Defining the order of assembly of the Clostridioides difficile divisome complex

14 Cell division is the ancient pathway by which bacteria synthesize a septum of peptidoglycan, 15 dividing the cell into two. Where all walled bacteria were previously thought to use FtsW -FtsI 16 orthologs to synthesize septal peptidoglycan during division, we recently discovered that the 17 major pathogen Clostridioides difficile is missing FtsW -FtsI and instead relies on the activity of 18 the bifunctional Class A PBP called PBP1 to synthesize the septal peptidoglycan. Furthermore, 19 C. difficile either does not encode or require the majority of canonical divisome proteins 20 described in model bacteria aside from the divisome protein orthologs FtsZ, SepF, and ZapA. 21 Indeed, unlike model systems, SepF and ZapA are essential in C. difficile, suggesting that they 22 have evolved to have a critical function in cell division without the redundant mechanisms 23 present in model organisms. Thus, C. difficile uses a fundamentally different division mechanism 24 compared to previously studied bacteria. To understand how this unusual complex is assembled 25 in C. difficile, we combine CRISPR interference (CRISPRi) -based knock-downs with fluorescent 26 fusions to determine that the hierarchical order of assembly occurs in three phases: (i ) 27 FtsZ/ZapA, ( ii) SepF, and (iii) PBP1. We further investigate the order of assembly of several 28 non-essential mid-cell localizing proteins and discover that MldA, MldC, DivIVA, FtsK, and PBP3 29 depend on FtsZ, SepF, and PBP1 for localization, whereas MldB localizes independently of 30 SepF and PBP1. Our work provides a model for divisome assembly in C. difficile and validates 31 genetic and cytological tools that can be used to mechanistically dissect this pathway in the 32 future. 33 34 IMPORTANCE: 35 Bacterial cell division has been extensively studied in model systems, but little is known about 36 how this essential process occurs in the clinically important pathogen Clostridioides difficile . 37 Studies in model systems have shown that cell division is carried out by a large multi -protein 38 complex called the “divisome.” While components of the divisome are widely conserved and can 39 be traced back to the last bacterial common ancestor billions of years ago, C. difficile uses a 40 unique mechanism of division that is independent of the majority of these canonical divisome 41 genes. In the current study, we characterize the core, essential divisome comprised of FtsZ, 42 ZapA, SepF, and PBP1, and build a model for the order of assembly of this unusual divisome 43 complex. 44 45 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 3

46 The essential process of bacterial cell division is driven by a large multi -protein complex known 47 as the divisome. Divisome assembly centers around the tubulin -like protein FtsZ, which 48 assembles into a ring at the mid -cell and serves as a scaffold to organize the divisome complex 49 (1–3). Divisome assembly is a multi -step process that broadly occurs in three stages: (i) FtsZ 50 and other early divisome proteins mark the site of division, ( ii) transmembrane proteins are 51 recruited, including the septal peptidoglycan (PG) synthases, and ( iii) when assembly is 52 complete, the PG synthases are activated, leading to septum synthesis, which drives 53 cytokinesis. 54 55 In the model rod -shaped Firmicute species Bacillus subtilis , the localization of FtsZ is 56 coordinated by multiple divisome -associated proteins that promote bundling of FtsZ filaments 57 and/or tethering of the filaments to the membrane, including ZapA, SepF, FtsA, EzrA, and GpsB 58 (4–14). These non-essential proteins have partially overlapping functions to promote FtsZ -ring 59 assembly and /or membrane anchoring and are therefore each individually dispensable ; 60 however, they become synthetically lethal in higher order mutant combinations (4, 6 –9, 11, 12, 61 15, 16). 62 63 In B. subtilis and other model systems, proper assembly of the FtsZ proto -ring in the cytosol 64 allows this cytoskeletal scaffold to recruit trans -envelope proteins, including the regulatory sub -65 complex comprised of FtsQ (DivIB), FtsL, and FtsB (DivIC) (17 –22). The FtsQ -FtsL-FtsB sub-66 complex promotes the recruitment and activation of the FtsW -FtsI septal peptidoglycan 67 synthase complex at the site of division (23 –36). Each of these proteins is essential for division 68 in standard laboratory conditions in B. subtilis (18, 20, 37, 38), and the genes encoding FtsQ, 69 FtsL, FtsW, and FtsI have been traced to the last bacterial common ancestor billions of years 70 ago (39). Indeed, the FtsW -FtsI synthase complex was previously considered essential for 71 septum synthesis in virtually all bacteria with a cell wall (3, 23, 39, 40). 72 73 Despite the extreme conservation of divisome components across most bacteria (39), w e 74 recently showed that C. difficile exhibits marked differences in the composition of its divisome 75 relative to B. subtilis and previously studied bacteria. Specifically , we found that C. difficile does 76 not encode orthologs of FtsW and FtsI for mediating vegetative cell division (41). Additionally, 77 the genes encoding the typically essential divisome regulators FtsQ, FtsL, and FtsB are 78 completely dispensable for vegetative cell division and have evolved to function specifically 79 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 4 during sporulation (41) . In lieu of FtsW -FtsI, we showed that C. difficile uses its sole class A 80 penicillin binding protein (aPBP), PBP1, to drive the synthesis of the septum during vegetative 81 cell division (41). The essential function for an aPBP during vegetative cell division contrasts 82 with previously studied bacteria, where aPBP enzymes are important for fortifying the cell wal l 83 (42–44), filling in gaps in the peptidoglycan mesh (45 –48), reinforcing the septum synthesized 84 by FtsW-FtsI (49–51), or driving cell elongation at the poles (52 –57), depending on the species. 85 Thus, the essential role of PBP1 in synthesizing vegetative division septa in C. difficile 86 represents to our knowledge the first example of an aPBP driving cell division in the absence of 87 the canonical FtsW-FtsI synthase complex. 