Introduction

Spores are metabolically dormant cells that can survive unfavorable conditions, including extremes of temperature, desiccation, and ionizing radiation (Huang & Hull, 2017, Hutchison et al., 2014). Sporulation, the process of spore development, is a strategy utilized by a wide variety of organisms, from bacteria and protozoa to fungi and plants. In addition to protective spore coats, spore resilience relies on cytosol dehydration and DNA compaction—key processes that decrease cell volume and alter cell shape (Higgins & Dworkin, 2012). Thus, morphological differentiation is a hallmark of sporulation. In bacterial spore formers, as peptidoglycan (PG) cell walls largely determine cell morphology, their sporulation requires profound PG remodeling.

PG is a rigid, mesh-like macromolecule that is composed of glycan strands of repeating units of N-acetyl glucosamine (GluNAc)-N-acetyl muramic acid (MurNAc) crosslinked by peptides (Egan et al., 2020). PG encloses the entire cytoplasmic membrane, and its rigidity provides mechanical support against osmotic stress. For this reason, PG is an essential structure for most bacteria and a major target for antibacterial treatments. Sporulation provides invaluable opportunities for understanding the dynamics of PG, which is controlled by multiple synthases and hydrolases. PG synthases include glycosyltransferases (GTases) that polymerize glycan strands and transpeptidases (TPases) that form peptide crosslinks (Egan et al., 2020). PG hydrolases, also known as autolysins, include glycosidases that break the glycan strands, and amidases and endo/carboxypeptidases that cleave the peptide crosslinks (Egan et al., 2020, Rohs & Bernhardt, 2021, van Heijenoort, 2011).

Firmicutes, including Gram-positive bacteria like Bacilli and Clostridia, produce endospores. Their morphological transition from rod-shaped vegetative cells to ovoid spores occurs through asymmetric cell division, resulting in the formation of a smaller forespore and a larger mother cell. Eventually, the forespore becomes an oval endospore after being engulfed by the mother cell (Higgins & Dworkin, 2012, McKenney & Eichenberger, 2012). A mature endospore contains two PG layers: the germ cell wall, derived from the forespore’s original PG, and a thickened PG cortex deposited by the mother cell (Popham & Bernhards, 2015). While a few Gram-negative spore formers also belong to the Firmicutes phylum, they share conserved sporulation genes with Bacilli and Clostridia, which suggests similar sporulation mechanisms (Tocheva et al., 2011, Yutin & Galperin, 2013).

Sporulation by non-firmicutes bacteria has been largely overlooked. Myxobacteria, a group of Gram-negative, non-firmicutes spore formers, were first assigned to the phylum δ-proteobacteria, but recently reclassified into the newly established phylum Myxococcota (Waite et al., 2020). Myxococcus xanthus is a model organism of myxobacteria. Rod-shaped M. xanthus cells can undergo sporulation via two distinct pathways, both resulting in spherical spores. First, in response to certain chemical signals, such as glycerol and dimethyl sulfoxide (DMSO), individual M. xanthus cells can rapidly transition into isolated spherical spores in aqueous environments within hours (Dworkin & Gibson, 1964). Second, millions of cells can aggregate on solid starvation media and develop into spore-filled fruiting bodies, a process that takes a few days to complete (O’Connor & Zusman, 1988). Both sporulation pathways are programmed, tightly controlled processes that involve over 1,000 genes (Muller et al., 2010, Munoz-Dorado et al., 2019). In contrast to endospore formation, cell division is not involved in either the M. xanthus sporulation mechanisms, and M. xanthus lacks the sporulation genes in Firmicutes (Aramayo & Nan, 2022). Thus, sporulating M. xanthus must accomplish the rod-to-sphere transition through yet to be discovered PG remodeling mechanisms, which provide invaluable opportunities to investigate PG dynamics (Zhang et al., 2021).

Distinct from Bacilli and Clostridia, M. xanthus spores lack cortex PG (Bui et al., 2009, Zhang et al., 2020, Voelz & Dworkin, 1962). Rather, their resistance is derived from polysaccharide coats (Kottel et al., 1975, Perez-Burgos et al., 2020, Saidi et al., 2021, Muller et al., 2012). Sporulating cells that fail to deposit coat polysaccharides on their surfaces produce spherical pseudospores that lack resistance (Wartel et al., 2013, Zhang et al., 2020, Holkenbrink et al., 2014). In a pioneering study, Bui et al. did not detect muropeptides in glycerol-induced spores, indicating that M. xanthus degrades its PG during rapid sporulation (Bui et al., 2009). However, it remains unclear whether starvation-induced spores within fruiting bodies still retain PG and how M. xanthus orchestrates programmed morphological changes.

In this report, we used ultra-performance liquid chromatography (UPLC) to demonstrate that M. xanthus degrades PG in both sporulation pathways. Through mutagenesis studies, we discovered that these sporulation pathways require different lytic transglycosylases (LTGs) for PG degradation. While the rapid sporulation relies on both LtgA and LtgB, the slow formation of fruiting body spores only requires LtgB. Remarkably, LtgB regulates the pace of PG degradation by LtgA during rapid sporulation, preventing the formation of non-resistant pseudospores due to abrupt PG breakdown. This research not only uncovers novel mechanisms of sporulation in non-firmicutes but also highlights the crucial role of cross-regulation between PG hydrolases in maintaining cell integrity.

