Robust control of replication initiation in the absence of DnaA-ATP ↔ DnaA-ADP regulatory elements inEscherichia coli

Abstract

Replication initiation, especially the widely conserved master initiator protein DnaA, is one of the most well-studied biological problems in bacteria. Specifically, conversion between the ATP and ADP form of DnaA during growth is critical for initiation control. In Escherichia coli, the regulatory elements for conversion have been discovered and extensively characterized over decades. However, they are not widely conserved in bacteria, raising questions on how to generalize the findings in E. coli to other organisms. In this work, we show that the intrinsic ATPase activity of DnaA itself is sufficient for robust and precise initiation control. We constructed and studied E. coli mutants lacking the extrinsic control of either DnaA-ATP → DnaA-ADP (by hda and datA) or DnaA-ADP → DnaA-ATP (by DARS1 and DARS2) at the single-cell and population levels. These cells showed distinct and opposing characteristics in initiation timing, the degree of initiation asynchrony, and cell-to-cell variability. Strikingly, when all four regulatory elements were deleted, E. coli exhibited a near wild-type phenotype, with only mildly increased intrinsic and extrinsic initiation noise. By further characterizing the DnaA variants with increased and decreased ATPase activity, we conclude that DnaA is the only requirement for robust initiation, shedding new evolutionary light on cell-cycle control in bacteria.

Controlling DNA replication initiation: from living to synthetic cells

“…The effective Hill coefficient lies therefore in the range n eff = 29−38 (vertical dotted grey lines), corresponding to a coefficient of variation of CV=0.05-0.07. This agrees well with the experimentally measured precision of the initiation volume of CV≤0.1 [22,23,135]. Interestingly, the average degree of synchrony at n eff = 30 and n eff = 40, respectively, is given by 〈s〉(n eff = 30) = 0.975 and 〈s〉(n eff = 40) = 0.996, corresponding to the probabilities to fire all origins synchronously of 〈P (∆t < τ l )〉(n eff = 30) ≡ P s (n eff = 30) = 95.5% and P s (n eff = 40) = 98.9%.…”

“…Later, Skarstad et al confirmed the prediction of Cooper and Helmstetter by counting the numbers of origins in rapidly growing cultures: They found that most cells have 2 n (n = 1, 2, 3) copies of their chromosome and only a small fraction of cells (2-7%) contained 3, 5, 6 or 7 chromosomes [134]. Recent single cell measurements indeed show that E. coli initiates replication synchronously at up to eight origins with a very high precision in the fast growth regime [22,135]. It remains however an open question how E. coli achieves such a high degree of synchrony.…”

“…[22]) and would thus only result in a relatively low degree of synchrony of less than 〈s〉 = 0.92 corresponding to a probability of initiating synchronously of 〈P (∆t < τ l )〉 ≡ P s = 84%. Recent experiments show however that the contribution from the intrinsic noise in replication initiation to the CV is only about CV int =0.04 -0.05 [135]. Our model, which only concerns the effect of intrinsic noise, then predicts that n eff must be at least 40 (Fig.…”

Direct single cell observation of a keyE. colicell cycle oscillator

“…In summary, here we show that in single cells, for all the promoters studied here, the volume- ers [36] shows that the DNA replication cycle is very similar to the wild type strain when RIDA and the other DnaA activity regulators, datA and the DARS sites, are inactive, showing that the inherent ATP hydrolysis activity of DnaA can be sufficient to drive a stable replication cycle. Further work will be required to determine how the activity of these different regulatory factors maintains a robust replication program in the presence of perturbations such as growth shifts or DNA damage [75,76].…”

“…On the other hand, the increase in DnaA-ATP required for the initiation of a new DNA replication cycle depends in part on the timely accumulation of newly expressed protein [6,25,32] and the binding of DnaA to the DARS1 and DARS2 sites contributes to a further change in the DnaA-ATP to DnaA-ADP ratio by favoring the exchange of the DnaA-bound nucleotide [6,[33][34][35]. A recent study has shown that in the absence of RIDA, datA and DARS-dependent regulation, the intrinsic ATP hydrolysis activity of DnaA suffices to maintain a near-wild type replication cycle [36].…”