ABSTRACT Archaeal molecular biology has been a topic of intense research in recent decades as their role in global ecosystems, nutrient cycles, and eukaryotic evolution comes to light. The hypersaline-adapted archaeal species Halobacterium salinarum and Haloferax volcanii serve as important model organisms for understanding archaeal genomics, genetics, and biochemistry, in part because efficient tools enable genetic manipulation. As a result, the number of strains in circulation among the haloarchaeal research community has increased in recent decades. However, the degree of genetic divergence and effects on genetic integrity resulting from the creation and inter-lab transfer of novel lab stock strains remain unclear. To address this, we performed whole-genome re-sequencing on a cross-section of wild-type, parental, and knockout strains in both model species. Integrating these data with existing repositories of re-sequencing data, we identify mutations that have arisen in a collection of 60 strains, sampled from two species across eight different labs. Independent of sequencing, we construct strain lineages, identifying branch points and significant genetic events in strain history. Combining this with our sequencing data, we identify small clusters of mutations that definitively separate lab strains. Additionally, an analysis of gene knockout strains suggests that roughly one in three strains currently in use harbors second-site mutations of potential phenotypic impact. Overall, we find that divergence among lab strains is thus far minimal, though as the archaeal research community continues to grow, careful strain provenance and genomic re-sequencing are required to keep inter-lab divergence to a minimum, prevent the compounding of mutations into fully independent lineages, and maintain the current high degree of reproducible research between lab groups. IMPORTANCE Archaea are a domain of microbial life whose member species play a critical role in the global carbon cycle, climate regulation, the human microbiome, and persistence in extreme habitats. In particular, hypersaline-adapted archaea are important, genetically tractable model organisms for studying archaeal genetics, genomics, and biochemistry. As the archaeal research community grows, keeping track of the genetic integrity of strains of interest is necessary. In particular, routine genetic manipulations and the common practice of sharing strains between labs allow mutations to arise in lab stocks. If these mutations affect cellular processes, they may jeopardize the reproducibility of work between research groups and confound the results of future studies. In this work, we examine DNA sequences from 60 strains across two species of archaea. We identify shared and unique mutations occurring between and within strains. Independently, we trace the lineage of each strain, identifying which genetic manipulations lead to observed off-target mutations. While overall divergence across labs is minimal so far, our work highlights the need for labs to continue proper strain husbandry.

The Min system is a key spatial regulator of cell division in rod-shaped bacteria and the first FtsZ negative modulator to be recognized. Nevertheless, despite extensive genetic and in vitro studies, the molecular mechanism used by MinC to inhibit Z-ring formation remains incompletely understood. The crystallization of FtsZ in complex with other negative regulators such as SulA and MciZ has provided important structural information to corroborate in vitro experiments and establish the mechanism of Z-ring antagonism by these modulators. However, MinC and FtsZ have so far eluded co-crystallization, probably because their complex is too unstable to be crystallized. To gain structural insight into the mechanism of action of MinC, we determined the solution structure of the N-terminal domain of B. subtilis MinC, and through NMR titration experiments and mutagenesis identified the binding interfaces involved in the MinC N -FtsZ interaction. By using our experimental results as restraints in docking, we also constructed a molecular model for the FtsZ:MinC N complex and validated it by molecular dynamics. The model shows that MinC N binding overlaps with the FtsZ polymerization interface on the C-terminal globular subdomain of FtsZ and, thus, provides a structural basis for MinC N inhibition of FtsZ filament formation. Given that the C-terminal polymerization interface of FtsZ corresponds to the plus end of FtsZ filaments, we propose that capping is the main mechanism employed by MinC to antagonize FtsZ polymerization.

Archaea play indispensable roles in global biogeochemical cycles, yet many crucial cellular processes, including cell-shape determination, are poorly understood. Haloferax volcanii , a model haloarchaeon, forms rods and disks, depending on growth conditions. Here, we used a combination of iterative proteomics, genetics, and live-cell imaging to identify mutants that only form rods or disks. We compared the proteomes of the mutants with wild-type cells across growth phases, thereby distinguishing between protein abundance changes specific to cell shape and those related to growth phases. The results identified a diverse set of proteins, including predicted transporters, transducers, signaling components, and transcriptional regulators, as important for cell-shape determination. Through phenotypic characterization of deletion strains, we established that rod-determining factor A (RdfA) and disk-determining factor A (DdfA) are required for the formation of rods and disks, respectively. We also identified structural proteins, including an actin homolog that plays a role in disk-shape morphogenesis, which we named volactin. Using live-cell imaging, we determined volactin’s cellular localization and showed its dynamic polymerization and depolymerization. Our results provide insights into archaeal cell-shape determination, with possible implications for understanding the evolution of cell morphology regulation across domains.