Results

PG is degraded in both spore types

To investigate if starvation-induced spores in M. xanthus fruiting bodies retain PG, we broke the fruiting bodies after 120 h of starvation and purified saculli from spores. For comparison, we also purified the saculli from vegetative cells and glycerol-induced spores. PG contents in all three cell types were analyzed using UPLC after muramidase digestion (see materials and Methods). The two spore types showed similar profiles of a discernible presence of muropeptides that resembled those found in vegetative cells, albeit in significantly reduced quantities (Fig. 1). Importantly, the most abundant muropeptides identified in the vegetative cells, MurNAc-tetrapeptide monomer (M4) and MurNAc-tetrapeptide dimer (D44), were nearly absent in both spore types (Fig. 1). Such residual muropeptides in both spore types are inadequate for forming continuous PG layers (Bui et al., 2009). This observation indicates that spore development in both pathways involves the breakdown of the vegetative PG cell wall.

Fig. 1.

Despite the overall decline in muropeptides, anhydro-muropeptides markedly increased in both spore types, comprising over 90% of the total muropeptides (Fig. 1). Especially, anhydro-MurNAc-tetrapeptide trimer (T444^Anh) that was only detected in trace amounts in vegetative cells, became a prominent muropeptide in both spore types (Fig. 1). Moreover, new anhydro-muropeptides, including anhydro, anhydro-MurNAc-tetrapeptide dimer (D44^Anh,Anh), anhydro, anhydro-MurNAc-tetrapeptide trimer (T444^Anh,Anh), and an anhydrodimer without a NAG (D44^Anh(-NAG)), were only present in spores (Fig. 1). Because anhydro-muropeptides are the signature products of LTGs (Dik et al., 2017, Williams et al., 2018), their abundance in spores indicates that certain LTGs must play essential roles in M. xanthus sporulation.

The two sporulation pathways require different LTGs

The genome of M. xanthus encodes 14 putative LTGs (Aramayo & Nan, 2022, Ramirez Carbo et al., 2024). We imaged the cells of the 14 knockout mutants (Ramirez Carbo et al., 2024) at different time points after glycerol induction and used their length/width (L/W) ratios to monitor potential defects in the sporulation process. Wild-type cells initiated sporulation within 1 h of glycerol induction, which was evidenced by the increase of cell width and decrease of cell length (Fig. 2A). Consistent with the change in cell morphology, their L/W ratios decreased continuously and stabilized after 2 h of induction, when sporulation completed (Fig. 2B). In contrast, the cells that carry the deletion of open reading frame (ORF) K1515_20820 (MXAN_RS16290) (Aramayo & Nan, 2022) were able to shorten cell length slightly but retained rod shape after prolonged induction.

Fig. 2.

Surprisingly, another mutant that carries the deletion of ORF K1515_17460 (MXAN_RS19615) started to lose rod shape immediately after glycerol induction and became spherical within 1 h (Fig. 2A, 2B). However, different from the wild-type spores that appeared dark and heterogeneous under differential interference contrast (DIC) microscopy, the spheres formed by this mutant appeared bright and homogenous, similar to the pseudospores from the ΔaglQS mutant that lack the motor for depositing spore coat polysaccharides onto cell surfaces (Zhang et al., 2020) (Fig. 2A). To test whether these spheres are real spores, we subjected them to sonication and quantified their survival rate using a Helber bacterial counting chamber. After sonication, while 91 ± 6% (calculated from three independent experiments, n > 1,000, same below) of the wild-type spores appeared intact, only 3 ± 1% of the spheres produced by the K1515_17460 deletion mutant remained. Thus, this mutant indeed formed pseudospores that lacked the resistance against sonication. As the products of both K1515_20820 and K1515_17460 show homology to MltE, an LTG in Escherichia coli (encoded by emtA, Fig. 2D), we named them ltgA and ltgB, respectively. Both LtgA and LtgB are required for forming resistant spores via the rapid sporulation pathway, albeit playing opposite roles in the rod-to-sphere transition.

To identify the LTGs that are required for slow sporulation, we grew the 14 knockout strains in rich liquid media and spotted cells on solid starvation (CF) media. After 96 h of incubation, when wild-type cells formed dark fruiting bodies on the agar surface, the ΔltgB strain only formed flat aggregations (Fig. 2C). To test if ΔltgB cells could generate starvation-induced spores without forming fruiting bodies, we scraped such cell aggregations from the agar surface, suspended cells in rich liquid media, subjected them to sonication, then plated them on solid rich media. These cells failed to form colonies after five days of incubation, indicating that LtgB is essential for forming starvation-induced spores. In contrast, deleting ltgA did not affect fruiting body formation (Fig. 2C). Thus, the slow sporulation pathway only requires LtgB.

