilis, ParC was shown to form foci close to the cell poles, away from the nucleoids, while ParE was present throughout the cells, the relevance of which is unknown (13).
In this work, we have explored the localization of B. subtilis topoisomerases in live cells. We found that Topo I, II, and IV localize to the nucleoids, but each in a distinct pattern, most likely reflecting their different roles in vivo. While gyrase formed dynamic accumulations that mostly colocalized with the replication machinery, Topo I formed discrete centers in the proximity of the SMC complex, revealing a spatial connection between topoisomerases, the SMC complex, and the replisome in vivo.
MATERIALS AND METHODS
Growth conditions. Escherichia coli XL1-Blue (Stratagene) or B. subtilis strains were grown in Luria-Bertani rich medium supplemented with 50 μg/ml ampicillin or other antibiotics where appropriate. For microscopic analysis, Bacillus strains were grown in S750 defined medium (14) complemented with 1% Casamino Acids. For induction of the xylose promoter, 0.5% xylose was added to the medium, and glucose was exchanged for fructose and xylose in S750 medium. Antibiotics were used at the following final concentrations: chloramphenicol (Cm) at 5 μg/ml, spectinomycin (Spec), 100 μg/ml; Mls (erythromycin, 1 μg/ml, and lincomycin, 25 μg/ml); kanamycin (Kan), 10 μg/ml; and tetracycline (Tet), 10 μg/ml.
Construction of plasmids and bacterial strains. Plasmid pSG1164 (19) was modified to carry cerulean cyan fluorescent protein (CFP), a novel version of CFP having much brighter fluorescence (25), or yellow fluorescent protein (YFP). Green fluorescent protein (GFP) was removed by digestion with EcoRI and SpeI and was exchanged for PCR-amplified cerulean CFP or YFP to yield pCCFP or p64YFP, respectively. To create C-terminal fusions of genes encoding the Topo IV subunits, parE and parC, with gfp, yfp or cfp, the 3' (561-bp) region of parE or the 3' (658-bp) parC region was amplified by PCR, and each was cloned into ApaI and EcoRI sites of pSG1164, pCCFP, or p64YFP, resulting in pgST1, pyST2, and pcST3 and pgST4, pyST5, and pcST6, respectively. The C-terminal fusions of topA (encoding Topo I) with gfp, yfp, and cfp were created by cloning the 3' (688-bp) region of the topA gene into KpnI and XhoI sites of pSG1164 or of pCCFP, resulting in pgST7, pyST8, and pcST9. The C-terminal GFP fusions of gyrA (encoding one subunit of gyrase) were created by cloning the 3' (678-bp) region of the gyrA gene into the KpnI and XhoI sites of pSG1164, pCCFP, or p64YFP, creating pgST10, pcST11, and pyST12. B. subtilis PY79 cells were transformed with the different plasmids using a two-step method (10) and selecting for Cm resistance, yielding the strains listed in Table 1. The expression of full-length GFP fusions was verified by Western blotting for all constructs using anti-GFP antibodies.
Strain ST15 (gyrA-yfp dnaX-cfp) was generated by transformation of strain PG28 with chromosomal DNA from strain ST11, selecting for Cm and Spec resistance. Strain ST14 (topA-cfp smc-yfp) was generated by transformation of strain ST9 with chromosomal DNA from strain JM25 (22), selecting for Cm and Mls resistance. Strains ST22 (smc::kan gyrA-gfp) and ST23 (smc::kan topA-gfp) were generated by transformation of strain ST10 or strain ST7, respectively, with chromosomal DNA from strain PG388 (8), selecting for Cm and Kan resistance. Strains ST21 and ST24 were generated by transformation of strain ST9 with chromosomal DNA from strain JM8 or JM9 (22), respectively. Strain ST26 (gyrA-yfp Venus YFP) was generated by substitution of YFP for Venus YFP (23) in plasmid pyST12 (3' gyrA-yfp) and transformation into PY79, selecting for Cm resistance. Strain ST27 (gyrA-cfp smc-yfp) was generated by transformation of strain ST12 with chromosomal DNA from JM25, selecting for Cm and Mls resistance. Strain ST28 (dnaX-cfp topA-yfp) was generated by transformation of strain ST8 with chromosomal DNA from PG28, selecting for Cm and Spec resistance.
Image acquisition. Fluorescence microscopy was performed on an Olympus AX70 microscope. Cells were mounted on agarose gel pads containing S750 growth medium on microscope slides. Images were acquired with a digital charge-coupled device camera; signal intensities and cell lengths were measured using the Metamorph 5.0 program (Universal Imaging Corp.). DNA was stained with 4',6-diamidino-2-phenylindole (DAPI; final concentration, 0.2 ng/ml), and membranes were stained with FM4-64 (final concentration, 1 nM).
The average maximal signal intensity (the mean of the measurement of high values of signal intensities as measured using Metamorph software) for GyrA-YFP foci was 115 units (after subtraction of background fluorescence, which was 85 units on average, with 50 cells analyzed), and 90 units for nucleoid background staining (i.e., staining of nucleoids lacking foci or of nucleoid areas not containing foci) of cells expressing GyrA-YFP, so the difference in maximal intensity was 22% between foci and general nucleoid staining. For Topo I-YFP-expressing cells, the maximum intensity of foci was 177 units, and that of nucleoid background staining was 149 units (with 52 cells analyzed), yielding a 17% difference.
Three-dimensional (3D) time lapse microscopy was performed by taking an image at the focal plane that was defined as home, followed by successive acquisitions at 0.2 mm above and below the home plane, as defined in the Metamorph software.
RESULTS
We wished to localize all essential B. subtilis topoisomerases in live cells. We created various C-terminal fusions of topoisomerases to GFP or to its fluorescent variants. All of the fusions were the only copies of the respective topoisomerase or enzyme subunit and were driven by the original promoter. Western blot analysis showed that all protein fusions were expressed as full-length proteins (data not shown). All engineered cells grew indistinguishably from wild-type cells and showed wild-type nucleoid morphology. In the case of the parC-gfp fusion, we found a mild condensation defect, as shown by decondensed nucleoids in 1 to 2% of the cells, but for the purposes of this study, the parC-gfp fusion was effectively fully functional. All labeled proteins localized to the nucleoids, but in considerably different manners.
Dynamic localization of topoisomerase II (DNA gyrase) in live cells. To visualize DNA gyrase, we tagged the GyrA subunit with GFP, which showed a clear nucleoid stain in 96% of the cells (Fig. 1A). Interestingly, a subfraction (34%) of gyrA-gfp-expressing cells showed fluorescent patches or discrete foci on the nucleoids (Fig. 1A). These foci did not simply coincide with a larger amount of DNA, because foci were apparent at sites that did not show a higher concentration of DNA as visualized with DAPI (Fig. 1A, compare first and second panels). Thus, fluorescent patches and foci most likely represented an accumulation of gyrase independent of the actual amount of DNA at the site. To understand why the foci were present in only a subpopulation of the cells, and to gain further insight into the dynamics of GyrA-GFP localization, we performed 3D time lapse microscopy experiments. At each time interval, three images were taken, one in the central (focal) plane of the cell, called "home"; one 0.2 μm above the home plane (called "top"); and one 0.2 μm below the home plane (called "bottom"). The rationale of this technique was to investigate if GyrA-YFP foci (YFP was used because of its lower bleaching properties) are static, and indeed only present in a subset of the cells, or are dynamic and move to a different focal plane (so that they are not visible in many cells) or disassemble between time intervals, i.e., cannot be found in different focal planes in different time intervals. When the spacing between the home and top or bottom plane was set to 0.3 μm, fluorescence signals were completely out of focus, while with a spacing of 0.2 μm, relatively clear top and bottom signals could still be obtained from a variety of fluorescent-protein fusions, showing that 0.2 μm above or below the home plane can capture out-of-focus signals away from the central cell axis. Figure 2A and B shows images taken in 2-min time intervals, with top, home, and bottom sections shown below each other. It is apparent from Fig. 2A that the localization of GyrA-YFP is different in most of the cells between each pair of time intervals. Moreover, a GyrA-YFP focus in one cell can be seen to assemble in the cell center at minute 0 and to disassemble at minute 2 in all three planes, to reassemble in the cell center at minute 4, while at minute 6, GyrA-YFP is again more uniformly distributed throughout the nucleoid. Similar assembly and disassembly kinetics were observed in 15 cells, while in only three cases did GyrA-YFP assemblies remain visible for more than four time intervals (i.e., 8 min). In most movies, only assembly or disassembly of foci could be captured, due to the rapid bleaching of the signal (because of the threefold exposure at different Z planes) and the need for perfect focusing throughout the time lapse. However, in 60% of the cells, GyrA-YFP changed from a dispersed localization to an accumulated form, or conversely, the accumulations dispersed within few minutes. Unlike GyrA-YFP, the DnaX-CFP subunit of DNA polymerase did not show comparable dynamics: the signal could be well traced over 8 min without a single disassembly/assembly occurrence seen in 55 cells observed (Fig. 2B). A clear disassembly and reassembly reaction was seen in only four cells. Some focal drift is apparent, as in Fig. 2B: in the 4-min panel, one focus has moved to the bottom plane, while in the 6-min panel, one focus is more clearly seen in the top panel, but the foci did not disassemble between time intervals. Thus, these experiments show that GyrA-YFP assembles and disassembles in a dynamic manner in most of the cells analyzed, which explains in part why only a third of the cells show GyrA-GFP foci in a given population. In addition, Fig. 2A shows that GyrA-GFP assemblies can be observed only in the home plane, but rarely in the top or bottom plane (and thus in a very narrow focal window), while DnaX-CFP foci can often be captured in all three planes (Fig. 2B). Thus, because GyrA-GFP foci assemble and disassemble in a majority of cells observed, it is clear that during the 120-min cell cycle under the experimental conditions used, most cells have gyrase centers that are visible only in a subset at any given time.