Because archaea are the evolutionary ancestors of eukaryotes, archaeal molecular biology has been a topic of intense recent research. The hypersaline adapted archaeal species Halobacterium salinarum and Haloferax volcanii serve as important model organisms because facile tools enable genetic manipulation. As a result, the number of strains in circulation among the haloarchaeal research community has increased over the last few decades. However, the degree of genetic divergence and effects on genetic integrity during inter-lab transfers remain unclear. To address this question, we performed whole genome re-sequencing on a cross-section of wild-type, parental, and knockout strains in both model species. Integrating these data with existing repositories of re-sequencing data, we identify mutations that have arisen in a collection of 60 strains, sampled from 2 species across 8 different labs. Independent of sequencing, we construct strain lineages, identifying branch points and significant genetic effects in strain history. Combining this with our sequencing data, we identify small clusters of mutations that definitively separate lab strains. Additionally, an analysis of gene knockout strains suggests that roughly 1 in 3 strains currently in use harbors second-site mutations of potential phenotypic impact. Overall, we find that divergence among lab strains is thus far minimal, though as the archaeal research community continues to grow, careful strain provenance and genomic re-sequencing are required to keep inter-lab divergence to a minimum, prevent the compounding of mutations into fully independent lineages, and maintain the current high degree of reproducible research between lab groups in the haloarchaeal research community. Data Summary Novel sequencing data for this project was submitted to the National Center for Biotechnology Information (NCBI) Sequence Read Archive (SRA) and can be found under bioproject accession PRJNA1120443. SRA accessions for previously published sequencing data are available in supplementary table 1. R code for performing analysis and generating figures is available at [https://github.com/andrew-soborowski/halophile\_genome\_resequencing][1]. Impact Statement Archaea are important due to their shared evolutionary history with eukaryotes. As the archaeal research community grows, keeping track of the genetic integrity of archaeal strains of interest is necessary. In particular, routine genetic manipulations and the common practice of sharing strains between labs allow mutations to arise in lab stocks. If these mutations affect cellular processes, they may jeopardize the reproducibility of work between research groups and confound the results of future studies. In this work, we examine DNA sequences from 60 strains across two species of archaea. We identify shared and unique mutations occurring between and within strains. Independently, we trace the lineage of each strain, identifying which genetic manipulations lead to observed off-target mutations. While overall divergence across labs is minimal so far, our work highlights the need for labs to continue proper strain husbandry. ### Competing Interest Statement The authors have declared no competing interest. [1]: https://github.com/andrew-soborowski/halophile_genome_resequencing

ABSTRACT Across the domains of life, actin homologs are integral components of many essential processes, such as DNA segregation, cell division, and cell shape determination. Archaeal genomes, like those of bacteria and eukaryotes, also encode actin homologs, but much less is known about these proteins’ in vivo dynamics and cellular functions. We identified and characterized the function and dynamics of Salactin, an actin homolog in the hypersaline archaeon Halobacterium salinarum . Live-cell time-lapse imaging revealed that Salactin forms dynamically unstable filaments that grow and shrink out of the cell poles. Like other dynamically unstable polymers, Salactin monomers are added at the growing filament end, and its ATP-bound critical concentration is substantially lower than the ADP-bound form. When H. salinarum’s chromosomal copy number becomes limiting under low-phosphate growth conditions, cells lacking Salactin show perturbed DNA distributions. Taken together, we propose that Salactin is part of a previously unknown chromosomal segregation apparatus required during low-ploidy conditions. IMPORTANCE Protein filaments play important roles in many biological processes. We discovered an actin homolog in halophilic archaea, which we call Salactin. Just like the filaments that segregate DNA in eukaryotes, Salactin grows out of the cell poles towards the middle, and then quickly depolymerizes, a behavior known as dynamic instability. Furthermore, we see that Salactin affects the distribution of DNA in daughter cells when cells are grown in low-phosphate media, suggesting Salactin filaments might be involved in segregating DNA when the cell has only a few copies of the chromosome.

Faculty at research institutions play a central role in advancing knowledge and careers, as well as promoting the well-being of students and colleagues in research environments. Mentorship from experienced peers has been touted as critical for enabling these myriad roles to allow faculty development, career progression, and satisfaction. However, there is little information available on who supports faculty and best ways to structure a faculty mentorship programme for early- and mid-career academics. In the interest of advocating for increased and enhanced faculty mentoring and mentoring programmes, we surveyed faculty around the world to gather data on whether and how they receive mentoring. We received responses from 457 early- and mid-career faculty and found that a substantial portion of respondents either reported having no mentor or a lack of a formal mentoring scheme. Qualitative responses on the quality of mentorship revealed that the most common complaints regarding mentorship included lack of mentor availability, unsatisfactory commitment to mentorship, and non-specific or non-actionable advice. On these suggestions, we identify a need for training for faculty mentors as well as strategies for individual mentors, departments, and institutions for funding and design of more intentional and supportive mentorship programmes for early- and mid-career faculty.

Mentorship is vital for early career researchers in training positions, allowing them to navigate the challenges of work and life in research environments. However, the quality of mentorship received by trainees can vary by investigator and by institution. One challenge faced by those hoping to improve trainee mentorship is that the extent to which mentorship is offered to and experienced by research trainees is not well characterized. To address this knowledge gap, we conducted a survey to examine the quality of mentorship received by trainees in research environments, to identify characteristics of positive and negative mentorship, and to highlight best practices to improve trainee mentorship. We received 2,114 responses from researchers at graduate and postdoctoral career stages worldwide. Quantitative analysis showed that at least ∼25-45% of respondents were dissatisfied with some aspects of their mentorship. Qualitative responses revealed that common issues in mentorship include unclear expectations in research and mentoring interactions, lack of guidance, and inadequate support of trainee independence and career goals. Our findings also identified key mentorship elements desired by trainee mentees. Based on trainee suggestions, we describe strategies for individual mentors, departments, and institutions to improve the training experience for graduate and postdoctoral researchers.

Helicobacter pylori is a gram-negative bacterium commonly found in the human stomach. Its most famous feature is its helical shape, which is essential for invading the mucoid tissue and establishing infection that can cause gastrointestinal cancer. This study explored the regulatory role of a bactofilin, a cytoskeletal protein present in many bacteria and known to control cell shape and the assembly of subcellular structures. In vitro and in vivo experiments showed that the H. pylori bactofilin CcmA controls the curvature of cells by interacting with factors that build up or degrade the cell wall. The new mechanisms of CcmA self-inhibition and binding to specific cell shape factors represents a breakthrough in how we understand the role of bactofilins. It also suggests bactofilins as a possible therapeutic target against a broad range of pathogenic spiral-shaped bacteria.

