DNA replication and cell division are fundamental processes that are essential for the survival and proliferation of all cells. Successful cell division and chromosome replication require the careful coordination of multiple factors, including accurate DNA replication, precise regulation of the cell cycle, and proper segregation of chromosomes to the daughter cells. Bacterial DNA replication is a tightly regulated process that involves the initiation and progression of the replication fork, the elongation of the nascent DNA strands, and the termination of DNA synthesis, which are coordinated by a network of proteins and regulatory factors (reviewed in Reyes-Lamothe et al., 2012). The process of bacterial cell division, or cytokinesis, requires the formation of a cell wall and the generation of a septum at the midcell. The division septum then acts as a scaffold for the recruitment of the protein complex required for cell division, also known as the divisome (reviewed in Mahone and Goley, 2020).

The divisome complex is assembled around the FtsZ protein, which forms the Z-ring. The Z-ring formation starts with FtsZ polymerization, a tubulin homolog that recruits ZapAB and other divisome proteins to the septum (Adams and Errington, 2009; Galli and Gerdes, 2012). The Z-ring-associated protein ZapB plays a central role in the cell division process as it also coordinates chromosome segregation by interacting with MatP, a factor required for the terminus region of E. coli chromosome (Männik et al., 2016; see Figure 1A). MatP-ZapB interaction is necessary to anchor the Ter macrodomain at midcell (Espéli et al., 2012), thus positioning the Z-ring at the right place (Bailey et al., 2014). As expected for septation protein, mutant of zapB gene results in longer cells (Ebersbach et al., 2008). There are still many questions remaining concerning the molecular mechanisms of cell division, such as specific recruitment of divisome proteins in time and space. Moreover, different factors can influence bacterial cell division rates like nutrient availability (reviewed in Kellogg and Anne Levin, 2022).

Iron is an essential element for optimal growth and fitness of bacterial species. Iron is used in the electron transport chain to generate ATP and is central to the synthesis of amino acids and nucleotides. It is key to the transport of oxygen and contributes to cellular respiration (Messenger and Barclay, 1983; Andrews et al., 2003; Frawley and Fang, 2014). Iron also enables the expression of genes involved in iron uptake, storage, and utilization through the transcriptional regulator Fur (Hantke, 2001). Furthermore, iron is critical for DNA replication and repair by providing efficient synthesis of dNTPs through the action of ribonucleotide reductase. For instance, certain ribonucleotide reductases employ a variety of co-factors, ranging from di-iron cofactor (Fe-O-Fe) to Fe-S cluster, that catalyzes the conversion of ribonucleotide into deoxyribonucleotide (Puig et al., 2017). Moreover, DNA processing requires a series of Fe-containing protein to maintain genomic integrity. Many DNA-binding protein involved in base excision repair, endonuclease III, helicase, DNA primase, as well as DNA and RNA polymerases require Fe-S cluster for their activity (Puig et al., 2017; Grodick et al., 2014).

Previously, it was demonstrated that iron starvation, induced by various chelators, effectively blocked bacterial cell division (Santos et al., 2018). The observed phenotype indicated a significant increase (30–40%) in E. coli cell length. While the mechanism is not clear, formation of the cell division machinery was clearly affected in absence of iron. These results suggest a coordinated response between iron availability and the cytokinesis process. Nevertheless, iron starvation may create an unfavorable condition for the function of many iron-dependent enzymes, which prevents growth.

Absence of iron is one of the many stresses to which bacterial cells must adapt to thrive. To efficiently adjust their metabolism according to environmental insults, bacteria use small regulatory RNAs (sRNAs). These molecules extend usually between 100 and 500 nucleotides and can rapidly regulate gene expression to overcome different environmental stress. sRNAs that regulate gene expression at post-transcriptional level typically bind to their mRNA targets and either induce degradation of the mRNA by recruiting the RNA degradosome or activate their target mRNA translation by disrupting secondary structures and releasing the RBS (Hör et al., 2020; Lalaouna et al., 2013).

In E. coli, RyhB sRNA is expressed under iron-limiting conditions and mainly regulates iron-using proteins that are specifically involved in metabolism (sodB and cysE), TCA cycle (acnB, sdhCDAB, and fumA), iron metabolism (iscRSUA and erpA), and iron transport (shiA and cirA; Massé et al., 2005; Salvail et al., 2010; Salvail and Massé, 2012; Salvail et al., 2013; Chareyre and Mandin, 2018). Our group previously observed that cells lacking RyhB exhibited severe growth limitation during iron starvation (Jacques et al., 2006). However, the mechanism underlying this observation remains unclear.

