Multiple ParA/MinD ATPases coordinate the positioning of disparate cargos in a bacterial cell

Pulianmackal, Lisa T.; Limcaoco, Jose Miguel I.; Ravi, Keerthikka; Yang, Sinyu; Zhang, Jeffrey; Tran, Mimi K.; Ghalmi, Maria; O’Meara, Matthew J.; Vecchiarelli, Anthony G.

doi:10.1038/s41467-023-39019-x

Download PDF

Article
Open access
Published: 05 June 2023

Multiple ParA/MinD ATPases coordinate the positioning of disparate cargos in a bacterial cell

Nature Communications volume 14, Article number: 3255 (2023) Cite this article

3481 Accesses
1 Citations
105 Altmetric
Metrics details

Subjects

Abstract

In eukaryotes, linear motor proteins govern intracellular transport and organization. In bacteria, where linear motors involved in spatial regulation are absent, the ParA/MinD family of ATPases organize an array of genetic- and protein-based cellular cargos. The positioning of these cargos has been independently investigated to varying degrees in several bacterial species. However, it remains unclear how multiple ParA/MinD ATPases can coordinate the positioning of diverse cargos in the same cell. Here, we find that over a third of sequenced bacterial genomes encode multiple ParA/MinD ATPases. We identify an organism (Halothiobacillus neapolitanus) with seven ParA/MinD ATPases, demonstrate that five of these are each dedicated to the spatial regulation of a single cellular cargo, and define potential specificity determinants for each system. Furthermore, we show how these positioning reactions can influence each other, stressing the importance of understanding how organelle trafficking, chromosome segregation, and cell division are coordinated in bacterial cells. Together, our data show how multiple ParA/MinD ATPases coexist and function to position a diverse set of fundamental cargos in the same bacterial cell.

CryoEM structures reveal how the bacterial flagellum rotates and switches direction

Article Open access 17 April 2024

Pooled multicolour tagging for visualizing subcellular protein dynamics

Article Open access 19 April 2024

Bridging structural and cell biology with cryo-electron microscopy

Article 03 April 2024

Introduction

Actin filaments, microtubules, and the linear motor proteins that walk along them, are well known for spatial organization in eukaryotic cells. In bacteria, however, where linear motors involved in positioning are absent, a widespread family of ParA/MinD (A/D) ATPases spatially organize plasmids, chromosomes, and an array of protein-based organelles, many of which are fundamental to cell survival and pathogenesis. By far the two best studied ATPases, and family namesake, are ParA involved in plasmid partition and chromosome segregation^1,2, and MinD involved in divisome positioning³. Less studied is the growing list of A/D ATPases, widespread across prokaryotes, involved in spatially regulating diverse protein-based organelles, such as Bacterial Microcompartments (BMCs)^4,5, flagella^6,7, chemotaxis clusters^8,9, and conjugation machinery¹⁰.

Despite the cargos being so diverse, A/D ATPases share a number of features: (i) all form ATP-sandwich dimers¹¹, (ii) dimerization forms an interface for binding a positioning matrix—the nucleoid for ParA-like ATPases^12,13 or the inner membrane for MinD-like ATPases^14,15, and (iii) dimerization also forms a binding site for a cognate partner protein that connects an ATPase to its cargo and stimulates its release from the positioning matrix. For example, in chromosome segregation, the ParA partner is ParB, which loads onto a centromere-like site, called parS, to form a massive complex on the chromosome near the origin of replication (OriC)². This ParB-parS complex locally stimulates ParA ATPase activity and nucleoid release, which generates ParA gradients on the nucleoid. Segregation of sister chromosomes ensues as the ParB-parS complex chases nucleoid-bound ParA gradients in opposite directions¹⁶. Therefore, unlike the mitotic-spindle apparatus used in eukaryotic chromosome segregation, prokaryotes use a fundamentally different mode of spatial organization—A/D ATPases make waves on biological surfaces to position their respective cargos.

Chromosome segregation, cell division positioning, and organelle trafficking reactions have been independently investigated to varying degrees in several prokaryotes. Yet, it remains unknown how many A/D ATPases can be encoded in a single bacterium to position multiple disparate cargos, or how bacteria spatiotemporally coordinate the positioning of such a diverse set of fundamental cargos in the same cell. Additionally, the mechanistic variations and specificity determinants that govern the positioning of such a diverse set of cellular cargo remain unclear. This is because A/D-based positioning reactions are typically studied independently of one another and in model bacteria with few A/D ATPases.

Here, we find that a third of sequenced bacteria encode multiple A/D ATPases. Among these bacteria, we identified Halothiobacillus neapolitanus (H. neapolitanus hereafter), with seven putative A/D ATPases. Neighborhood analysis of the A/D ATPase genes in H. neapolitanus implicate several putative cargos. We use genetics and cell biology to assign five of the A/D ATPases in H. neapolitanus to their cargos. Our findings show that each ATPase is directly dedicated to the positioning and faithful inheritance of a specific cargo type. The tractability and number of A/D ATPases made H. neapolitanus a valuable tool for also investigating how bacteria coordinate the processes of chromosome segregation and cell division with organelle trafficking—a well-studied question in eukaryotic cells that remains unaddressed in prokaryotes. We show how the deletion of one A/D ATPase can have indirect effects on the inheritance of disparate cargos positioned by other A/D ATPases via defects in DNA replication, chromosome segregation, and/or cell division. We also provide evidence that flagella positioning influences the spatial regulation of chemotaxis clusters. Finally, we identify putative sequence- and structural-determinants that uniquely link each A/D ATPase to a specific cargo, ultimately allowing these related ATPases to coexist and function in the same cell. Together, our study probes mechanistic commonality and variation in the most widespread ATPase family used in the spatial regulation of diverse cellular cargos across prokaryotes, and all within a single cell.

Results

A third of sequenced bacteria encode multiple ParA/MinD family ATPases

It is unclear how many ParA/MinD (A/D) family ATPases are encoded in a single organism. To answer this question, we performed an extensive tBLASTn analysis using a consensus protein sequence, generated from well-studied A/D ATPases, as the query (see methods). As already established¹⁷, we found that ~ 96% of bacteria from the NCBI Reference Sequence (RefSeq) database encode at least one A/D ATPase (Supplementary Data 1 and 2). These hits were binned by bacterial species, which were then ordered by the number of A/D hits. From this initial list, we found many bacterial genomes encoding 10 to 20 A/D ATPases. However, these bacteria with the most A/D ATPases had their genomes encoded on multiple plasmids and chromosomes, each of which encode its own ParA-based DNA segregation system¹⁸. For this study, we were specifically focused on understanding how multiple A/D ATPases coexist and coordinate the positioning of disparate cargos in the same cell. Therefore, we further filtered our dataset to identify bacteria encoding multiple A/D ATPases, but only one chromosome and no stable plasmids (Supplementary Data 1 and 2). Even after accounting for bacteria with genomes encoded on multiple genetic elements, our bioinformatic analysis revealed that more than a third of sequenced bacteria encode multiple A/D ATPases (Fig. 1a, Supplementary Data 1 and 2).

**Fig. 1: ParA/MinD-like ATPases are widespread in bacteria.**

Gene neighborhood analysis of A/D ATPases in H. neapolitanus implicate cellular cargos

We next set out to determine how multiple A/D ATPases can coexist in the same cell to position disparate cargos. To address this question, we identified an organism from our list of bacteria encoding multiple A/D ATPases (Supplementary Data 1 and 2). Among the top 1% of bacteria (encoding six or more A/D ATPases), we identified several pathogens including Clostridia, Burkholderia, Mycobacteria, Vibrio, Pseudomonas, and Xanthomonas species. We also identified the non-pathogenic and experimentally tractable organism, H. neapolitanus—a slow-growing, sulfur-oxidizing chemoautotroph that encodes seven putative A/D ATPases on one chromosome (Fig. 1b). Spatial regulation by A/D ATPases has largely been studied in fast-growing model bacteria. We intentionally chose a slow growing bacterium (6 h doubling time) because the infrequent DNA segregation and cell division events allowed for larger observation windows and a more direct view into the dynamics of organelle trafficking. The tractability, slow-growth rate, and abundance of A/D ATPases made H. neapolitanus an ideal choice for further study.

To determine whether the A/D ATPase hits in H. neapolitanus were indeed spatial regulators, we performed a gene neighborhood analysis (GNA) to identify putative cargos (Fig. 1b). GNA allows us to infer function because A/D ATPase genes are often encoded near cargo-associated loci. For example, the ParABS chromosome segregation system is typically encoded near OriC¹⁹. Strikingly, the putative cargos we identified using this approach includes the chromosome and all known protein-based cargos of the A/D ATPase family^20,21. While spatial regulation of these diverse cargos has been individually studied in many different model bacteria, their coordinated positioning by multiple A/D ATPases has never been investigated together in one organism.