LtgA and LtgB play distinct roles in different sporulation pathways

Do LtgA and LtgB degrade PG during sporulation? To answer this question, we purified cell sacculi and imaged PG using immunofluorescence and an anti-PG serum (de Pedro et al., 1997). The sacculi of vegetative cells from the wild-type, ΔltgA, and ΔltgB strains were not visible under DIC microscope, likely due to their flattened structures minimizing light interference. However, PG from all three strains was readily detected in the fluorescence channel (Fig. 3A). After 6 h of glycerol-induction, the sacculi of both the wild-type and ΔltgA cells remained visible under DIC microscope (Fig. 3A), likely due to the deposition of spore coat polysaccharides that sustained unflattened cell structures. After the sacculus purification process, only background PG signals were detected in glycerol-induced wild-type spores (Fig. 3A). In contrast, the ΔltgA cells retained PG in substantial quantities (Fig. 3A). Sacculi from the ΔltgB pseudospores only contained a small amount of PG (Fig. 3A). Thus, compared to LtgA, the enzymatic activity of LtgB plays a more limited role in PG degradation during rapid sporulation. Different from the induced wild-type and ΔltgA cells, sacculi from the ΔltgB pseudospores were undetectable under DIC microscopy (Fig. 3A), reflecting the absence of rigid polysaccharide coats. Consequently, the residual PG in these pseudospores lost integrity during purification, with many sacculi displaying irregular shapes and ruptures in the fluorescence channel (Fig. 3A). Collectively, LtgB appears to be a pace-keeper that prevents abrupt PG degradation and thus allows sporulating cells to assemble polysaccharide coats during rapid sporulation.

Fig. 3.

To determine the roles of LtgA and LtgB in slow sporulation, we first scraped the wild-type and ΔltgA fruiting bodies from CF agar surface (Fig. 2C), dispersed spores by sonication, purified their sacculi, and visualized PG using immunofluorescence. Spherical spores from both the wild-type and ΔltgA cells were visible under DIC microscope, indicating that they retained unflattened shapes. These fruiting body spores lacked PG-specific fluorescence (Fig. 3A), likely due to both the absence of PG (Fig. 1) and the thickened polysaccharide coats (Voelz & Dworkin, 1962) preventing PG antibody access. Second, we investigated if ΔltgB cells degraded PG during starvation-induced sporulation. Because the flat aggregates formed by ΔltgB cells on CF agar did not contain mature spores, we scrapped cell aggregates from agar surface and purified their sacculi without sonication. Instead of forming spores, most of the ΔltgB cells were still rod-shaped, indistinguishable from vegetative ones (Fig. 3A). These observations indicate that LtgB is the only essential LTG for PG degradation during slow sporulation.

The two sporulation pathways vary greatly in duration: glycerol-induced rapid sporulation is completed within two hours, while starvation-induced slow sporulation unfolds over several days. Based on this distinction, we hypothesized that LtgA and LtgB could degrade PG at different rates. We used a vanillate-inducible promoter (Iniesta et al., 2012) to overexpress LtgA as a merodiploid. Induced by 200 µM vanillate, cells showed heterogeneous morphology, reflecting the variation in LtgA production. Many cells lost rod shape even in the absence of glycerol, indicating that these cells over-degraded their PG by LtgA (Fig. 2A, 3B). In contrast, the overexpression of LtgB using the same method did not affect the morphology of vegetative cells (Fig. 3B).

To further investigate the activities of these LTGs, we expressed the periplasmic domains of wild-type LtgA (amino acids 21 – 243) and LtgB (amino acids 26 – 710) in E. coli (Fig. 3C). We purified PG from wild-type M. xanthus cells, labeled it with Remazol brilliant blue (RBB), and tested if the purified AgmT variants hydrolyze labeled PG in vitro and release the dye (Jorgenson et al., 2014, Ramirez Carbo et al., 2024, Uehara et al., 2010). Wild-type LtgA solubilized dye-labeled PG, which absorbed light at 595 nm, demonstrating stronger hydrolytic activity than lysozyme, an enzyme that specifically cleaves β-1,4-glycosidic bonds in PG. In contrast, PG incubated with LtgB only showed minimum release of the dye, indicating slow PG hydrolysis (Fig. 3C). While we cannot rule out the possibility that our purification and reaction conditions were suboptimal for LtgB, both the in vitro RBB assay and the in vivo phenotype resulting from LtgB overexpression support our hypothesis that LtgA degrades PG at significantly higher rates than LtgB.

LtgB regulates LtgA during glycerol-induced sporulation

In line with its function in PG degradation, ltgA transcription increases approximately twofold during rapid sporulation but remains unchanged in slow sporulation. In contrast, ltgB shows a slight increase in expression during slow sporulation but remains unchanged in rapid sporulation (Muller et al., 2010, Munoz-Dorado et al., 2019). However, transcriptomic data alone do not account for the opposing roles of LtgA and LtgB in the rod-to-sphere transition during rapid sporulation. The resemblance in morphology between glycerol-induced ΔltgB cells and uninduced cells overproducing LtgA suggests that the slower-acting LtgB may offset the rapid activity of LtgA during sporulation, allowing sufficient time for spore coat polysaccharide deposition.

To test if LtgB regulates LtgA, we constructed a ΔltgA ΔltgB double deletion strain. Cells from this strain shortened their length slightly but failed to abolish rod shape after prolonged glycerol induction, phenocopying the sporulation defect of the ΔltgA strain (Fig. 2A, 2B). Hence, LtgB is a regulator upstream of LtgA.