Interestingly, of those cells containing visible GyrA-GFP foci, most contained a single focus close to the middle of the cell (Fig. 1A). To obtain a statistical representation of the localization of GyrA-GFP foci, we measured the distances from foci to the nearest cell pole relative to the size of the cell in a large number of cells that were representative of a growing population of cells, as explained above. In Fig. 3A, it is apparent that only large cells (<3 μm) contained two bipolar GyrA-GFP foci, while small cells predominantly possessed a single central focus. Therefore, bipolar gyrase centers arise relatively late during the cell cycle compared with the early separation of SMC foci (22) or of Topo I-GFP foci (see below). We were able to capture three movies of cells (out of 125 cells monitored) in which a single central GyrA-GFP focus split into two bipolar foci within a few minutes. In Fig. 2C, a central GyrA-GFP focus resolves into two foci located within each cell half between minutes 1 and 3. At minute 2, two foci appear to be present, one still close to the cell center, while the other is positioned within one cell half. Due to the lack of a larger number of movies with foci that are clearly visible through an extended period of time (since they are so dynamic), we could not determine whether the central foci split into two foci, one of which moved earlier than the other (similar to the SMC complex [31]) or whether they moved simultaneously (similar to the movement of replication origins [32]). Also, we could not conclude that the rapid movement is a general property of GyrA-YFP foci. However, it is clear that within 2 to 4 min, gyrase foci can change from a central to a bipolar pattern of localization.
Dynamic localization of topoisomerase I in live cells. A Topo I-GFP fusion localized to the nucleoids in all 460 of the cells monitored. In 66% of these cells, Topo I-GFP localized rather uniformly throughout the nucleoids, but in 34% of the cells, distinct concentrations, or foci, of Topo I-GFP were apparent (Fig. 1B). Similar to gyrase, Topo I-GFP foci were seen at sites that did not show a higher degree of DNA staining (Fig. 1B). To investigate why only a fraction of the cells contained visible foci, we performed time lapse microscopy. From several experiments, it is clear that like GyrA-GFP, Topo I-GFP foci were highly dynamic; they were absent or present at 1-min intervals (Fig. 2D). In the left-hand cell in Fig. 2D, a focus can be seen to disappear at minute 1, to reappear at minute 2, and to disappear again at minute 5. In the right-hand cell, the indicated focus disappears at minute 3 and reappears 1 minute later. Occasionally, foci were seen to split into two foci that united again a few minutes later (Fig. 2D). Similar time lapse experiments with the SMC complex (e.g., with ScpA-YFP) that also forms discrete bipolar foci showed that ScpA-YFP signals were visible at each time interval through 15 to 20 exposures at 1-min intervals (Fig. 2E) (22), showing that the fluctuations in Topo I localization were not caused by simple focal drift during capturing of the images. Because Topo I-GFP showed even greater bleaching than GyrA-YFP, we were not able to obtain extended 3D time lapse movies. However, similar to gyrase, Topo I foci were found to assemble or disassemble within a few minutes (data not shown), so that most likely over time Topo I foci are present in many more cells than just the observed subset.
Of the cells containing Topo I foci, 26% had a single focus, while 66% had two foci and 8% showed multiple foci. To investigate the positioning of the Topo I foci, we measured the distances of the foci to the nearest cell poles, which were plotted relative to cell size in Fig. 3B. It is apparent that in small cells, a single focus is generally present close to the middle of the cells, while in middle-size (>2 μm) and large (>3 μm) cells, one focus was present mostly close to one cell pole, or more frequently, two foci were located in a bipolar manner (that is, one close to each cell pole). These experiments suggest that Topo I centers are formed in a cell cycle-dependent manner, but in a manner different from that of gyrase. Middle-size cells contained two (bipolar) Topo I centers, but usually a central GyrA focus (Fig. 2A and B), showing that separation of Topo I-GFP foci occurred earlier in the cell cycle than with GyrA-GFP foci. Thus, Topo I and gyrase form discrete centers on the nucleoids; however, they are distinct from each other. The distinct localization patterns were further supported by dual-labeling experiments, which are described below.
DNA gyrase centers generally coincide with DNA polymerase, while Topo I centers and SMC protein localize to similar regions on the nucleoids. The formation of patches, or foci, whose positions were cell cycle dependent for gyrase and Topo I was highly similar to patterns of localization described for DNA polymerase, which is a stationary complex at the cell center, or the SMC complex, which is usually located in a bipolar fashion, respectively. While two bipolar DNA polymerase centers form within each daughter cell late in the cell cycle, the SMC complex moves from a central position toward opposite cell poles early in the cell cycle, so that most frequently, two bipolar SMC centers are present in growing cells (21, 22). The similarity of the patterns and localizations of gyrase and Topo I prompted us to generate dually labeled strains for simultaneous visualization of topoisomerases relative to DnaX, the tau subunit of the replisome, or to SMC, ScpA, or ScpB, which form the SMC complex. From Fig. 1C, it is apparent that gyrase centers were often coincident with the replication machinery or localized in close proximity. Of 380 cells investigated, 72 GyrA-YFP and DnaX-CFP foci were coincident (that is, in 57% of the cells showing GyrA-YFP foci or patches) and 46 were in close proximity (i.e., within 0.2 μm in 37% of the cells showing GyrA-YFP foci), while 8 foci (6%) were clearly separate. In the remaining cells, no clear GyrA-YFP foci were apparent. Thus, gyrase centers are most frequently located close to the central replisome. In agreement with this, DnaX-CFP shows a cell 3cycle-dependent localization similar to that of gyrase: small and medium-size cells generally contain a single central DnaX-CFP focus, while large cells (between 2.5 and 3 μm; compare Fig. 3A) contain two bipolar foci (3, 17).
The cell cycle-dependent localization of gyrase centers implies that at higher growth rates, cells should contain more foci than the slow-growing cells used for the studies described above, because at higher growth rates, the numbers of replication forks and of replication centers increase (17). To test this idea, we grew GyrA-YFP-expressing cells at 25°C in minimal or rich medium, where single or multiple rounds of replication take place, respectively. Indeed, similar to DnaX-CFP (17), the number of cells containing two GyrA-GFP foci increased at the expense of cells having one GyrA-GFP focus at the higher growth rate. In minimal medium, of all cells showing GyrA-YFP foci, 48.2% contained one focus, 44% two foci, and 4.7 or 3.1% three or four foci (Fig. 4A; 350 cells analyzed), whereas in rich medium, 31.3% contained one focus, 52.7% two foci, 8.9% three foci, and 7.1% four foci (Fig. 4B; 340 cells analyzed). These experiments support a close cell cycle-dependent association of gyrase foci or patches with the replication machinery.
Simultaneous visualization of Topo I-CFP and SMC-YFP revealed that in 30% of the cells containing Topo I-CFP foci or patches, both signals colocalized, while in 60% of the cells, both signals were in close proximity, or clearly apart in 10% of the cells (with 332 cells analyzed) (Fig. 1D). As before, 68% of the cells did not show clear Topo I foci or patches. Simultaneous labeling of Topo I and ScpA or ScpB yielded similar results (data not shown). Therefore, Topo I centers are frequently close to, or to a lesser extent, coincident with the SMC condensation centers. Of note, in about 20% of cells expressing both Topo I-CFP and SMC-YFP, SMC foci showed irregular positioning or were difficult to detect (data not shown) compared to about 5% of cells expressing SMC-YFP only, in which no foci were detectable. Additionally, cells with the dual label occasionally showed aberrant nucleoids (about 5%) (data not shown), indicating that the GFP tags, when present on both Topo I and SMC, slightly interfere with the function of SMC protein, showing a possible direct interaction.
To verify the specific localization patterns of Topo I and gyrase, both fluorescently labeled fusions were combined with other GFP fusions that were expected not to colocalize with the respective topoisomerase. Indeed, GyrA-CFP foci were generally separate from SMC-YFP foci (Fig. 1E; 85% separate versus 15% close/coincident in 90 cells containing well-defined foci), and similar to SMC-YFP (22, 31), Topo I-YFP foci were generally separate from DnaX-CFP foci (Fig. 1F; 82% separate versus 18% close/coincident in 115 cells containing well-defined foci). These experiments support the similar localization patterns of gyrase and the DNA polymerase, and of Topo I and the SMC complex.