All living systems evolved molecular mechanisms to respond to their physical surroundings, such as by tuning cell cycle progression as well as transport of different cargo across intracellular compartments. However, it is difficult to determine if such responses are genetically programmed or purely resultant of physical forces. This study deploys a new confinement device that modulates chamber stiffness, curvature, and compression over epithelial tissue. Compression specifically regulated cell cycle progression, while increasing stiffness changed cell aspect ratio. The authors propose a new independent function of previous cooperative pathways that regulate long-range tissue scales (β-catenin) and local spatial changes (YAP/TAZ). This work represents a methodological advance in creating specific physical cues while measuring cellular processes, enabling many future studies combining other chemical and genetic perturbations..

Precise control of the cell cycle is central to the physiology of all cells. In prior work we demonstrated that archaeal cells maintain a constant size; however, the regulatory mechanisms underlying the cell cycle remain unexplored in this domain of life. Here we use genetics, functional genomics, and quantitative imaging to identify and characterize the novel CdrSL gene regulatory network in a model species of archaea. We demonstrate the central role of these ribbon-helix-helix family transcription factors in the regulation of cell division through specific transcriptional control of the gene encoding FtsZ2, a putative tubulin homolog. Using time lapse fluorescence microscopy in live cells cultivated in microfluidics devices, we further demonstrate that FtsZ2 is required for cell division but not elongation. The cdrS-ftsZ2 locus is highly conserved throughout the archaeal domain, and the central function of CdrS in regulating cell division is conserved across hypersaline adapted archaea. We propose that the CdrSL-FtsZ2 transcriptional network coordinates cell division timing with cell growth in archaea. Importance Healthy cell growth and division are critical for individual organism survival and species long-term viability. However, it remains unknown how cells of the domain Archaea maintain a healthy cell cycle. Understanding archaeal cell cycle is of paramount evolutionary importance given that an archaeal cell was the host of the endosymbiotic event that gave rise to eukaryotes. Here we identify and characterize novel molecular players needed for regulating cell division in archaea. These molecules dictate the timing of cell septation, but are dispensable for growth between divisions. Timing is accomplished through transcriptional control of the cell division ring. Our results shed light on mechanisms underlying the archaeal cell cycle, which has thus far remained elusive.

Precise control of the cell cycle is central to the physiology of all cells. In prior work we demonstrated that archaeal cells maintain a constant size; however, the regulatory mechanisms underlying the cell cycle remain unexplored in this domain of life. Here, we use genetics, functional genomics, and quantitative imaging to identify and characterize the novel CdrSL gene regulatory network in a model species of archaea. We demonstrate the central role of these ribbon-helix-helix family transcription factors in the regulation of cell division through specific transcriptional control of the gene encoding FtsZ2, a putative tubulin homolog. Using time-lapse fluorescence microscopy in live cells cultivated in microfluidics devices, we further demonstrate that FtsZ2 is required for cell division but not elongation. The cdrS-ftsZ2 locus is highly conserved throughout the archaeal domain, and the central function of CdrS in regulating cell division is conserved across hypersaline adapted archaea. We propose that the CdrSL-FtsZ2 transcriptional network coordinates cell division timing with cell growth in archaea. IMPORTANCE Healthy cell growth and division are critical for individual organism survival and species long-term viability. However, it remains unknown how cells of the domain Archaea maintain a healthy cell cycle. Understanding the archaeal cell cycle is of paramount evolutionary importance given that an archaeal cell was the host of the endosymbiotic event that gave rise to eukaryotes. Here, we identify and characterize novel molecular players needed for regulating cell division in archaea. These molecules dictate the timing of cell septation but are dispensable for growth between divisions. Timing is accomplished through transcriptional control of the cell division ring. Our results shed light on mechanisms underlying the archaeal cell cycle, which has thus far remained elusive.

Supplementary text, figures, and tables for The ribbon-helix-helix domain proteins CdrS and CdrL regulate cell division in archaea. Cynthia L. Darnell, Jenny Zheng, Sean Wilson, Ryan M. Bertoli, Alexandre W. Bisson-Filho, Ethan C. Garner, Amy K. Schmid. Submitted to mBio

Many postdoctoral researchers apply for faculty positions knowing relatively little about the hiring process or what is needed to secure a job offer. To address this lack of knowledge about the hiring process we conducted a survey of applicants for faculty positions: the survey ran between May 2018 and May 2019, and received 317 responses. We analyzed the responses to explore the interplay between various scholarly metrics and hiring outcomes. We concluded that, above a certain threshold, the benchmarks traditionally used to measure research success – including funding, number of publications or journals published in – were unable to completely differentiate applicants with and without job offers. Respondents also reported that the hiring process was unnecessarily stressful, time-consuming, and lacking in feedback, irrespective of outcome. Our findings suggest that there is considerable scope to improve the transparency of the hiring process.

This first molecular biological study of archaeal immersed liquid biofilms advances our basic biological understanding of the model archaeon Haloferax volcanii . Data gleaned from this study also provide an invaluable foundation for future studies to uncover components required for immersed liquid biofilms in this haloarchaeon and also potentially for liquid biofilm formation in general, which is poorly understood compared to the formation of biofilms on surfaces.

One mechanism for achieving accurate placement of the cell division machinery is via Turing patterns, where nonlinear molecular interactions spontaneously produce spatiotemporal concentration gradients. The resulting patterns are dictated by cell shape. For example, the Min system of Escherichia coli shows spatiotemporal oscillation between cell poles, leaving a mid-cell zone for division. The universality of pattern-forming mechanisms in divisome placement is currently unclear. We examined the location of the division plane in two pleomorphic archaea, Haloferax volcanii and Haloarcula japonica, and showed that it correlates with the predictions of Turing patterning. Time-lapse analysis of H. volcanii shows that divisome locations after successive rounds of division are dynamically determined by daughter cell shape. For H. volcanii, we show that the location of DNA does not influence division plane location, ruling out nucleoid occlusion. Triangular cells provide a stringent test for Turing patterning, where there is a bifurcation in division plane orientation. For the two archaea examined, most triangular cells divide as predicted by a Turing mechanism; however, in some cases multiple division planes are observed resulting in cells dividing into three viable progeny. Our results suggest that the division site placement is consistent with a Turing patterning system in these archaea.

