Here is pasted the response provided by the authors to the report of referee #2. Experiments aimed at better characterizing DNA replication in suppressed and non suppressed cells at high temperature are under way.

Response to reviewer #2 :

The main claim of this paper is that the three-carbon portion of the glycolysis/gluconeogenesis pathway acts in part to regulate lagging strand DNA replication in Bacillus subtilis. This claim is based on the finding that mutations mapping to pgk, pgm, eno, pykA, or gapA act to suppress temperature sensitive growth conferred by various mutations in one of three different genes encoding essential replication proteins thought to act largely on lagging strand, including dnaE, dnaG, and dnaC. Although not much has been reported about the relationship between bacterial replication and glucose metabolism, the authors do discuss their results in the context of what has been described regarding this relationship in other model systems, primarily yeast. The work described in this manuscript will be of interest to those working on both replication/repair, as well as glucose metabolism as it illustrates a possible intersection between these two important biological processes.

Most of the manuscript is clearly written, and is fairly easy to follow, even for non-experts. However, there are a fair number of typos throughout the text that need to be corrected. Although some do not impair the clarity of the text, some do add confusion. For example, on page 5, line 135, the authors refer to dnaE2.6 ∆gapA in reference to Figure 1E
(There is no typo here: Fig1E presents the lack of filament in the gapA dnaE2.6 strain);

however, the Figure shows dnaE2.10 pgm25
(This is shown Fig1C and not Fig1E).

In addition, although the materials and methods employed are clearly spelled out, they are restricted to just the materials and methods section. Importantly, there are a few sections in the results where a few simple statements regarding the methodology employed would help greatly to clarify both the results, as well as the approaches used. For example, in the legend to Figure 1, the authors state that transformed strains were grown in LB with or without xylose, but I had to dig around in the methods to confirm that they were actually using a xylose-dependent promoter It would help greatly if the authors clarified this in the legend and the results section.
The text now includes more technical information. It is now specified (lines 134-5, 1013-7) that the cloned WT metabolic genes mentioned in the text (line 134 and in Fig. 1B) are expressed from the xylose dependent Pxyl promoter.

With respect to the genetic data, the authors appear to have demonstrated very convincingly that mutations affecting genes whose products act in the three-carbon portion of the glycolysis/gluconeogenesis pathway suppress the temperature sensitive growth phenotype of specific replication mutants. What is unclear, however, is the mechanistic basis by which the suppressor mutations achieve this. The authors do test the possibility that it is a general stress response, and provide evidence that suppression is specific to these combinations of mutations. However, what is currently known about the replication mutants that they examined? Have the various dnaE, dnaG, and dnaC Ts alleles been characterized previously? Is the mechanistic basis for their Ts phenotypes understood? Are these mutant known to be folding mutants?
To answer the referee’s questions, we now mention data in the discussion section indicating that dnaE, dnaC and dnaG mutants are likely affected in DNA chain elongation (lines 336-350).
Also, how can the authors be certain that suppression does not involve a change in the requirements for dnaE, dnaG, or dnaC

I understand that some suppressed strains still display a Ts phenotype at very high temperatures, and that growth was still dependent upon dnaE function, but that does not necessarily mean that these proteins are functioning in a similar manner in both the wild type and the suppressed strains.
Mechanisms of suppression relying on changes in the requirement for DnaE, DnaG and DnaC at the fork are now presented in details. In lines 431-450, we address the possibility that suppression results from compositional changes in the fork that make dispensable the DnaTs proteins in glycolytic mutants. In lines 452–469, we discuss the possibility that suppression results from functional changes in the fork.

Thus, perhaps the authors should consider relaxing their wording on the top of page 9.
(This has been done, lines 286-287)

Also, along these lines, I think that the authors may want to soften their enthusiasm for their model. Although it is a reasonable model, it is quite speculative at this point, and many other possible explanations exist. Importantly, the data presented is focused exclusively on growth phenotypes; the authors did not examine replication efficiency in vivo using labeled DNA precursors, etc., and as noted above, they do not discuss evidence that alterations in the structures of the respective Dna-Ts proteins contribute to their temperature resistance
I do not mean to imply that I disagree with the model presented: indeed, I think that it is a plausible model, but just one of many. Clearly, considerable effort may be required in order to establish the actual mechanism(s) of suppression.

Finally, I wonder why the authors chose to not show a complete set of data demonstrating suppression by a single method for all the suppressor strains? I think that this is important to show data indicating suppression for all strains. They could then illustrate the rigor they employed by showing data for a single suppressor demonstrating the various methods discussed, and simply indicate that others were screen similarly, citing data not shown.
During the writing process we decided to present data in two ways. In the first part entitled “ Genetic link between DNA replication and glycolysis” (lines 120-178, Fig. 1-3), we deliberately choose to illustrated the robustness of the suppression phenotype by presenting different type of results obtained with different combination of Ts and metabolic mutations. This part of the paper also contains, as requested by the referee, results obtained with a single assay for all the double mutant tested (the plating assay, Fig2 and 3A). In the remaining sections (“Identification of metabolic alterations causing suppression” and “ Impact of metabolic alterations on DnaTs protein activity » lines 180-305), the presented results (Fig 4-5) are all derived from strains carrying a pykA mutation. So we believe that the actual organization of the result section answer the referee’s comment.