As a second line of bioinformatic evidence linking each A/D ATPase to a specific cargo, we investigated the conservation of the A/D ATPase gene neighborhoods using FlaGs (Flanking Genes) analysis²². FlaGs analysis predicts functional associations by taking a list of NCBI protein accessions as input and clusters neighborhood-encoded proteins into homologous groups. Homologs of each A/D ATPase in H. neapolitanus were identified using BLASTp, and top hits were used as input for FlaGs analysis (Supplementary Fig. 1). The analysis shows strong conservation of the A/D ATPase gene neighborhoods across multiple bacterial phyla, further implicating the putative cargos. Due to the limited data on A/D ATPases associated with conjugation¹⁰ or nitrogen metabolism, we excluded these hits from further investigation in this study.

With the remaining five A/D ATPases, we performed a multiple sequence alignment against ParA/MinD family members that have been previously established to position a cellular cargo (Fig. 1c). Each A/D ATPase in H. neapolitanus clustered with a specific family member known to position a specific cellular cargo in other bacteria—the chromosome (ParA²), the divisome (MinD²³), the carboxysome (McdA^4,5), the flagellum (FlhG^6,7) and the chemotaxis cluster (ParC^8,9). These data provide a third line of evidence that further implicates the putative cargos identified by our GNA. We next sought to directly identify the role of each A/D ATPase in positioning the cargos implicated by bioinformatics.

The ParAB system is required for chromosome segregation in H. neapolitanus

Chromosome segregation prior to cell division is critical for cellular survival. In most bacteria, faithful chromosome segregation and inheritance are mediated by a ParAB system encoded near OriC¹⁹. Without ParAB, DNA is asymmetrically inherited, resulting in anucleate and polyploid cells, and reduced cell fitness or death². In H. neapolitanus, there is a putative parAB system encoded near OriC (Fig. 2a). To image chromosome segregation, the ParB homolog, encoded downstream of the putative parA gene (Hn2335), was fused to the fluorescent protein monomeric Neon Green (mNG). ParB-mNG was observed as one or two puncta per cell (Fig. 2b). Population analysis showed that shorter cells had a single focus at mid-cell, whereas longer cells had two foci at the quarter positions of the cell (Fig. 2b). When the putative parA gene (Hn2335) was deleted, ParB-mNG foci were completely absent in 25% of cells (Supplementary Fig. 2d). When ParB foci were present, they were randomly positioned regardless of cell length (Fig. 2c) and significantly brighter compared to that of WT (wild-type) cells (Supplementary Fig. 2e). Complementation with Hn2335 at an exogenous locus restored foci positioning (Supplementary Fig. 2h). These data suggest that replicated chromosomes in ∆Hn2335 were no longer faithfully segregated, resulting in anucleate and polyploid cells.

**Fig. 2: Each A/D ATPase in *H. neapolitanus* positions a specific cargo.**

We then performed time-lapse microscopy of ParB-mNG foci and SYTOX-stained nucleoids to observe chromosome segregation in real time. Newborn WT cells have a single ParB focus at mid-cell, which then splits into two foci that bidirectionally segregate towards the quarter positions of the growing cell (Supplementary Fig. 2f and Supplementary Movie 1A). Foci positioning at the cell quarters was maintained, which then became the mid-cell position of each daughter cell following division. Faithful chromosome segregation and inheritance were lost when the putative parA gene (Hn2335) was deleted (Supplementary Fig. 2g and Supplementary Movie 1B). Many cells had a single large ParB focus at a cell pole that did not split and all DNA was concentrated into a single daughter cell upon division. These polyploid daughters continued to divide, while the anucleate daughters were no longer viable. These data confirmed that the increased ParB foci intensities in the deletion mutant represents an increase in chromosome copy number in these cells. In summary, the A/D ATPase encoded by Hn2335 is required for faithful chromosome segregation (Fig. 2d) and we will henceforth refer to this protein as ParA.

The MinCDE system aligns cell division at mid-cell in H. neapolitanus

Proper positioning of the divisome ensures that when a cell divides, both daughter cells are roughly equal in length. Without the Min system, division occurs at any nucleoid-free region²⁴, producing anucleate minicells, the products of polar divisions. Our bioinformatic analyses showed that the protein encoded by Hn1364 is a MinD homolog, and immediately downstream of this gene in the same operon is a gene encoding a MinE homolog (Fig. 2e). To determine if Hn1364 was indeed involved in divisome positioning, we analyzed dividing cells and identified their constriction sites relative to cell length (Fig. 2f, g). WT cells divided close to mid-cell (Fig. 2f and Supplementary Fig. 3d) while the ΔHn1364 cells divided asymmetrically (Fig. 2g and Supplementary Fig. 3e). Population analysis of dividing cells identified mid-cell constriction sites in 97% of WT cells compared to only 41% of ΔHn1364 cells (Supplementary Fig. 3f and Supplementary Movie 2). In addition to single asymmetric division events, 9% of dividing cells in the ΔHn1364 mutant population formed multiple division sites simultaneously along the cell length (Supplementary Fig. 3e, Supplementary Fig. 3g and Supplementary Movie 2C). The unequal divisions resulted in greater variation in cell length (Supplementary Fig. 3h). Complementation with Hn1364 at an exogenous locus restored mid-cell constriction and single division events (Supplementary Fig. 3j). Overall, our findings show that Hn1364 is critical for positioning the divisome at mid-cell (Fig. 2h) and we will henceforth refer to this protein as MinD.

The A/D ATPase, McdA, encoded in the carboxysome operon positions carboxysomes

Bacterial microcompartments, or BMCs, are large icosahedral protein-based organelles that encapsulate sensitive metabolic reactions to provide prokaryotes with distinct growth advantages²⁵. Despite their importance, little is known about how BMCs are spatially regulated. The model BMC is the carbon-fixing carboxysome found in cyanobacteria and proteobacteria²⁶. It was recently found that an A/D ATPase, we termed Maintenance of carboxysome distribution protein A (McdA), is widespread among carboxysome-containing bacteria, including H. neapolitanus (Fig. 2i)⁴. McdA spaces carboxysomes on the nucleoid along with its partner protein, McdB^27,28. To demonstrate its requirement for carboxysome positioning, we visualized carboxysomes by labelling the small subunit of the encapsulated Rubisco enzyme (cbbS) with mTurquoise2 to form CbbS-mTQ. As previously shown, in WT cells, carboxysomes are distributed down the cell length (Fig. 2j, Supplementary Fig. 4d). In the deletion mutant (ΔmcdA), carboxysome aggregates form a large bright polar focus at one or occasionally both poles (Fig. 2k, Supplementary Fig. 4d, e). Complementation with mcdAB at an exogenous locus restored carboxysome positioning (Supplementary Fig 4h). These data are consistent with our previous observations using TEM, which showed that foci in the mutant population represent an aggregation of assembled carboxysomes²⁸.

We extend our previous findings here with long-term time-lapse microscopy. In WT cells, carboxysomes are dynamically positioned along the cell length across multiple generations (Supplementary Fig. 4f and Supplementary Movie 3A). In the deletion mutant, polar carboxysome aggregates were stagnant (Supplementary Fig. 4g and Supplementary Movie 3B). Together, our findings show that the A/D ATPase encoded within the carboxysome operon, we termed McdA, is essential for distributing carboxysomes across the cell length and ensuring organelle homeostasis following division (Fig. 2l).

Hn0716 is required for regulating flagella number and positioning

Flagella are external filamentous structures that allow for bacterial motility. Bacteria vary in flagella location, number, and pattern. Many bacteria encode an A/D ATPase called FlhG in their flagellar operon, which is essential for diverse flagellation patterns in many bacteria⁷, yet the mechanisms remain unclear (Supplementary Discussion). In polar flagellates, deletion of flhG typically results in changes in flagella number and motility^{6,29,30,31,32,33,34}. Meanwhile, in the peritrichous organism, Bacillus subtilis, deletion of flhG results in changes to flagella location³⁵. Surprisingly, deletion of flhG in the amphitrichous organism, Campylobacter jejuni, also results in cell division defects^36,37.

Our bioinformatic analysis suggests that Hn0716 within the flagella operon encodes an FlhG homolog (see Fig. 1c). To determine if Hn0716 is involved in flagellar spatial regulation, we first visualized a mNG fusion of FliN, which is a component of the flagellar basal body that assembles at the cytoplasmic face of the membrane (Fig. 2m)³⁸. WT cells had a single mNG-FliN focus at the extreme cell pole (Fig. 2n). In ΔHn0716 cells, FliN foci were no longer faithfully positioned (Fig. 2o) and cells were more likely to have zero or multiple foci (Supplementary Fig. 5d). Complementation with Hn0716 at an exogenous locus restored FliN foci localization to the poles (Supplementary Fig. 5i). These data suggest that Hn0716 is required for positioning a single flagellum at a single pole in H. neapolitanus.

