Cell cycle coordination and regulation of bacterial chromosome segregation dynamics by polarly localized proteins

Abstract

What regulates chromosome segregation dynamics in bacteria is largely unknown. Here, we show in Caulobacter crescentus that the polarity factor TipN regulates the directional motion and overall translocation speed of the parS/ParB partition complex by interacting with ParA at the new pole. In the absence of TipN, ParA structures can regenerate behind the partition complex, leading to stalls and back-and-forth motions of parS/ParB, reminiscent of plasmid behaviour. This extrinsic regulation of the parS/ParB/ParA system directly affects not only division site selection, but also cell growth. Other mechanisms, including the pole-organizing protein PopZ, compensate for the defect in segregation regulation in ΔtipN cells. Accordingly, synthetic lethality of PopZ and TipN is caused by severe chromosome segregation and cell division defects. Our data suggest a mechanistic framework for adapting a self-organizing oscillator to create motion suitable for chromosome segregation.

Figure 1

FtsZ and MipZ dynamics in wild-type and ΔtipN cells. (A) Time-lapse microscopy of FtsZ–YFP in wild-type (MT199) and ΔtipN (CJW2563) cells after synchrony. The expression of ftsZ–yfp was induced with 0.5 mM vanillic acid 2.5 h before synchronization and imaging. Images were acquired every 1.5 min and the cells were identified using MicrobeTracker. The FtsZ–YFP signal in representative wild-type and ΔtipN cells is shown for selected time points as an overlay with the MicrobeTracker cell outline (the old pole is marked by the arrow). The graphs show the trace of the relative FtsZ–YFP position along the cell length over time. (B) Time-lapse microscopy of MipZ–mCFP and FtsZ–YFP in a wild-type background (strain CJW3455). FtsZ–YFP expression was induced with 0.5 mM vanillic acid 1 h before cell synchronization. Time-lapse results from a representative cell are shown as an overlay with MicrobeTracker cell outlines in white (the old pole is marked by the arrow.). The relative position of FtsZ–YFP (red trace) is indicated over a kymograph of the MipZ–mCFP signal profile (green) along the cell length as a function of time after synchrony. Yellow arrows show first instance of MipZ–mCFP and FtsZ–YFP colocalization at the new pole. (C) Same as (B) except in a ΔtipN background (strain CJW3612). Purple arrowheads show backwards motion of MipZ–mCFP. (D) Time-lapse recordings of MipZ–YFP in wild-type (CJW2022) and ΔtipN (CJW3366) cells after synchrony. Images were acquired every 1.5 min and the cells were identified using MicrobeTracker. Shown are profiles of the mean MipZ–YFP signal along the cell long axis of 94 wild-type cells (green) and 63 ΔtipN cells (blue) from the time point when MipZ–YFP becomes bipolar (yellow arrow in (B) and (C) to the onset of cell constriction). The arrows show the minima in MipZ–YFP intensity for each strain.

Figure 2

ParA dynamics in wild-type, ΔtipN and parA-overexpressing cells. (A) Kymograph of the average ParA–YFP signal intensity along the cell length as a function of time after synchrony. Wild-type cells (CJW3010) were imaged every 1.5 min by time-lapse microscopy and analysed using MicrobeTracker. (B) Same as (A) except that the average spatial distribution of ParA–YFP over time was obtained from imaging ΔtipN cells (CJW3011). (C) Time-lapse microscopy of ParA–YFP and CFP–ParB in a wild-type background (strain CJW3367). The results from a representative cell are shown as an overlay between the relative position of CFP–ParB (green trace) and a kymographic representation of the ParA–YFP signal profile (red) along the cell length as a function of time after synchrony. CFP–ParB synthesis was induced with 0.03% xylose for 1 h before synchronization and imaging. (D) Profiles of the mean CFP–ParB and ParA–YFP intensity profiles of wild-type cells (n=72) from time-lapse sequences described in (C). (E) Same as (C) except in a ΔtipN background (strain CJW3376). (F) Time-lapse microscopy of YFP–ParA and MipZ–mCFP in cells slightly overproducing YFP–ParA (strain CJW3373). YFP–ParA overproduction was induced with 0.03% xylose for 2 h before cell synchronization and imaging. The results from a representative cell are shown as an overlay between the relative position of MipZ–mCFP (green trace) and a kymographic representation of the ParA–YFP signal profile (red) along the cell length as a function of time after synchrony. (G) Same as (F) except that the overproduction of YFP–ParA was induced with 10 times more xylose (0.3%) for 2 h before cell synchronization and imaging.

