Review 2: "Chalkophore Mediated Respiratory Oxidase Flexibility Controls M. Tuberculosis Virulence"

RR:C19 Evidence Scale rating by reviewer:

• Potentially informative. The main claims made are not strongly justified by the methods and data, but may yield some insight. The results and conclusions of the study may resemble those from the hypothetical ideal study, but there is substantial room for doubt. Decision-makers should consider this evidence only with a thorough understanding of its weaknesses, alongside other evidence and theory. Decision-makers should not consider this actionable, unless the weaknesses are clearly understood and there is other theory and evidence to further support it.

***************************************

Review: This preprint is focused on chalkophores and their role in in supporting the function of the heme-copper respiratory oxidase. The authors use a mutant strain of M. tuberculosis that cannot synthetize chalkophores (Δnpr), in presence or absence of copper-independent oxidase system (cydAB), to probe gene expression, growth, survival, oxygen utilization, ATP production, and bacterial burden in mice. This study is continuation of the previously published study by the same group (ref #21), in which they already demonstrated that chalkophores are required for optimal growth in vitro under copper starvation.

Although this study could contribute to better understanding of M. tuberculosis pathogenesis, several concerns regarding the way results were presented and interpreted are noted. The authors claim that the Δnpr mutant “stimulates gene expression that mimics inhibition of the bcc:aa₃” (copper-dependent respiratory oxidase), but judging by the heatmap showed in Fig. 1A, this characterization may be exaggerated. It was stated that the growth of wild-type M. tuberculosis was not affected by increasing concentrations of TTM, a copper chelator. However, it appears the wild-type strain’s growth was inhibited by 50-90% when exposed to 50 µM TTM compared to DMSO (Figure 1D). They also state that complementation with npr or cydABDC in the double mutant (ΔnprΔcydAB) restored growth and survival. The latter only had partial complementation (Figures 2B an 2C). Therefore, all data with the npr and cydAB double mutant must be interpreted with caution and results should be accurately presented and discussed.

It is curious that the authors claim that “chalkophore deficient M. tuberculosis cannot survive, respire to oxygen, or produce ATP under copper deprivation”, while oxygen consumption is the same in WT and Δnpr mutant under copper limitation (Fig. 3C) and the phenotype is more prominent only in the ΔnprΔcydAB double mutant, with partial complementation with npr (Fig. 3D). Furthermore, the lack of difference in ATP production between ΔnprΔcydAB and ΔnprΔcydAB+nrp strains in presence of TTM (Fig. 3E) does not fit the model in which chalkophores support ATP production. It is essential to also include individual mutants (i.e., Δnpr and ΔcydAB) in testing ATP production under copper limitation, in order to investigate the role of chalkophores in ATP synthesis more directly. As with the oxygen consumption, it is likely that ATP production is only affected in the ΔcydAB mutant, which should be addressed.

In summary, although the authors convincingly show that the Δnpr mutant has a growth defect in presence of copper chelators in vitro and in the mouse infection model (both observations already reported previously), the lack of complementation of the double mutant with npr for ATP production (or the lack of phenotype of the Δnpr regarding oxygen consumption) when grown under copper limitation contradicts one of the main conclusions. Changing statements that claim that the Δnpr mutant cannot efficiently respire oxygen or produce ATP under copper deprivation (such as the one in the abstract) to specify that this is true only in the strains lacking copper-independent oxidase as well, would greatly improve this manuscript.

Note for the authors:

• RNA-seq data are shown in the way that cannot be properly evaluated. While the authors provide normalized count data in the supplementary information and the heatmap in Fig. 1A, the differential expression analysis of the RNA-seq data (i.e., log2FC of mutant vs. wild type, and p-values) should be included to convince the readers that the differences in expression are real and meaningful. Judging by the heatmap colors, it is not clear if the Δnpr mutant indeed significantly upregulates copper-independent oxidase and ATP synthase genes, as claimed.

• The authors alternative between “diisonitrile chalkophores” and “chalkophores”, and use copper deprivation, starvation and limitation interchangeably throughout the text. They should choose one term and consistently use it throughout the text.

• More details for each experiment should be provided, e.g., exposure times, control conditions/strains in the figure captions, and comparisons (i.e., if significantly different or not) between conditions/strains should be consistently shown on bar graphs. 

• The authors complemented the cydAB mutant with cydABDC without providing reasoning for it.

• An alternative chelator BCS was used in some experiments without explanation.

• Some figures were not referenced correctly in the text.

• For the mouse study, it would be useful to note that the mutants were attenuated early on during infection, but reached similar lung burden as the wild-type strain at 112 days post-infection. In contrast, bacterial burden in spleen remained lower for the mutant strains, suggesting that oxidase systems may have an effect on dissemination.

• Inclusion of SCID mice is not clearly justified.   

• It is critical to expand the discussion section to better understand the impact of this study and address weaknesses of their model.