We next set out to determine whether these alterations to FliN positioning affected cell motility. Motility assays in soft agar found that ΔHn0716 cells were not motile (Supplementary Fig. 5g). Loss of motility could be due to flagellar mispositioning, a loss of flagella, or hyper-flagellation. To image flagella, we engineered flagellin^T185C, which allows for fluorescent labeling of flagella via the addition of a cysteine-reactive maleimide stain to the media³⁹. We found that WT H. neapolitanus cells are monotrichous, with a single polar flagellum emanating from the FliN focus (Supplementary.Fig. 5h). ΔHn0716 cells also had flagella emanating from FliN foci, however, the cells were hyper-flagellated, with multiple flagella often bundled together as tufts, emanating from several FliN foci. We conclude that Hn0716 is required for regulating flagella number and position in H. neapolitanus (Fig. 2p) and we will henceforth refer to this protein as FlhG. Future studies will investigate the unique pleiotropic effects of flhG in flagella assembly, number, and location, as well as cell division in H. neapolitanus (Supplementary Discussion).

Hn0722 is required for chemotaxis cluster assembly and positioning

Directing bacterial motility are large hexagonal arrays called chemotaxis clusters, comprised of chemoreceptors, an adaptor protein (CheW), and a kinase (CheA). Several mechanisms have evolved to control both the number and positioning of chemotaxis clusters, including the use of A/D ATPases (called ParC in Vibrio species or PpfA in R. sphaeroides). In Vibrio species, ParC directs chemotaxis arrays to a polar landmark protein called HubP^8,40. As a consequence, daughter cells inherit an array at their old pole upon cell division. In R. sphaeroides, there are no polar landmarks and PpfA distributes chemotaxis clusters over the nucleoid^9,41. Where studied, deletion of the A/D ATPase alters chemotaxis cluster number and positioning in cells, which results in a reduction in spreading motility in soft agar.

Our bioinformatics analysis showed that the protein encoded by Hn0722 is a ParC/PpfA homolog within the chemotaxis operon of H. neapolitanus (see Fig. 1c). To determine if Hn0722 is important for spatially regulating chemotaxis clusters, we first imaged chemotaxis clusters by fusing mNG to CheY (Fig. 2q). CheY is a response regulator that is phosphorylated by CheA, and has previously been shown to colocalize with chemotaxis clusters in E. coli^42,43. Consistent with electron micrographs of chemotaxis clusters in H. neapolitanus⁴⁴, CheY-mNG formed a single focus immediately proximal to one cell pole in ~ 85% of WT cells (Fig. 2r, Supplementary Fig. 6d). However, when Hn0722 was deleted, the CheY-mNG signal was diffuse in ~ 80% of the cell population (Fig. 2s, Supplementary Fig. 6d). In the ~20% of ΔHn0722 cells that had a CheY-mNG focus, the foci were significantly lower in intensity (Supplementary Fig. 6e). Complementation with the overlapping genes Hn0722 and Hn0723 recovered CheY foci at the poles (Supplementary Fig. 6f). Together, we conclude that Hn0722 is required for chemotaxis cluster assembly and positioning in H. neapolitanus (Fig. 2t) and we will henceforth refer to this protein as ParC.

Cargo positioning is not directly controlled by A/D ATPases encoded at distant loci

We have thus far provided direct evidence showing that five A/D ATPases position five disparate cellular cargos in H. neapolitanus (Fig. 2). We next asked if each of the five positioning reactions occurred independently from each other. To answer this question, we individually deleted each A/D ATPase in every cargo-labeled background strain (Fig. 3). We found carboxysome (Fig. 3c) and flagella (Fig. 3d) positioning were largely unaffected by the deletion of distant A/D ATPases. Intriguingly, chromosome (Fig. 3a), divisome (Fig. 3b), and chemotaxis cluster (Fig. 3e) positioning or focus intensity were all influenced in ΔflhG cells, albeit with intermediate phenotypes when compared to deleting the dedicated A/D ATPase (Fig. 3, bold boxes). Together our data show that each A/D ATPase is dedicated to the positioning of a specific cargo type, and not directly involved in the positioning of other cargos. However, our data at the cell-population level also unveiled potential coordination, crosstalk, and/or interdependencies among certain positioning reactions. In the next three sections, we dissect how cargo positioning by one A/D ATPase can be indirectly coordinated with the positioning and inheritance of a cargo positioned by another A/D ATPase.

Deletion of parA, minD, or flhG results in anucleate cells via three different mechanisms

We have shown how the deletion of parA results in a significant fraction of anucleate cells because ParA is directly required for chromosome segregation and inheritance following cell division (Fig. 4a). Deleting minD or flhG also resulted in anucleate cells, but the mechanism was different in each case. In ΔminD cells, chromosome positioning (Fig. 4b) and ParB foci intensities (Fig. 4c) were similar to that of WT, showing that chromosome segregation was still active. Instead, it was divisome mispositioning in ΔminD cells and subsequent asymmetric cell division (Fig. 4d) that indirectly caused asymmetric chromosome inheritance and anucleate cell formation.

**Fig. 4: Anucleate cells form via three different mechanisms.**

In ΔflhG cells, chromosome positioning (Fig. 4b) and ParB foci intensities (Fig. 4c) were also similar to that of WT, again suggesting that chromosome segregation was unaffected. But, divisome mispositioning was not as severe as ΔminD cells (Fig. 4d, e), suggesting that anucleate cells were forming via a third distinct mechanism. Intriguingly, ΔflhG cells were less likely to have two ParB foci compared to WT cells (Fig. 4a). Instead, pre-divisional cells still had a single ParB focus (Fig. 4b) with intensities that suggested the presence of only a single chromosome copy (Fig. 4c). Time-lapse microscopy confirmed that ParB foci positioning was actively maintained (Supplementary Movie 4). Therefore, anucleate ΔflhG cells are likely formed due to defects in chromosome replication and/or premature cell division; a question of future study.

Together, the findings emphasize the importance of probing the functional relationships of A/D ATPases with each other and the bacterial cell cycle, when occupying the same organism.

Anucleate cells inherit carboxysomes

McdA uses the nucleoid as a positioning matrix for distributing carboxysomes^27,28. Therefore, it was surprising that in the ΔparA mutant population, all anucleate cells retained carboxysomes (Fig. 5a). To determine if anucleate cells synthesized carboxysomes de novo or carboxysomes were somehow still inherited following division, we performed time-lapse microscopy on ΔparA cells with fluorescent carboxysomes (Fig. 5b and Supplementary Movie 5A). Intriguingly, anucleate cells indeed inherited carboxysomes, but in the most unexpected fashion. In dividing ΔparA cells that failed to split their ParB foci, carboxysomes in the to-be-anucleate cell bundled up immediately adjacent to the division plane. Carboxysome bundling was coincident with the chromosome spooling action that occurred at the invaginating septum just prior to complete division and asymmetric chromosome inheritance (see Supplementary Fig. 2g and Supplementary Movie 1B). After septation was complete, the massive carboxysome bundle was explosively liberated from the new pole of the anucleate cell, resulting in multiple freely diffusing carboxysome foci (Fig. 5b and Supplementary Movie 5A). Anucleate cells harboring carboxysomes did not divide further. We found that anucleate cells in the ΔminD (Fig. 5c and Supplementary Movie 5B) and ΔflhG (Fig. 5d and Supplementary Movie 5C) cell populations also inherited carboxysomes via the same mechanism. Together the data show how carboxysome trafficking and distribution by McdA are dependent upon faithful chromosome segregation. However, anucleate cells can still inherit carboxysomes in parA, minD, or flhG deletion strains because carboxysomes are scraped off of missegregated chromosomes that are spooled through the invagination site during septation. We speculate that all mesoscale cargos using asymmetrically inherited nucleoids as a positioning matrix would show the same mode of inheritance.

Deletion of parA, minD or flhG influence chemotaxis cluster assembly

The A/D ATPase that positions chemotaxis clusters in Rhodobacter sphaeroides uses the nucleoid as its positioning matrix⁴¹. Therefore, we suspected that A/D ATPase deletions resulting in anucleate cells would indirectly impact the spatial regulation of chemotaxis clusters in these strains. Indeed, we found that parA, minD, and flhG deletion strains, all of which form anucleate cells (see Fig. 4a), had a corresponding increase in cells lacking chemotaxis clusters (Fig. 6a), and when cells had foci, they were notably dimmer compared to WT (Fig. 6b).