Topoisomerase I and gyrase are also present in anucleate cells. The SMC chromosome condensation complex is strongly associated with chromosomal DNA, as it is present only in nucleated and not in anucleate cells (22). We wished to investigate if Topo I and gyrase have similar properties. Therefore, the smc gene was deleted in Topo I-GFP- or GyrA-GFP-expressing cells, and the localization was monitored at 25°C in minimal medium, which is permissive for growth of smc mutant cells. In both cases, the pattern of localization of the topoisomerases was disturbed. GyrA-GFP was present in all 131 of the anucleate cells monitored as staining of the entire cell, either smooth staining (Fig. 5A) or with brighter patches at random places in the cell. Likewise, 23 anucleate cells contained weak but detectable Topo I-GFP fluorescence, while 11 anucleate cells had high levels similar to nucleated cells, and 4 cells had only background staining (Fig. 5B). Likewise, GyrA-GFP or Topo I-GFP was present in anucleate cells of a spo0J mutant strain (data not shown). We conclude that Topo I and gyrase are also present in anucleate cells and that therefore, not all topoisomerase molecules are bound to DNA, in marked contrast to the SMC complex. In a few anucleate cells (10%), GyrA-GFP and Topo I-GFP formed foci, which were different from those in normal cells because they formed at random positions within the cell, indicating that in the absence of DNA, topoisomerase molecules may form nonspecific aggregates.
Localization of topoisomerase IV in live cells. Unlike Topo I and gyrase, both subunits of Topo IV, ParC and ParE, showed rather uniform staining of the nucleoids, with no apparent concentration at any particular site on the nucleoids (Fig. 5C and D). To find out if localization on the nucleoid depends on the presence of both subunits, we depleted ParC in cells expressing ParE-GFP. Removal of xylose (which drives transcription of parC in the parE-gfp strain) led to strong nucleoid decondensation, cell elongation, and formation of anucleate cells until the cells lysed (Fig. 5E). Under these conditions, ParE-GFP localized mostly throughout the cells (in 86% of 180 cells analyzed) (Fig. 5E) and was also present in anucleate cells, although 14% of the cells retained ParE staining on the nucleoids. Thus, Topo IV is present throughout the nucleoid, depending on the presence of both subunits, which suggests that it interacts with many if not all regions of the chromosome. This pattern of localization is in marked contrast to a previous investigation, in which ParC was found to be present close to the cell poles away from the nucleoids (13). Although we have no easy explanation for this discrepancy, we note that our parC-gfp fusion is almost fully active, while the fusion used in the study by Huang et al. was cold sensitive below 30°C and was visible in only 6% of all cells (13).
DISCUSSION
This work describes the subcellular localization of B. subtilis topoisomerases and their spatial connection with the SMC complex and with the DNA replication machinery, providing several important conclusions. First, all three essential topoisomerases have distinct patterns of localization, each of which suggests a spatial specialization for each class of enzymes. Gyrase was present on the nucleoids and showed discrete accumulations in a large fraction of growing cells. Strikingly, these gyrase accumulations assembled and disassembled within few minutes, showing that gyrase localization is not static but highly dynamic. Assembly and disassembly dynamics were rather irregular and distinct from oscillation movements described for Min proteins or ParA proteins, the latter of which appear to oscillate from one end of the nucleoid to the other and back (5, 24). We speculate that the assemblies are due to an accumulation of supercoils, leading to the recruitment of gyrase to its substrate on the chromosome. The dynamics of gyrase localization explain why the accumulations are detectable in only a subset of the cells. Until late in the cell cycle, usually a single GyrA-GFP fluorescent focus was detectable close to the middle of the cells, and during late steps in the cell cycle, two foci were seen, one within each cell half. The accumulation of gyrase did not simply reflect an increased concentration of chromosomal DNA at this position, reinforcing the idea that the foci represent gyrase centers that form due to physical properties independent of the amount of DNA. Dual labeling of the replication machinery and gyrase verified that most frequently, gyrase centers coincide with the centrally located DNA polymerase. This is entirely consistent with biochemical and electron microscopical data showing a tight link between gyrase and the replication machinery (1, 2). During most of the cell cycle, the replication machinery is a stationary complex at midcell, and only late in the cell cycle do two new centers form, one within each cell pole (18). Likewise, two distinct gyrase foci were detectable during late stages of the cell cycle, suggesting that gyrase centers are associated with the replication machinery throughout the cell cycle. As the chromosome is moved through the replisome, the DNA duplex is unwound, which generates positive supercoils ahead of the replication fork (7). The localization of gyrase suggests that a large fraction of the active molecules are dealing with the removal of the positive supercoils. It is even possible (but yet to be investigated) that there is very little need to remove precatenanes behind the replication fork if gyrase is efficient enough to largely abrogate positive supercoils in front of the fork. It is also possible that gyrase centers reintroduce negative supercoils behind the forks, but this issue cannot yet be resolved with light microscopy.
Like gyrase, Topo I formed discrete centers on the nucleoids, which appeared to also assemble and disassemble rapidly and continuously. However, Topo I accumulations were mostly present within one or both cell halves, away from the cell middle, and were often closely associated with SMC condensation centers. Bacterial SMC protein has been shown to influence DNA supercoiling in E. coli (here, MukB is the SMC analog) and in B. subtilis, and a genetic interaction with DNA Topo I and gyrase has been established (20, 26). Loss of SMC function can be compensated for by reduction of Topo I activity (which relaxes negative supercoiling) and is exacerbated by reduction of gyrase activity, showing that SMC introduces overall negative supercoiling into DNA. Our data show that SMC is also spatially associated with high Topo I activity, in that Topo I and the SMC complex localize to similar areas on the nucleoids. SMC protein binds to DNA as a ringlike structure (9, 11, 31), in a highly cooperative and repetitive manner (29). This is thought to generate positive writhe within the DNA, which is translated into negative supercoils that are a substrate for Topo I (16). Our findings imply that the SMC condensation centers generate a high degree of negative supercoiling, which is at least partially removed by Topo I in the vicinity of the SMC centers.
Unlike Topo I and gyrase, and E. coli cells, both subunits of Topo IV localized throughout the nucleoids, without any apparent concentration at distinct sites. The pattern of localization suggests not only a role for Topo IV in the decatenation of chromosomes at the end of the cell cycle, as seems to be the case in E. coli (6), but also a general role in chromosome supercoiling. This is supported by our recent finding that the depletion of Topo IV affects global chromosome compaction and protein synthesis and that overproduction of Topo IV can rescue the condensation defect and the defect in global protein synthesis caused by the loss of SMC activity (30). Conversely, our results suggest that the amount of Topo IV that is uniformly present throughout the nucleoids is sufficient to decatenate the termini without the need for an increase in concentration at the termini.
In toto, our results suggest that regions with different degrees of supercoiling exist on the nucleoids in B. subtilis cells, which fluctuate within a 1-min time frame and are thus highly dynamic. These regions correspond to the region of active replication at midcell, where positive supercoils need to be removed, and to the regions close to the bipolarly located SMC complex, in which newly replicated DNA is thought to be condensed and organized for efficient chromosome segregation, where positive writhe might be relaxed by Topo I. So, although Topo I, II, and IV are present throughout the nucleoids, higher levels of Topo I and gyrase appear to be required at specific places on the nucleoids. Topo IV interacts with many, if not all, sites on the chromosomes throughout the cell cycle and thus appears to perform a more general role in supercoiling than Topo IV in E. coli. The SMC complex is involved in general supercoiling, but also in active separation of sister chromosomes, most likely through its specific localization within each cell half. A close spatial connection with Topo I and genetic interactions with Topo I and gyrase underline the intricate interplay of the SMC complex with topoisomerases, whose molecular bases will be interesting for future studies.
ACKNOWLEDGMENTS
We thank Judita Mascarenhas for help and advice in this project.
The work was supported by the Deutsche Forschungsgemeinschaft (Emmy Noether and Heisenberg Programm).
REFERENCES
Baker, T. A., B. E. Funnell, and A. Kornberg. 1987. Helicase action of DnaB protein during replication from the Escherichia coli chromosomal origin in vitro. J. Biol. Chem. 262:6877-6885.
Baker, T. A., K. Sekimizu, B. E. Funnell, and A. Kornberg. 1986. Extensive unwinding of the plasmid template during staged enzymatic initiation of DNA replication from the origin of the Escherichia coli chromosome. Cell 45:53-64.
Defeu Soufo, H. J., and P. L. Graumann. 2005. Bacillus subtilis actin-like protein MreB influences the positioning of the replication machinery and requires membrane proteins MreC/D and other actin-like proteins for proper localization. BMC Cell Biol. 6:10.
Dekker, N. H., V. V. Rybenkov, M. Duguet, N. J. Crisona, N. R. Cozzarelli, D. Bensimon, and V. Croquette. 2002. The mechanism of type IA topoisomerases. Proc. Natl. Acad. Sci. USA 99:12126-12131.