LtgB exhibits a response to glycerol induction earlier than LtgA

Biochemical reactions on PG are unique for the stark size difference between the enzymes and their substrates. While the enzymes are in nanometer scales, their substrates, the PG sacculi, span several micrometers. Under microscope, PG remains stationary but PG-related enzymes are free to move. Even for M. xanthus that moves on surfaces, PG-related enzymes move at least two orders of magnitude faster than cell/PG (Zhang et al., 2023b, Ramirez Carbo et al., 2024). Thus, when diffusive enzymes bind to PG, their mobility decreases (Lee et al., 2016, Zhang et al., 2023b). By tracking single fluorescently-labeled enzyme particles, we can approximate their PG-binding in different physiological conditions and genetic backgrounds (Ramirez Carbo et al., 2024, Zhang et al., 2023b). We individually expressed photo-activatable mCherry (PAmCherry)-labeled LtgA and LtgB using their native loci and promoters. Both labeled LTGs accumulated as full-length proteins (Fig. S1A) and the PAmCherry tags did not affect the formation of either glycerol– or starvation-induced spores, indicating that these fusion proteins were fully functional (Fig. S1B, C). Consistent to the transcriptomic data (Muller et al., 2010, Munoz-Dorado et al., 2019), vegetative cells produce significantly less LtgA than LtgB (Fig. S1A).

We used a 405-nm excitation laser (0.3 kW/cm², 0.1 s) to activate the fluorescence of a few labeled LTG particles randomly in each cell and quantified their localization using a 561-nm laser at 10 Hz using single particle tracking photo-activated localization microscopy (sptPALM, see Materials and Methods) (Fu et al., 2018, Nan et al., 2015, Nan et al., 2013). As free PAmCherry particles diffuse extremely fast in the cytoplasm, entering and exiting the focal plane frequently, they usually appear as blurry objects that cannot be followed at 10 Hz close to the cell surface (Fu et al., 2018, Zhang et al., 2023b). For this reason, the noise from free PAmCherry due to potential protein degradation was negligible.

Single-particles of PAmCherry-labeled LtgA and LtgB can be categorized into two populations, immobile and mobile. The immobile particles remained within a single pixel (160 nm × 160 nm) before photobleach, and the mobile ones displayed typical diffusion (Fig. 4A). For the particles that switched between mobile and immobile states, our algorithm categorized them as mobile and calculated their diffusion coefficients (D) from their entire trajectories that contained both mobile and immobile segments. Hence, binding to PG not only increases the immobile population of the enzyme particles but also decreases the D of the mobile particles. In vegetative cells where large-scale PG degradation does not occur, 23.9% (n = 1175) and 18.2% (n = 824) of LtgA and LtgB particles were immobile, respectively (Fig. 4B). D values of the mobile population are 2.62 × 10^-2 ± 2.0 × 10^-3 µm²/s (n = 894) for LtgA-PAmCherry and 2.80 × 10^-2 ± 2.50 × 10^-3 µm²/s (n = 674) for LtgB-PAmCherry (Fig. 4C).

Fig. 4.

We then determined how the dynamics of both LTGs respond to glycerol induction in the rapid sporulation pathway. Immediately (1 min) after adding glycerol, the dynamics of LtgA remained little changed, when 21.0% (n = 928) of particles were immobile and the D of the mobile ones was 2.53 × 10^-2 ± 0.19 × 10^-3 µm²/s (n = 733) (Fig. 4B, 4C). In contrast, the mobility of LtgB decreased significantly, with the immobile population increased to 28.0% (n = 1359) and D of the mobile population decreased to 2.17 × 10^-2 ± 2.75 × 10^-3 µm²/s (n = 978) (Fig. 4B, 4C). Therefore, these results indicate that upon glycerol induction, LtgB rapidly enhances its binding to PG.

We then chose 30 min after glycerol induction as a time point for rapid PG degradation, which is reflected by the dramatic decrease of L/W during the first hour of sporulation (Fig. 2A, 2B). Compared to its inert response 1 min after glycerol induction, the mobility of LtgA particles decreased significantly at 30 min, when the immobile population increased to 31.5% (n = 1163) and D of the mobile population decreased to 1.80 × 10^-2 ± 2.12 × 10^-3 µm²/s (n = 797) (Fig. 4B, 4C). The reduced mobility of LtgA aligns with its function as the primary LTG in glycerol-induced sporulation. Strikingly different from LtgA, as sporulation advanced to the 30-min mark, LtgB’s mobility returned to its pre-induction level, with the immobile population decreased to 10.2% (n = 1549) and D of the mobile population increased to 2.87 × 10^-2 ± 2.76 × 10^-3 µm²/s (n = 1391) (Fig. 4B, 4C). Therefore, the PG-binding of LtgB shows a negative correlation with PG degradation. Taken together, LtgA and LtgB display opposite responses to glycerol induction, in which LtgB rapidly binds to PG before yielding to LtgA, whose PG-binding is concurrent with the rod-to-sphere transition. The sequential PG-binding by these two LTGs controls the pace of PG-degradation during glycerol-induced sporulation.