88 89 Intriguingly, C. difficile encodes orthologs of ZapA and SepF, but it lacks identifiable orthologs of 90 the FtsA, EzrA, or GpsB divisome proteins found in the Firmicutes, suggesting that regulation of 91 FtsZ in C. difficile differs substantially from B. subtilis . Since the majority of the expected 92 divisome proteins in C. difficile are either missing from the genome (FtsA, EzrA, GpsB, FtsW, 93 FtsI), or are dispensable for division (FtsQ, FtsL, FtsB), the molecular details of how C. difficile 94 carries out cell division remain largely unknown. 95 96 Recent work has shown that at least 24 different proteins localize to mid -cell in C. difficile (41, 97 58–65), of which 8 were predicted to be essential by transposon -insertion sequencing (63, 66) . 98 Four of these proteins are likely essential for reasons unrelated to cell division (RodA, PBP2, 99 MreC, and MreD), although it is possible they play an auxiliary function in the division apparatus 100 (41, 63). Therefore, of the essential, mid -cell localizing proteins, we are primarily interested in 101 the predicted proto-ring components, FtsZ, ZapA, and SepF, and the primary septum synthase, 102 PBP1. 103 104 While FtsZ has been localized to mid -cell and is essential for division in C. difficile (60, 67), its 105 regulation remains poorly understood. C. difficile ZapA also localizes to mid-cell and is essential 106 for division (61, 63), although th is essentiality appears to be unique to C. difficile because it is 107 dispensable in most previously studied bacteria. While ZapA has been shown to promote FtsZ 108 filament bundling and FtsZ -ring stability by mediating lateral interactions between filaments in 109 model systems (4, 68 –70), it is likely non-essential due to the presence of alternative 110 mechanisms to bundle FtsZ filaments and promote FtsZ -ring stability (4, 71 –77). While these 111 redundant mechanisms presumably do not exist in C. difficile, the precise mechanism by which 112 ZapA regulates the divisome complex has not yet been examined. 113 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 5 114 In addition to ZapA, the C. difficile SepF ortholog localizes to mid-cell (63) and is predicted to be 115 essential (63, 66) , but its requirement for cell division in C. difficile has not yet been examined . 116 In model systems, SepF interacts with FtsZ and associates with the membrane, serving as an 117 FtsZ-membrane anchoring protein (5, 6, 11) . However, in Firmicutes species that encode other 118 membrane anchors such as FtsA, SepF is not required for stable FtsZ -ring assembly, although 119 SepF-deficient cells exhibit septal irregularities (5, 78) . Interestingly, SepF is considered the 120 ancestral FtsZ membrane anchor, as bacterial and archaeal species that lack other membrane 121 anchors, such as FtsA, often encode a SepF ortholog (79); for example, it is essential for FtsZ -122 ring assembly in Actinobacteria species that lack FtsA (80 –82). Since SepF is surprisingly not 123 required for stable FtsZ -ring assembly in the archeal species in which it has been stud ied, 124 additional unknown mechanisms for stabiliz ing and tethering the FtsZ ring likely exist in those 125 Archaea (83). 126 127 In this study, we use CRISPR -interference (CRISPRi) and fluorescent fusion proteins to FtsZ, 128 ZapA, SepF, and PBP1 to investigate the assembly of the C. difficile divisome. Using CRISPRi -129 compatible trans -complementation, we develop genetic tools to assess the function of 130 fluorescently-tagged divisome proteins. Additionally, w e investigate the order of assembly of C. 131 difficile’s core essential divisome proteins and examine the role of several non -essential 132 auxiliary cell division proteins in this pathway. These studies provide insight into the non-133 canonical divisome used by C. difficile to mediate vegetative cell division and establish a 134 framework for future studies of this important process. 135 136

137 SepF is essential for septum synthesis in C. difficile 138 Based on Tn-seq data, there are four genes encoding known or predicted divisome proteins that 139 are predicted to be essential in C. difficile: ftsZ, zapA, sepF, and pbp1 (63, 66). Prior work using 140 CRISPRi-based knock-down (KD) show ed that ftsZ-KD, zapA-KD, and pbp1-KD each result in 141 cells with a long, filamentous morphology (41, 63, 67), which is a hallmark phenotype of cells 142 that cannot divide (84 –87). Since ftsZ, zapA, and pbp1 are each either transcribed as a sole 143 gene or are the last gene in their operon, based on RNA sequencing and transcription start site 144 mapping (88), the filamentation phenotypes can be attributed to the specific knock -down of 145 these genes . However, because the sepF gene is embedded within an operon that contains 146 ylmD, ylmE, sepF, ylmG, ylmH, and divIVA (Fig. 1A), the role of sepF in C. difficile division has 147 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 6 yet to be directly addressed thus far . While prior work that CRISPRi -KD of the entire sepF-148 containing operon using a sgRNA targeting ylmD induced cell filamentation (63) , the specific 149 gene(s) in this operon that are required for cell division remain unclear. 150 151 While sepF, ylmG, and divIVA all encode proteins known to localize to the site of division (63, 152 65), we focused on the potential role of sepF due to its known role in division in other species 153 (5–7, 79 –81, 83, 89) . To directly test if sepF is required for cell division, we used a xylose-154 inducible CRISPRi system targeting sepF combined with CRISPRi-compatible complementation 155 (62, 90, 91) . We have used this type of approach to complement CRISPRi -KD of essential 156 genes in C. difficile (62), and similar approaches have been used in Borrelia burgdorferi and 157 Mycobacterium tuberculosis (90, 91). 158 159 CRISPRi-KD of sepF caused a block in cell division, resulting in long filamentous cells defective 160 in septum formation, as indicated by an absence of HADA incorporation into septa within the 161 filaments ( Fig. 1B ). Since repression of sepF is also expected to knock- down expression of 162 ylmG, ylmH, and divIVA based on the predicted operon structure , we then test ed if 163 complementing with sepF alone could restore septum synthesis in the KD strain. To do this, we 164 integrated a sepF complementation construct into the genome under the control of an 165 anhydrotetracycline (aTc)-inducible Ptet promoter (92) containing synonymous point mutations in 166 the sgRNA-targeted sequence, rendering the construct “immune” to CRISPRi-targeting (sepFim). 