The Conserved Sporulation Protein YneE Inhibits DNA Replication in Bacillus subtilis

Abstract

Cells of Bacillus subtilis triggered to sporulate under conditions of rapid growth undergo a marked decrease in chromosome copy number, which was partially relieved by a mutation in the sporulation-induced gene yneE. Cells engineered to express yneE during growth were impaired in viability and produced anucleate cells. We conclude that YneE is an inhibitor of DNA replication.

The ability of bacteria to adapt to a variety of growth conditions demands mechanisms for controlling DNA replication and coordinating chromosome duplication with the cell cycle. Bacteria initiate one and only one round of DNA replication at the origin of chromosomal DNA replication, oriC, for each cycle of cell division (1). During slow growth, the single copy of oriC is duplicated to create two copies of oriC. In contrast, under conditions of rapid growth, when the time required to duplicate a chromosome is longer than the generation time, bacteria inherit chromosomes undergoing active replication with multiple replication forks. In this case, replication initiation occurs synchronously across all origins, leading to multifork replication with 2ⁿ (n = 1,2,3) copies of oriC present in each cell (15). Multifork replication ensures that each newly divided cell will receive at least one chromosome. The transition from one growth condition to another, such as when nutrients become limiting and cells enter stationary phase, requires mechanisms to adjust the frequency of replication initiation (10). One noteworthy but poorly understood example of replication control is exhibited by Bacillus subtilis during the entry into sporulation. Sporulating cells complete a final round of replication before undergoing a process of asymmetric division that results in a forespore and a mother cell (11, 14). Each of these progeny cells ordinarily inherits a single chromosome, and mature spores, which are derived from the forespore, are known to contain only one chromosome (9, 14).

Cells normally enter the pathway to sporulate in response to nutrient limitation. We have previously shown, however, that artificial induction of synthesis of the kinase KinA causes efficient sporulation even during growth in rich medium (7). KinA transfers phosphate groups into a multicomponent phosphorelay that phosphorylates the response regulator Spo0A, the master regulator for entry into sporulation (2). The discovery that rapidly growing cells can be triggered to sporulate with high efficiency implies that B. subtilis must have an active mechanism to shut down DNA replication, given that rapidly growing cells contain multiple replication forks. An appealing hypothesis for how such a mechanism might work stems from the observation that the B. subtilis origin of replication contains multiple binding sites for phosphorylated Spo0A (Spo0A∼P) and that many of these sites overlap with binding sites for the replication initiation protein DnaA (5). This suggests that Spo0A∼P may impede replication by directly blocking the action of DnaA. It has not been established, however, whether Spo0A∼P binds to the origin region in vivo and, if so, whether this binding blocks the initiation of replication. We wondered, therefore, whether Spo0A∼P might additionally, or alternatively, block replication indirectly by turning on the expression of a gene or genes whose product is an inhibitor of replication.

In previous work, we identified genes under the direct control of Spo0A, a small number of which encode proteins of unknown function and mutants of which do not exhibit conspicuous defects in spore formation (7, 13). We used this as a starting point to investigate whether any previously uncharacterized genes under Spo0A∼P control inhibit DNA replication. Here we report the discovery that a Spo0A∼P-induced gene, yneE, for which we introduce the name sirA (for sporulation inhibitor of replication) encodes a protein that inhibits DNA replication.

To determine if the induction of sporulation by actively growing cells blocks replication, we turned on the synthesis of KinA in cells in the midexponential phase of growth by placing kinA under the control of an IPTG (isopropyl-β-d-thiogalactopyranoside)-inducible promoter and treating the cells with IPTG. We measured chromosome content by staining for DNA and analyzing the cells with a flow cytometer. Either samples were fixed immediately for a “snapshot” of the total amount of DNA present at the time that the sample was collected or the cells were allowed to continue growing in the presence of chloramphenicol. Because chloramphenicol blocks protein synthesis, it allows ongoing rounds of replication to reach completion but prevents new rounds of replication from initiating (15). These chloramphenicol-treated samples provide a “resolved” picture of how many chromosomes had initiated replication at the time the samples were collected.

As shown in Fig. 1, in the absence of IPTG, the cells contained multiple chromosomes. Even at the end of the time course, many cells contained four chromosome equivalents of DNA after resolution with chloramphenicol. Chromosome content decreased, however, when KinA synthesis was turned on with IPTG. Indeed, 4 hours after the induction of KinA, most cells contained two fully replicated chromosomes. To ensure that the effect of the addition of IPTG was dependent on Spo0A, we repeated the experiments in cells mutant for the master regulator. In the absence of Spo0A, the addition of IPTG had little or no effect on DNA content (data not shown).