LtgB blocks LtgA from binding PG in the early stage of glycerol-induced sporulation

Does LtgB’s early PG-binding suppress PG degradation by LtgA? To answer this question, we expressed LtgA-PAmCherry using the native ltgA locus and promoter in the ΔltgB background. In vegetative cells, the absence of LtgB significantly reduced the mobility of single LtgA-PAmCherry particles, suggesting that LtgB does affect LtgA’s binding to the PG (Fig. 4B, 4C). Similarly, we expressed LtgB-PAmCherry using the native ltgB locus and promoter in the ΔltgA background. In contrast, the mobility of single LtgB-PAmCherry particles only decreased slightly in the absence of LtgA (Fig. 4B, 4C). These results suggest that while LtgB significantly reduces LtgA’s binding to PG, probably due to its higher expression level, LtgA has only a modest impact on LtgB’s ability to bind PG.

Strikingly different from its slow response to glycerol in wild-type cells, LtgA increased its PG-binding immediately (1 min) after glycerol induction in the ΔltgB background, confirming that LtgB restricts LtgA access to the PG in the early stage of sporulation (Fig. 4B, 4C). In contrast, at 30 min of induction, a time point when LtgB dissociated from PG, its absence no longer affected LtgA’s PG-binding (Fig. 4B, 4C). Similar to the observation in wild-type cells, the lack of LtgA did not affect the PG-binding of LtgB at either 1 min or 30 min after glycerol induction (Fig. 4B, 4C). In summary, LtgB, a slow LTG, regulates the progression of rapid sporulation by preventing LtgA, a fast LTG, from excessively degrading PG, thereby ensuring that sporulating cells can deposit polysaccharide spore coats before PG disintegration occurs.

Upregulated PG synthesis negatively affects PG degradation

Since LtgA is less mobile in vegetative ΔltgB cells, i.e. strongly binds to PG (Fig. 4B, C), why do these cells retain their rod shape rather than losing it spontaneously before glycerol induction? In addition to PG-binding, LtgA may require additional stimuli to initiate PG degradation. We hypothesized that, similar to certain antibiotics targeting PG synthases, which induce wild-type cells to degrade PG and form pseudospores in rich liquid media (Zhang et al., 2023b, O’Connor & Zusman, 1997), glycerol may alter cellular metabolism, leading to reduced PG synthesis. If this is the case, cells with upregulated PG synthesis should remain rods even in the presence of glycerol.

To test our hypothesis, we used the vanillate-inducible promoter (Iniesta et al., 2012) to overexpress murA as a merodiploid in the wild-type background. Because MurA catalyzes the first committed step of PG synthesis that produces UDP-MurNAc (Egan et al., 2020, Rohs & Bernhardt, 2021), elevated MurA levels are expected to channel more cellular resources toward PG synthesis. Overproduction of MurA in the presence of 200 µM vanillate resulted in a heterogeneous cell population, with normal cells coexisting alongside elongated ones (Fig. 2A, B). This heterogeneity may reflect the heterogeneous induction of murA across the whole population. Overall, excessive MurA increased the average length of vegetative cells by 19.6%. Cells overproducing MurA also displayed heterogeneity during glycerol-induced sporulation. While some cells transitioned into spheres, many retained their rod shape even after 6 h of induction (Fig. 2A, B). Notably, sporadic lysis of rod-shaped cells began 2 h post-induction and became increasingly frequent with continued incubation (Fig. 2A). Potentially, elevated muropeptide production may lead to the accumulation of toxic intermediates that the relatively insufficient LTGs failed to degrade (Weaver et al., 2022). Nevertheless, the capacity of MurA overexpression to enable certain cells to circumvent glycerol-induced sporulation implies that PG production acts as a regulatory cue for PG degradation.

To test if excessive MurA can also reduce PG degradation during slow sporulation, we induced MurA overexpression on solid CF agar containing 100 µM vanillate. After 96 h of incubation, these cells developed fruiting bodies that were noticeably larger and flatter compared to those of the wild-type strain (Fig. 2C). To test if such fruiting bodies contained mature spores, we scraped them from the agar surface and tested the germination rate of spores on CYE agar after sonication. The fruiting bodies formed by cells overexpressing MurA yielded 21.3% of the colonies produced by an equivalent number of wild-type cells, indicating that upregulated PG synthesis also prevents slow sporulation. In contrast, cells grown without vanillate progressed normally through both sporulation pathways, showing no differences compared to wild-type cells (Fig. S2). Therefore, the sporulation of M. xanthus is sensitive to PG synthesis and increased PG synthesis balances PG degradation in both sporulation pathways.

Discussion

It is still unknown how many bacteria outside of the Firmicutes phylum form true spores. Nevertheless, our findings elucidate a new mechanism of non-firmicute bacterial sporulation that depends on PG degradation. This mechanism presents a significant challenge for sporulating cells: preserving their structural integrity while simultaneously dismantling the primary framework that upholds it. Our findings revealed that besides regulating the production of LTGs, cells could control the pace of PG degradation in a rapid sporulation pathway through the regulation between two LTGs and hence allow the resistant polysaccharide coats to assemble and mature.

