Failsafe mechanisms couple division and DNA replication in bacteria

Abstract

The past 20 years have seen tremendous advances in our understanding of the mechanisms underlying bacterial cytokinesis, particularly the composition of the division machinery and the factors controlling its assembly [1]. At the same time, we understand very little about the relationship between cell division and other cell-cycle events in bacteria. Here we report that inhibiting division in Bacillus subtilis and Staphylococcus aureus quickly leads to an arrest in the initiation of new rounds of DNA replication, followed by a complete arrest in cell growth. Arrested cells are metabolically active but are unable to initiate new rounds of either DNA replication or division when shifted to permissive conditions. Inhibiting DNA replication results in entry into a similar quiescent state in which cells are unable to resume growth or division when returned to permissive conditions. Our data suggest the presence of two failsafe mechanisms: one linking division to the initiation of DNA replication and another linking the initiation of DNA replication to division. These findings contradict the prevailing view of the bacterial cell cycle as a series of coordinated but uncoupled events. Importantly, the terminal nature of the cell-cycle arrest validates the bacterial cell-cycle machinery as an effective target for antimicrobial development.

Figure 1. An extended block in cell division leads to a growth arrest and entry into a permanent, quiescent state at the Point of No Return (PONR)

(A-D) Cell growth as measured by OD600 of P_xyl-ftsZ (A), ftsZts (B), divICts (C) and PC190723 treated S. aureus (D). (A-G) Purple lines/double arrow (P_xyl-ftsZ, ftsZts) and blue lines/closed arrow (divICts) delineate the PONR boundary. Growth under permissive conditions (black squares), nonpermissive conditions (open circles), shifted to permissive conditions ~ 1-2 Mass Doubling Periods (MDP) before the PONR (3.9 MDP – P_xyl-ftsZ, 4.4 MDP – ftsZts, 3 MDP – divICts) (open diamonds), or shifted to permissive conditions ~1 generation after the PONR (6.5 MDP – P_xyl-ftsZ, 7 MDP – ftsZts, 5 MDP – divICts) (open triangles). (E-G) Cells are metabolically active but unrecoverable following the PONR. P_xyl-ftsZ (black bars), ftsZts (gray bars), divICts (white bars) and S. aureus (dashed bars). (E) Colony forming ability (plating efficiency) was determined after culturing cells at nonpermissive conditions prior to plating cells under permissive conditions. (F) Propidium iodide (PI) is a marker for loss of membrane integrity. PI negative (live) cells were quantified after cells were cultured for the indicated number of generations under nonpermissive conditions. (G) Percent MTT reduction serves as a direct readout of metabolic activity. Permissive levels were set to 100%. (A-G) Error bars = SD, n=3-5. (H) Cells are refractile to FtsZ assembly after the PONR. ftsZts cells were cultured under nonpermissive conditions for 1, 3, 5 and 7 MDP prior to a shift to permissive conditions for 1 MDP. Length-per-ring ratios were calculated by dividing total cell length by the number of FtsZ rings. ~200 wild-type cell lengths scored per replicate. n=3-4, SD in parentheses. Arrows indicate FtsZ rings. Bar = 5 μm. (I) Relative expression of genes in the SOS response or the Sigma W regulon. Transcript levels for 3 replicates are represented as the fold change for FtsZ depleted cells compared to FtsZ+ cells. See also Figure S1, S2 and Table S1.

Figure 2. Inhibiting cell division triggers an arrest in DNA replication

(A) WGA labeled cell wall and dapi-labeled DNA in wild-type and P_xyl-ftsZ cells. The average length per nucleoid ratio ±SD in μm are shown. n=3, ~200 cells scored per replicate. (B) Newly replicated DNA labeled with EdU (Methods). The fluorescence of induced P_xyl-ftsZ or untreated S. aureus was set to 100%. Average ±SD of fluorescence is shown. n=3, ~100 cells per replicate. (C) Marker frequency analysis of the origin to terminus (ori:ter) ratios of P_xyl-ftsZ after depleting FtsZ. Error bars = SD, n=3. (D) Quantitative immunoblots of DnaA in P_xyl-ftsZ and P_xyl-ftsZ, P_IPTG-dnaA cells depleted for FtsZ. The average ±SD is shown below a representative blot. Induced P_xyl-ftsZ (+xyl) was set to 100%, n=3. (A-B) Bar = 5 μm. See also Figure S3.

Figure 3. An extended block in DNA replication initiation results in a terminal cell cycle arrest

(A) Cell growth of dnaB134ts under permissive conditions (black squares), nonpermissive conditions (open circles), shifted to permissive conditions at 1 MDP (open diamonds), shifted to permissive conditions approximately at 3 MDP (X’s), or shifted to permissive conditions at 5 MDP (open triangles). (B) PI negative (live) cells and colony forming ability (plating efficiency) was determined after culturing cells under nonpermissive conditions for the indicated MDP (PI) and prior to plating cells at permissive conditions (CFU). (C) MTT reduction levels indicate metabolic activity of cells blocked for initiation. (A-C) Error bars = SD, n=3, dashed blue line indicates the PONR. (D) Cells were cultured under nonpermissive conditions for 1, 3, 5 and 7 MDP. Length-per-ring ratios were calculated as in Figure 1. n=3, SD in parentheses. Bar = 5 μm.

Fig 4. Failsafe mechanisms couple division and DNA replication in the bacterial cell cycle

DNA replication (blue) and growth (beige) are ongoing and the division ring is present for approximately the second half of each cell cycle. A division-dependent failsafe mechanism ensures one DNA replication initiation occurs per each division. Failure to divide leads to a block in replication initiation followed by entry into a terminal cell cycle arrest at the PONR, equivalent to a eukaryotic G₀ state. A second initiation-dependent failsafe mechanism ensures that cells impaired in the initiation of DNA replication do not assemble division rings. The dashed arrows indicate hypothetical localizations of the failsafe mechanisms.