Discovery of a new type of regulator of DNA replication : the glycolytic enzyme pyruvate kinase in Bacillus subtilis

It has been known for decades that the timing of DNA replication within the cell cycle in bacteria is coupled to nutrient richness. However, the mechanisms that enable this temporal gating have remained elusive. The goal of this work was to test whether the CCM enzyme PykA in the bacterium B. subtilis helps achieve this gating by moonlighting as a temporal effector of replication that communicates information about the metabolic state of the cell to the replication machinery. To demonstrate that PykA moonlights as a replication protein, the effect of pykA mutations on DNA replication parameters was assessed using run-out flow cytometry and marker frequency analysis by qPCR in growth conditions where its metabolic activity is dispensable.Deletion of pykA did not affect the growth rate during these conditions, but it affected the replication speed and timing of initiation, altering therefore the temporal control of replication. The deletion of either the catalytic domain or the PEP utilizer domain of PykA showed that both domains play a role in the temporal control of replication, identifying for the first time a replication function to these domains. Mutagenesis of specific amino acids inside the catalytic domain showed that some, but not all key catalytic amino acids are involved in replication. Mutagenesis of conserved amino acids in the TSH motif of the PEP utilizer on the other hand suggests an important role for threonine phosphorylation. Expression of the PEP utilizer detached from the catalytic domain in trans unexpectedly affected replication control at medium copy numbers. This phenotype depended strictly on H of the TSH motif, establishing a role for H in replication. Furthermore, the effect of expression of free PEP utilizer may depend of an unidentified inhibitor effector and is fully suppressed by a mutation in the catalytic domain. Moreover, expression of the catalytic domain in cis and the PEP utilizer in trans did not restore proper replication control. Finally, in vitro assays demonstrated that PykA can directly stimulate DnaE polymerase activity and inhibit DnaC helicase activity. Together, theseresults imply that PykA moonlights in replication.To demonstrate that PykA communicates information about the metabolic state of the cell to the replication machinery, the effect of pykA mutations on DNA replication parameters during a metabolic shift was assessed using the techniques described above.The adaptation of replication parameters during the metabolic shift in wild type cells unexpectedly consisted of three phases: (i) a decrease of replication fork speed between 15 and 45 min after the shift, (ii) an advancement of initiation timing and increase in initiation frequency from 30 min onwards and (iii) an adaptation of the growth rate sometime after 45 min after the shift. Mutations in the catalytic domain did not clearly demonstrate a change in the replication response to the shift. By contrast, the effect of the PEP utilizer domain (both in cis and in trans) on replication parameters did change during the shift. Moreover, the H amino acid, not T, was important for replication control during later time points of the shift. This suggests that the regulatory behaviour of the PEP utilizer and more generally PykA changes in response to metabolism.Overall, we conclude that PykA is a new type of regulator of DNA replication that helps the temporal positioning of replication in the cell cycle through processes that varies with the richness of the carbon sources provided in the environment. This work, combined with previous work in this field suggest that the temporal control of replication in a large range of metabolic growth conditions is achieved at least in part via a network of moonlighting CCM enzymes with important implications for our basic understanding of cell biology and human health.