Since the identification of the archaeal domain in the mid-1970s, we have collected a great deal of metagenomic, biochemical, and structural information from archaeal species. However, there is still little known about how archaeal cells organize their internal cellular components in space and time. In contrast, live-cell imaging has allowed bacterial and eukaryotic cell biologists to learn a lot about biological processes by observing the motions of cells, the dynamics of their internal organelles, and even the motions of single molecules. The explosion of knowledge gained via live-cell imaging in prokaryotes and eukaryotes has motivated an ever-improving set of imaging technologies that could allow analogous explorations into archaeal biology. Furthermore, previous studies of essential biological processes in prokaryotic and eukaryotic organisms give methodological roadmaps for the investigation of similar processes in archaea. In this perspective, we highlight a few fundamental cellular processes in archaea, reviewing our current state of understanding about each, and compare how imaging approaches helped to advance the study of similar processes in bacteria and eukaryotes.

MreB is essential for rod shape in many bacteria. Membrane-associated MreB filaments move around the rod circumference, helping to insert cell wall in the radial direction to reinforce rod shape. To understand how oriented MreB motion arises, we altered the shape of Bacillus subtilis. MreB motion is isotropic in round cells, and orientation is restored when rod shape is externally imposed. Stationary filaments orient within protoplasts, and purified MreB tubulates liposomes in vitro, orienting within tubes. Together, this demonstrates MreB orients along the greatest principal membrane curvature, a conclusion supported with biophysical modeling. We observed that spherical cells regenerate into rods in a local, self-reinforcing manner: rapidly propagating rods emerge from small bulges, exhibiting oriented MreB motion. We propose that the coupling of MreB filament alignment to shape-reinforcing peptidoglycan synthesis creates a locally-acting, self-organizing mechanism allowing the rapid establishment and stable maintenance of emergent rod shape.