167 We then titrated the aTc concentration to identify the lowest concentration of aTc that enabled 168 complementation of the KD . At 2.5 ng/mL aTc, induction of sepFim restored septum synthesis in 169 the sepF-KD strain (Fig. 1B). However, complemented cells developed a chaining phenotype . 170 Therefore, while our data demonstrates that sepF is specifically essential for septum synthesis , 171 cell chaining in the sepF-KD/sepFim complementation strain suggests that decreased ylmG, 172 ylmH, or divIVA expression likely prevents efficient cell separation. 173 174 Chaining has been reported in mutant strains of Streptococcus pneumoniae , S. suis , and 175 Listeria monocytogenes lacking divIVA (93–97). We therefore examined whether knocking down 176 divIVA is sufficient to induce cell chaining using CRISPRi. We found that divIVA-KD 177 phenocopied the cell chaining in the sepF-KD/sepFim complemented strain (Fig 1C), suggesting 178 that the chaining phenotype in this strain is likely caused by a lack of divIVA expression. 179 Consistent with this conclusion, we found that DivIVA protein levels are decreased in the sepF-180 KD similar to the divIVA-KD (Fig 1D), verifying that sepF-KD strain is depleted of DivIVA due to 181 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 7 polar effects on the operon. Together, our findings strongly suggest that SepF is critical for 182 septum synthesis, while reduced DivIVA levels in the complementation of sepF-KD with sepFim 183

in a secondary cell chaining phenotype. 184 185 The FtsZ-ZapA-SepF proto-ring assembles at mid-cell earlier than PBP1 186 To compare the localization of these proteins to the C. difficile divisome, we integrated the 187 expression constructs in to the genome under control of an aTc -inducible promoter and titrated 188 the aTc concentration to identify a level of fusion protein production that recapitulates their 189 septal localization without inducing morphological abnormalities. We found that low -level 190 expression of ftsZ-mScI3, mScI3-zapA, sepF-mScI3, and mScI3-pbp1 allowed all the fusion 191 proteins to localize to mid-cell (Fig. 2A), similar to prior studies (41, 60–63). When we quantified 192 the enrichment of the proteins at the mid -cell relative to the sidewall or cytosol, we found that 193 FtsZ, ZapA, and SepF fusions were highly enriched (≥5 -fold) at the mid -cell, whereas PBP1 194 exhibited a modest mid -cell enrichment of ~1.8 -fold (Fig S1). Transmembrane proteins that are 195 enriched at mid-cell are expected to be enriched by more than 2 -fold above the sidewall, as the 196 mid-cell will have two membranes once the septa are complete. Thus, our data suggests that 197 PBP1 is localized throughout the cell, which may be consistent with PBP1 having a role in both 198 the synthesis of the septum and the sidewall (41, 67). Additionally, by western blot analysis, we 199 found that there is some level of cleavage of the mScI3 fluorescent fusion from each of these 200 proteins (Fig S2), which likely also confounds our ability to precisely quantify protein localization 201 within the cell. With these minor limitations in mind, we used these fluorescent fusions to learn 202 about the relative order of assembly of the divisome complex in C. difficile. 203 204 We used the MicrobeJ plugin in FIJI to generate demographs, which sorts cells based on their 205 length, to visualize the medial fluorescence profile of each fluorescently -tagged protein across 206 hundreds of cells. This allowed us to visualize the mid- cell mScI3 signal in cells at various 207 stages of division. By monitoring HADA incorporation at mid -cell, it is possible to estimate 208 whether the fluorescently -tagged protein localizes to the mid -cell prior to or concurrently with 209 septum synthesis (98). The FtsZ -mScI3, mScI3-ZapA, and SepF -mScI3 fusions all localized to 210 the mid -cell prior to the onset of septum synthesis ( Fig. 2B ), consistent with these proteins 211 being early components of the divisome complex. In contrast, mScI3 -PBP1 localized to mid-cell 212 coincident with septum synthesis, as we reported in our prior work (62). These findings are 213 consistent with FtsZ, ZapA, and SepF comprising the early proto -ring of the C. difficile divisome 214 that assembles prior to recruitment of the trans-envelope proteins including PBP1 that ultimately 215 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 8 trigger septum synthesis. 216 217 Assessing the functionality of fluorescent protein fusions to essential divisome proteins 218 Although our fluorescent protein fusion constructs suggest that these divisome proteins localiz e 219 to mid-cell in C. difficile (Fig 2A), similar to prior work (41, 60 –63), it was unclear whether these 220 fluorescent protein fusions are functional. Assessing the functionality of these tagged proteins is 221 important for understanding whether there are limitations to using these fusions should they be 222 found to be non-functional. 223 224 To determine the functionality of FtsZ -mScI3, mScI3 -ZapA, SepF -mScI3, and mScI3 -PBP1 225 fluorescent fusions, we used the CRISPRi trans-complementation system described above . 226 Specifically, we combined xylose-inducible CRISPRi-KD cassettes with aTc inducible CRISPRi -227 “immune” complementation constructs encoding the fluorescent protein fusions. Each fusion 228 carries a (GGGGS)3 linker between the mScI3 and protein of interest. While complementation of 229 ftsZ-KD with the WT control ftsZim construct reversed the filamentation phenotype caused by 230 ftsZ-KD (Fig. 3A-B), complementation with ftsZim-mScI3 did not, indicating that the FtsZ -mScI3 231 fusion is not functional (Fig. 3B). Notably, FtsZ -fluorescent protein fusions have often been 232 found to be either non -functional (99) or only partially functional , with low temperatures being 233 required for their proper function (8) in many model systems, so the localization of these fusions 234 is typically analyzed in a merodiploid background. 