**Fig. 6: Deletion of *parA*, *minD*, or *flhG* influences chemotaxis cluster assembly.**

It is important to note that while ΔparA and ΔminD strains exhibited moderate effects on chemotaxis cluster assembly, deletion of flhG was more severe (Fig. 6a, b). Given that chemotaxis clusters communicate with the flagellum to move the bacterium towards favorable conditions, we hypothesize that chemotaxis cluster assembly and organization in H. neapolitanus is regulated by flagellum positioning. Interestingly, this effect was not reciprocal, as deletion of parC had minimal effect on flagella foci number (Fig. 6c) or flagella basal body intensity (Fig. 6d). Identifying the molecular players responsible for this crosstalk in the spatial regulation of the flagellum and chemotaxis cluster is a subject of future work. Together, our data demonstrate interdependencies in how A/D ATPases coordinate the positioning of protein-based organelles with each other as well as the processes of DNA segregation and cell division.

A/D ATPases have unique interfaces that confer cargo-positioning specificity

We have experimentally demonstrated that multiple A/D ATPases coexist and coordinate the positioning of multiple disparate cargos in the same cell. We also showed that A/D-based positioning is cargo specific. A/D ATPases have been shown to form very similar sandwich dimer structures^45,46,47,48, and AlphaFold2 (AF2) predictions suggest this is also the case for all five A/D ATPases of H. neapolitanus studied here (Supplementary Fig. 7a-b). The structural similarities suggest there are conserved interfaces unique to each A/D ATPase that provide specificity, linking an A/D ATPase to its particular cargo.

Using AF2 and Rosetta predictions, we determined the A/D ATPase structures from H. neapolitanus and identified the putative interaction interfaces that provide specificity to each positioning reaction. There are three interfaces on an A/D ATPase that confer specificity: (1) the dimerization interface, (2) the interface for interaction with its positioning matrix (nucleoid or membrane), and (3) the interface for interaction with its partner protein, which ultimately links the ATPase to its cargo. We identified these three interfaces for all five of the A/D ATPases in H. neapolitanus (Fig. 7) and predict key residues required for these associations (Supplementary Data 3).

**Fig. 7: A/D ATPases have unique interfaces that confer cargo-positioning specificity.**

Specificity at the dimer interface restricts A/D ATPases to homodimerization, and thereby allows each A/D ATPase to function independently in the same cell without cross-interference (Fig. 7a). The putative residues required for homodimerization of the five A/D ATPases in H. neapolitanus are provided in Supplementary Data 3, Tab. 1.

ParA-like ATPases use the nucleoid and MinD-like ATPases use the inner membrane for cargo positioning. ParA, McdA, and ParC have basic residues at their C-termini for non-specific binding to nucleoid DNA, whereas MinD and FlhG have membrane targeting sequences (MTSs) for membrane binding (Fig. 7b). We identified the predicted residues required for each A/D ATPase to associate with their respective positioning matrix (Supplementary Data 3, Tab 2).

Cargo specificity for A/D ATPases comes from interacting with a partner protein that either associates with the cargo, or is a physical component of the cargo itself. Partner proteins have a stretch of amino acids enriched in charged residues at the N-terminus that exclusively interacts with its A/D ATPase, while the rest of the partner protein is dedicated to cargo association⁴⁹. We generated docking models of each A/D ATPase with an N-terminal peptide from its partner protein (Fig. 7c). The peptides were defined as the first 30 residues of the putative partner protein from the N-terminus. The peptide docking simulations identified several putative residues that are key to system specificity between an A/D ATPase and its partner protein (Supplementary Data 3, Tab 3).

Finally, we performed in silico alanine-scanning mutagenesis across all residues comprising the interacting interface of the N-terminal peptides of partner proteins docked onto their cognate A/D ATPase (Supplementary Data 3, Tab 4). The resulting ΔΔG values identify the extent to which each residue contributes to the stability of the partner protein interaction with its cognate A/D ATPase. Importantly, our in silico alanine-substitution simulations identified all residues experimentally shown to be important for docking the Escherichia coli MinE peptide onto the MinD dimer^47,50,51 (Supplementary Fig. 7c). Together, our in silico data provide a roadmap for strategic mutagenesis and mechanistic probing of the specificity determinants involved in bacterial chromosome segregation, cell division positioning, and protein-based organelle trafficking by A/D ATPases across prokaryotes.

Discussion

The study of A/D ATPases has focused on a specific cargo of a certain biological process, and largely in model organisms that encode only one or two A/D ATPases. In these studies, two questions are typically posed: How does a specific cargo find its correct position, and how does this position change over time? Here, our focus was on the positioning systems, rather than a specific cargo-type or biological process. Our systems biology approach therefore addresses how multiple A/D ATPases coordinate the positioning of diverse cargos in the same cell.

Encoding multiple A/D ATPases is a shared feature across prokaryotes. We find that over a third of all sequenced bacterial genomes encode multiple A/D ATPases (Fig. 1a), with some bacteria encoding >10. Interestingly, although most bacteria have the same fundamental cargos, not all use dedicated A/D-based positioning systems. For example, many of the cellular cargos we found here to be positioned by A/D ATPases in certain bacteria, like H. neapolitanus, are not actively positioned by A/D ATPases in others, like E. coli. What necessitates an A/D ATPase for positioning a certain cellular cargo in one bacterium and not in another remains an open question. There does, however, seem to be a limit to the number of A/D ATPases that a bacterium can encode. A/D ATPases are also encoded in archaeal genomes⁵², but little is known about their roles in subcellular organization. A recent study showed that archaeal species across several phyla, Euryarchaeota in particular, encode multiple A/D ATPases⁵³. Several of these species contained more than a dozen, including H. volcanii with 13 A/D ATPases, four of which are MinD-homologs. Strikingly, all four MinD homologs were not required for cell division positioning, but one (MinD4) stimulated the formation of chemotaxis arrays and the archaella, which is the functional equivalent of the bacterial flagellum. This study stresses the importance of experimentally linking A/D ATPases to their cellular cargos as we have done here.

The ParA/MinD family of ATPases spatiotemporally regulates a growing list of diverse mesoscale complexes critical to fundamental processes in prokaryotes, including cell growth and division, DNA segregation, motility, conjugation, and pathogenesis^20,21. Therefore, understanding how A/D ATPases coordinate the positioning of cellular cargos at the correct location at the correct time is key to understanding bacterial cell function. Using H. neapolitanus as a non-pathogenic model, our findings strongly support the idea that each ATPase is dedicated to the positioning of a specific cellular cargo. Our model also unveiled the dependency of protein-based organelle trafficking on the faithful replication and segregation of the chromosome as well as positioning cell division at mid-cell. For example, while the direct consequence of deleting ParA was chromosome missegregation, the resulting asymmetric chromosome inheritance also led to indirect defects in the positioning of carboxysomes and chemotaxis clusters. We also identified a previously uncharacterized epistatic relationship in flagella positioning by FlhG and its downstream influence on the spatial regulation of the chemotaxis cluster by ParC. It is well known that chemotaxis clusters direct cell motility by controlling the direction of flagellar rotation. However, to our knowledge, crosstalk in the spatial regulation of flagella and chemotaxis clusters has yet to be documented. Future studies will determine the molecular players involved in the crosstalk between A/D-based positioning of these two cellular cargos both involved in cell motility.

Our bioinformatics also identified a sixth A/D ATPase in the H. neapolitanus genome. Hn1669 is an A/D ATPase that shows homology to VirC1 and is located near trb genes, which encode conjugation machinery components (Fig. 1b). A single study has shown that the VirC1 ATPase, encoded on the Ti plasmid of Agrobacterium tumefaciens, is involved in recruiting the conjugative Ti plasmid to the Type IV secretion machine at the cell poles¹⁰. Intriguingly, in H. neapolitanus, the VirC1 homolog is encoded on an Integrative Conjugative Element (ICE) and the downstream gene from this A/D ATPase encodes a ParB homolog. ICEs are major drivers of bacterial evolution and the spread of antibiotic resistance genes⁵⁴. We have yet to image conjugation in H. neapolitanus, but we suspect this ParB homolog binds and demarcates the ICE locus during conjugation. It is attractive to speculate that the VirC1-homolog and its downstream ParB homolog are involved in transporting and positioning the ICE locus to the Type IV secretion machine at the cell pole for conjugation.

While A/D ATPases share sequence, structural, and biochemical commonalities, the partner proteins linking these ATPases to their cargos are extremely diverse. Due to this diversity, the partner protein has not always been identified, and as a result, many A/D ATPases are called ‘orphans’²⁰. The extreme diversity is largely due to the partner proteins providing the specificity determinants linking an A/D ATPase to its cognate cargo. Despite their extreme diversity, data across the field supports the idea that partner proteins interact with and stimulate their A/D ATPases via a shared mechanism. Partner proteins involved in plasmid partition and chromosome segregation, as well as those required for positioning BMCs, flagella, chemotaxis clusters, and the divisome have all been shown, or suggested, to interact with their A/D ATPase via a positively charged and disordered N-terminus⁴⁹. Our in silico analysis of A/D ATPase dimers docked with N-terminal peptides of their partner proteins demonstrates how specific cargos are assigned, and how these related positioning systems coexist and function in the same cell without cross-interference.