Article Figures and data Abstract eLife digest Introduction Results Discussion Materials and methods Appendix 1 Appendix 2 References Decision letter Author response Article and author information Metrics Abstract MreB is essential for rod shape in many bacteria. Membrane-associated MreB filaments move around the rod circumference, helping to insert cell wall in the radial direction to reinforce rod shape. To understand how oriented MreB motion arises, we altered the shape of Bacillus subtilis. MreB motion is isotropic in round cells, and orientation is restored when rod shape is externally imposed. Stationary filaments orient within protoplasts, and purified MreB tubulates liposomes in vitro, orienting within tubes. Together, this demonstrates MreB orients along the greatest principal membrane curvature, a conclusion supported with biophysical modeling. We observed that spherical cells regenerate into rods in a local, self-reinforcing manner: rapidly propagating rods emerge from small bulges, exhibiting oriented MreB motion. We propose that the coupling of MreB filament alignment to shape-reinforcing peptidoglycan synthesis creates a locally-acting, self-organizing mechanism allowing the rapid establishment and stable maintenance of emergent rod shape. https://doi.org/10.7554/eLife.32471.001 eLife digest Many bacteria are surrounded by both a cell membrane and a cell wall – a rigid outer covering made of sugars and short protein chains. The cell wall often determines which of a variety of shapes – such as rods or spheres – the bacteria grow into. One protein required to form the rod shape is called MreB. This protein forms filaments that bind to the bacteria’s cell membrane and associate with the enzymes that build the cell wall. Together, these filament-enzyme complexes rotate around the cell to build and reinforce the cell wall in a hoop-like manner. But how do the MreB filaments know how to move around the circumference of the rod, instead of moving in any other direction? Using a technique called total internal reflection microscopy to study how MreB filaments move across bacteria cells, Hussain, Wivagg et al. show that the filaments sense the shape of a bacterium by orienting along the direction of greatest curvature. As a result, the filaments in rod-shaped cells orient and move around the rod, while in spherical bacteria they move in all directions. However, spherical bacteria can regenerate into rods from small surface ‘bulges’. The MreB filaments in the bulges move in an oriented way, helping them to generate the rod shape. Hussain, Wivagg et al. also found that forcing cells that lack a cell wall into a rod shape caused the MreB filaments bound to the cell membrane to orient and circle around the rod. This shows that the organization of the filaments is sufficient to shape the cell wall. In the future, determining what factors control the activity of the MreB filaments and the enzymes they associate with might reveal new targets for antibiotics that disrupt the cell wall and so kill the bacteria. This will require higher resolution microscopes to be used to examine the cell wall in more detail. The activity of all the proteins involved in building cell walls will also need to be extensively characterized. https://doi.org/10.7554/eLife.32471.002 Introduction Although many bacteria are rod shaped, the cellular mechanisms that construct and replicate this geometry have remained largely unknown. Bacterial shape is determined by the cell wall sacculus, a giant, encapsulating macromolecule that serves to resist internal osmotic pressure. One of the primary components of the cell wall is peptidoglycan (PG), which is created by the polymerization of single glycan strands linked by peptide crossbridges. Studies of isolated cell walls from rod-shaped bacteria suggest material is generally oriented circumferentially around the rod, perpendicular to the long axis of the cell (Gan et al., 2008; Hayhurst et al., 2008; Verwer et al., 1980), or in thick cables in others (Hayhurst et al., 2008). This mostly circumferential, hoop-like organization of cell wall material allows the cell wall to better resist the internal osmotic pressure, as this pressure causes a stress twice as large in the circumferential direction (on the rod sidewalls) than in the axial direction (on the poles) (Amir and Nelson, 2012; Chang and Huang, 2014). This organization of material confers a mechanical anisotropy to the cell wall, causing it to stretch more along its length than across its width for a given stress; this anisotropy may assist rod-shaped cells in preferentially elongating along their length (Baskin, 2005; Chang and Huang, 2014). Concordantly, atomic force microscopy (AFM) has shown that Escherichia coli sacculi are 2–3 times more elastic along their length than across their width (Yao et al., 1999). This rod-reinforcing circumferential organization is also observed in the cell walls of plants; hypocotyl and root axis cells rapidly elongate as rods by depositing cellulose fibrils in circumferential bands around their width, resulting not only in a similar dispersive rod-like growth, but also a similar anisotropic response to stress (Baskin, 2005). The organized deposition of cellulose arises from cortical microtubules self-organizing into a radial array oriented around the rod width, and this orients the directional motions of the cellulose synthases to insert material in circumferential bands (Paredez et al., 2006). In contrast to our understanding of the self-organization underlying rod-shaped growth in plants, how bacteria construct a circumferential organization of glycan strands is not known. This organization may arise via the actions of a small number of genes essential for the formation and maintenance of rod shape. Collectively termed the Rod complex, (or elongasome) these include MreB, MreC, MreD (encoded by the mreBCD operon) (Wachi et al., 1989), RodZ (Alyahya et al., 2009; Bendezú et al., 2009), and the glycosyltransferase/transpeptidase enzyme pair RodA/Pbp2 (Cho et al., 2016). These components are conserved across a wide range of rod shaped bacteria, and mostly absent in cocci (Alyahya et al., 2009; Chastanet and Carballido-Lopez, 2012), leading many to speculate they function as the central determinants of rod shape (Carballido-Lopez, 2006; Jones et al., 2001). The spatial coordination of RodA/Pbp2-mediated PG synthesis is conferred by MreB, an actin homolog (Jones et al., 2001; van den Ent et al., 2001). MreB polymerizes onto membranes as antiparallel double filaments, which have been observed to bend liposome membranes inward (Figure 1A) (Salje et al., 2011; van den Ent et al., 2014). Loss or depolymerization of MreB causes rod-shaped cells to grow as spheres (Gitai et al., 2005; Jones et al., 2001; Bendezú et al., 2009). B. subtilis contains 3 MreB paralogs (MreB, Mbl, and MreBH) that have been shown to co-polymerize into mixed filaments in vitro, and always colocalize in vivo (Defeu Soufo and Graumann, 2004; Soufo and Graumann, 2010; Dempwolff et al., 2011). Figure 1 with 3 supplements see all Download asset Open asset Curved MreB filament motions do not follow an ordered template (A–C). (A) The negative curvature of MreB filaments (arrowheads) aligns with the negative principal curvature of