Figure 3

Interaction between ParA and TipN. (A) Western blotting of pull-down eluates from lysates of cells carrying ParA–YFP and TipN-FLAG (TipN-FLAG; strain CJW3359) and lysates of control cells carrying ParA–YFP and untagged TipN (TipN; strain CJW3010). Elution was carried out in the presence (FLAG) or absence (Mock) of FLAG peptide. (B) Distributions of nFRET/YFP values for TipN–CFP and ParA–YFP (strain CJW3406) and for TipN–CFP and mYFP-PopZ (strain CJW3614) at the new pole. mYFP-PopZ was induced with 0.03% xylose for 2 h before imaging. The red line is the Gaussian fit of the distributions (see Supplementary data). (C) Overlays of phase-contrast images with fluorescent images of MipZ–mCFP, ParA–YFP and TipN–mCherry localization in CJW3407 cells in which TipN–mCherry overproduction was either uninduced (normal TipN) or induced with 0.5 mM vanillic acid for 6 h (TipN–mCherry overproduced). (D) Time-lapse sequence of MipZ–mCFP (green) and ParA–YFP (red) signals in CJW3408 cells during TipN overproduction, which was induced with 0.25 mM vanillic acid for 3 h before imaging on agarose pads containing vanillic acid.

Figure 4

Mutational analysis of ParA. (A–D) Kymographs of a specific ParA–YFP mutant (in red) in single cells. The traces of MipZ–mCFP are shown in green. The synthesis of each ParA mutant was induced for 1.5 h with 0.3% xylose before synchronization and imaging. (A) ParA_G16V–YFP; strain CJW3337. (B) ParA_R195E–YFP; strain CJW3369. (C) ParA_D44A–YFP; strain CJW3346. (D) ParA_{D44A R195E}–YFP; strain CJW3507. (E) ParA_G16V–YFP in a ΔtipN cell (strain CJW3338) as in (A–D). (F) Mean integrated intensity of ParA_G16V–YFP signal in wild-type (CJW3337) and ΔtipN (CJW3338) cells during MipZ–mCFP segregation (top) and after MipZ–mCFP reaches the new pole until the onset of cell constriction (bottom). (G) Western blotting of pull-down eluates from lysates of cells carrying TipN-FLAG and ParA_D44AR195E–YFP or ParA_G16V–YFP (strains CJW3538 or CJW3537). Elution was carried out in the presence (FLAG) or absence (Mock) of FLAG peptide.

Figure 5

Mechanisms compensating for a loss of TipN function. (A) Schematic showing the different stages plotted in (B) and (C). (B, C) Synchronized predivisional cells of wild-type (CJW2022) and ΔtipN (CJW3366) strains carrying mipZ–yfp were grown on agarose pads containing M2G and imaged every 1.5 min. The time right after DNA replication in the swarmer and stalked progeny was determined by the appearance of a second MipZ–YFP focus. Cell identification and image analysis were performed using MicrobeTracker. (B) Distributions of (1) the length of wild-type swarmer and stalked cells at birth, (2) times between birth and right after DNA replication and (3) cell lengths right after DNA replication. (C) Distribution of (1) cell lengths at birth, (2) times between birth and right after DNA replication and (3) cell length right after DNA replication, starting with wild-type or ΔtipN swarmer cells. (D) Overlays of phase contrast and MipZ–mCFP images of ΔpopZ cells carrying mipZ–mcfp (strain CJW3599) and ΔpopZ cells carrying mipZ–mcfp and depleted of TipN (growth without xylose for 17h; strain CJW3543). Purple arrows show minicells lacking MipZ–mCFP signal and yellow arrows show the creation of these cells. (E) Percentage of the cell population lacking MipZ–mCFP from wild-type cells (CB15N), ΔpopZ cells (CJW3599) and ΔpopZ cells (CJW3543) either depleted of TipN (growth without xylose for 17 h) or overproducing TipN (growth in 0.3% xylose). Shown is the mean±s.d. from three independent experiments. (F) Top, phase-contrast images of ΔpopZ cells (CJW3512) either producing TipN–mYFP at a normal level (growth in PYE with 0.2% glucose) or overproducing TipN–mYFP (growth in PYE with 0.3% xylose for 17 h). Scale bar, 5 μm. Bottom, distribution of cell lengths under each condition. (G) Phase-contrast and overlays of ParA–YFP (red) and MipZ–mCFP (green) in a ΔpopZ background carrying tipN under Pxyl on a low-copy-number plasmid (strain CJW3043). Overexpression of tipN was either uninduced with 0.2% glucose or induced with 0.3% xylose for 5 h.