235 236 In contrast, expression of mScI3-zapAim rescued synthesis of septa and partially rescued the 237 filamentation phenotype caused by zapA-KD (Fig. 3C). Since complementation with the WT 238 zapAim only partially rescued the filamentation phenotype, our ectopic zapA complementation 239 cassette may not perfectly match the expression levels and/or regulation found at the 240 endogenous gene locus. Notably, the mScI3-ZapA did not form distinct foci in the cells , perhaps 241 due to over -expression of the fusion, mask ing the discrete mid -cell localization. We also 242 observed the occasional formation of unusual spiral -like septa with the HADA label during 243 conditional expression of mScI3-zapAim (Fig. 3C, yellow arrow)(Fig. S3). Thus, while our data 244 indicate that the mScI3 -ZapA fusion can partially restore septum synthesis , it can also cause 245 abnormal septum synthesis, perhaps due to incomplete bundling of the FtsZ-ring that directs the 246 septum synthesis machinery. Regardless, these data indicate that mScI3 -ZapA localization to 247 mid-cell must be studied in a merodiploid background. 248 249 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 9 Notably, c onditional expression of sepFim-mScI3 in the sepF-KD strain enabled synthesis of 250 division septa ( Fig. 3D), similar to the sepF-KD/sepFim complementation mutant ( Fig. 3D)(Fig. 251 1B), although neither sepFim construct was able to rescue the chaining phenotype that is 252 presumably caused by the polar effects on divIVA expression. Therefore, these data indicate 253 that SepF-mScI3 is at least partially functional. Finally, conditional expression of mScI3-pbp1im 254 also restored normal septum synthesis in the pbp1-KD strain (Fig. 3E), strongly suggesting that 255 the mScI3 -PBP1 protein fusion is also functional. Nevertheless, since fluorescent protein 256 fusions to ZapA, SepF, or PBP1 undergo low levels of cleavage of mScI3 from the fusions, it is 257 possible that the small amount of untagged protein produced is responsible for driving the 258 complementation phenotype (Fig S2). While ruling out this possibility will require additional work 259 to identify fluorescent fusions that are not cleaved, our data thus far suggest that mScI3-ZapA, 260 SepF-mScI3, and mScI3-PBP1 are partially or fully functional in C. difficile, whereas FtsZ-mScI3 261 is clearly not functional. 262 263 Hierarchical recruitment of C. difficile divisome proteins 264 We next sought to determine the order of assembly of these essential divisome proteins 265 because in many systems the recruitment of the divisome proteins is hierarchical, where the 266 recruitment of each subsequent protein to the complex is dependent on the proper assembly of 267 the earlier proteins (22, 100). To examine the order of assembly of C. difficile’s core divisome 268 complex, we analyzed the localization of the fluorescent fusion proteins in which ftsZ, zapA, 269 sepF, or pbp1 expression was knocked-down using CRISPRi. While our analyses indicated that 270 ZapA, SepF, and PBP1 fluorescent fusions are at least partially functional (Fig 3 ), we 271 nevertheless decided to visualize these fusions in the context of a fully functional divisome 272 complex by employ ing the commonly -used dilute -label approach, in which the localization of 273 fluorescently-tagged proteins is studied in a merodiploid background (101, 102). 274 275 When we knocked -down ftsZ expression, the mScI3 -ZapA, SepF -mScI3, and mScI3 -PBP1 276 fusions all failed to assemble into distinct foci , unlike in control cells (Fig. 4A-B). Therefore, the 277 assembly of ZapA, SepF, and PBP1 at mid -cell depends on the presence of FtsZ , similar to 278 previously studied model systems (4, 5, 103) . When zapA expression was knocked-down, FtsZ-279 mScI3 and mScI3-PBP1 also failed to localize to distinct foci, suggesting that ZapA is critical for 280 FtsZ and PBP1 localization (Fig. 4B ). This is somewhat surprising, as ZapA in well -studied 281 model systems is dispensable for division due to the presence of redundant mechanisms for 282 promoting FtsZ-ring stability (4, 71 –77). Thus, ZapA appears to have evolved a uniquely critical 283 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 10 function in driving assembly of the divisome complex in C. difficile. This unique function likely 284 explains why this gene is essential in C. difficile despite being dispensable in previously studied 285 bacterial systems. Intriguingly, we observed that SepF -mScI3 localized to distinct foci in zapA-286 KD cells (Fig. 4B ). Since the SepF-mScI3 puncta were irregularly spaced and less well defined 287 than the crisp foci observed in control cells ( Fig. 4A ), SepF appears to only partially depend 288 upon ZapA to assemble into foci at mid -cell, while FtsZ and ZapA are co -dependent for 289 assembly into foci (Fig. 4C). 290 291 We next analyzed the localization dependency of divisome proteins in the absence of SepF. W e 292 found that FtsZ -mScI3 and mScI3 -ZapA could assemble into foci in sepF-KD cells, albeit at a 293 lower frequency along the cell length than in control cells (Fig. 4B ). These data indicate that 294 SepF depends on FtsZ and partially on ZapA for assembly at mid -cell, but both FtsZ and ZapA 295 can assemble in the absence of SepF. Thus, SepF is downstream of FtsZ and ZapA in the 296 divisome assembly pathway (Fig. 4C ). We also found that mScI3 -PBP1 localized solely to the 297 sidewall in the absence of SepF and failed to form the distinct foci observed in control cells (Fig. 298 4B). Therefore, SepF is required for PBP1 localization to the proto -ring complex, despite being 299 dispensable for the assembly of the underlying FtsZ -ZapA complex. Th is loss of PBP1 300 localization likely explains why septum synthesis is blocked in the absence of SepF. Finally, we 301 assessed the localization of divisome proteins during pbp1-KD. These analyses revealed that 302 depletion of PBP1 does not prevent the assembly of FtsZ -mScI3, mScI3-ZapA, and SepF -303 mScI3 into foci, confirming that PBP1 recruitment occurs downstream of the proto -ring 304 components FtsZ, ZapA, and SepF in the assembly pathway (Fig. 4C). 