the liposome surface (arrow). Scale bar is 50 nm. (B) Angular distribution of GFP-MreB and GFP-Mbl trajectories relative to the long axis of B subtilis cells indicates that while the distributions have a mode of 90°, they are broad (Mean Deviation = 34°, n = 1041 for GFP-MreB and Mean Deviation = 26°, n = 1772 for GFP-Mbl). (C) Particle tracking of GFP-Mbl (left) and Halo-JF549-MreB (right) during 120 s (~1 rotation) indicates trajectories close in time frequently cross paths (white arrows). Scale bar is 1 μm. See corresponding Figure 1—video 1. (D) Strains with tagO under inducible control display a teichoic acid-dependent decrease in width. Strain BEG300 was grown in LB supplemented with 20 mM Mg2+. (E) Plot of cell width as a function of tagO induction in LB supplemented with 20 mM Mg2+, calculated from rod-shaped cells (error bars are Standard Error of the Mean (SEM), n = 33, 56, 104, 175, 228, 489, 119). Areas not plotted at lower xylose levels (red dashed rectangle) are regions where cells are round (no width axis). Color scheme for D-F: red indicates round cells (no width axis), blue indicates rods (measurable width axis), and green indicates intermediate regimes where both rods and round cells are observed. (F) BEG300 at an intermediate level of tagO induction (15 mM xylose) shows a Mg2+ dependent decrease in width. All scale bars are 5 μm. See also Figure 1—figure supplement 1. https://doi.org/10.7554/eLife.32471.003 Figure 1—source data 1 Figure 1B – Raw GFP-MreB and GFP-Mbl track angle values from the cell midline and their associated frequency distributions. Figure 1E – Mean cell width, Standard Deviation (SD) and number of cells (n) analyzed at various xylose concentrations to create the graph in Figure 1E. https://doi.org/10.7554/eLife.32471.006 Download elife-32471-fig1-data1-v3.xlsx Figure 1—source data 2 Figure 1—figure supplement 1B-left – Mean cell width, Standard Deviation (SD) and number of cells (N) analyzed at various magnesium and xylose concentrations. Figure 1—figure supplement 1B-right – Tangential correlation along the cell contours at various xylose and magnesium concentrations. This data was used to calculate the correlation curves in Figure 1—figure supplement 1B. https://doi.org/10.7554/eLife.32471.007 Download elife-32471-fig1-data2-v3.xlsx MreB filaments move circumferentially around the width of rod-shaped cells (Domínguez-Escobar et al., 2011; Garner et al., 2011; van Teeffelen et al., 2011).Super-resolution imaging has demonstrated that MreB filaments always translocate along their length, moving in the direction of their orientation (Olshausen et al., 2013). MreB filaments move in concert with MreC, MreD, and RodA/Pbp2 (Domínguez-Escobar et al., 2011; Garner et al., 2011), and loss of any one component stops the motion of the others. The directional motion of MreB filaments and associated Rod complexes depends on, and thus likely reflects, the insertion of new cell wall, as this motion halts upon the addition of cell wall synthesis-inhibiting antibiotics (Domínguez-Escobar et al., 2011; Garner et al., 2011; van Teeffelen et al., 2011), or specific inactivation or depletion of Pbp2 (Garner et al., 2011; van Teeffelen et al., 2011) or RodA (Cho et al., 2016). It is not known how MreB and the rest of the Rod complex construct rod-shaped cells. As the motions of the Rod complexes reflect the insertion of new cell wall, their circumferential motions could deposit glycans in the hoop-like organization required to both build and reinforce rod shape. Therefore, we worked to understand the origin of this circumferential organization, seeking to determine what orients the motions of MreB and associated enzymes around the rod width in Bacillus subtilis. Results Oriented MreB motion is unlikely to arise from an ordered cell wall template The mechanism by which MreB filaments and associated PG synthases orient their motion around the rod circumference is not known. Each filament-synthase complex is disconnected from the others, moving independently of proximal neighbors (Garner et al., 2011). The organized, circumferential motion of these independent filament-synthase complexes could arise in two ways: (1) A templated organization, where cell wall synthetic complexes move along an existing pattern of ordered glycan strands in the cell wall as they insert new material into it (Höltje, 1998), or (2) A template-independent organization, where each synthetic complex has an intrinsic mechanism that orients its motion and resultant PG synthesis around the rod circumference. To explore the extent of order within the motions of the Rod complex, we analyzed the trajectories of GFP-Mbl and GFP-MreB with respect to the cell body using total internal reflection fluorescence microscopy (TIRFM) (Figure 1B). Overall, these motions are circumferentially oriented, but not perfectly aligned, a characteristic reflected by the broad distribution of angles that GFP-MreB, its homologs, and the other components of the Rod complex move relative to the long axis of the cell (Domínguez-Escobar et al., 2011; Garner et al., 2011). However, examination of TIRFM time lapse movies revealed that both MreB and Mbl trajectories close in time (within the period of one revolution) frequently cross (Figure 1C, Figure 1—video 1), making it unlikely that MreB filaments move along a perfectly ordered template. As MreB movement reflects the insertion of new glycan strands, these motions indicate that the siacculus is built from somewhat disorganized, yet predominantly circumferential strands. This conclusion is in agreement with X-ray diffraction (Balyuzi et al., 1972; Labischinski et al., 1979) and cryoelectron microscopy studies of E. colisacculi (Gan et al., 2008) which found that, while glycans are oriented circumferentially around the rod width on average, they are not ordered, running at variable angles to each other. Similarly, atomic force microscopy of B. subtilis sacculi has observed a generally oriented, but unaligned arrangement of 50 nm thick cables oriented roughly perpendicular to the long axis (Hayhurst et al., 2008). Thus, both the motions of MreB and ultrastructural studies indicate the sacculus is not highly ordered, making it unlikely that it can serve as a self-propagating spatial template for rod shape. Furthermore, given that preexisting cell wall is not necessary for the regeneration of rod shape from wall-less B. subtilis L-forms (Kawai et al., 2014), it is likely that both oriented MreB motion and rod shape can arise without an ordered template. MreB motions become isotropic in the absence of rod shape As it appeared that organized MreB motion does not arise from patterns in the cell wall, we hypothesized there was an intrinsic mechanism orienting the motion of each MreB filament-cell wall synthetic complex. To test this hypothesis, we examined MreB motions as we changed the shape of cells from rods to spheres. As the internal osmotic pressure and stiffness of B. subtilis resists external mechanical perturbations to its shape (Renner et al., 2013), we first altered the shape of cells by controlling the level of wall teichoic acids (WTAs). WTAs are negatively charged cell wall polymers believed to increase the rigidity of the sacculus (Matias and Beveridge, 2005), a process that could occur