FIG. 1.

SirA causes a reduction in chromosome content. Flow cytometry was done with a strain wild type for sirA (BG197; P_hyperspank-kinA ΔsinI) grown in the presence or absence of IPTG and a strain mutant for sirA (BG221; P_hyperspank-kinA ΔsinI ΔsirA) grown in the presence of IPTG. (A) Cells were grown in CH medium (13a) at 37°C to an optical density at 600 nm of 0.25, at which point 1 mM IPTG was either added (+IPTG) or not added (−IPTG). Samples were then collected at hourly intervals as indicated and fixed in 70% ethanol. (B) Cells were grown as described for panel A except that chloramphenicol (CM; 200 μg/ml) was added at hourly intervals as indicated and the samples were further incubated at 37°C for about 4 hours to allow the completion of ongoing rounds of replication before the cells were fixed in 70% ethanol. Samples were stained with 90 μg/μl mitomycin and 20 μg/μl ethidium bromide to visualize DNA and then run through an Apogee A10 flow cytometer equipped with a 100-W mercury lamp as described previously (11). A total of 20,000 to 40,000 cells were analyzed per sample, and data were normalized and analyzed in Microsoft Excel. Guidelines are drawn at two and four genome equivalents of DNA and were determined by analyzing the dnaB(Ts) strain BG164 grown at the permissive temperature and then shifted to the nonpermissive temperature to create one-chromosome cells. Strains were mutant for sinI to prevent cell chaining. Strains are described in Table S1 in the supplemental material.

Next, we investigated the hypothesis that Spo0A was turning on a gene(s) responsible for the observed inhibition of replication. Two attractive candidates were sirA (yneE) and yttP, which are directly activated by Spo0A and which have not been previously characterized (7, 13). When subjected to flow cytometry analysis as described above, cells with a deletion of yttP behaved similarly to wild-type cells (data not shown), but cells with a deletion of sirA were impaired in their ability to reduce their DNA content upon KinA induction (Fig. 1).

As an independent assay of the effect of sirA on replication, we used a cassette of tandem operators for the tetracycline resistance gene that had been inserted into the chromosome near the origin of replication. Green florescent protein (GFP) fused to the tetracycline resistance repressor was used to visualize the origin region (6, 12). In these cells, the number of GFP foci in each cell represents the number of origins that have duplicated and moved far enough apart to be distinguishable by microscopy. An example of the results with mutant and wild-type cells is shown in Fig. 2; an arrow indicates a sirA mutant cell with four fluorescent foci. Figure 3 summarizes the results quantitively, with the percentages of cells that contained one, two, three, or four resolved origins shown at each hour after the induction of KinA. In agreement with the flow cytometry data, fewer sirA⁺ cells had three or four origins at later times than they did at time zero. The sirA mutant, on the other hand, while able to reduce its origin count somewhat, was not as efficient as the wild type in doing so (Fig. 3A). That the sirA mutant was able to reduce its chromosome number to some extent may indicate that additional sporulation-specific factors, such as Spo0A∼P binding at the origin of replication (5), contribute to blocking replication after kinA induction.

FIG. 2.

Visualization of the effect of SirA on the number of fluorescently tagged origin regions. Fluorescent microscopy was performed with wild-type cells (left; LR40) and ΔsirA mutant cells (right; LR74) 3 hours after IPTG addition. The cells contained a cassette of tandem tet operators near the origin of replication and constitutively synthesized TetR-GFP. The arrow indicates an example of a cell with four origin foci. Cells were grown exponentially in CH medium (13a) at 37°C to an optical density at 600 nm of 0.25, at which point 1 mM IPTG was added. Membranes were stained with 1 μg/ml FM4-64, and micrographs were obtained essentially as previously described (4), except that cells were immobilized with a poly-l-lysine-treated coverslip and a filter for GFP was included. Images were processed with Simple PCI and Adobe Photoshop.

FIG. 3.