305 306 Localization profile for non -essential divisome proteins MldA, MldB, MldC, DivIVA, FtsK, 307 and PBP3 308 While these analyses revealed the localization dependencies of the essential divisome proteins 309 FtsZ, ZapA, SepF, and PBP1, numerous other proteins have been shown to localize to the site 310 of division that are not predicted to be essential in C. difficile (59, 63–65). Notably, non-essential 311 genes may still play important roles during C. difficile cell division despite being dispensable for 312 division in standard laboratory conditions for multiple reasons, including having a subtle 313 regulatory function, being redundant with other genes, or exhibiting conditional essentiality. We 314 therefore generated mScI3 fusions to several non -essential mid -cell localizing proteins, 315 including MldA, MldB, MldC, DivIVA, FtsK, and PBP3. 316 317 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 11 MldA is single -pass transmembrane protein that carries an extracellular SPOR domain and a 318 cytosolic globular domain and is encoded in an operon with MldB and MldC (64). These proteins 319 are only encoded in close relatives of C. difficile, and their function remains unclear (64). 320 321 DivIVA is a late-stage division protein widely conserved across the Firmicutes, and it is known to 322 localize to sites of negative membrane curvature (95, 104 –106). C. difficile DivIVA has been 323 shown to localize to septa (65) , although heterologous expression in B. subtilis and biochemical 324 interaction studies suggest that C. difficile DivIVA behaves differently from previously 325 characterized systems (107, 108). Despite these analyses, DivIVA function in C. difficile has not 326 been systematically examined. 327 328 FtsK is a single-pass transmembrane protein important for divisome assembly and chromosome 329 segregation in many model systems (109 –114). FtsK has also been localized to the division 330 apparatus in C. difficile (59, 63), although it is not strictly required for C. difficile cell division (63, 331 66). Finally, PBP3 is a non -essential class B PBP that we recently found non -catalytically 332 promotes the activity of PBP1 (62). While each of these proteins have been shown to localize to 333 the site of division (59, 62 –65), whether they localize to the mid -cell prior to, during, or after the 334 onset of septum synthesis has not yet been determined, with the exception of PBP3, which we 335 found localizes concurrently with septum synthesis (62). 336 337 To analyze the localization of MldA, MldB, MldC, DivIVA, FtsK, and PBP3 in C. difficile, we used 338 a similar dilute -labeling approach where constructs encoding mScI3 fusions to each of these 339 proteins were integrated in to the genome under control of the aTc -inducible Ptet promoter (Fig. 340 5A). Each fusion construct encodes either a GSAGSAAGSGKL linker (for MldA, MldB, and 341 MldC) or (GGGGS)3 linker (for DivIVA, FtsK, and PBP3). To visualize the mid-cell localization for 342 each of these non -essential mid -cell localizing proteins in relation to the onset of septum 343 synthesis, we again generated demographs analyzing the localization of these proteins as a 344 function of cell l ength (Fig 5B). We found that all six proteins appear to localize to the mid -cell 345 later in the cell cycle than FtsZ -mScI3, mScI3-ZapA, or SepF-mScI3 (Fig 2B)(Fig 5B), with the 346 localization of mScI3 -MldB, mScI3-MldC, DivIVA-mScI3, FtsK-mScI3, and mScI3 -PBP3 to mid -347 cell being largely coincident with the appearance of septa, as detected by HADA labeling ( Fig 348 5B). In contrast, mScI3-MldA showed strong mid-cell localization prior to clear HADA labeling of 349 septa (Fig 5B), suggesting that MldA localization to mid-cell occurs either prior to or early at the 350 onset of septum synthesis. 351 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 12 352 Dependence of MldA, MldB, MldC, DivIVA, FtsK, and PBP3 on FtsZ, SepF, and PBP1 for 353 mid-cell localization 354 Since the four, core essential divisome proteins are recruited stepwise in a hierarchical order 355 consisting of (i) FtsZ and ZapA, ( ii) SepF, and (iii) PBP1 (Fig 4C), we next sought to identify the 356 localization dependencies for the non -essential divisome proteins . To this end, we localiz ed 357 fluorescent protein fusions of the non -essential divisome proteins upon KD of ftsZ, sepF, or 358 pbp1, which represent the three stages of C. difficile divisome complex assembly. We found that 359 none of the non -essential divisome proteins localized in ftsZ-KD cells, and upon sepF-KD, no 360 foci were observed for mScI3-MldA, mScI3-MldC, DivIVA-mScI3, mScI3-FtsK, and mScI3-PBP3 361 (Fig 6A). Intriguingly, mScI3-MldB formed foci in sepF-KD cells, suggesting that MldB depends 362 on FtsZ but not SepF for localization. 363 364 The localization hierarchy was more difficult to assess during pbp1 KD because some pbp1-KD 365 cells form septa with associated divisome protein foci. We reasoned that septal co -366 colocalization was most likely due to incomplete depletion of PBP1, since only ~75% PBP1 is 367 depleted during the CRISPRi KD based on prior results (62). To determine if a protein was 368 capable of localizing independently of PBP1, we therefore looked specifically for divisome 369 protein foci that did not co-localize with septa. It should be noted that this approach was further 370 complicated by the heterogeneity in HADA labeling observed during pbp1 KD, with a proportion 371 of pbp1-KD cells failing to label with HADA entirely (Fig 6A)(62); thus, the presence or absence 372 of septa in some cells could not be determined . Despite these limitations, we found that mScI3 -373 MldA, mScI3-MldC, DivIVA-mScI3, mScI3-FtsK, and mScI3-PBP3 all formed mid -cell foci in the 374 pbp1-KD cells that were either co-localized with septa or in cells that did not label with HADA 375 (Fig 6A). Only mScI3 -MldB could form foci that were not clearly associated with septa (yellow 376 arrows, Fig 6A ), consistent with MldB not being dependent on PBP1 for recruitment to the 377 divisome complex. Collectively, our data support that MldA, MldC, DivIVA, FtsK, and PBP3 are 378 likely dependent on the presence of FtsZ, SepF, and PBP1 to be recruited to the divisome, 379 whereas MldB is dependent only on FtsZ ( Fig 6B). Thus, with the exception of MldB, the non -380 essential divisome proteins MldA, MldC, DivIVA, FtsK, and PBP3 appear to depend on septum 381 synthesis to localize to mid-cell. 382 383 To more precisely analyze the role of these non -essential proteins during cell division, we 384 generated clean, in -frame deletion mutants for each of the genes encoding non -essential 385 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 13 divisome proteins analyzed above. This genetic strategy allowed us to avoid the potentially off -386 target or polar effects caused by the genetic strategies previously used to study many of the se 387 non-essential divisome genes, including Targetron disruption (MldA, MldB, or MldC) (64) , 388 antisense RNA (FtsK) (59), or transposon-insertion (63, 66). 