Studying the regulation between two LTGs is especially challenging both in vivo and in vitro for a few reasons. First, it is difficult to monitor their instantaneous activities in vivo. Secondly, muropeptide analysis may not distinguish their roles because their products bear the same signature, anhydro-MurNAc. Thirdly, some enzymes require special substrates or activators, which would be difficult to study in vitro. To overcome these technical hurdles, we leveraged single-particle mobility to quantify the PG-binding of these enzymes. The simultaneous occurrence of reduced LtgA mobility and PG degradation during glycerol-induced sporulation indicates that the molecular dynamics of LtgA accurately mirrors its enzymatic activity. Such correlation between decreased particle mobility and increased enzymatic activity applies to many other PG-related enzymes, including multiple PG polymerases in E. coli and the endopeptidase DacB in M. xanthus (Lee et al., 2016, Zhang et al., 2023b, Yang et al., 2021). Because the size difference between PG and its related enzymes is a conserved feature in all bacteria, our method using single-particle tracking to quantify PG-binding can be applied in many organisms.

We discovered that LtgB, a slower-acting LTG, limits the access of the faster LtgA to the PG during the early phase of rapid sporulation, thereby regulating the rate of its degradation. Then how does LtgA breach the blockage of LtgB and gain access to PG as sporulation progresses? Because the overexpression of LtgA is sufficient to damage PG in uninduced cells (Fig. 3B), we propose that the relative abundance of these two LTGs plays a critical role in determining the fate during fast sporulation. While LtgB nearly saturates PG binding sites in vegetative cells where few LtgA are produced, LtgA— upregulated after glycerol induction (Muller et al., 2010) —gradually outcompetes LtgB and thereby causes LtgB to disassociate from PG.

As increased PG synthesis enables many cells to bypass both sporulation pathways, it is reasonable to propose that diminished PG synthesis initiates PG degradation by LTGs. While functional coordination between PG synthesis and hydrolysis has been speculated for over 50 years (Koch, 1985, Koch, 1990), its mechanisms only came to light recently. Some endopeptidases are proposed to serve as the space-makers for PG synthases (Dorr et al., 2013, Singh et al., 2012). In M. xanthus, we reported that moenomycin, an antibiotic that specifically binds to the GTase domains of class A penicillin-binding proteins (aPBPs) (Sung et al., 2009, Lovering et al., 2007), specifically activates a PG endopeptidase DacB through PBP1a2, an aPBP (Zhang et al., 2023b). Because moenomycin mimics a growing glycan strand in the GTase domains of aPBPs, it actually locks aPBPs in their glycan-charged conformations (Sung et al., 2009, Lovering et al., 2007). Thus, similar to the moenomycin-bound form, glycan-charged PBP1a2 in physiological conditions recruits DacB to the PG assembly sites, which in turn, generates openings in the existing PG network as the crosslink sites for the TPase activity of PBP1a2. Such “make-before-break” mechanism (Koch & Doyle, 1985) could prevent unneeded hydrolysis and thus maintain PG integrity.

Compared to the well-characterized space-making functions of endopeptidases, the roles of LTGs, beyond PG recycling, remain ill-defined. Even less is known on their potential coordination with PG synthases (Weaver et al., 2023). Because LTGs from bacterial and phage origins are inhibited by L,D-crosslinks, they may functionally connect to PBPs that form D,D-crosslinks (Alvarez et al., 2024). Additionally, some LTGs, such as MltD and MltG in E. coli and MltG in Vibrio cholerae, prefer to degrade nascent, uncrosslinked glycan strands (Kaul et al., 2024, Weaver et al., 2022, Bohrhunter et al., 2021). Our recent report indicates that AgmT, an M. xanthus LTG homologous to E. coli MltG, detoxifies uncrosslinked glycan strands under antibiotic stress and modifies crosslinked PG scaffolds to attach a motility machinery to PG (Ramirez Carbo et al., 2024). Then how does reduced PG synthesis activate LtgA in M. xanthus? Although MltE, the closest homolog of LtgA in E. coli, degrades glycan strands regardless of their crosslinking status or peptide content, it shows a marked preference for uncrosslinked substrates (Dik et al., 2017, Fibriansah et al., 2012). Similar to AgmT and MltE, LtgA might also bind to both crosslinked and uncrosslinked glycan strands. Consistent with its high mobility, LtgA may coordinate with diffusive PG polymerases and cleave the uncrosslinked glycan strands. During rapid sporulation, as nascent glycan strands decline due to muropeptide depletion, LtgA increasingly binds to the crosslinked PG scaffold, which is evidenced by its reduced mobility. This binding intensifies further as PG degradation continues, establishing a positive feedback loop. In this case, LtgB serves as a brake that dampens this positive feedback, moderates PG degradation, and thus allows sporulating cells to maintain integrity by assembling polysaccharide coats. Such cross-regulation between PG hydrolases, which is currently under investigation, provides yet another mechanism by which bacteria dynamically maintain their PG cell walls.

Materials and methods

Bacterial strains and growth conditions

Vegetative M. xanthus cells were grown in liquid CYE medium (10 mM MOPS pH 7.6, 1% (w/v) Bacto™ casitone (BD Biosciences), 0.5% yeast extract and 8 mM MgSO₄) at 32 °C, in 125-ml flasks with vigorous shaking, or on CYE plates that contains 1.5% agar. We used strain DZ2 as the wild-type M. xanthus strain (Campos & Zusman, 1975). Knock-out mutants were constructed by electroporating DZ2 cells with 4 µg of plasmid DNA. Transformed cells were plated on CYE plates supplemented with 100 mg/ml sodium kanamycin sulfate or 10 mg/ml tetracycline. The strains and plasmids used in this study are listed in Table 1.