In this study, our results suggest that the sRNA RyhB directly downregulates ZapB translation during iron starvation. Notably, ZapB is most likely the first target of RyhB not involved in iron metabolism, does not carry an iron co-factor, and the first involved in cytokinesis and DNA segregation. Reduction of ZapB expression by RyhB in wild-type cells correlates with longer cells, as compared to ryhB mutant, potentially due to reduced cell division. Our observations on the growth of ΔryhB and ΔryhBΔzapB mutants suggest that absence of ZapB helps recover the reduced growth of ΔryhB mutant during iron starvation. Furthermore, data from flow cytometry indicate that ZapB repression by RyhB resulted in increased number of chromosomes in cells growing under iron starvation. Additional observations suggest that helicase paralogs DinG and YoaA, both containing Fe-S clusters, might depend on RyhB activity for successful chromosome replication. Overall, our study suggests a possible mechanism of coordination between intracellular iron and cell division mediated by the sRNA RyhB, potentially by reducing cytokinesis when iron is limiting. This mechanism might also create safeguards for facilitating DNA replication and chromosome segregation by keeping Fe-S-dependent DNA helicases operational.

Most organisms growing under iron starvation require extensive metabolic adaptation to palliate and optimize growth in the absence of this key element. Previous studies indicated that the Krebs cycle, respiratory chain complex, and metabolite trans-membrane transport are particularly affected during iron starvation (reviewed in Chareyre and Mandin, 2018). A more recent study added cell division as another cell function impacted by absence of iron (Santos et al., 2018). Our observations corroborate with the description that iron starvation decreases E. coli cytokinesis and suggest a mechanism in which the sRNA RyhB reduces the expression of ZapB. Remarkably, repression of the factor ZapB through RyhB-zapB interaction not only impacts cell length but also chromosome distribution within a bacterial population. Moreover, our data showed that RyhB-deficient cells growing in absence of iron exhibit increased sensitivity to AZT, a drug targeting Fe-S enzymes involved in the DNA replication machinery. Interestingly, deletion of the zapB gene leads ΔryhB strain to recover normal growth in iron depleted conditions. What's more, the ΔryhBΔzapB strain shows similar recovery of both chromosome distribution and AZT resistance as compared to ΔryhB strain. Altogether, these results suggest a possible link between cytokinesis, chromosome segregation, and iron availability based on RyhB-zapB interaction (Figure 7).

Santos et al. (2018) previously demonstrated that both assembly and localization of the cell divisome, at the late stages of cytokinesis, were affected during iron starvation. Moreover, iron chelation caused a reduction of fluorescent signals in ZapA, FtsN, FtsK, and FtsI proteins tagged with GFP, suggesting a reduced expression in parts of the divisome machinery. Although the authors did not specifically investigate ZapB protein, it had already been shown that ZapB directly interacts with ZapA (Galli, 2011; Buss et al., 2017). Our own results suggest that, under restricted iron levels, the sRNA RyhB reduces the expression of ZapB, thereby reducing cell division and affecting the functional divisome machinery. Our present work supports the idea that observations made by Santos et al. were due to ZapB repression by RyhB in the absence of iron.

Our laboratory has previously shown that a ΔryhB strain had a deficient growth in minimal media without iron (Jacques et al., 2006). While this suggests that RyhB is essential for optimal growth under iron restricted condition, it remained unclear why cells struggled in absence of the sRNA. We assumed that absence of regulation of several target mRNAs by RyhB could result in growth struggle. Our new data suggests that this growth phenotype could be the consequence of a single target, zapB, being misregulated in absence of iron, which results in premature cell division, giving rise to shorter cells and abnormal chromosome segregation.

The ΔryhB growth defect under restricted iron could also be resulting from fewer chromosomes and DNA deficiencies. Data from flow cytometry (Figure 4B) indicate that the ΔryhB mutant exhibits the most cells with only one or two chromosomes while having the least cells with 3 or 4 chromosomes as compared to WT, ΔzapB, and ΔryhBΔzapB backgrounds. Additional observations we made on the ΔzapB strain in Figures 4, 5 also correlate with previous observations made by Ebersbach et al. (2008), which stated that a deletion of zapB increases the average cell length. Our results suggest that a ΔzapB mutant presents longer cells and additional chromosomes per cell, which can be observed in Figure 5A by a shift in the chromosome profiles in absence of iron. Moreover, our control experiment in the presence of rifampicin and cephalexin to stop DNA replication (through RNA polymerase inhibition) and cell division supports this observation (Supplementary Figure 5).