via their coordination of extracellular Mg2+ (Thomas and Rice, 2014; Kern et al., 2010), or modulation of hydrolase activity (Atilano et al., 2010). Knockouts of tagO, the first gene in the WTA synthesis pathway, create large, slow-growing, round cells that still synthesize PG, building extremely thick and irregular cell walls (D'Elia et al., 2006). We placed tagO under xylose-inducible control and grew cells at different induction levels. As expected, at high TagO inductions, cells displayed normal widths. As we reduced TagO levels, rods became gradually wider (Figure 1D–E) until, beneath a given induction, cells were no longer able to maintain rod shape, growing as spheres (or clumps of spheres) with no identifiable long axis. At intermediate induction levels, we observed a transition region between the two states, with cells growing as steady state populations of interconnected rods and spheres (Figure 1D). In agreement with models that (A) WTAs work with PG to bind Mg2+ (Thomas and Rice, 2014; Kern et al., 2010), and (B) are required for cell wall rigidity (Matias and Beveridge, 2005), both the cell width and the amount of TagO induction determining the rod/sphere transition could be modulated by Mg2+ levels (Figure 1F, Figure 1—figure supplement 1B). Likewise, decreasing extracellular Mg2+ or tagO induction resulted in increasingly curved cell contours (Figure 1—figure supplement 1B), suggesting the wall was becoming more flexible. By tracking the motion of GFP-MreB filaments in these differing cell shapes, we found that motion is always oriented in rods, moving predominantly circumferentially at all induction levels above the rod/sphere transition. However, in round cells (those induced beneath the rod/sphere transition point or in tagO knockouts) MreB filaments continued to move directionally, but their motions were isotropic, moving in all directions (Figure 2A, Movie Figure 2—video 1). To quantify the relative alignment of MreB under each condition, we calculated the angle between trajectory pairs less than 1 μm apart (Figure 2B, Figure 2—figure supplement 1A). This analysis revealed that MreB motions are more aligned when cells are rods: above the rod/sphere transition, trajectories have a median angle difference of 26°; while at low TagO inductions, where cells are round, the angle difference increases to 42°, close to that of randomly oriented trajectories (45°). Figure 2 with 3 supplements see all Download asset Open asset Oriented MreB motion correlates with rod shape. (A) BEG300 at maximum tagO induction (30 mM) is rod-shaped, and MreB tracks are largely oriented perpendicular to the midline of the cell (left). ∆tagO cells show round morphologies with unaligned MreB motion (right). (B) Median inter-track angle difference for track pairs ≤ 1 μm apart, plotted for BEG300 at several tagO induction levels, ∆tagO cells, and a simulation of randomly oriented angles (sim). n > 1300 for all data points. For spherical cells width is not measurable, indicated with a dashed red line. (C) ∆pbpH cells with pbpA under IPTG control display aligned MreB motion when pbpA is fully induced and cells are rods (left), but display unaligned MreB motion as Pbp2a levels reduce and cells become round (right). (D) Median inter-track angle difference for track pairs 1 μm apart during Pbp2a depletion with cell widths at each time point. Error bars are SEM, n > 600 for all data points. (E) Median angle from the midline (white circles) calculated for all rod-shaped cells from experiments in 2A-D plotted as a function of cell width. MreB filament alignment falls off rapidly beyond 2 μm, a point corresponding to where cells become round, as shown by the ratio of principal curvatures (blue squares) approaching 1. Error bars are SEM, n = 2993. See Figure 2—figure supplement 1E for further explanation. (F) Schematic showing the difference between the 2D surface curvature profile of rods and spheres. On the inside surface of spheres, all points have negative, yet equal values for both principal curvatures. In rods, however, one principal curvature is negative (the radius), while the other is 0 (the flat axis along the rod). All scale bars are 1 μm. All error bars are SEM. See also Figure 2—figure supplement 1. https://doi.org/10.7554/eLife.32471.009 Figure 2—source data 1 Figure 2B – Median angle difference between track pairs at 2 mM, 6 mM, 15 mM and 30 mM xylose and in the tagO knockout, along with Standard Deviation (SD) values and the number of track pairs analyzed (N). This data was used to create Figure 2B. Figure 2D – Changes in (i) the median angle difference between track pairs and (ii) mean cell width over time during a Pbp2 depletion, along with Standard Deviation (SD) and the number of track pairs (N). This data was used to create Figure 2D. Figure 2E – Median track angles from the cell midline binned as a function of cell width and the ratio of principal curvatures k1/k2 also binned over the same cell width. This data was used to create Figure 2E. https://doi.org/10.7554/eLife.32471.011 Download elife-32471-fig2-data1-v3.xlsx Figure 2—source data 2 Figure 2—figure supplement 1A – Angular correlation between track pairs binned as a function of the distance between the pair at various xylose concentrations and in the tagO knockout. Figure 2—figure supplement 1B – Mean curvature of cell contours (curvature) and the median track angle (angle) over time during a Pbp2 depletion, along with Standard Deviation and the number of tracks used in the analysis at each time point. Figure 2—figure supplement 1C – Data from Figure 2E is split here to show the median angle from the cell midline as a function of cell width during a Pbp depletion and at various steady state TagO levels, along with Standard Deviation (SD) and the number of tracks (N). This data was used to create Figure 2—figure supplement 1C. Figure 2—figure supplement 1D – Raw track angle values of GFP-MreB from the cell midline in rod-shaped cells at various xylose levels (8, 12 and 30 mM), along with their associated frequency distributions. https://doi.org/10.7554/eLife.32471.012 Download elife-32471-fig2-data2-v3.xlsx To verify that the loss of oriented MreB motion was due to the changes in cell shape, and not from some other effect of reduced WTA levels, we created round cells by alternate means. Depletion of both elongation PG transpeptidases (Pbp2a and PbpH) causes rod-shaped cells to become wider over time as they convert to spheres (Garner et al., 2011). We used this gradual transition of rods into spheres to examine both the width and overall shape dependence of MreB motion. At initial points of depletion (1–2 hr) the rods widened but maintained circumferential MreB motion. At 2.5 hr of PbpA depletion, cells were a mix of spheres and rods of differing widths. These cells displayed the same pattern of MreB orientation observed with tagO depletions: round cells contained unoriented MreB, while nearby rod-shaped cells showed circumferential motion. Identical behavior was observed for GFP-Mbl during PbpA depletions (Figure 2C, Movie Figure 2—video 2). Quantitation of trajectories from all cells (both rods and spheres) at each time point of depletion indicated an increase in the median angle between trajectories as the population grew wider and