Figure 6

Effects of ΔtipN mutation on the temporal and spatial regulation of cell division and cell growth. (A) Top, distributions of the time from synchrony to the appearance of the FtsZ–YFP ring. Bottom, distributions of the time from synchrony to the appearance of cell constriction. These distributions were obtained from time-lapse sequences of wild-type (MT199; n=190) and ΔtipN (CJW2563; n=114) cells as described in Figure 1A. (B) Distributions of rates of cell-length elongation during the time-lapse sequences described in Figure 1A. (C) Rates of cell-length growth were plotted as a function of the cell length at time=0 min in Figure 1A. (D) Mean degree of cell constriction ±s.e.m. of wild-type (CB15N) and ΔtipN (CJW1407) cells after the time of cell constriction initiation. Following synchronization, cells were imaged every 1.5 min by time-lapse microscopy and analysed by MicrobeTracker. (E) Distributions of time gaps between initiation of cell constriction initiation and cell separation for cell populations in (D).

Figure 7

Proposed models for chromosome segregation and cell cycle coordination in C. crescentus. (A) Model for parS/ParB segregation. In wild-type cells, after initiation of DNA replication, one of the duplicated parS/ParB partition complexes comes in contact with the leading edge of a DNA-bound ParA-ATP structure (red circles) for which ParB has a strong binding affinity (step 1). (Only a small portion of the ParA structure is shown for simplicity.) ParB then stimulates ParA-ATPase activity, resulting in dissociation from the nucleoprotein filament and release of ParA molecules (blue circles), presumably ParA monomers, into the cytoplasm (step 2). This shortens the DNA-bound ParA-ATP structure and the released ParA is sequestered by TipN at the new pole to prevent regeneration of DNA-associated ParA-ATP dimers (step 3). Presumably through Brownian motion, the segregating parS/ParB reaches the new leading edge of the DNA-bound ParA-ATP structure, thereby getting closer to the new pole. Successive rounds of this cycle result in net parS/ParB translocation to the new pole. See text for more details. (B) Model for cell cycle coordination. Before DNA replication and segregation are initiated, the parS/ParB/MipZ complex is attached at the old pole by the PopZ matrix, promoting the establishment of a DNA-bound ParA-ATP structure biased towards the new pole (that is, away from the destabilizing activity of ParB at the old pole) (i). During segregation of the partition complex (described in panel A), the TipN-mediated condensation of the ParA cloud structure ensures rapid and unidirectional translocation of the partition complex (ii). Rapid segregation, which affects the timing of FtsZ ring formation, ensures an early switch between a dispersed/helical mode of growth and a primarily zonal, FtsZ ring-dependent mode of growth. Unidirectional translocation, on the other hand, affects the proper positioning of the FtsZ ring along the cell length, which dictates where the cell will divide (iii). TipN and PopZ functions maintain ParA and the partitioning complexes at the cell poles (iv) such that only a single round of segregation occurs per cell cycle. Near the end of the cell cycle, TipN delocalizes from the new cell pole, contributing to the release of ParA (v). This event, coupled with cytokinesis, which creates new cell poles free of PopZ and ParB, resets the cycle and allows for the reestablishment of the DNA-bound ParA-ATP structure in the daughter cells (vi).