Notably, WT cells growing in the absence of iron display an increase in chromosome number as compared with cells grown in the presence of iron (Figure 5A, WT +/- iron). This increase in chromosome number in absence of iron probably stems from RyhB repressing ZapB expression as the zapB mutant also shows increase in chromosome number (Figure 5A, ΔzapB strain, -iron). Previous reports described how the chromosome number usually follows a trend correlating with growth rate (reviewed in Reyes-Lamothe et al., 2012; Sanders et al., 2023). This higher number of chromosomes would allegedly prepare rapidly growing cells to divide at higher rate. In our observations however, the higher number of cells with four chromosomes correlates with a situation where iron is scarce and growth rate is low. This situation is similar to UPEC infecting the host bladder cells where significantly longer bacterial cells were observed (Justice et al., 2004, 2006). Surprisingly, septation of these filamented cells on the bladder surface favored infection. Moreover, a recent study reported that overproducing the iron-using IscA protein leads E. coli cells to filament (Wei et al., 2024). These observations can be explained by sequestration of free intracellular iron when overproducing the protein IscA, thus effectively inducing cellular iron starvation and RyhB expression. Notably, the zapA transcript, encoding a factor interacting with ZapB, was strongly induced during these conditions, which indicate that other proteins of the divisome can potentially affect cytokinesis during iron starvation. Overproduction of ZapA was previously shown to generate filamentation of E. coli cells (Galli and Gerdes, 2012).

Repression of another septation protein has been shown in previous work, where the sRNA DicF can repress translation of the ftsZ transcript (Tétart et al., 1992; Balasubramanian et al., 2016). Although the physiological reason for this regulation remains unclear, it was shown that anaerobic conditions increased DicF stability and permitted the repression of ftsZ translation (Murashko and Lin-Chao, 2017). These results indicate how the bacterial cell can change its shape to adapt to environmental challenges similar to our results in absence of iron.

To our knowledge, YoaA, DinG, and QueE are the only proteins involved in cell division or chromosome segregation which are using iron. Notably in E. coli, paralogs YoaA, and DinG encode for DNA helicases, both of which use 4Fe-4S clusters as prosthetic groups (Bak and Weerapana, 2023; Ren et al., 2009). We previously demonstrated that a ΔryhB mutant contains less intracellular iron than WT cells growing under iron starvation conditions (Salvail et al., 2010). From this perspective it is tempting to hypothesize that, under these conditions (iron restriction or absence of RyhB), intracellular iron becomes too limited to sustain YoaA and DinG enzymatic activity. This reduced YoaA and DinG enzymatic activity in absence of iron can potentially lead to sensitivity to azidothymidine drug (AZT), an analog of thymidine which blocks DNA replication, as previously demonstrated in ΔyoaA and ΔdinG strains (Brown et al., 2015). Our results demonstrated that a ΔryhB strain growing in absence of iron is more sensitive to AZT. This suggests that in conditions of iron starvation, and potentially reduced YoaA and DinG activities, cells might benefit from reduced division through repression of ZapB protein by RyhB. Even though zapB is potentially the first target of RyhB not directly involved in iron metabolism, reducing its expression might create an environment facilitating DNA replication and chromosome segregation. Another reason why RyhB could downregulate ZapB in an iron starved environment would be that longer cells expose more surface for iron acquisition machinery and siderophores transport.

Finally, this regulation of RyhB on zapB mRNA seems peculiar as RyhB usually rapidly degrades its targets within minutes of iron depletion induction (Massé et al., 2003; Massé and Gottesman, 2002). It is interesting to note that in the RIL-seq experiment (Melamed et al., 2016), the interaction between RyhB and zapB depends on Hfq pulldown, suggesting that Hfq is involved in this interaction. In the case of RyhB-zapB regulation, it is mostly at translational level and the mRNA level remains stable throughout a 30-min induction of iron depletion. This could be explained by the fact that RyhB has two binding sites on zapB mRNA, similar to RprA regulation of hdeD mRNA (Lalaouna et al., 2018). We also hypothesize that two RyhB molecules are needed to bind zapB mRNA due to the opposite 5′-3′ sense of its binding regions. The first binding site explains the translation block since it's located near the translation start codon of ZapB. The second binding site, located in zapB CDS, could block a nuclease site which could explain the regulatory effect on zapB transcription and zapB mRNA stability when RyhB is expressed. This hypothesis is also supported with the RNase E-CLASH experiment which indicated an interaction of RyhB::zapB along with RNase E at the second binding site (Waters et al., 2017). Notably, this suggests a functional regulation of RyhB on zapB transcript beyond the five-codon window already described by Bouvier et al. (2008). Altogether, our results demonstrate that RyhB not only regulates iron metabolism but DNA replication, chromosome segregation, and cell division as well.