Going forward, we aim to use H. neapolitanus as a model to define the general mode of transport shared among the entire A/D ATPase family and to determine how positioning reactions are altered for disparate cargos. These findings are significant because A/D ATPases spatially organize essentially all aspects of bacterial cell function. An additional future direction is to experimentally verify the specificity determinants we identified here for each partner protein and cargo, and leverage this knowledge in the rational design of positioning systems in bacteria. These contributions are expected to be significant because minimal self-organizing systems are vital tools for synthetic biology⁵⁵. We aim to design Minimal Autonomous Positioning Systems (MAPS) consisting of a positioning ATPase and their partner-protein N-terminal peptide as a “luggage tag” to be used as spatial regulators for natural- and synthetic-cargos in heterologous bacteria.

Methods

tBLASTn analysis

tBLASTn analysis was done using a ParA/MinD consensus sequence (xKGGxxK[T/S]), as a query against RefSeq Representative genomes database with max target sequences as 5000 and E value threshold at 0.0001. Sequences were filtered for those that shared sequence homology and had one of the identified putative cargo genes, confirmed using webFlaGs (https://pubmed.ncbi.nlm.nih.gov/32956448/). The consensus query was generated using COBOLT (https://pubmed.ncbi.nlm.nih.gov/17332019/).

FlaG analysis

A few representative genomes were selected to display gene neighborhood conservation. Identification of replication origins (OriC’s) were performed using Ori-Finder (https://bmcbioinformatics.biomedcentral.com/articles/10.1186/1471-2105-9-79). FlaGs analysis figure was generated using Gene Graphics (https://katlabs.cc/genegraphics/).

Multiple sequence alignment

Sequences were aligned using Clustal Omega. The resulting tree was imported into iTOL to generate an unrooted tree. (NCBI Accession numbers: Hn2335—ACX97145.1; Hn1364—ACX96198.1; Hn0912—ACX95755.1; Hn0716—ACX95565.1; Hn0722—ACX95571.1; Hn1669—ACX96495.1; Hn0255—ACX95118.1).

Media and growth conditions

All mutants described in this study were constructed using WT Halothiobacillus neapolitanus (Parker) Kelly and Wood (ATCC® 23641™) purchased from ATCC. Cultures were grown in ATCC® Medium 290: S-6 medium for Thiobacilli (Hutchinson et al., 1965) and incubated at 30 °C, while shaken at 130 RPM in air supplemented with 5% CO₂. Strains were preserved frozen at −80 °C in 10% DMSO.

Construct designs and cloning

All constructs were generated using Gibson Assembly and verified by sequencing. Fragments for assembly were synthesized by PCR or ordered as a gBlock (IDT). Constructs contained flanking DNA that ranged from 750 to 1100 bp in length upstream and downstream of the targeted insertion site to promote homologous recombination into target genomic loci. Cloning of plasmids was performed in chemically competent E. coli Top10 or Stellar cells (Takara Bio).

Making competent cells in H. neapolitanus C2

Competent cells of H. neapolitanus were generated as previously reported. In short, 1 L of culture was grown to an OD of 0.1–0.15. Cultures were harvested by centrifugation at 5000 x g for 20 min at 4 °C. Pellets were resuspended and washed twice with 0.5 volumes of ice-cold nanopore water. All wash centrifugation steps were performed at 3000xg for 30 min at 4 °C. The resulting pellet after washing was resuspended in 1 × 10⁻³ volumes of ice-cold nanopore water. These competent cells were used immediately or frozen at −80 °C for future use. Frozen competent cells were thawed at 4 °C before use.

Transformation in H. neapolitanus C2

50-100 µL of competent cells were mixed with 5 µL plasmid DNA (1–5 µg) and incubated on ice for 5 min. This mixture was then transferred to a tube containing 5 mL ice-cold S6 medium without antibiotics and incubated on ice for 5 min. Transformations were recovered for 16–36 h, while shaken at 130 RPM, at 30 °C, in air supplemented with 5% CO₂. Clones were selected by plating on selective medium with antibiotics. Colonies were restreaked. Restreaked colonies were verified for mutation by PCR.

Native fluorescent fusions

For the native fluorescent fusion of ParB-mNG, mNG-FliN, and CheY-mNG, the sequence encoding the fluorescent protein mNeonGreen (mNG) was attached to the 3' or 5' region of the native coding sequences, separated by a GSGSGS linker. For the native fluorescent fusion of Cbbs-mTQ, the sequence encoding the fluorescent protein mTurquoise (mTQ) was attached to the 3' region of the native coding sequence, separated by a GSGSGS linker. A kanamycin resistance cassette was inserted before the gene for N-terminal tags or after the gene for C-terminal tags. When necessary, the promoter was duplicated. The mutant was selected by plating on S6 agar plates supplemented with 50 μg/mL of kanamycin. All fusions were verified by PCR.

Deletion mutants

For deletions of Hn2335, Hn1364, Hn0912, and Hn0722, the genes were replaced with a spectinomycin resistance cassette, followed by a duplicated promoter for the downstream gene. Deletion of Hn0716 was obtained by codon-optimizing the downstream gene and inserting the spectinomycin resistance cassette after this codon-optimized gene. Mutants were selected by plating on S6 agar plates supplemented with 50 μg/mL of spectinomycin. All mutations were verified by PCR.

Complementation mutants

The deleted genes (Hn2335, Hn1364, Hn0912, Hn0716 and Hn0722) were individually placed under the expression of a P_trc promoter, and inserted into a neutral site, located between genes Hn0933 and Hn0934. Mutants were selected by plating on S6 agar plates supplemented with 25 µg/mL chloramphenicol. The insertion was verified by PCR. For imaging, cells were grown to an OD of 0.1, induced with 0, 0.25, 1, 5, 10, and/or 50 µM IPTG for up to 6 h, and imaged for complementation at various time points. Hn2335 complemented with 50 µM IPTG for 3 h. Hn1364 complemented with 50 µM IPTG for 6 h. Hn0716 complemented with leaky expression of the P_trc promoter and induction was not necessary. Hn0912 and Hn0722 did not complement with those genes alone. For these two mutants, additional constructs were made to include the neighboring overlapping genes: Hn0911-Hn0912 and Hn0722-Hn0723. Constructs were generated, verified, induced, and imaged as detailed above. Hn0911-Hn0912 complemented with 50 µM IPTG for 3 h. Hn0722-Hn0723 complemented with 50 µM IPTG for 3 h.

Microscopy

All live-cell microscopy was performed using exponentially growing cells. 3-5 µL of cells were dropped onto a piece of 2% UltraPure agarose (Invitrogen, catalog number 16500) + S6 pad and imaged on a 35-mm glass-bottom dish (MatTek, catalog number P35G-1.5-14-C). All fluorescence and phase contrast imaging were performed using a Nikon Ti2-E motorized inverted microscope controlled by NIS Elements software with a SOLA 365 LED light source, a 100X Objective lens (Oil CFI Plan Apochromat DM Lambda Series for Phase Contrast), and a Photometrics Prime 95B Back-illuminated sCMOS camera or a Hamamatsu Orca Flash 4.0 LT + sCMOS camera. ParB-mNG, mNG-FliN, and CheY-mNG were imaged using a “GFP” filter set (C-FL GFP, Hard Coat, High Signal-to-Noise, Zero Shift, Excitation: 470/40 nm [450-490 nm], Emission: 525/50 nm [500-550 nm], Dichroic Mirror: 495 nm). CbbS-mTQ labeled carboxysomes were imaged using a “CFP” filter set (C-FL CFP, Hard Coat, High Signal-to-Noise, Zero Shift, Excitation: 436/20 nm [426-446 nm], Emission: 480/40 nm [460-500 nm], Dichroic Mirror: 455 nm). DAPI fluorescence was imaged using a standard “DAPI” filter set (C-FL DAPI, Hard Coat, High Signal-to-Noise, Zero Shift, Excitation: 350/50 nm [325-375 nm], Emission: 460/50 nm [435-485 nm], Dichroic Mirror: 400 nm). Alexa Fluor 594 C₅ maleimide-conjugated flagella were imaged using a “TexasRed” filter set (C-FL Texas Red, Hard Coat, High Signal-to-Noise, Zero Shift, Excitation: 560/40 nm [540-580 nm], Emission: 630/75 nm [593-668 nm], Dichroic Mirror: 585 nm).

Long-term microscopy

For multigenerational time-lapse microscopy, 1.5% UltraPure agarose (Invitrogen, catalog number 16500) + S6 pads were cast in 35-mm glass-bottom dishes (MatTek, catalog number P35G-1.5-14-C). Dishes were preincubated at 30 °C in 5% CO2 for at least 24 h. 4 µl of exponentially growing cells were spotted onto the agar pad. Temperature, humidity, and CO₂ concentrations were controlled with a Tokai Hit Incubation System. NIS Elements software was used for image acquisition. Cells were preincubated in the stage top for at least 30 min before image acquisition. Videos were taken at one frame per 2.5–5 min for a duration of 12–24 h.