Ebersbach, G., and K. Gerdes. 2004. Bacterial mitosis: partitioning protein ParA oscillates in spiral-shaped structures and positions plasmids at mid-cell. Mol. Microbiol. 52:385-398.
Espeli, O., C. Levine, H. Hassing, and K. J. Marians. 2003. Temporal regulation of topoisomerase IV activity in E. coli. Mol. Cell 11:189-201.
Espeli, O., and K. J. Marians. 2004. Untangling intracellular DNA topology. Mol. Microbiol. 52:925-931.
Graumann, P. L., R. Losick, and A. V. Strunnikov. 1998. Subcellular localization of Bacillus subtilis SMC, a protein involved in chromosome condensation and segregation. J. Bacteriol. 180:5749-5755.
Gruber, S., C. H. Haering, and K. Nasmyth. 2003. Chromosomal cohesin forms a ring. Cell 112:765-777.
Harwood, C. R., and S. M. Cutting. 1990. Molecular biological methods for Bacillus. Wiley, New York, N.Y.
Hirano, M., and T. Hirano. 2002. Hinge-mediated dimerization of SMC protein is essential for its dynamic interaction with DNA. EMBO J. 21:5733-5744.
Hirano, T. 2002. The ABCs of SMC proteins: two-armed ATPases for chromosome condensation, cohesion, and repair. Genes Dev. 16:399-414.
Huang, W. M., J. L. Libbey, P. van der Hoeven, and S. X. Yu. 1998. Bipolar localization of Bacillus subtilis topoisomerase IV, an enzyme required for chromosome segregation. Proc. Natl. Acad. Sci. USA 95:4652-4657.
Jaacks, K. J., J. Healy, R. Losick, and A. D. Grossman. 1989. Identification and characterization of genes controlled by the sporulation regulatory gene spo0H in Bacillus subtilis. J. Bacteriol. 171:4121-4129.
Kato, J., Y. Nishimura, R. Imamura, H. Niki, S. Hiraga, and H. Suzuki. 1990. New topoisomerase essential for chromosome segregation in E. coli. Cell 63:393-404
Kimura, K., V. V. Rybenkov, N. J. Crisona, T. Hirano, and N. R. Cozzarelli. 1999. 13S condensin actively reconfigures DNA by introducing global positive writhe: implications for chromosome condensation. Cell 98:239-248.
Lemon, K. P., and A. D. Grossman. 1998. Localization of bacterial DNA polymerase: evidence for a factory model of replication. Science 282:1516-1519.
Lemon, K. P., and A. D. Grossman. 2000. Movement of replicating DNA through a stationary replisome. Mol. Cell 6:1321-1330.
Lewis, P. J., and A. L. Marston. 1999. GFP vectors for controlled expression and dual labelling of protein fusions in Bacillus subtilis. Gene 227:101-110.
Lindow, J. C., R. A. Britton, and A. D. Grossman. 2002. Structural maintenance of chromosomes protein of Bacillus subtilis affects supercoiling in vivo. J. Bacteriol. 184:5317-5322.
Lindow, J. C., M. Kuwano, S. Moriya, and A. D. Grossman. 2002. Subcellular localization of the Bacillus subtilis structural maintenance of chromosomes (SMC) protein. Mol. Microbiol. 46:997-1009.
Mascarenhas, J., J. Soppa, A. Strunnikov, and P. L. Graumann. 2002. Cell cycle dependent localization of two novel prokaryotic chromosome segregation and condensation proteins in Bacillus subtilis that interact with SMC protein. EMBO J. 21:3108-3118.
Nagai, T., K. Ibata, E. S. Park, M. Kubota, K. Mikoshiba, and A. Miyawaki. 2002. A variant of yellow fluorescent protein with fast and efficient maturation for cell-biological applications. Nat. Biotechnol. 20:87-90.
Raskin, D. M., and P. A. de Boer. 1999. Rapid pole-to-pole oscillation of a protein required for directing division to the middle of Escherichia coli. Proc. Natl. Acad. Sci. USA 96:4971-4976.
Rizzo, M. A., G. H. Springer, B. Granada, and D. W. Piston. 2004. An improved cyan fluorescent protein variant useful for FRET. Nat. Biotechnol. 22:445-449.
Sawitzke, J. A., and S. Austin. 2000. Suppression of chromosome segregation defects of Escherichia coli muk mutants by mutations in topoisomerase I. Proc. Natl. Acad. Sci. USA 97:1671-1676.
Soppa, J., K. Kobayashi, M. F. Noirot-Gros, D. Oesterhelt, S. D. Ehrlich, E. Dervyn, N. Ogasawara, and S. Moriya. 2002. Discovery of two novel families of proteins that are proposed to interact with prokaryotic SMC proteins, and characterization of the Bacillus subtilis family members ScpA and ScpB. Mol. Microbiol. 45:59-71.
Stone, M. D., Z. Bryant, N. J. Crisona, S. B. Smith, A. Vologodskii, C. Bustamante, and N. R. Cozzarelli. 2003. Chirality sensing by Escherichia coli topoisomerase IV and the mechanism of type II topoisomerases. Proc. Natl. Acad. Sci. USA 100:8654-8659.
Strick, T. R., T. Kawaguchi, and T. Hirano. 2004. Real-time detection of single-molecule DNA compaction by condensin I. Curr. Biol. 14:874-880.
Tadesse, S., J. Mascarenhas, B. Ksters, A. Hasilik, and P. L. Graumann. 2005. Genetic interaction of the SMC complex with topoisomerase IV in Bacillus subtilis. Microbiology 151:3729-3737.
Volkov, A., J. Mascarenhas, C. Andrei-Selmer, H. D. Ulrich, and P. L. Graumann. 2003. A prokaryotic condensin/cohesin-like complex can actively compact chromosomes from a single position on the nucleoid and binds to DNA as a ring-like structure. Mol. Cell. Biol. 23:5638-5650.
Webb, C. D., P. L. Graumann, J. Kahana, A. A. Teleman, P. Silver, and R. Losick. 1998. Use of time-lapse microscopy to visualize rapid movement of the replication origin region of the chromosome during the cell cycle in Bacillus subtilis. Mol. Microbiol. 28:883-892.
Zhu, Q., P. Pongpech, and R. J. DiGate. 2001. Type I topoisomerase activity is required for proper chromosomal segregation in Escherichia coli. Proc. Natl. Acad. Sci. USA 98:9766-9771.(Serkalem Tadesse and Pete)
ABSTRACT
Visualization of topoisomerases in live Bacillus subtilis cells showed that Topo I, Topo IV, and DNA gyrase differentially localize on the nucleoids but are absent at cytosolic spaces surrounding the nucleoids, suggesting that these topoisomerases interact with many regions of the chromosome. While both subunits of Topo IV were uniformly distributed throughout the nucleoids, Topo I and gyrase formed discrete accumulations, or foci, on the nucleoids in a large fraction of the cells, which showed highly dynamic movements. Three-dimensional time lapse microscopy showed that gyrase foci accumulate and dissipate within a 1-min time scale, revealing dynamic assembly and disassembly of subcelluar topoisomerase centers. Gyrase centers frequently colocalized with the central DNA replication machinery, suggesting a major role for gyrase at the replication fork, while Topo I foci were frequently close to or colocalized with the structural maintenance of chromosomes (SMC) chromosome segregation complex. The findings suggest that different areas of supercoiling exist on the B. subtilis nucleoids, which are highly dynamic, with a high degree of positive supercoiling attracting gyrase to the replication machinery and areas of negative supercoiling at the bipolar SMC condensation centers recruiting Topo I.
INTRODUCTION
Chromosomes must be compacted to fit into the nucleus or the bacterial cell, and DNA topology must be regulated to allow transcription and replication to take place. DNA compaction is regulated by histones or histone-like proteins, which wind DNA around themselves; by topoisomerases, which affect supercoiling directly through changes in the linking number (the number of times DNA strands pass around each other); and by SMC proteins. Homeostasis of global supercoiling of the chromosome (which is negatively supercoiled in mesophilic bacteria) is regulated by the antagonistic action of topoisomerase (Topo) II (DNA gyrase), which actively introduces negative supercoils, and by Topo I, which partially relaxes negative supercoils (7). There are two additional topoisomerases, whose functions in global supercoiling are less clear. Topo III is not essential (in contrast to the other three enzymes) and appears to be involved in DNA recombination (33), while Topo IV is required for the separation of interlinked sister chromosomes (15), which arise during replication and DNA recombination and would resist complete segregation of sister chromosomes. Topo I (encoded by the topA gene) and III belong to the type I family of topoisomerases. They are composed of a single polypeptide and change supercoiling through breakage, strand passage, and resealing of a single DNA strand. Conversely, type II topoisomerases (gyrase and Topo IV) break both strands of the DNA duplex and move an unbroken DNA strand through this gap before resealing it and are composed of two different polypeptides (encoded by gyrA and gyrB for gyrase and by parC and parE for Topo IV). In vitro, gyrase can relax positive supercoils and actively introduce negative supercoils and can decatenate interlinked DNA molecules (28). Topo IV preferentially relaxes positive supercoils and has a much higher decatenation activity than gyrase. Topo I and III can relax negative supercoils, and Topo III can also decatenate circular DNA molecules (4). However, their detailed contributions to chromosomal processes in vivo are still unclear.