389 390 The ΔmldA, ΔmldB, ΔmldC, ΔdivIVA, ΔftsK, and Δpbp3 mutants all complete d cell division and 391 form septa (Fig 7A, yellow arrows), consistent with these genes being dispensable for division. 392 However, we found that clean deletion of ΔmldA, ΔmldB, and ΔmldC led to a mild chaining 393 phenotype (Fig 7A ), similar to prior Targetron mutants targeting mldA and mldB (64) . 394 Importantly, we could complement the ΔmldA mutant and reverse its chaining phenotype by 395 expressing mldA from an ectopic site in the genome ( Fig 7B). The ΔdivIVA deletion mutant also 396 grew as short chains ( Fig 7A ), phenocopying the CRISPRi divIVA-KD mutant ( Fig 1C ). The 397 chaining phenotype could be complement ed by expressing divIVA under the control of the 398 constitutive P cwp2 promoter from an ectopic site. Intriguingly, although an ftsK-antisense KD 399 mutant had previously been shown to exhibit a filamentation phenotype (59), we found that a 400 ΔftsK clean deletion mutant had a WT morphology ( Fig 7A ). Together, our data support that 401 these non-essential divisome proteins are not required for septum synthesis and likely play an 402 accessory and/or regulatory role in C. difficile division. 403 404

405 Despite the highly conserved nature of cell division in bacteria (39), the majority of canonical 406 divisome proteins characterized in bacteria are either missing from C. difficile’s genome (FtsW, 407 FtsI, FtsA, EzrA, GpsB) or their function has not been preserved during vegetative cell division 408 (FtsQ, FtsL, FtsB). Since this essential process remain ed poorly characterized in C. difficile, in 409 this study, we gained insight into how C. difficile assembles its divisome by determining the 410 order of assembly of four essential divisome proteins FtsZ, ZapA, SepF, and PBP1, along with 6 411 non-essential mid-cell localizing proteins. We also genetically defined the specific requirement 412 for SepF and show that several non -essential mid-cell localization proteins are dispensable for 413 septum synthesis. 414 415 These analyses revealed that ZapA plays a critical role during C. difficile FtsZ-ring formation 416 (Fig 4), in contrast with most model systems studied to date, where ZapA is dispensable for this 417 process due to the presence of multiple redundant FtsZ -bundling mechanisms (4, 71 –77). 418 Similarly, while SepF is often dispensable in Firmicutes species due to the presence of 419 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 14 additional FtsZ membrane tethers (5, 6, 78, 115), we found that SepF is essential for septum 420 synthesis in C. difficile (Fig 1 ). Our studies further revealed that FtsZ/ZapA rings form in the 421 absence of SepF in C. difficile (Fig 4 ). Since the septum synthase PBP1 depends on SepF to 422 be recruited to marked division sites (Fig 4), these analyses suggest that the primary function of 423 SepF in C. difficile is to link the cytoskeletal FtsZ/ZapA scaffold to late-stage divisome proteins, 424 including PBP1 . Thus, rather than promot ing FtsZ -ring stability or membrane tethering, SepF 425 may play an important role in licensing the progression of cell division . The localization 426 dependence of SepF on FtsZ is similar to other Firmicutes such as S. suis and B. subtilis, where 427 SepF localization depends on FtsZ, but FtsZ can still assemble without SepF (Hamoen 2006 428 Mol Micro )(Gao 2025 BMC Microl). These findings are also reminiscent of the localization 429 dependency observed in archaeal species that encode SepF as the ancestral FtsZ membrane 430 anchor, where SepF depends on FtsZ1 to assemble at mid -cell, but FtsZ1 and FtsZ2 can still 431 assemble into rings when SepF is depleted, at least in Haloferax volcani (83). Notably, in 432 addition to its proposed FtsZ membrane -anchoring function, SepF functions as an interaction 433 hub for divisome proteins in several organisms. I n C. glutamicum , SepF assembles into a 434 complex with the FtsZ -interacting gephyrin -like protein Glp and its transmembrane receptor 435 GlpR (116), and in Archaea, SepF organizes the photosynthesis reaction center proteins 436 CdpB1/2/3, which are important for division in Archaea (117, 118), and the archaeal cell division 437 protein CdpA (119). Thus, understanding how C. difficile SepF contributes to the recruitment of 438 late divisome proteins, such as PBP1, should advance our understanding of how this pathogen 439 assembles a functional divisome. 440 441 Our work also provides insight into the recruitment and function of the non-essential mid -cell 442 localizing proteins MldA, MldB, MldC, DivIVA, FtsK, and PBP3. W hile most of these proteins 443 localize to mid-cell coincident with septum synthesis, MldB can localize to mid -cell independent 444 of septum synthesis, including in the absence of SepF and PBP1 ( Fig 6 ). Since our clean 445 deletion mutant analyses showed that ΔmldA, ΔmldB, ΔmldC, and ΔdivIVA mutants all exhibit a 446 chaining phenotype, presumably due to a defect in cell separation after septum synthesis is 447 complete (Fig 7), we propose that most of these non-essential proteins play a role in late -stage 448 cell division. These genetic analyses build on prior work using Targetron insertions and plasmid -449 based complementation to individually analyze MldA, MldB, and MldC function during cell 450 division, reinforcing the previous conclusion that each of these genes is required for proper cell 451 separation (64)(Fig 7A ). Our analyse s also establish a role for DivIVA during C. difficile cell 452 separation. While the