Image analysis

Several fields of view were captured for each mutant. All fluorescence channels were subjected to background subtraction on Fiji with a rolling ball radius of 50 μm. Background-subtracted fluorescence images were merged with phase contrast images to create composites used for image analysis. Image analysis including cell identification, quantification of cell length, foci localization, foci number, foci fluorescence intensity, and identification of constriction sites were performed using Fiji plugin MicrobeJ 5.13I^56,57. Cell perimeter detection and segmentation were done using the rod-shaped descriptor with default threshold settings at a tolerance of 56. Maxima detection parameters were individually set for each cargo. For ParB-mNG (cargo = chromosome) foci detection, tolerance and z-score were both set to 100. For CheY-mNG (cargo = chemotaxis) foci detection, tolerance was set to 150 and z-score was set to 100. For mNG-FliN (cargo = flagella) foci detection, tolerance was set to 710 and z-score was set to 69. For CbbS-mTQ (cargo = carboxysome) foci detection, point detection was used instead of foci detection, and tolerance was set to 1010. Results were manually verified using the experiment editor, and non-segmented cells were cut using a particle cutter. Associations, shape descriptors, profiles, and localization were recorded for each strain. Localization graphs were automatically generated through MicrobeJ. Fluorescence intensity graphs and foci number count graphs were made in GraphPad Prism (GraphPad Software, San Diego, CA, www.graphpad.com).

Nucleoid staining for live imaging

Cells were harvested by centrifugation at 5000 g for 5 min. Following centrifugation, the cells were washed in PBS, pH 7.4. The resulting cell pellet was resuspended in 100 μL PBS and stained with SytoxBlue (Invitrogen, catalog number S34862) at a final concentration of 500 nM. The samples were incubated in the dark, at room temperature for 5–10 min. Stained cells were directly loaded onto an S6 agar pad that had been infused with 500 nM SytoxBlue.

Carboxysome count

Eleven frames were captured over the course of 2 min. Carboxysomes were counted in each frame. The highest number counted per cell was used for carboxysome count.

Motility assay

Motility assays were run in S6 in 0.4% agar. Cells were grown on a plate of S6 medium. Individual colonies were inoculated into tubes or plates of motility media and incubated at 30 °C in air supplemented with 5% CO₂. Tubes and plates were checked daily for motility for 2 weeks.

Cysteine-labelling of flagellin

H. neapolitanus flagellin was identified using a BLAST homology search with Hag from Bacillus subtilis as the query. Only a single flagellin gene was found. Several threonine and serine residues were considered for cysteine mutagenesis. Residues were mutated using the Q5 site-directed mutagenesis kit (NEB, E0554S). Clones were verified for cysteine mutation by sequencing.

Flagella stain

Alexa Fluor 594 C₅ maleimide dye (Invitrogen, A10256) was resuspended in DMSO to the working concentration of 10 mM. H. neapolitanus cultures were grown to an OD of 0.1–0.2. The cultures were adjusted to a pH of 7.0 using PBS, pH 11.7. Cells at an adjusted pH of 7.0 were then stained with Alexa Fluor 594 C₅ maleimide dye at a final dye concentration of 100 µM. Staining cultures were incubated overnight at 4 °C. The stained cells were washed 4+ times in PBS, pH 7.4. All centrifugation steps were performed at 5000 × g for 3 min.

Statistical analysis

For whole population analyses, such as fluorescence intensity and cell length measurements, we performed a non-parametric Wilcoxon (Kruskal-Wallis) test followed by a Dunn’s multiple comparisons test. For percent of population analyses, such as foci number or the presence of mid-cell constriction sites, populations were analyzed as separate field of views and the summary values were plotted as separate data points. We then performed a Brown-Forsythe and Welch ANOVA test followed by Dunnett’s T3 multiple comparisons test. GraphPad Prism (version 9.5.1) was used to perform all analyses. P-value style: 0.1234 (ns), 0.0332 (*), 0.0021 (**), 0.0002 (***), < 0.0001 (****).

Movie editing

Movies were cropped using Fiji. Movies were stabilized using the multi-channel hyperstack alignment plug-in called HyperStackReg. Time stamp and scale bar annotations were added using Fiji. Arrows and pauses were added using Adobe Premier Pro.

Structure prediction and peptide docking using AlphaFold2 and Rosetta

For MinD and McdA, we generated the N-terminus peptide-ATPase docking models using the CollabFold implementation of AlphaFold2^58,59. The N-terminal peptides were defined as the first 30 residues of the putative partner protein from the N-terminus. Multiple sequence alignments were constructed using MMseqs2. For each peptide-ATPase pair, we generated five structures with the default CollabFold/AlphaFold2 hyperparameters saved for the number of recycles for each model being increased to 12. As default to Alphafold2, the structures were energetically minimized with AMBER using the Amber99sb force field. We selected docked peptide models based on the pLDDT scores of the binding interface residues and similarity to previously resolved ParA-like ATPase/partner-protein crystal structures.

For ParA, FlhG, and ParC we generated docked peptide models using Rosetta’s FlexPepDock protocol⁶⁰. First, we equilibrated the ATPase homodimer structures generated from AlphaFold2 to the ref2015_cart_cst Rosetta force field with the FastRelax full-atom refinement protocol with cartesian coordinate space minimization using the lbfgs_armijo_nonmonotone minimizer. To preserve the position of the backbone atoms predicted by AlphaFold2, a backbone atom coordinate constraint was added. For this initial step, we generated 20 trajectories with the lowest scoring structure being used for the peptide docking step. We used the score3 with the docking_cen.wts_patch and the REF2015 force fields for the low-resolution and high-resolution docking steps of the FlexPepDock protocol, respectively. For each putative partner protein/ATPase pair, we simulated 50,000 docking trajectories. The top two thousand trajectories defined by the lowest energetics were then clustered using Calibur⁶¹. We then chose the final models based on cluster information, energetics, and mechanistic plausibility. From the final docked models chosen from the Rosetta and Alphafold2 simulations, we sought to identify key binding residues through an in-silico alanine mutation scan and ΔΔG calculations. We iteratively mutated all interface residues of the docked peptide and calculated the ΔΔG using the FlexDDG protocol in Rosetta with the talaris2014 forcefield⁶². In FlexDDG, for each mutation, the backbone and side chain conformations were sampled 35,000 times using Rosetta’s Monte Carlo backrub method. At every 2500 sample interval the ΔΔG of mutation was calculated. The final reported ΔΔG is the average of 35 such trajectories.

(NCBI Accession numbers: ParA partner protein (ParB)—ACX97144.1; MinD partner protein (MinE)—ACX96199.1; McdA partner protein (McdB)—ACX95754.1; FlhG partner protein (FliA)—ACX95566.1; ParC partner protein (CheW)—ACX95572.1). The first 30 amino acids from the N-terminus of each partner protein were docked onto the dimer structures in silico using AlphaFold2 (MinD, McdA) or Rosetta (ParA, FlhG, ParC). Binding interface was defined by those residues that shared a minimum of 1 Angstrom² of surface area.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

Bacterial genomes obtained from the NCBI RefSeq DataBase (https://www.ncbi.nlm.nih.gov/refseq/). Experimentally determined ParA/MinD ATPase structures obtained from the Protein Data Bank (PDB): ParA (5U1G); McdA (6NOP); ParC (5U1G); MinD (3Q9L); FlhG (4RZ3). All data generated or analyzed during this study are included in this published article and its supplementary information files. Source data are also provided with this paper. Source data are provided with this paper. All data generated or analyzed during this study are included in this published article and its supplementary information files. Source data are also provided with this paper. Source data are provided with this paper.

Code availability

The code generated during this study are available at GitHub: Identifying ParA/MinD ATPases in bacteria using BLAST (https://github.com/krthkkrv/Multiple-ParA-MinD-ATPases-coordinate-the-positioning-of-disparate-cargos-in-a-bacterial-cell)⁶³, Identifying specificity determinants of each ParA/MinD ATPase in H. neapolitanus using AlphaFold2 and Rosetta (https://github.com/jilimcaoco/Multiple-ParA-MinD-ATPases)⁶⁴.