SMC proteins are key players in various chromosome dynamics in most organisms. Their functions include cohesion of sister chromosomes, DNA double-strand break repair, chromosome compaction, and maintenance of chromosome structure (12). SMC proteins invariably act within protein complexes, containing an SMC homodimer (in prokaryotic cells) or a heterodimer, and additionally, non-SMC proteins. In prokaryotes, an SMC homodimer forms a complex with ScpA, belonging to the kleisin protein family, and with ScpB, both of which are widely conserved in bacteria and archaea (22, 27). In vitro, condensin can introduce positive writhe into DNA through an as yet ill-defined mechanism, that is, it bends the DNA into a right-handed superhelix (16). This tension can be relieved by Topo I, the net result being that negative supercoiling is introduced into DNA. In vivo, condensin and the prokaryotic SMC complex are thought to mediate chromosome compaction through the introduction of negative supercoiling into DNA. This is based on two important observations, (i) that reduction of Topo I activity is a suppressor of a mukB deletion in E. coli (the organism's equivalent of smc) (26), and (ii) that smc mutant cells are hypersensitive to inhibitors of DNA gyrase (20). The bacterial SMC complex localizes in two discrete centers, one within each cell half (21, 22), while DNA is replicated by centrally located replication machinery (18). DNA replication depends on removal of positive supercoils that arise ahead of the replication fork because of opening of the DNA duplex. These supercoils could be removed by gyrase and Topo IV, based on their biochemical properties. Alternatively, or additionally, the supercoils could be converted into precatenanes, when duplicated sister strands wind around each other behind the replication fork and become intertwined (7). In vitro, such entanglements can be removed by gyrase, Topo III, or Topo IV, but it is unclear which enzyme(s) mediates these functions in vivo.
In Escherichia coli, Topo IV has an unusual pattern of localization: one of its subunits (ParC) is associated with the replication machinery that is localized close to the cell center, whereas the other subunit, ParE, was found in DNA-free spaces, that is, mostly close to the cell poles or in the middle of cells containing two separated nucleoids (6). Topo IV activity was found to be repressed until late in the cell cycle, when DNA replication is almost or fully complete, suggesting that ParC is sequestered by the replication machinery and released to form a functional complex with ParE during the final steps of chromosome segregation, when DNA decatenation is most needed (6). Conversely, in Bacillus subtilis, ParC was shown to form foci close to the cell poles, away from the nucleoids, while ParE was present throughout the cells, the relevance of which is unknown (13).
In this work, we have explored the localization of B. subtilis topoisomerases in live cells. We found that Topo I, II, and IV localize to the nucleoids, but each in a distinct pattern, most likely reflecting their different roles in vivo. While gyrase formed dynamic accumulations that mostly colocalized with the replication machinery, Topo I formed discrete centers in the proximity of the SMC complex, revealing a spatial connection between topoisomerases, the SMC complex, and the replisome in vivo.
MATERIALS AND METHODS
Growth conditions. Escherichia coli XL1-Blue (Stratagene) or B. subtilis strains were grown in Luria-Bertani rich medium supplemented with 50 μg/ml ampicillin or other antibiotics where appropriate. For microscopic analysis, Bacillus strains were grown in S750 defined medium (14) complemented with 1% Casamino Acids. For induction of the xylose promoter, 0.5% xylose was added to the medium, and glucose was exchanged for fructose and xylose in S750 medium. Antibiotics were used at the following final concentrations: chloramphenicol (Cm) at 5 μg/ml, spectinomycin (Spec), 100 μg/ml; Mls (erythromycin, 1 μg/ml, and lincomycin, 25 μg/ml); kanamycin (Kan), 10 μg/ml; and tetracycline (Tet), 10 μg/ml.
Construction of plasmids and bacterial strains. Plasmid pSG1164 (19) was modified to carry cerulean cyan fluorescent protein (CFP), a novel version of CFP having much brighter fluorescence (25), or yellow fluorescent protein (YFP). Green fluorescent protein (GFP) was removed by digestion with EcoRI and SpeI and was exchanged for PCR-amplified cerulean CFP or YFP to yield pCCFP or p64YFP, respectively. To create C-terminal fusions of genes encoding the Topo IV subunits, parE and parC, with gfp, yfp or cfp, the 3' (561-bp) region of parE or the 3' (658-bp) parC region was amplified by PCR, and each was cloned into ApaI and EcoRI sites of pSG1164, pCCFP, or p64YFP, resulting in pgST1, pyST2, and pcST3 and pgST4, pyST5, and pcST6, respectively. The C-terminal fusions of topA (encoding Topo I) with gfp, yfp, and cfp were created by cloning the 3' (688-bp) region of the topA gene into KpnI and XhoI sites of pSG1164 or of pCCFP, resulting in pgST7, pyST8, and pcST9. The C-terminal GFP fusions of gyrA (encoding one subunit of gyrase) were created by cloning the 3' (678-bp) region of the gyrA gene into the KpnI and XhoI sites of pSG1164, pCCFP, or p64YFP, creating pgST10, pcST11, and pyST12. B. subtilis PY79 cells were transformed with the different plasmids using a two-step method (10) and selecting for Cm resistance, yielding the strains listed in Table 1. The expression of full-length GFP fusions was verified by Western blotting for all constructs using anti-GFP antibodies.
Strain ST15 (gyrA-yfp dnaX-cfp) was generated by transformation of strain PG28 with chromosomal DNA from strain ST11, selecting for Cm and Spec resistance. Strain ST14 (topA-cfp smc-yfp) was generated by transformation of strain ST9 with chromosomal DNA from strain JM25 (22), selecting for Cm and Mls resistance. Strains ST22 (smc::kan gyrA-gfp) and ST23 (smc::kan topA-gfp) were generated by transformation of strain ST10 or strain ST7, respectively, with chromosomal DNA from strain PG388 (8), selecting for Cm and Kan resistance. Strains ST21 and ST24 were generated by transformation of strain ST9 with chromosomal DNA from strain JM8 or JM9 (22), respectively. Strain ST26 (gyrA-yfp Venus YFP) was generated by substitution of YFP for Venus YFP (23) in plasmid pyST12 (3' gyrA-yfp) and transformation into PY79, selecting for Cm resistance. Strain ST27 (gyrA-cfp smc-yfp) was generated by transformation of strain ST12 with chromosomal DNA from JM25, selecting for Cm and Mls resistance. Strain ST28 (dnaX-cfp topA-yfp) was generated by transformation of strain ST8 with chromosomal DNA from PG28, selecting for Cm and Spec resistance.
Image acquisition. Fluorescence microscopy was performed on an Olympus AX70 microscope. Cells were mounted on agarose gel pads containing S750 growth medium on microscope slides. Images were acquired with a digital charge-coupled device camera; signal intensities and cell lengths were measured using the Metamorph 5.0 program (Universal Imaging Corp.). DNA was stained with 4',6-diamidino-2-phenylindole (DAPI; final concentration, 0.2 ng/ml), and membranes were stained with FM4-64 (final concentration, 1 nM).
The average maximal signal intensity (the mean of the measurement of high values of signal intensities as measured using Metamorph software) for GyrA-YFP foci was 115 units (after subtraction of background fluorescence, which was 85 units on average, with 50 cells analyzed), and 90 units for nucleoid background staining (i.e., staining of nucleoids lacking foci or of nucleoid areas not containing foci) of cells expressing GyrA-YFP, so the difference in maximal intensity was 22% between foci and general nucleoid staining. For Topo I-YFP-expressing cells, the maximum intensity of foci was 177 units, and that of nucleoid background staining was 149 units (with 52 cells analyzed), yielding a 17% difference.
Three-dimensional (3D) time lapse microscopy was performed by taking an image at the focal plane that was defined as home, followed by successive acquisitions at 0.2 mm above and below the home plane, as defined in the Metamorph software.
RESULTS
We wished to localize all essential B. subtilis topoisomerases in live cells. We created various C-terminal fusions of topoisomerases to GFP or to its fluorescent variants. All of the fusions were the only copies of the respective topoisomerase or enzyme subunit and were driven by the original promoter. Western blot analysis showed that all protein fusions were expressed as full-length proteins (data not shown). All engineered cells grew indistinguishably from wild-type cells and showed wild-type nucleoid morphology. In the case of the parC-gfp fusion, we found a mild condensation defect, as shown by decondensed nucleoids in 1 to 2% of the cells, but for the purposes of this study, the parC-gfp fusion was effectively fully functional. All labeled proteins localized to the nucleoids, but in considerably different manners.