mechanism by which DivIVA regulates this process in C. difficile remains 453 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 15 unclear, in some Firmicutes, DivIVA appears to promote cell separation often by regulating the 454 export of cell separation enzymes (93 –97, 120) . Interestingly, our clean deletion analyses 455 showed that deletion of ftsK does not result in a cell division defect ( Fig 7), in contrast with prior 456 work indicating that antisense RNA -based KD of ftsK causes filamentation (59). Deletion of ftsK 457 may produce a different phenotype from antisense RNA KD due to compensatory transcriptional 458 changes in the deletion mutant, off-target or polar effects of the antisense RNA, or differences in 459 culture conditions or CD630Δerm isolates between labs. 460 461 Our analyses provide further insight into how the orphan Class B PBP enzyme PBP3 modulates 462 both vegetative cell division and asymmetric division during sporulation (62, 121) . We previously 463 implicated PBP3 in promoting PBP1 glycosyltransferase activity during vegetative cell division 464 (62) and asymmetric division (121). Since our prior work indicated that in both cases PBP3 465 function does not depend on its own catalytic activity, PBP3 likely regulat es C. difficile cell 466 division via protein -protein interactions with the divisome complex (62, 121) . We build upon 467 these prior findings by demonstrating that PBP3 localization to the site of division depends on 468 FtsZ, SepF, and PBP1 (Fig 6). It should be noted that significant processing of the mScI3 -PBP3 469 fusion protein limits the sensitivity of these localization assays. Still, precise ly how PBP3 470 regulates the C. difficile divisome remains to be determined. 471 472 Finally, our study optimizes CRISPRi -compatible trans -complementation methods for rapidly 473 testing the function of FtsZ, ZapA, and SepF variants in C. difficile. Indeed, our understanding of 474 FtsZ, ZapA, and SepF function in C. difficile lags behind model systems, where dozens of 475 mutant variants have been evaluated to dissect the function of these proteins at the molecular 476 level. To our knowledge, not a single point mutant has yet been evaluated in C. difficile for FtsZ, 477 ZapA, or SepF. Our newly established conditional expression system will facilitate rapid 478 structure-function analyses of these core divisome proteins in the future, bypassing more 479 laborious strategies for generating chromosomally -encoded conditional expression strains. 480 Indeed, our trans-complementation system allowed the functionality of tagged divisome proteins 481 and their potential processing to be directly addressed. These analyses revealed that an FtsZ -482 mScI3 fusion is inactive in C. difficile when expressed as the sole copy of FtsZ, whereas mScI3 -483 ZapA, SepF -mScI3, and mScI3 -PBP1 each have at least partial activity ( Fig 3 ). It should be 484 noted that it has been challenging to identify functional fusions to FtsZ, with N - and C -terminal 485 FtsZ-fluorescent protein fusions either being non-functional or causing suppressor mutations in 486 E. coli (99, 122), or requiring low temperatures for function, as in B. subtilis (8). Furthermore, for 487 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 16 E. coli FtsZ, it took nearly 20 years before functional sandwich fluorescent protein fusions to 488 FtsZ were identified, and still these fusions cause subtle abnormalities at 37° C (123). Since 489 fluorescent protein fusions to the C -terminus of FtsZ are functional in Streptococcus 490 pneumoniae (93, 124), exploring different C -terminal tags other than mScI3 or implementing a 491 sandwich fusion strategy for C. difficile FtsZ may lead to functional fusions in the future. 492 493 Altogether, our work supports a core function for FtsZ, ZapA, SepF, and PBP1 in C. difficile cell 494 division, and supports prior work showing that the C. difficile divisome differs substantially from 495 model systems (41, 63). In addition to the genes analyzed in this study, there are likely 496 unidentified divisome proteins important for this process. For example, how PBP1 is recruited to 497 the divisome complex and how its activity is licensed at mid -cell to drive septum synthesis and 498 cytokinesis remains to be determined. The methods for conditional gene expression and protein 499 localization optimized here will facilitate future mechanistic analyses of this unique and essential 500 pathway. 501 502

503 504 Bacterial strains and growth conditions 505 C. difficile strains are listed in Table S1 and derived from the 630Δerm strain background. 506 Chromosomally-encoded mutations, including insertion of mScI3-tagged constructs and gene 507 deletions were generated in a ΔpyrE strain using pyrE-based allele coupled exchange as 508 previously described (125) . C. difficile strains were cultured in brain heart infusion medium 509 supplemented with 0.5% yeast extract and 0.1% L -cysteine (BHIS) with thiamphenicol (10 510 μg/mL) to maintain plasmids, and/or kanamycin (50 μg/mL) and cefoxitin (8 μg/mL), and/or 511 spectinomycin (500 μg/mL) as needed for genetic manipulation. C. difficile defined medium 512 (CDDM) (126) was used to select for pyrE restoration during allele -coupled exchange, or was 513 supplemented with 5 -fluoroorotic acid (2 mg/mL) and uracil (5 μg/mL) to counter- select against 514 pyrE for making clean deletion mutants, with spectinomycin (500 μg/mL) included to select for 515 the aad9 marker when making a marked deletion (e.g. ΔftsK::aad9). Deletion mutants were 516 verified by PCR with locus -specific primers. C. difficile cultures were grown at 37°C in an 517 anaerobic chamber containing a gas mixture comprised of 85% N2, 5% CO2, and 10% H2. 518 519 Plasmids used in this study were cloned by Gibson assembly, propagated in E. coli DH5α, and 520 sequence-confirmed by Sanger or nanopore sequencing. E. coli strains and plasmids used in 521 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 17 this study are listed in Table S2, with links to plasmid maps that include the primer sequences 522 used for cloning. To introduce constructs into C. difficile, plasmids were first transformed into E. 523 coli HB101/pRK24 and then conjugated into C. difficile as previously described (127) . E. coli 524 cultures were grown in LB at 37°C supplemented as needed with chloramphenicol (20 μg/mL), 525 ampicillin (100 μg/mL), or kanamycin (30 μg/mL). 