References

Baxter, J. C. & Funnell, B. E. Plasmid partition mechanisms. Microbiol. Spectr. https://doi.org/10.1128/microbiolspec.PLAS-0023-2014 (2014).
Jalal, A. S. B. & Le, T. B. K. Bacterial chromosome segregation by the ParABS system. Open Biol. 10, 200097 (2020).
Article CAS PubMed PubMed Central Google Scholar
Lutkenhaus, J. Assembly dynamics of the bacterial MinCDE system and spatial regulation of the Z ring. Annu. Rev. Biochem. 76, 539–562 (2007).
Article CAS PubMed Google Scholar
Maccready, J. S. & Vecchiarelli, A. G. Positioning the model bacterial organelle, the carboxysome. MBio 12, e02519–e02519 (2021).
Article CAS PubMed PubMed Central Google Scholar
Savage, D. F., Afonso, B., Chen, A. H. & Silver, P. A. Spatially ordered dynamics of the bacterial carbon fixation machinery. Science 327, 1258–1261 (2010).
Article ADS CAS PubMed Google Scholar
Schuhmacher, J. S. et al. MinD-like ATPase FlhG effects location and number of bacterial flagella during C-ring assembly. Proc. Natl Acad. Sci. USA 112, 3092–3097 (2015).
Article ADS CAS PubMed PubMed Central Google Scholar
Schuhmacher, J. S., Thormann, K. M. & Bange, G. How bacteria maintain location and number of flagella? FEMS Microbiol. Rev. 39, 812–822 (2015).
Article CAS PubMed Google Scholar
Ringgaard, S., Schirner, K., Davis, B. M. & Waldor, M. K. A family of ParA-like ATPases promotes cell pole maturation by facilitating polar localization of chemotaxis proteins. Genes Dev. 25, 1544–1555 (2011).
Article CAS PubMed PubMed Central Google Scholar
Thompson, S. R., Wadhams, G. H. & Armitage, J. P. The positioning of cytoplasmic protein clusters in bacteria. Proc. Natl Acad. Sci. Usa. 103, 8209–8214 (2006).
Article ADS CAS PubMed PubMed Central Google Scholar
Atmakuri, K., Cascales, E., Burton, O. T., Banta, L. M. & Christie, P. J. Agrobacterium ParA/MinD-like VirC1 spatially coordinates early conjugative DNA transfer reactions. EMBO J. 26, 2540–2551 (2007).
Article CAS PubMed PubMed Central Google Scholar
Shan, S. ATPase and GTPase tangos drive intracellular protein transport. Trends Biochem. Sci. 41, 1050–1060 (2016).
Article CAS PubMed PubMed Central Google Scholar
Hester, C. M. & Lutkenhaus, J. Soj (ParA) DNA binding is mediated by conserved arginines and is essential for plasmid segregation. Proc. Natl Acad. Sci. USA 104, 20326–20331 (2007).
Article ADS CAS PubMed PubMed Central Google Scholar
Kiekebusch, D., Michie, K. A., Essen, L.-O., Löwe, J. & Thanbichler, M. Localized dimerization and nucleoid binding drive gradient formation by the bacterial cell division inhibitor MipZ. Mol. Cell 46, 245–259 (2012).
Article CAS PubMed PubMed Central Google Scholar
Szeto, T. H., Rowland, S. L., Rothfield, L. I. & King, G. F. Membrane localization of MinD is mediated by a C-terminal motif that is conserved across eubacteria, archaea, and chloroplasts. Proc. Natl Acad. Sci. USA 99, 15693–15698 (2002).
Article ADS CAS PubMed PubMed Central Google Scholar
Hu, Z. L. & Lutkenhaus, J. A conserved sequence at the C-terminus of MinD is required for binding to the membrane and targeting MinC to the septum. Mol. Microbiol 47, 345–355 (2003).
Article CAS PubMed Google Scholar
Vecchiarelli, A. G. et al. ATP control of dynamic P1 ParA-DNA interactions: a key role for the nucleoid in plasmid partition. Mol. Microbiol. 78, 78–91 (2010).
CAS PubMed PubMed Central Google Scholar
Koonin, E. V. A superfamily of ATPases with diverse functions containing either classical or deviant ATP-binding Motif. J. Mol. Biol. 229, 1165–1174 (1993).
Article CAS PubMed Google Scholar
Tilly, K., Checroun, C. & Rosa, P. A. Requirements for Borrelia burgdorferi plasmid maintenance. Plasmid 68, 1–12 (2012).
Article CAS PubMed PubMed Central Google Scholar
Livny, J., Yamaichi, Y. & Waldor, M. K. Distribution of centromere-like parS sites in bacteria: Insights from comparative genomics. J. Bacteriol. 189, 8693–8703 (2007).
Article CAS PubMed PubMed Central Google Scholar
Lutkenhaus, J. The ParA/MinD family puts things in their place. Trends Microbiol. 20, 411–418 (2012).
Article CAS PubMed PubMed Central Google Scholar
Vecchiarelli, A. G., Mizuuchi, K. & Funnell, B. E. Surfing biological surfaces: exploiting the nucleoid for partition and transport in bacteria. Mol. Microbiol. 86, 513–523 (2012).
Article CAS PubMed PubMed Central Google Scholar
Saha, C. K., Sanches Pires, R., Brolin, H., Delannoy, M. & Atkinson, G. C. FlaGs and webFlaGs: discovering novel biology through the analysis of gene neighbourhood conservation. Bioinformatics 37, 1312–1314 (2021).
Article CAS PubMed Google Scholar
Raskin, D. M. & De Boer, P. A. J. Rapid pole-to-pole oscillation of a protein required for directing division to the middle of Escherichia coli. Proc. Natl Acad. Sci. USA 96, 4971–4976 (1999).
Article ADS CAS PubMed PubMed Central Google Scholar
Yu, X. & Margolin, W. FtsZ ring clusters in min and partition mutants: role of both the Min system and the nucleoid in regulating FtsZ ring localization. Mol. Microbiol 32, 315–326 (1999).
Article CAS PubMed Google Scholar
Kerfeld, C. A., Aussignargues, C., Zarzycki, J., Cai, F. & Sutter, M. Bacterial microcompartments. Nat. Rev. Microbiol. 16, 277–290 (2018).
Article CAS PubMed PubMed Central Google Scholar
Turmo, A., Gonzalez-Esquer, C. R. & Kerfeld, C. A. Carboxysomes: metabolic modules for CO2 fixation. FEMS Microbiol. Lett. 364, 18 (2017).
MacCready, J. S. et al. Protein gradients on the nucleoid position the carbon-fixing organelles of cyanobacteria. Elife 7, e39723 (2018).
Article PubMed PubMed Central Google Scholar
MacCready, J. S., Tran, L., Basalla, J. L., Hakim, P. & Vecchiarelli, A. G. The McdAB system positions α-carboxysomes in proteobacteria. Mol. Microbiol. 00, 1–21 (2021).
Google Scholar
Campos-García, J., Nájera, R., Camarena, L. & Soberón-Chávez, G. The Pseudomonas aeruginosa motR gene involved in regulation of bacterial motility. FEMS Microbiol. Lett. 184, 57–62 (2000).
Article PubMed Google Scholar
Dasgupta, N., Arora, S. K. & Ramphal, R. fleN, a gene that regulates flagellar number in Pseudomonas aeruginosa. J. Bacteriol. 182, 357–364 (2000).
Article CAS PubMed PubMed Central Google Scholar
Gao, T., Shi, M., Ju, L. & Gao, H. Investigation into FlhFG reveals distinct features of FlhF in regulating flagellum polarity in Shewanella oneidensis. Mol. Microbiol. 98, 571–585 (2015).
Article CAS PubMed Google Scholar
Kusumoto, A. et al. Collaboration of FlhF and FlhG to regulate polar-flagella number and localization in Vibrio alginolyticus. Microbiol 154, 1390–1399 (2008).
Article CAS Google Scholar
Kusumoto, A. et al. Regulation of polar flagellar number by the flhF and flhG genes in Vibrio alginolyticus. J. Biochem. 139, 113–121 (2006).
Article CAS PubMed Google Scholar
Arroyo-Pérez, E. E. & Ringgaard, S. Interdependent Polar Localization of FlhF and FlhG and their importance for flagellum formation of vibrio parahaemolyticus. Front. Microbiol. 0, 557 (2021).
Google Scholar
Guttenplan, S. B., Shaw, S. & Kearns, D. B. The cell biology of peritrichous flagella in Bacillus subtilis. Mol. Microbiol. 87, 211–229 (2013).
Article CAS PubMed Google Scholar
Balaban, M. & Hendrixson, D. R. Polar flagellar biosynthesis and a regulator of flagellar number influence spatial parameters of cell division in Campylobacter jejuni. PLOS Pathog. 7, e1002420 (2011).
Article CAS PubMed PubMed Central Google Scholar
Gulbronson, C. J. et al. FlhG employs diverse intrinsic domains and influences FlhF GTPase activity to numerically regulate polar flagellar biogenesis in Campylobacter jejuni. Mol. Microbiol. 99, 291–306 (2016).
Article CAS PubMed Google Scholar
Chang, Y. et al. Molecular mechanism for rotational switching of the bacterial flagellar motor. Nat. Struct. Mol. Biol. 27, 1041–1047 (2020).
Article CAS PubMed PubMed Central Google Scholar
Kühn, M. J. et al. Spatial arrangement of several flagellins within bacterial flagella improves motility in different environments. Nat. Commun. 9, 1–12 (2018).
Article Google Scholar
Ringgaard, S. et al. ParP prevents dissociation of CheA from chemotactic signaling arrays and tethers them to a polar anchor. Proc. Natl Acad. Sci. USA 111, E255–E264 (2014).
Article CAS PubMed Google Scholar
Roberts, M. A. J., Wadhams, G. H., Hadfield, K. A., Tickner, S. & Armitage, J. P. ParA-like protein uses nonspecific chromosomal DNA binding to partition protein complexes. Proc. Natl Acad. Sci. USA 109, 6698–6703 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Kentner, D. & Sourjik, V. Dynamic map of protein interactions in the Escherichia coli chemotaxis pathway. Mol. Syst. Biol. 5, 238 (2009).
Article PubMed PubMed Central Google Scholar
Sourjik, V. & Berg, H. C. Localization of components of the chemotaxis machinery of Escherichia coli using fluorescent protein fusions. Mol. Microbiol. 37, 740–751 (2000).
Article CAS PubMed Google Scholar
Briegel, A. et al. Universal architecture of bacterial chemoreceptor arrays. Proc. Natl Acad. Sci. USA 106, 17181–17186 (2009).
Article ADS CAS PubMed PubMed Central Google Scholar
Leonard, T. A., Butler, P. J. & Löwe, J. Bacterial chromosome segregation: structure and DNA binding of the Soj dimer—a conserved biological switch. EMBO J. 24, 270–282 (2005).
Article CAS PubMed PubMed Central Google Scholar
Dunham, T. D., Xu, W., Funnell, B. E. & Schumacher, M. A. Structural basis for ADP‐mediated transcriptional regulation by P1 and P7 ParA. EMBO J. 28, 1792–1802 (2009).
Article CAS PubMed PubMed Central Google Scholar
Wu, W., Park, K.-T., Holyoak, T. & Lutkenhaus, J. Determination of the structure of the MinD-ATP complex reveals the orientation of MinD on the membrane and the relative location of the binding sites for MinE and MinC. Mol. Microbiol. 79, 1515–1528 (2011).
Article CAS PubMed PubMed Central Google Scholar
Schumacher, M. A. et al. Structural mechanism of ATP-induced polymerization of the partition factor ParF. J. Biol. Chem. 287, 26146–26154 (2012).
Article CAS PubMed PubMed Central Google Scholar
Schumacher, D., Harms, A., Bergeler, S., Frey, E. & Søgaard-Andersen, L. Pomx, a para/mind atpase activating protein, is a triple regulator of cell division in myxococcus xanthus. Elife 10, e66160 (2021).
Article CAS PubMed PubMed Central Google Scholar
Park, K.-T., Wu, W., Lovell, S. & Lutkenhaus, J. Mechanism of the asymmetric activation of the MinD ATPase by MinE. Mol. Microbiol. 85, 271–281 (2012).
Article CAS PubMed PubMed Central Google Scholar
Park, K.-T. et al. The min oscillator uses MinD-dependent conformational changes in MinE to spatially regulate cytokinesis. Cell 146, 396–407 (2011).
Article CAS PubMed PubMed Central Google Scholar
Leipe, D. D., Wolf, Y. I., Koonin, E. V. & Aravind, L. Classification and evolution of P-loop GTPases and related ATPases. J. Mol. Biol. 317, 41–72 (2002).
Article CAS PubMed Google Scholar
Nußbaum, P. et al. An oscillating MinD protein determines the cellular positioning of the motility machinery in Archaea. Curr. Biol. 30, 4956–4972.e4 (2020).
Article PubMed Google Scholar
Norman, A., Hansen, L. H. & Srensen, S. J. Conjugative plasmids: vessels of the communal gene pool. Philos. Trans. R. Soc. B Biol. Sci. 364, 2275–2289 (2009).
Article CAS Google Scholar
Schwille, P. & Diez, S. Synthetic biology of minimal systems. Crit. Rev. Biochem. Mol. Biol. 44, 223–242 (2009).
Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nat. Methods 9, 676–682 (2012).
Article CAS PubMed Google Scholar
Ducret, A., Quardokus, E. M. & Brun, Y. V. MicrobeJ, a tool for high throughput bacterial cell detection and quantitative analysis. Nat. Microbiol. 1, 1–7 (2016).
Article Google Scholar
Mirdita, M. et al. ColabFold—making protein folding accessible to all. Nat. Methods 19, 679–682 (2022).
Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Raveh, B., London, N., Zimmerman, L. & Schueler-Furman, O. Rosetta FlexPepDockab-initio: Simultaneous folding, docking and refinement of peptides onto their receptors. PLoS One 6, e18934 (2011).
Article ADS CAS PubMed PubMed Central Google Scholar
Li, S. C. & Ng, Y. K. Calibur: A tool for clustering large numbers of protein decoys. BMC Bioinforma. 11, 25 (2010).
Article Google Scholar
Barlow, K. A. et al. Flex ddG: Rosetta ensemble-based estimation of changes in protein-protein binding affinity upon mutation. J. Phys. Chem. B 122, 5389–5399 (2018).
Article CAS PubMed PubMed Central Google Scholar
Pulianmackal, L. T. et al. Multiple ParA/MinD ATPases coordinate the positioning of disparate cargos in a bacterial cell. GitHub https://doi.org/10.5281/zenodo.7941697 (2023).
Pulianmackal, L. T. et al. Multiple ParA/MinD ATPases coordinate the positioning of disparate cargos in a bacterial cell. GitHub https://doi.org/10.5281/zenodo.7941403 (2023).

