SpoIID-Mediated Peptidoglycan Degradation Is Required throughout Engulfment during Bacillus subtilis Sporulation

SpoIID is a membrane-anchored enzyme that degrades peptidoglycan and is essential for engulfment and sporulation in Bacillus subtilis. SpoIID is targeted to the sporulation septum, where it interacts with two other proteins required for engulfment: SpoIIP and SpoIIM. We changed conserved amino acids in SpoIID to alanine to determine whether there was a correlation between the effect of each substitution on the in vivo and in vitro activities of SpoIID. We identified one amino acid substitution, E88A, that eliminated peptidoglycan degradation activity and one, D210A, that reduced it, as well as two substitutions that destabilized the protein in B. subtilis (R106A and K203A). Using these mutants, we show that the peptidoglycan degradation activity of SpoIID is required for the first step of engulfment (septal thinning), as well as throughout membrane migration, and we show that SpoIID levels are substantially above the minimum required for engulfment. The inactive mutant E88A shows increased septal localization compared to the wild type, suggesting that the degradation cycle of the SpoIID/SpoIIP complex is accompanied by the activity-dependent release of SpoIID from the complex and subsequent rebinding. This mutant is also capable of moving SpoIIP across the sporulation septum, suggesting that SpoIID binding, but not peptidoglycan degradation activity, is needed for relocalization of SpoIIP. Finally, the mutant with reduced activity (D210A) causes uneven engulfment and time-lapse microscopy indicates that the fastest-moving membrane arm has greater concentrations of SpoIIP than the slower-moving arm, demonstrating a correlation between SpoIIP protein levels and the rate of membrane migration.Endospore formation is an evolutionarily conserved process that allows Bacillus subtilis and related Gram-positive bacteria to adapt to changes in the environment, such as nutrient depletion. Many dramatic morphological changes occur during sporulation, each requiring a multitude of specialized proteins (reviewed in references 13 and 17). First, a sporulation septum is formed near one of the cell poles, forming two separate compartments of unequal sizes and with differing fates (Fig. ​(Fig.1A).1A). The smaller of the two, the forespore, will eventually become the spore, while the larger, the mother cell, will ultimately lyse. Next, the mother cell membranes move up and around the forespore in the poorly understood process of engulfment. Although this process is superficially similar to eukaryotic engulfment, it is complicated by the thick cell wall that surrounds and separates the two compartments. After engulfment, the migrating membranes pinch off from the mother cell membrane, thereby releasing the forespore into the cytoplasm of the mother cell, where it can be enveloped with protective coat proteins and eventually released into the environment as a mature spore. Sporulation provides an ideal, nonessential system for understanding how bacterial cells are capable of undergoing dramatic morphological changes.Open in a separate windowFIG. 1.Engulfment in B. subtilis. (A) (i) Engulfment begins with formation of an asymmetric septum that divides the cell into the forespore (FS) and mother cell (MC). SpoIID (orange pacman) and SpoIIP (green pacman) peptidoglycan degradation enzymes localize to the center of the septum. (ii) SpoIID and SpoIIP thin the septal peptidoglycan, starting from the center and moving toward the cell edges. SpoIIQ (purple ball) and SpoIIIAH (red ball) form a zipper across the septum, assembling foci behind the leading edges. (iii) The peptidoglycan degradation enzymes localize to the leading edges during membrane migration, while additional SpoIIQ-SpoIIIAH complexes assemble around the forespore. (iv) Engulfment membrane fission occurs at the top of the forespore, releasing the forespore into the mother cell cytoplasm. (B) Burnt-bridge Brownian ratchet model for membrane migration, adapted from earlier studies (1, 7). (C) Schematic representation of the SpoIID domain structure. The transmembrane domain (TM) and putative enzymatic domain, as defined by Pfam (14), are indicated. Amino acid numbers are below the schematic, and mutations causing in vivo phenotypes are indicated by an “X”.Engulfment involves dynamic protein localization and large-scale rearrangements of cellular membranes and peptidoglycan to accommodate internalization of the forespore. The physical basis for engulfment remains unclear, but two separate protein machineries that contribute to engulfment have been discovered. The first module involves the only three proteins known to be required for engulfment under all physiological conditions: SpoIID, SpoIIM, and SpoIIP (16, 24, 35). Zymography assays have demonstrated that both SpoIID and SpoIIP degrade peptidoglycan in vitro (1, 8), and this function is thought to be essential for engulfment in wild-type cells (1, 2, 8). SpoIID and SpoIIP are membrane-spanning proteins that directly interact both in vivo and in vitro, as demonstrated by coimmunoprecipitation and affinity chromatography techniques (2, 8). These studies failed to demonstrate an interaction between SpoIIM and either SpoIID or SpoIIP, perhaps because SpoIIM is an integral membrane protein. However, SpoIIM is required for localization of SpoIID and SpoIIP (2, 8), suggesting that all three proteins interact to form a peptidoglycan degradation