Table 1.

Sporulation and spore purification

Cells were grown in 25 ml liquid cell culture to OD₆₀₀ 0.8 – 1.2. Glycerol was added to 1 M to induce sporulation. The liquid culture was incubated at 32 °C with vigorous shaking. After 24 h, spores and cells were harvested by centrifugation (10 min, 10,000 g and 25 °C). To further purify the wild-type spores, the pellet was resuspended in 5 ml water, the remaining vegetative cells were eliminated by sonication (Cole Palmer 4710 ultrasonic homogenizer, 30% output, 10 cycles), and sonication-resistant spores were washed three times with water and collected by centrifugation (5 min, 10,000 g and 4 °C). The elimination of vegetative cells was confirmed by DIC microscopy.

For phenotypic assays on starvation-induced fruiting body formation, vegetative cells (10 ul), at a concentration of 4 × 10⁹ colony-forming units (cfu) ml^-1, were spotted on CF (0.015% Casitone, 0.2% sodium citrate, 0.1% sodium pyruvate, 0.02% (NH₄)₂SO₄, 10 mM MOPS (pH 7.6), 8 mM MgSO₄ and 1 mM KH₂PO₄) plates containing an agar concentration of 1.5%, incubated at 32°C. To purify fruiting body spores, cells were grown in 25 ml liquid cell culture to OD₆₀₀ 0.8 – 1.2 and harvested by centrifugation (10 min, 10,000 g and 25 °C). The pellet was resuspended in 1 ml water and plated on two 150 mm CF plates containing 1.5% agar using a spreader. The plates were air-dried and incubated at 32 °C for 120 h. For the wild-type and ΔltgA strains, fruiting bodies were scraped from the plates, suspended in 2 ml water, and subjected to sonication (Cole Palmer 4710 ultrasonic homogenizer, 60% output, 20 cycles). Spores were wash three times in water and collected by centrifugation (5 min, 10,000 g and 4 °C). For the ΔltgB strains, the sonication step was omitted.

Peptidoglycan purification and UPLC analysis

For vegetative cell peptidoglycan analysis, samples were processed as previously described for Gram negative bacteria (Alvarez et al., 2016, Desmarais et al., 2013). In brief, M. xanthus cells were grown until mid-stationary phase and harvested by centrifugation (30 min, 8,000 g). Supernatant was discarded and the pellet was resuspended and boiled in 1x PBS with 5% SDS for 2 h. The saccules were repeatedly washed with MilliQ water by ultracentrifugation (150,000 g, 10 min, 20°C) to remove the remaining SDS. The samples were then treated with muramidase (100 µg/mL) for 16 hours at 37 °C. Muramidase digestion was stopped by boiling and coagulated proteins were removed by centrifugation (10 min, 15,000 g). The supernatants were first adjusted to pH 8.5-9.0 with sodium borate buffer and then sodium borohydride was added to a final concentration of 10 mg/mL. After reducing the samples at room temperature for 30 mins, the pH was adjusted to pH 3.5 with orthophosphoric acid.

UPLC analyses of muropeptides were performed on a Waters UPLC system (Waters Corporation, USA) equipped with an ACQUITY UPLC BEH C18 Column, 130Å, 1.7 µm, 2.1 mm X 150 mm (Waters, USA) and a dual wavelength absorbance detector. Elution of muropeptides was detected at 204 nm. Muropeptides were separated at 45 °C using a linear gradient from buffer A (formic acid 0.1% in water) to buffer B (formic acid 0.1% in acetonitrile) in a 25-min run, with a 0.30 mL/min flow. For vegetative cells analysis, relative total PG amounts were calculated by comparison of the total intensities of the chromatograms (total area) from three biological replicas normalized to the same initial biomass and extracted with the same volumes. Quantification of muropeptides was based on their relative abundances (relative area of the corresponding peak).

Imaging and data analysis

For all imaging experiments on isolated spores/cells, we spotted 5 μl of cells grown in liquid CYE medium to OD₆₀₀ ∼1 on agar (1.5%) pads and imaged using a Andor iXon Ultra 897 EMCCD camera (effective pixel size 160 nm) on an inverted Nikon Eclipse-Ti™ microscope with a 100✕ 1.49 NA TIRF objective. Fruiting bodies were photographed after a 120-h incubation on CF agar using a Nikon SMZ1000 microscope and an OMAX A3590U digital camera. Cell morphology was imaged at indicated time points using DIC microscopy. The geometric aspect ratios (L/W) of spores/cells were determined using a custom algorithm written in MATLAB, which is available in the GitHub repository, https://github.com/NanLabMyxo/Rod_shape_paper (Zhang et al., 2020, Zhang et al., 2023a, Zhang et al., 2023b).