Download references

Acknowledgements

We thank Joshua S. MacCready, Christopher A. Azaldegui, Joseph L. Basalla, Ajai J. Pulianmackal, and Claudia Mak for their thoughtful discussions. We thank Holly Turula and Miles Mckenna for their technical assistance. This work was supported by the National Science Foundation Award No. 1817478 (A.G.V.), the National Science Foundation Graduate Research Fellowship Program DGE 1841052 (L.T.P.), and from research initiation funds provided by the MCDB Department (A.G.V.).

Author information

Authors and Affiliations

Department of Microbiology and Immunology, University of Michigan, Ann Arbor, MI, 48109, USA
Lisa T. Pulianmackal
Department of Computational Medicine & Bioinformatics, University of Michigan, Ann Arbor, MI, 48109, USA
Jose Miguel I. Limcaoco & Matthew J. O’Meara
Department of Molecular, Cellular, and Developmental Biology, University of Michigan, Ann Arbor, MI, 48109, USA
Keerthikka Ravi, Sinyu Yang, Jeffrey Zhang, Mimi K. Tran, Maria Ghalmi & Anthony G. Vecchiarelli

Authors

Lisa T. Pulianmackal
View author publications
You can also search for this author in PubMed Google Scholar
Jose Miguel I. Limcaoco
View author publications
You can also search for this author in PubMed Google Scholar
Keerthikka Ravi
View author publications
You can also search for this author in PubMed Google Scholar
Sinyu Yang
View author publications
You can also search for this author in PubMed Google Scholar
Jeffrey Zhang
View author publications
You can also search for this author in PubMed Google Scholar
Mimi K. Tran
View author publications
You can also search for this author in PubMed Google Scholar
Maria Ghalmi
View author publications
You can also search for this author in PubMed Google Scholar
Matthew J. O’Meara
View author publications
You can also search for this author in PubMed Google Scholar
Anthony G. Vecchiarelli
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

Conceptualization, L.T.P. and A.G.V.; Methodology, L.T.P., M.J.L., K.R.; Formal Analysis, L.T.P. M.J.L., K.R., S.Y., M.G., M.J.O. and A.G.V.; Investigation, L.T.P., J.Z., and M.K.T.; Resources, A.G.V.; Writing—Original Draft, L.T.P. and A.G.V.; Visualization, L.T.P., J.M.I.L., K.R., and A.G.V.; Supervision, L.T.P., M.J.O., and A.G.V.; Funding Acquisition, L.T.P. and A.G.V.

Corresponding author

Correspondence to Anthony G. Vecchiarelli.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks Kai Thormann and the other, anonymous, reviewers for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Description of Additional Supplementary Files

Supplementary Data 1

Supplementary Data 2

Supplementary Data 3

Supplementary Movie 1

Supplementary Movie 2

Supplementary Movie 3

Supplementary Movie 4

Supplementary Movie 5

Reporting Summary

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Pulianmackal, L.T., Limcaoco, J.M.I., Ravi, K. et al. Multiple ParA/MinD ATPases coordinate the positioning of disparate cargos in a bacterial cell. Nat Commun 14, 3255 (2023). https://doi.org/10.1038/s41467-023-39019-x

Download citation

Received: 12 January 2023
Accepted: 22 May 2023
Published: 05 June 2023
DOI: https://doi.org/10.1038/s41467-023-39019-x

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.