module that is essential for engulfment.The second system influencing membrane migration is the SpoIIQ/SpoIIIAH zipper, which is required for engulfment only under certain conditions (2, 7, 38). SpoIIQ is produced in the forespore (23) and SpoIIIAH is produced in the mother cell (19). SpoIIQ and SpoIIIAH interact both in vitro and in vivo via their extracellular domains (6, 7, 10). Because these two proteins are produced in separate compartments, the only possible place for an interaction is the intermembrane space between the mother cell and forespore, forming a protein-protein zipper between the two cells. This zipperlike interaction is necessary for septal localization of SpoIIIAH and other mother cell proteins (6, 10, 18) and is capable of holding the two cells together when peptidoglycan is removed with lysozyme (7). Surprisingly, digestion of the peptidoglycan with lysozyme also allows membrane migration in about half of treated cells, in a process requiring the SpoIIQ/SpoIIIAH zipper but not the SpoIIDMP peptidoglycan degradation module. The SpoIIQ-SpoIIIAH zipper also contributes to engulfment in living cells, since strains lacking SpoIIQ or SpoIIIAH complete engulfment more slowly than the wild type and have synergistic engulfment defects when certain secondary mutations are introduced (2, 7, 38). Together, these results strongly support a role for the SpoIIQ/SpoIIIAH module in engulfment, demonstrating that the zipper contributes to the efficiency of membrane migration even when the SpoIIDMP module is present and functional. They also suggest that the engulfment machinery displays functional redundancy and that the zipper module provides a backup machinery for membrane migration.The precise role of the SpoIIDMP module during engulfment remains unclear. One model proposes that SpoIID and SpoIIP act as a burnt-bridge Brownian ratchet (Fig. ​(Fig.1B)1B) (1, 7). This model asserts that as SpoIID and SpoIIP degrade peptidoglycan, they eliminate their own enzymatic targets, resulting in the absence of substrate in one direction and therefore, overall movement in the opposite direction. As the enzymes move forward toward new targets, the mother cell membranes are dragged along with them because they are anchored in the membrane. This hypothesis predicts that SpoIID and SpoIIP are processive enzymes and that the SpoIIDMP complex could function as a motor, moving along peptidoglycan as a track and pulling the membranes with it (1, 7). A second model predicts that peptidoglycan degradation could simply remove a steric hindrance to membrane migration (such as links between the forespore membrane and the cell wall) and that some other mechanism provides the force required for membrane migration. Although the SpoIIQ-SpoIIIAH module can contribute to membrane migration, these proteins are not always essential for engulfment in intact cells (7, 38), suggesting that another unidentified system must generate the force required for membrane movement if the DMP module does not act as a burnt-bridge Brownian ratchet. Recent evidence suggests that peptidoglycan biosynthesis, which is localized to the leading edge of the engulfing membrane and necessary for membrane migration in the absence of the SpoIIQ-SpoIIIAH proteins, might be this missing force generating mechanism (26).Both models predict that the activities of SpoIID and SpoIIP are essential for membrane migration. This requirement has been demonstrated for SpoIIP (8) and, while this work was under review, for SpoIID. SpoIID shows no sequence similarity to any characterized enzyme that degrades peptidoglycan and thus constitutes the founding member of a new class of enzymes that remodel peptidoglycan (1, 27). However, SpoIID does show some similarity to B. subtilis LytB (24), a protein that enhances the activity of the amidase LytC (5, 20, 34), while SpoIIP is related to LytC (14). A recent study demonstrated that SpoIIP is both an amidase and endopeptidase and that SpoIID both activates SpoIIP and functions as a lytic transglycosylase, cleaving peptidoglycan between NAG and NAM (27). Together, these two enzymes degrade peptidoglycan into its smallest repeating subunits. However, it remains unclear which of the demonstrated or suggested biochemical functions of SpoIID are required for its various in vivo activities (interaction with SpoIIP, localization, septal thinning, and membrane migration), and it is unclear whether peptidoglycan degradation activity is required throughout engulfment or only for the initial stage of septal thinning.We use site-directed mutagenesis to test the role of 56 conserved amino acids in SpoIID, focusing on hydrophilic amino acids that might be involved in protein-protein interactions and peptidoglycan degradation. We identified one mutation (E88A) that eliminates and three others (R106A, K203A, and D210A) that reduce peptidoglycan degradation activity and show that SpoIID activity is required for the earliest stage of engulfment (septal thinning), as well as throughout membrane migration. Our results confirm and extend those of Morlot et al. (27) and also demonstrate that SpoIID activity is required throughout engulfment. Furthermore, our data indicate that the enzymatically inactive mutant protein (E88A) shows increased septal localization compared to the wild-type protein, suggesting that peptidoglycan degradation contributes to the release of SpoIID from the septum. We propose a modified model for the enzymatic cycle of the SpoIID and SpoIIP complex.