rounder over time (Figure 2D, Figure 2—figure supplement 1B). In E. coli, the angle of mutant MreB filaments relative to the long axis has been reported to increase with cell width (Ouzounov et al., 2016). To test if the angle of MreB movement changes with respect to cell width in B. subtilis, we calculated the angle of each trajectory to the long axis for all cells in our data with an identifiable width axis. At the same time, we also measured the curvature of each cell to determine how the overall shape of the cell affected the orientation of motion. This revealed that MreB motion in rods remained equivalently oriented over a wide range of rod widths, up to ~2 μm (Figure 2E, Figure 2—figure supplement 1C–E). Beyond a 2 μm width, cells began to lose their rod shape as they became more spherical, and the predominantly circumferential orientation of MreB motion was lost (Figure 2E, Figure 2—figure supplement 1E). This suggested that oriented MreB motion does not sense or rely on a specific cell radius; rather the orientation relies on differences between the two principal curvatures of the membrane. It appears that the motion of MreB filaments is oriented along the direction of greatest principal curvature: In rods, there is zero curvature along the rod length, and high curvature around the rod circumference, along which filaments orient. In contrast, in round cells where MreB motion is isotropic, the two principal curvatures are equal (Figure 2F). MreB aligns within round cells and protoplasts forced into rod shape To further verify that MreB filaments orient in response to overall cell shape, we externally imposed rod shape on cells with unoriented MreB motion. We loaded TagO-induced cells into long 1.5 ×1.5 μm microfluidic chambers, then reduced TagO expression to levels insufficient to produce rods in liquid culture (Figure 3A, Figure 3—figure supplement 1A). After TagO depletion, cells expanded to fill the chamber indicating that WTA-depletion caused shape changes just as in bulk culture (Figure 3A, Figure 3—figure supplement 1A). Within these chambers, cells grew as rods, but at a wider width (1.5 μm) than wild-type cells, set by the chamber. When cells grew out of the chamber they swelled just as in bulk culture, showing confinement was required for rod shape at this induction level (Figure 3B, Figure 3—figure supplement 1A). In the TagO-depleted cells confined into rod shapes, MreB moved circumferentially (Figure 3C, Figure 3—video 1), confirming that MreB orients in response to the cells having rod shape. This experiment demonstrates that the isotropic MreB motion observed in round cells arises from the lack of rod shape, and not from some other effect of our genetic perturbations. This experiment also showed another unexpected result: the doubling time of free (unconstrained) cells induced at similar TagO levels is long (53 ± 10 min), but confining them into rod shape restored their doubling time (44 ± 4 min) toward wild-type times (39 ± 9 min) (Figure 3—figure supplement 1B). Figure 3 with 3 supplements see all Download asset Open asset MreB filaments orient when rod shape is induced by external confinement. (A) Phase contrast images of BEG300 grown in LB supplemented with 2 mM xylose and 20 mM Mg2+ in bulk culture (left) or confined into microfluidic channels of 1.5 ×1.5 μm (right). (B) Confined cells induced at 3 mM xylose in 20 mM Mg2+ progressively swell upon escaping confinement into free culture. See also Figure Figure 3—figure supplement 1A. (C) (Left) Fluorescence microscopy montage of MreB filaments moving across a confined cell of BEG300 induced at 2 mM xylose in 20 mM Mg2+. (Right) Maximal intensity projection of montage, kymographs of marked points and a phase contrast image of the cell. Scale bars for a-c = 5 μm. (D) Phase contrast images of protoplasts contained in agar crosses. Cells in the center grow to be round while cells in arms grow as elongated rods. (E) (left) Short GFP-Mbl filaments orient circumferentially in rod-shaped protoplasts (2 μm) but lack orientation in round protoplasts (spheres). (right) Long GFP-MreB filaments orient in rod-shaped protoplasts (2 μm); GFP-MreB filaments are still oriented in wider rod-shaped protoplasts (5 μm), but not to the same extent. In round protoplasts, GFP-MreB filaments are unoriented (spheres). Scale bar is 2 μm. (F) (i)The angular distribution of filaments within protoplasts (Protoplast filaments) is peaked at 90° (mean deviation = 34°, n = 147), similar to that of MreB motion in TagO-depleted, confined cells (Mother machine tracks) (mean deviation = 36°, n = 359) and MreB motion in wild-type cells (Wt. tracks) (mean deviation = 34°, n = 1041). (ii) In channels of varying widths (2, 4 and 5 μm), the orientation of GFP-MreB filaments remains circumferential, peaking at 90° but the filament angles deviate more from 90° as channel width increases (mean deviation = 34°, n = 258 at 2 μm), (mean meviation = 35°, n = 260 at 4 μm) and (Mean Deviation = 41°, n = 203 at 5 μm.). All mean deviation”values are calculated as the mean deviation from 90°. https://doi.org/10.7554/eLife.32471.015 Figure 3—source data 1 Figure 3Fi – Raw angle values from the cell midline of wildtype MreB tracks (tracks), MreB tracks in confined cells (mother machine) and MreB filaments in protoplasts (filaments), along with their associated frequency distributions. Figure 3Fii – Raw angle values from the cell midline of GFP-MreB filaments at varying cells widths (1–2 um, 3–4 um and 4–5 um) and GFP-Mbl filaments (mbl), along with their associated frequency distributions. https://doi.org/10.7554/eLife.32471.017 Download elife-32471-fig3-data1-v3.xlsx Figure 3—source data 2 Figure 3—figure supplement 1B – Doubling times (min) and Standard Deviation (SD) of rod-shaped and spherical cells measured by taking bulk OD600 measurements (Bulk), using single cell measurements (Single Cell), single cell measurements in cells recovering rod shape (Recovery Single), single cell measurements normalized to the cell volume (Single/Volume) and single cell measurements in spherical cells confined to rod shape (Confined). This data was used to create Figure 3—figure supplement 1B. https://doi.org/10.7554/eLife.32471.018 Download elife-32471-fig3-data2-v3.xlsx We next attempted to minimize any contribution to MreB filament orientation from (A) the directional motion of filaments, and (B) any pre-existing order within the sacculus. To accomplish this, we examined filament orientation in protoplasts (cells that had their cell wall enzymatically removed) confined into different shapes, using highly expressed GFP-MreB to assay long filaments, and GFP-Mbl to assay short filaments. We protoplasted cells in osmotically stabilized media (Wyrick and Rogers, 1973), then grew them under agar pads containing micro-patterned cross shapes. Cells in the center of these crosses (~5 μm diameter) were forced to grow as spheres, whereas cells in the arms were constrained to grow into rods of various widths ranging from 2 to 5 μm (Figure 3D). As reported previously (Domínguez-Escobar et al., 2011), MreB filaments within protoplasts did not move directionally (Figure 4—video 1), likely because the cell wall provides the fixed surface along which the PG synthesis enzymes move. Within the protoplasts confi