Dynamic localization of topoisomerase II (DNA gyrase) in live cells. To visualize DNA gyrase, we tagged the GyrA subunit with GFP, which showed a clear nucleoid stain in 96% of the cells (Fig. 1A). Interestingly, a subfraction (34%) of gyrA-gfp-expressing cells showed fluorescent patches or discrete foci on the nucleoids (Fig. 1A). These foci did not simply coincide with a larger amount of DNA, because foci were apparent at sites that did not show a higher concentration of DNA as visualized with DAPI (Fig. 1A, compare first and second panels). Thus, fluorescent patches and foci most likely represented an accumulation of gyrase independent of the actual amount of DNA at the site. To understand why the foci were present in only a subpopulation of the cells, and to gain further insight into the dynamics of GyrA-GFP localization, we performed 3D time lapse microscopy experiments. At each time interval, three images were taken, one in the central (focal) plane of the cell, called "home"; one 0.2 μm above the home plane (called "top"); and one 0.2 μm below the home plane (called "bottom"). The rationale of this technique was to investigate if GyrA-YFP foci (YFP was used because of its lower bleaching properties) are static, and indeed only present in a subset of the cells, or are dynamic and move to a different focal plane (so that they are not visible in many cells) or disassemble between time intervals, i.e., cannot be found in different focal planes in different time intervals. When the spacing between the home and top or bottom plane was set to 0.3 μm, fluorescence signals were completely out of focus, while with a spacing of 0.2 μm, relatively clear top and bottom signals could still be obtained from a variety of fluorescent-protein fusions, showing that 0.2 μm above or below the home plane can capture out-of-focus signals away from the central cell axis. Figure 2A and B shows images taken in 2-min time intervals, with top, home, and bottom sections shown below each other. It is apparent from Fig. 2A that the localization of GyrA-YFP is different in most of the cells between each pair of time intervals. Moreover, a GyrA-YFP focus in one cell can be seen to assemble in the cell center at minute 0 and to disassemble at minute 2 in all three planes, to reassemble in the cell center at minute 4, while at minute 6, GyrA-YFP is again more uniformly distributed throughout the nucleoid. Similar assembly and disassembly kinetics were observed in 15 cells, while in only three cases did GyrA-YFP assemblies remain visible for more than four time intervals (i.e., 8 min). In most movies, only assembly or disassembly of foci could be captured, due to the rapid bleaching of the signal (because of the threefold exposure at different Z planes) and the need for perfect focusing throughout the time lapse. However, in 60% of the cells, GyrA-YFP changed from a dispersed localization to an accumulated form, or conversely, the accumulations dispersed within few minutes. Unlike GyrA-YFP, the DnaX-CFP subunit of DNA polymerase did not show comparable dynamics: the signal could be well traced over 8 min without a single disassembly/assembly occurrence seen in 55 cells observed (Fig. 2B). A clear disassembly and reassembly reaction was seen in only four cells. Some focal drift is apparent, as in Fig. 2B: in the 4-min panel, one focus has moved to the bottom plane, while in the 6-min panel, one focus is more clearly seen in the top panel, but the foci did not disassemble between time intervals. Thus, these experiments show that GyrA-YFP assembles and disassembles in a dynamic manner in most of the cells analyzed, which explains in part why only a third of the cells show GyrA-GFP foci in a given population. In addition, Fig. 2A shows that GyrA-GFP assemblies can be observed only in the home plane, but rarely in the top or bottom plane (and thus in a very narrow focal window), while DnaX-CFP foci can often be captured in all three planes (Fig. 2B). Thus, because GyrA-GFP foci assemble and disassemble in a majority of cells observed, it is clear that during the 120-min cell cycle under the experimental conditions used, most cells have gyrase centers that are visible only in a subset at any given time.
Interestingly, of those cells containing visible GyrA-GFP foci, most contained a single focus close to the middle of the cell (Fig. 1A). To obtain a statistical representation of the localization of GyrA-GFP foci, we measured the distances from foci to the nearest cell pole relative to the size of the cell in a large number of cells that were representative of a growing population of cells, as explained above. In Fig. 3A, it is apparent that only large cells (<3 μm) contained two bipolar GyrA-GFP foci, while small cells predominantly possessed a single central focus. Therefore, bipolar gyrase centers arise relatively late during the cell cycle compared with the early separation of SMC foci (22) or of Topo I-GFP foci (see below). We were able to capture three movies of cells (out of 125 cells monitored) in which a single central GyrA-GFP focus split into two bipolar foci within a few minutes. In Fig. 2C, a central GyrA-GFP focus resolves into two foci located within each cell half between minutes 1 and 3. At minute 2, two foci appear to be present, one still close to the cell center, while the other is positioned within one cell half. Due to the lack of a larger number of movies with foci that are clearly visible through an extended period of time (since they are so dynamic), we could not determine whether the central foci split into two foci, one of which moved earlier than the other (similar to the SMC complex [31]) or whether they moved simultaneously (similar to the movement of replication origins [32]). Also, we could not conclude that the rapid movement is a general property of GyrA-YFP foci. However, it is clear that within 2 to 4 min, gyrase foci can change from a central to a bipolar pattern of localization.
Dynamic localization of topoisomerase I in live cells. A Topo I-GFP fusion localized to the nucleoids in all 460 of the cells monitored. In 66% of these cells, Topo I-GFP localized rather uniformly throughout the nucleoids, but in 34% of the cells, distinct concentrations, or foci, of Topo I-GFP were apparent (Fig. 1B). Similar to gyrase, Topo I-GFP foci were seen at sites that did not show a higher degree of DNA staining (Fig. 1B). To investigate why only a fraction of the cells contained visible foci, we performed time lapse microscopy. From several experiments, it is clear that like GyrA-GFP, Topo I-GFP foci were highly dynamic; they were absent or present at 1-min intervals (Fig. 2D). In the left-hand cell in Fig. 2D, a focus can be seen to disappear at minute 1, to reappear at minute 2, and to disappear again at minute 5. In the right-hand cell, the indicated focus disappears at minute 3 and reappears 1 minute later. Occasionally, foci were seen to split into two foci that united again a few minutes later (Fig. 2D). Similar time lapse experiments with the SMC complex (e.g., with ScpA-YFP) that also forms discrete bipolar foci showed that ScpA-YFP signals were visible at each time interval through 15 to 20 exposures at 1-min intervals (Fig. 2E) (22), showing that the fluctuations in Topo I localization were not caused by simple focal drift during capturing of the images. Because Topo I-GFP showed even greater bleaching than GyrA-YFP, we were not able to obtain extended 3D time lapse movies. However, similar to gyrase, Topo I foci were found to assemble or disassemble within a few minutes (data not shown), so that most likely over time Topo I foci are present in many more cells than just the observed subset.
Of the cells containing Topo I foci, 26% had a single focus, while 66% had two foci and 8% showed multiple foci. To investigate the positioning of the Topo I foci, we measured the distances of the foci to the nearest cell poles, which were plotted relative to cell size in Fig. 3B. It is apparent that in small cells, a single focus is generally present close to the middle of the cells, while in middle-size (>2 μm) and large (>3 μm) cells, one focus was present mostly close to one cell pole, or more frequently, two foci were located in a bipolar manner (that is, one close to each cell pole). These experiments suggest that Topo I centers are formed in a cell cycle-dependent manner, but in a manner different from that of gyrase. Middle-size cells contained two (bipolar) Topo I centers, but usually a central GyrA focus (Fig. 2A and B), showing that separation of Topo I-GFP foci occurred earlier in the cell cycle than with GyrA-GFP foci. Thus, Topo I and gyrase form discrete centers on the nucleoids; however, they are distinct from each other. The distinct localization patterns were further supported by dual-labeling experiments, which are described below.
DNA gyrase centers generally coincide with DNA polymerase, while Topo I centers and SMC protein localize to similar regions on the nucleoids. The formation of patches, or foci, whose positions were cell cycle dependent for gyrase and Topo I was highly similar to patterns of localization described for DNA polymerase, which is a stationary complex at the cell center, or the SMC complex, which is usually located in a bipolar fashion, respectively. While two bipolar DNA polymerase centers form within each daughter cell late in the cell cycle, the SMC complex moves from a central position toward opposite cell poles early in the cell cycle, so that most frequently, two bipolar SMC centers are present in growing cells (21, 22). The similarity of the patterns and localizations of gyrase and Topo I prompted us to generate dually labeled strains for simultaneous visualization of topoisomerases relative to DnaX, the tau subunit of the replisome, or to SMC, ScpA, or ScpB, which form the SMC complex. From Fig. 1C, it is apparent that gyrase centers were often coincident with the replication machinery or localized in close proximity. Of 380 cells investigated, 72 GyrA-YFP and DnaX-CFP foci were coincident (that is, in 57% of the cells showing GyrA-YFP foci or patches) and 46 were in close proximity (i.e., within 0.2 μm in 37% of the cells showing GyrA-YFP foci), while 8 foci (6%) were clearly separate. In the remaining cells, no clear GyrA-YFP foci were apparent. Thus, gyrase centers are most frequently located close to the central replisome. In agreement with this, DnaX-CFP shows a cell 3cycle-dependent localization similar to that of gyrase: small and medium-size cells generally contain a single central DnaX-CFP focus, while large cells (between 2.5 and 3 μm; compare Fig. 3A) contain two bipolar foci (3, 17).