SirA reduces the number of fluorescent foci associated with the origin region of the chromosome. (A) The percentages of cells containing one, two, three, or four resolved fluorescent foci at 0, 1, 2, 3, or 4 h after the addition of IPTG are shown. Open bars, wild type (LR40); filled bars, ΔsirA mutant (LR74). Data are the average of results of five experiments; error bars represent the standard deviations. (B) For each class of cells (one, two, three, or four foci), the percentage for wild-type cells was subtracted from the percentage for ΔsirA mutant cells (top), ΔyneF mutant cells (middle; LR76), or ΔsirA mutant cells carrying a complementing copy of wild-type sirA (bottom; LR86) over time. Symbols: squares, one origin; circles, two origins; triangles, three origins; diamonds, four origins. Data are the average of results of several experiments (for ΔsirA cells, n = 5; ΔyneF cells, n = 3; SirA-complemented cells, n = 2).

As an alternative way to present the results, Fig. 3B shows the difference between the percentages of sirA mutant cells and wild-type cells with each number (one, two, three, or four) of fluorescent foci. In comparison, deletion of the downstream gene yneF had no effect on the number of resolved origins (Fig. 3B). Although yneF is annotated as being in an operon with sirA, neither our previous microarray experiments (7, 13) nor reporter fusion experiments (data not shown) indicate that sirA and yneF are cotranscribed or that yneF is under Spo0A∼P control. To test further the idea that SirA was indeed responsible for the chromosome-reducing phenotype, we complemented our sirA deletion by inserting a wild-type copy of the gene at the distal locus thrC. When the sirA deletion was complemented, there was no significant difference from the wild type in origin number after IPTG addition (Fig. 3B).

Next, we asked whether sirA was capable of inhibiting replication in growing cells, independent of other Spo0A-induced genes. To address this question, we engineered cells to express sirA during growth using an IPTG-inducible promoter. When plated on solid minimal medium in the presence of IPTG, the engineered cells grew extremely slowly if at all, occasionally producing tiny colonies after prolonged incubation (data not shown). When plated on solid rich (LB) medium in the presence of inducer, the engineered cells produced colonies but the colonies were noticeably smaller than for cells plated on medium lacking IPTG (data not shown). The engineered cells also exhibited a modest decrease in growth rate in liquid LB medium in the presence of inducer. These growth inhibitory effects were also observed in the absence of spo0A (data not shown), reinforcing the view that SirA-mediated replication inhibition does not depend on Spo0A∼P. When the cells grown in liquid were stained with DAPI (4′,6-diamidino-2-phenylindole) and examined by fluorescence microscopy, we observed many anucleate cells (∼14%) (Fig. 4). These anucleate cells were often adjacent to each other in chains, suggesting that the cells were capable of undergoing an additional round of cell division in the absence of a chromosome.

FIG. 4.

SirA causes the production of anucleate cells when artificially synthesized during growth. Microscopy was done with a strain engineered to overexpress sirA (LR150) grown in the absence (left) or presence (right) of IPTG. Arrows indicate examples of anucleate cells. Cells were grown in LB medium at 37°C for 4 hours and then stained with 2 μg/ml DAPI to visualize DNA. Micrographs were obtained essentially as previously described (4), except that cells were immobilized with a poly-l-lysine-treated coverslip and a filter for DAPI was included. Images were processed with Simple PCI and Adobe Photoshop.

Finally, we note that SirA is conserved among endospore-forming members of the genus Bacillus (Fig. 5). On the other hand, it is absent in Listeria, a non-spore-forming genus whose members are closely related to B. subtilis. These findings reinforce the view that SirA plays a significant role in sporulation, at least under conditions in which rapidly growing cells enter the pathway to form spores.

FIG. 5.

SirA is conserved among Bacillus species. Sequences were aligned with ClustalW.

Interestingly, when replication is impaired, the replication initiation protein DnaA turns on the synthesis of Sda, a protein that blocks sporulation by inhibiting the action of KinA and hence the phosphorylation of Spo0A (3). Here we have demonstrated, conversely, that when sporulation is induced in cells that have a high chromosome copy number, a protein (SirA), whose synthesis is turned on by Spo0A∼P, acts to inhibit replication and hence reduce chromosome copy number. Just how SirA inhibits replication is not yet known, but SirA likely acts at the initiation stage of replication, with DnaA being an appealing candidate for the target of its action. That SirA is an inhibitor of DNA replication was also independently discovered by J. Wagner and D. Rudner (personal communication).

In closing, we note that SirA is the second example of a protein produced during sporulation that acts to inhibit the cell cycle, the other being MciZ, which inhibits the cytokinetic protein FtsZ (8).