For sptPALM, M. xanthus cells were grown in CYE to 4 ×10⁸ cfu/ml, spotted on 1.5% agar pads and subjected to highly inclined and laminated optical sheet (HILO) illumination (Fu et al., 2018, Nan et al., 2015, Nan et al., 2013, Tokunaga et al., 2008). PAmCherry was activated using a 405-nm laser (0.3 kW/cm²), excited and imaged using a 561-nm laser (0.2 kW/cm²). Images were acquired at 10 Hz. For each sptPALM experiment, single PAmCherry particles were localized in at least 100 individual cells from three biological replicates. sptPALM data were analyzed using a MATLAB (MathWorks) script, which is available in the GitHub repository, https://github.com/NanLabMyxo/Rod_shape_paper (Zhang et al., 2020, Zhang et al., 2023a, Zhang et al., 2023b). Briefly, cells were identified using differential interference contrast images. Single PAmCherry particles inside cells were fit by a symmetric 2D Gaussian function, whose center was assumed to be the particle’s position (Fu et al., 2018). Particles that explored areas smaller than 160 nm × 160 nm (within one pixel) in 0.4 – 1.2 s were considered immotile (Fu et al., 2018, Zhang et al., 2023a, Zhang et al., 2023b). D of all the mobile particles was determined from a linear fit to the first four points of the MSD using a formula MSD = 4DΔt (Fu et al., 2018, Lee et al., 2016). Error bars were the standard derivation of 1,000 bootstrap samples using the published method (Morgenstein et al., 2015). Sample trajectories were generated using the TrackMate (Ershov et al., 2022) plugin in the ImageJ suite (https://imagej.net).

Immunofluorescence

Cell samples were resuspended and boiled in 1x PBS with 5% sodium dodecyl sulfate (SDS) for 2 h. The saccules were repeatedly washed with water by centrifugation (5 min, 15,000 g and 25 °C) to remove the remaining SDS. Remaining proteins were further removed by incubation with proteinase K (1 µg/ml) for 2 h at 37 °C. Microscope cover slides were prepared by coating with 100 µl 0.1% poly-L-lysine for 5 min at room temperature (RT), washing with ddH₂O and air drying. 100 µl samples were spotted onto the slides and incubated for 20 min at RT. The slides were then washed three times with phosphate buffer saline (PBS; 81 mM Na₂HPO₄, 15 mM KH₂PO₄, 1.37 M NaCl, 27 mM KCl, pH 7.4), and then incubated with anti-PG primary antibody¹⁶ diluted 1:1000 in PBS at RT for 1 h. Slides were washed 6 times with PBS and then incubated with Alexa Fluor 488 conjugated goat-anti-rabbit secondary antibody (Invitrogen) at a 1:1000 dilution in PBS at RT for 1 h in the dark. Slides were washed 6 times with PBS. Each cover glass was applied to a supporting slide, fixed with nail polish, and stored at 4 °C, if necessary. Images were recorded using a 488-nm laser.

Protein expression and purification

DNA sequences encoding amino acids 21 – 243 of LtgA and 26 – 710 of LtgB were amplified by polymerase chain reaction (PCR) and inserted into the pET28a vector (Novagen) between the restriction sites of EcoRI and HindIII and used to transform E. coli strain BL21(DE3). Transformed cells were cultured in 20 ml LB (Luria-Bertani) broth at 37 °C overnight and used to inoculate 1 L LB medium supplemented with 1.0% glucose. Protein expression was induced by 0.1 mM IPTG (isopropyl-h-d-thiogalactopyranoside) when the culture reached an OD₆₀₀ of 0.8. Cultivation was continued at 16 °C for 10 h before the cells were harvested by centrifugation at 6,000 × g for 20 min. Proteins were purified using a NGC™ Chromatography System (BIO-RAD) and 5-ml HisTrap™ columns (Cytiva) (Pogue et al., 2018, Nan et al., 2010). Purified proteins were concentrated using Amicon™ Ultra centrifugal filter units (Millipore Sigma) with a 10-kDa molecular weight cutoff and stored at –80 °C.

LTG activity (RBB) assay

PG was purified following the published method (Zhang et al., 2023b, Alvarez et al., 2016, Ramirez Carbo et al., 2024). In brief, M. xanthus cells were grown until mid-stationary phase and harvested by centrifugation (6,000 × g, 20 min, 25 °C). Supernatant was discarded and the pellet was resuspended and boiled in 1× PBS with 5% SDS for 2 h. SDS was removed by repetitive wash with water and centrifugation (21,000 × g, 10 min, 25 °C). Purified PG from 100 ml culture was suspended into 1 ml 1× PBS and stored at –20 °C. RBB labelling of PG was performed essentially as previously described (Uehara et al., 2010, Jorgenson et al., 2014). Purified sacculi were incubated with 20 mM RBB in 0.25 M NaOH overnight at 37 °C. Reactions were neutralized by adding equal volumes of 0.25 M HCl and RBB-labeled PG was collected by centrifugation at 21,000 × g for 15 min. Pellets were washed repeatedly with water until the supernatants became colorless. RBB-labelled sacculi were incubated with purified LtgA and LtgB (1 mg/ml) at 25 °C for 12 h. lysozyme (1 mg/ml) was used as a positive control. Dye release was quantified by the absorption at 595 nm from the supernatants after centrifugation (21,000 × g, 10 min, 25 °C).

Acknowledgements

This work was supported by the National Institutes of Health grants GM129000 to B. N. Research in the F. C. laboratory is supported by the Swedish Research Council, the Laboratory for Molecular Infection Medicine Sweden (MIMS), Umeå University, the Knut and Alice Wallenberg Foundation (KAW) and the Kempe Foundation.

Additional files

Fig. S1, S2