The cell cycle-dependent localization of gyrase centers implies that at higher growth rates, cells should contain more foci than the slow-growing cells used for the studies described above, because at higher growth rates, the numbers of replication forks and of replication centers increase (17). To test this idea, we grew GyrA-YFP-expressing cells at 25°C in minimal or rich medium, where single or multiple rounds of replication take place, respectively. Indeed, similar to DnaX-CFP (17), the number of cells containing two GyrA-GFP foci increased at the expense of cells having one GyrA-GFP focus at the higher growth rate. In minimal medium, of all cells showing GyrA-YFP foci, 48.2% contained one focus, 44% two foci, and 4.7 or 3.1% three or four foci (Fig. 4A; 350 cells analyzed), whereas in rich medium, 31.3% contained one focus, 52.7% two foci, 8.9% three foci, and 7.1% four foci (Fig. 4B; 340 cells analyzed). These experiments support a close cell cycle-dependent association of gyrase foci or patches with the replication machinery.
Simultaneous visualization of Topo I-CFP and SMC-YFP revealed that in 30% of the cells containing Topo I-CFP foci or patches, both signals colocalized, while in 60% of the cells, both signals were in close proximity, or clearly apart in 10% of the cells (with 332 cells analyzed) (Fig. 1D). As before, 68% of the cells did not show clear Topo I foci or patches. Simultaneous labeling of Topo I and ScpA or ScpB yielded similar results (data not shown). Therefore, Topo I centers are frequently close to, or to a lesser extent, coincident with the SMC condensation centers. Of note, in about 20% of cells expressing both Topo I-CFP and SMC-YFP, SMC foci showed irregular positioning or were difficult to detect (data not shown) compared to about 5% of cells expressing SMC-YFP only, in which no foci were detectable. Additionally, cells with the dual label occasionally showed aberrant nucleoids (about 5%) (data not shown), indicating that the GFP tags, when present on both Topo I and SMC, slightly interfere with the function of SMC protein, showing a possible direct interaction.
To verify the specific localization patterns of Topo I and gyrase, both fluorescently labeled fusions were combined with other GFP fusions that were expected not to colocalize with the respective topoisomerase. Indeed, GyrA-CFP foci were generally separate from SMC-YFP foci (Fig. 1E; 85% separate versus 15% close/coincident in 90 cells containing well-defined foci), and similar to SMC-YFP (22, 31), Topo I-YFP foci were generally separate from DnaX-CFP foci (Fig. 1F; 82% separate versus 18% close/coincident in 115 cells containing well-defined foci). These experiments support the similar localization patterns of gyrase and the DNA polymerase, and of Topo I and the SMC complex.
Topoisomerase I and gyrase are also present in anucleate cells. The SMC chromosome condensation complex is strongly associated with chromosomal DNA, as it is present only in nucleated and not in anucleate cells (22). We wished to investigate if Topo I and gyrase have similar properties. Therefore, the smc gene was deleted in Topo I-GFP- or GyrA-GFP-expressing cells, and the localization was monitored at 25°C in minimal medium, which is permissive for growth of smc mutant cells. In both cases, the pattern of localization of the topoisomerases was disturbed. GyrA-GFP was present in all 131 of the anucleate cells monitored as staining of the entire cell, either smooth staining (Fig. 5A) or with brighter patches at random places in the cell. Likewise, 23 anucleate cells contained weak but detectable Topo I-GFP fluorescence, while 11 anucleate cells had high levels similar to nucleated cells, and 4 cells had only background staining (Fig. 5B). Likewise, GyrA-GFP or Topo I-GFP was present in anucleate cells of a spo0J mutant strain (data not shown). We conclude that Topo I and gyrase are also present in anucleate cells and that therefore, not all topoisomerase molecules are bound to DNA, in marked contrast to the SMC complex. In a few anucleate cells (10%), GyrA-GFP and Topo I-GFP formed foci, which were different from those in normal cells because they formed at random positions within the cell, indicating that in the absence of DNA, topoisomerase molecules may form nonspecific aggregates.
Localization of topoisomerase IV in live cells. Unlike Topo I and gyrase, both subunits of Topo IV, ParC and ParE, showed rather uniform staining of the nucleoids, with no apparent concentration at any particular site on the nucleoids (Fig. 5C and D). To find out if localization on the nucleoid depends on the presence of both subunits, we depleted ParC in cells expressing ParE-GFP. Removal of xylose (which drives transcription of parC in the parE-gfp strain) led to strong nucleoid decondensation, cell elongation, and formation of anucleate cells until the cells lysed (Fig. 5E). Under these conditions, ParE-GFP localized mostly throughout the cells (in 86% of 180 cells analyzed) (Fig. 5E) and was also present in anucleate cells, although 14% of the cells retained ParE staining on the nucleoids. Thus, Topo IV is present throughout the nucleoid, depending on the presence of both subunits, which suggests that it interacts with many if not all regions of the chromosome. This pattern of localization is in marked contrast to a previous investigation, in which ParC was found to be present close to the cell poles away from the nucleoids (13). Although we have no easy explanation for this discrepancy, we note that our parC-gfp fusion is almost fully active, while the fusion used in the study by Huang et al. was cold sensitive below 30°C and was visible in only 6% of all cells (13).
DISCUSSION
This work describes the subcellular localization of B. subtilis topoisomerases and their spatial connection with the SMC complex and with the DNA replication machinery, providing several important conclusions. First, all three essential topoisomerases have distinct patterns of localization, each of which suggests a spatial specialization for each class of enzymes. Gyrase was present on the nucleoids and showed discrete accumulations in a large fraction of growing cells. Strikingly, these gyrase accumulations assembled and disassembled within few minutes, showing that gyrase localization is not static but highly dynamic. Assembly and disassembly dynamics were rather irregular and distinct from oscillation movements described for Min proteins or ParA proteins, the latter of which appear to oscillate from one end of the nucleoid to the other and back (5, 24). We speculate that the assemblies are due to an accumulation of supercoils, leading to the recruitment of gyrase to its substrate on the chromosome. The dynamics of gyrase localization explain why the accumulations are detectable in only a subset of the cells. Until late in the cell cycle, usually a single GyrA-GFP fluorescent focus was detectable close to the middle of the cells, and during late steps in the cell cycle, two foci were seen, one within each cell half. The accumulation of gyrase did not simply reflect an increased concentration of chromosomal DNA at this position, reinforcing the idea that the foci represent gyrase centers that form due to physical properties independent of the amount of DNA. Dual labeling of the replication machinery and gyrase verified that most frequently, gyrase centers coincide with the centrally located DNA polymerase. This is entirely consistent with biochemical and electron microscopical data showing a tight link between gyrase and the replication machinery (1, 2). During most of the cell cycle, the replication machinery is a stationary complex at midcell, and only late in the cell cycle do two new centers form, one within each cell pole (18). Likewise, two distinct gyrase foci were detectable during late stages of the cell cycle, suggesting that gyrase centers are associated with the replication machinery throughout the cell cycle. As the chromosome is moved through the replisome, the DNA duplex is unwound, which generates positive supercoils ahead of the replication fork (7). The localization of gyrase suggests that a large fraction of the active molecules are dealing with the removal of the positive supercoils. It is even possible (but yet to be investigated) that there is very little need to remove precatenanes behind the replication fork if gyrase is efficient enough to largely abrogate positive supercoils in front of the fork. It is also possible that gyrase centers reintroduce negative supercoils behind the forks, but this issue cannot yet be resolved with light microscopy.
Like gyrase, Topo I formed discrete centers on the nucleoids, which appeared to also assemble and disassemble rapidly and continuously. However, Topo I accumulations were mostly present within one or both cell halves, away from the cell middle, and were often closely associated with SMC condensation centers. Bacterial SMC protein has been shown to influence DNA supercoiling in E. coli (here, MukB is the SMC analog) and in B. subtilis, and a genetic interaction with DNA Topo I and gyrase has been established (20, 26). Loss of SMC function can be compensated for by reduction of Topo I activity (which relaxes negative supercoiling) and is exacerbated by reduction of gyrase activity, showing that SMC introduces overall negative supercoiling into DNA. Our data show that SMC is also spatially associated with high Topo I activity, in that Topo I and the SMC complex localize to similar areas on the nucleoids. SMC protein binds to DNA as a ringlike structure (9, 11, 31), in a highly cooperative and repetitive manner (29). This is thought to generate positive writhe within the DNA, which is translated into negative supercoils that are a substrate for Topo I (16). Our findings imply that the SMC condensation centers generate a high degree of negative supercoiling, which is at least partially removed by Topo I in the vicinity of the SMC centers.
Unlike Topo I and gyrase, and E. coli cells, both subunits of Topo IV localized throughout the nucleoids, without any apparent concentration at distinct sites. The pattern of localization suggests not only a role for Topo IV in the decatenation of chromosomes at the end of the cell cycle, as seems to be the case in E. coli (6), but also a general role in chromosome supercoiling. This is supported by our recent finding that the depletion of Topo IV affects global chromosome compaction and protein synthesis and that overproduction of Topo IV can rescue the condensation defect and the defect in global protein synthesis caused by the loss of SMC activity (30). Conversely, our results suggest that the amount of Topo IV that is uniformly present throughout the nucleoids is sufficient to decatenate the termini without the need for an increase in concentration at the termini.
In toto, our results suggest that regions with different degrees of supercoiling exist on the nucleoids in B. subtilis cells, which fluctuate within a 1-min time frame and are thus highly dynamic. These regions correspond to the region of active replication at midcell, where positive supercoils need to be removed, and to the regions close to the bipolarly located SMC complex, in which newly replicated DNA is thought to be condensed and organized for efficient chromosome segregation, where positive writhe might be relaxed by Topo I. So, although Topo I, II, and IV are present throughout the nucleoids, higher levels of Topo I and gyrase appear to be required at specific places on the nucleoids. Topo IV interacts with many, if not all, sites on the chromosomes throughout the cell cycle and thus appears to perform a more general role in supercoiling than Topo IV in E. coli. The SMC complex is involved in general supercoiling, but also in active separation of sister chromosomes, most likely through its specific localization within each cell half. A close spatial connection with Topo I and genetic interactions with Topo I and gyrase underline the intricate interplay of the SMC complex with topoisomerases, whose molecular bases will be interesting for future studies.
ACKNOWLEDGMENTS
We thank Judita Mascarenhas for help and advice in this project.
The work was supported by the Deutsche Forschungsgemeinschaft (Emmy Noether and Heisenberg Programm).
REFERENCES
Baker, T. A., B. E. Funnell, and A. Kornberg. 1987. Helicase action of DnaB protein during replication from the Escherichia coli chromosomal origin in vitro. J. Biol. Chem. 262:6877-6885.
Baker, T. A., K. Sekimizu, B. E. Funnell, and A. Kornberg. 1986. Extensive unwinding of the plasmid template during staged enzymatic initiation of DNA replication from the origin of the Escherichia coli chromosome. Cell 45:53-64.
Defeu Soufo, H. J., and P. L. Graumann. 2005. Bacillus subtilis actin-like protein MreB influences the positioning of the replication machinery and requires membrane proteins MreC/D and other actin-like proteins for proper localization. BMC Cell Biol. 6:10.
Dekker, N. H., V. V. Rybenkov, M. Duguet, N. J. Crisona, N. R. Cozzarelli, D. Bensimon, and V. Croquette. 2002. The mechanism of type IA topoisomerases. Proc. Natl. Acad. Sci. USA 99:12126-12131.
Ebersbach, G., and K. Gerdes. 2004. Bacterial mitosis: partitioning protein ParA oscillates in spiral-shaped structures and positions plasmids at mid-cell. Mol. Microbiol. 52:385-398.
Espeli, O., C. Levine, H. Hassing, and K. J. Marians. 2003. Temporal regulation of topoisomerase IV activity in E. coli. Mol. Cell 11:189-201.
Espeli, O., and K. J. Marians. 2004. Untangling intracellular DNA topology. Mol. Microbiol. 52:925-931.
Graumann, P. L., R. Losick, and A. V. Strunnikov. 1998. Subcellular localization of Bacillus subtilis SMC, a protein involved in chromosome condensation and segregation. J. Bacteriol. 180:5749-5755.
Gruber, S., C. H. Haering, and K. Nasmyth. 2003. Chromosomal cohesin forms a ring. Cell 112:765-777.
Harwood, C. R., and S. M. Cutting. 1990. Molecular biological methods for Bacillus. Wiley, New York, N.Y.
Hirano, M., and T. Hirano. 2002. Hinge-mediated dimerization of SMC protein is essential for its dynamic interaction with DNA. EMBO J. 21:5733-5744.
Hirano, T. 2002. The ABCs of SMC proteins: two-armed ATPases for chromosome condensation, cohesion, and repair. Genes Dev. 16:399-414.
Huang, W. M., J. L. Libbey, P. van der Hoeven, and S. X. Yu. 1998. Bipolar localization of Bacillus subtilis topoisomerase IV, an enzyme required for chromosome segregation. Proc. Natl. Acad. Sci. USA 95:4652-4657.
Jaacks, K. J., J. Healy, R. Losick, and A. D. Grossman. 1989. Identification and characterization of genes controlled by the sporulation regulatory gene spo0H in Bacillus subtilis. J. Bacteriol. 171:4121-4129.
Kato, J., Y. Nishimura, R. Imamura, H. Niki, S. Hiraga, and H. Suzuki. 1990. New topoisomerase essential for chromosome segregation in E. coli. Cell 63:393-404
Kimura, K., V. V. Rybenkov, N. J. Crisona, T. Hirano, and N. R. Cozzarelli. 1999. 13S condensin actively reconfigures DNA by introducing global positive writhe: implications for chromosome condensation. Cell 98:239-248.
Lemon, K. P., and A. D. Grossman. 1998. Localization of bacterial DNA polymerase: evidence for a factory model of replication. Science 282:1516-1519.
Lemon, K. P., and A. D. Grossman. 2000. Movement of replicating DNA through a stationary replisome. Mol. Cell 6:1321-1330.
Lewis, P. J., and A. L. Marston. 1999. GFP vectors for controlled expression and dual labelling of protein fusions in Bacillus subtilis. Gene 227:101-110.
Lindow, J. C., R. A. Britton, and A. D. Grossman. 2002. Structural maintenance of chromosomes protein of Bacillus subtilis affects supercoiling in vivo. J. Bacteriol. 184:5317-5322.
Lindow, J. C., M. Kuwano, S. Moriya, and A. D. Grossman. 2002. Subcellular localization of the Bacillus subtilis structural maintenance of chromosomes (SMC) protein. Mol. Microbiol. 46:997-1009.
Mascarenhas, J., J. Soppa, A. Strunnikov, and P. L. Graumann. 2002. Cell cycle dependent localization of two novel prokaryotic chromosome segregation and condensation proteins in Bacillus subtilis that interact with SMC protein. EMBO J. 21:3108-3118.
Nagai, T., K. Ibata, E. S. Park, M. Kubota, K. Mikoshiba, and A. Miyawaki. 2002. A variant of yellow fluorescent protein with fast and efficient maturation for cell-biological applications. Nat. Biotechnol. 20:87-90.
Raskin, D. M., and P. A. de Boer. 1999. Rapid pole-to-pole oscillation of a protein required for directing division to the middle of Escherichia coli. Proc. Natl. Acad. Sci. USA 96:4971-4976.
Rizzo, M. A., G. H. Springer, B. Granada, and D. W. Piston. 2004. An improved cyan fluorescent protein variant useful for FRET. Nat. Biotechnol. 22:445-449.
Sawitzke, J. A., and S. Austin. 2000. Suppression of chromosome segregation defects of Escherichia coli muk mutants by mutations in topoisomerase I. Proc. Natl. Acad. Sci. USA 97:1671-1676.
Soppa, J., K. Kobayashi, M. F. Noirot-Gros, D. Oesterhelt, S. D. Ehrlich, E. Dervyn, N. Ogasawara, and S. Moriya. 2002. Discovery of two novel families of proteins that are proposed to interact with prokaryotic SMC proteins, and characterization of the Bacillus subtilis family members ScpA and ScpB. Mol. Microbiol. 45:59-71.
Stone, M. D., Z. Bryant, N. J. Crisona, S. B. Smith, A. Vologodskii, C. Bustamante, and N. R. Cozzarelli. 2003. Chirality sensing by Escherichia coli topoisomerase IV and the mechanism of type II topoisomerases. Proc. Natl. Acad. Sci. USA 100:8654-8659.
Strick, T. R., T. Kawaguchi, and T. Hirano. 2004. Real-time detection of single-molecule DNA compaction by condensin I. Curr. Biol. 14:874-880.
Tadesse, S., J. Mascarenhas, B. Ksters, A. Hasilik, and P. L. Graumann. 2005. Genetic interaction of the SMC complex with topoisomerase IV in Bacillus subtilis. Microbiology 151:3729-3737.
Volkov, A., J. Mascarenhas, C. Andrei-Selmer, H. D. Ulrich, and P. L. Graumann. 2003. A prokaryotic condensin/cohesin-like complex can actively compact chromosomes from a single position on the nucleoid and binds to DNA as a ring-like structure. Mol. Cell. Biol. 23:5638-5650.
Webb, C. D., P. L. Graumann, J. Kahana, A. A. Teleman, P. Silver, and R. Losick. 1998. Use of time-lapse microscopy to visualize rapid movement of the replication origin region of the chromosome during the cell cycle in Bacillus subtilis. Mol. Microbiol. 28:883-892.
Zhu, Q., P. Pongpech, and R. J. DiGate. 2001. Type I topoisomerase activity is required for proper chromosomal segregation in Escherichia coli. Proc. Natl. Acad. Sci. USA 98:9766-9771.(Serkalem Tadesse and Pete)