526 527 CRISPRi-KD and CRISPRi-compatible complementation 528 A plasmid-encoded CRISPRi-KD system based on the pIA33 plasmid was used as previously 529 described (67). Briefly, sgRNAs were designed to target the non -targeting strand, and were 530 chosen using the Benchling sgRNA design tool. These were cloned downstream of the 531 constitutive P gluD promoter. Plasmids were maintained in C. difficile by selection with 10 μg/mL 532 chloramphenicol. CRISPRi-KD was induced by culturing strains in the presence of 2.5% xylose, 533 back-diluting as needed to maintain logarithmic growth in the presence of inducer for 5 -6 hr. 534 When conditional expression of CRISPRi -resistant complementation constructs was required, 535 constructs containing synonymous mutations in the sgRNA -targeted sequence were integrated 536 into the genome downstream of the pyrE locus under the control of an aTc -inducible P tet 537 promoter. Cultures were maintained in the presence of both xylose and aTc to conditionally 538 express the CRISPRi -resistant complementation construct while knocking down expression of 539 the endogenous gene. 540 541 Fluorescent probes 542 The fluorescent D-amino acid HADA (Tocris) was used to label sites of peptidoglycan synthesis 543 and/or remodeling. 500 μL of logarithmically growing C. difficile was exposed to 50 μM HADA for 544 10 min, then fixed with a mixture of 100 μL 16% paraformaldehyde and 20 μL 1 M NaPO 4 buffer 545 (pH 7.4) for 30 min at room temperature and 30 min on ice (61). Fixed cells were washed three 546 times with 1 mL 1XPBS prior to imaging. 547 548 Protein localization with mScarlet-I3 fusions 549 C. difficile strains were engineered to express chromosomally -encoded divisome protein-mScI3 550 fusions under control of the aTc -inducible Ptet promoter, integrated downstream of pyrE. Strains 551 were cultured to the logarithmic phase in BHIS and then exposed to the indicated concentration 552 of aTc for 1 hr. Induced cultures were labeled with HADA and fixed as described above. After 553 cell fixation, cells were incubated overnight at room temperature in the dark to allow 554 chromophore maturation, as previously described (61). 555 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 18 556 Fluorescence microscopy 557 Microscopy samples were imaged on agarose pads (1% agarose in 1XPBS). Phase -contrast 558 and fluorescence micrographs were acquired with a Leica DMi8 inverted microscope equipped 559 with a 63X 1.4 NA Plan Apochromat oil -immersion phase -contrast objective, a high precision 560 motorized stage (Pecon), and in a 37°C incubator (Pecon). Excitation light was generated by a 561 Lumencor Spectra -X multi -LED light source with integrated excitation filters. An XLED -QP 562 quadruple-band dichroic beam -splitter (Leica) was used (transmission: 415, 470, 570, and 563 660 nm) with an external filter wheel for all fluorescent channels. HADA was excited at 395/25, 564 and emitted light was filtered using a 440/40 -nm emission filter, with a 120 ms exposure time ; 565 mScarlet-I3 was excited at 550/28 nm, and emitted light was filtered using a 590/50 -nm 566 emission filter, with a 150 ms exposure time. Light was detected using a Leica DFC 9000 GTC 567 sCMOS camera. 1 –2 μm z -stacks were taken with 0.21 μm z -slices. Images were acquired 568 using the LASX software, and fluorescence images were deconvolved using Leica Small 569 Volume Computational Clearing with the following settings: refractive index 1.33, strength 60%, 570 and regularization 0.05. 571 Images were processed using FIJI to select the best -focused z- plane for each channel and 572 adjust the image brightness and contrast of images. Identical minimum and maximum values 573 were applied to a given channel across all images in a figure panel so that direct comparisons 574 could be made across the images within a figure panel, unless otherwise specified in the figure 575 legend. Demographs were generated with MicrobeJ in FIJI, using segmentation masks 576 generated by SuperSegger (128, 129). 577 578 Western blot analysis 579 A volume of 1.4 -2.8 mL of logarithmically growing C. difficile culture (OD ~0.4 -.07) were 580 pelleted, resuspended in 25 μL 1X PBS, freeze -thawed three times, mixed with 25 μL EBB 581 buffer (9 M urea, 2 M thiourea, 4% SDS, 2 mM β-mercaptoethanol), and boiled for 20 min to lyse 582 cells. SDS -polyacrylamide gel electrophoresis (SDS -PAGE) was performed on 10% 583 polyacrylamide gels. Proteins were transferred to polyvinylidene difluoride membranes, which 584 were subsequently probed with the following primary antibodies: rabbit polyclonal anti -DivIVA 585 TF142 (62) at 1:1,000 dilution; chicken polyclonal anti -GDH (aCdGDH; Thermo) at 1:10,000 586 dilution; and/or a goat polyclonal anti -mScarlet at 1:5,000 dilution (MBS448290; MyBioSource) . 587 Anti-rabbit, anti -chicken, and anti -goat IR800 or IR680 secondary antibodies (LI -COR 588 Biosciences, 1:30,000) were used to detect bands with a LI-COR Odyssey CLx imaging system. 589 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 19 590

591 This work was supported by the National Institute of Allergy and Infectious Diseases grant R01 592 AI122232 to AS. Additionally, GAH was supported by the Tufts University Institutional Research 593 Career and Academic Development Award Program (K12GM133314) and a National Institute of 594 Allergy and Infectious Diseases training fellowship (F32AI191529). 595 596 The content is solely the responsibility of the authors and does not necessarily represent the 597 official views of the National Institutes of Health. The funders had no role in study design, data 598 collection and analysis, decision to publish, or preparation of the manuscript. 599 600

601 1. den Blaauwen T, Hamoen LW, Levin PA. 2017. The divisome at 25: the road ahead. Curr 602 Opin Microbiol 36:85–94. 603 604 2. Levin PA, Janakiraman A. 2021. Localization, Assembly, and Activation of the Escherichia 605 coli Cell Division Machinery. EcoSal Plus 9:eESP-0022-2021. 606 607 3. Cameron TA, Margolin W. 2023. Insights into the assembly and regulation of the bacterial 608 divisome. Nat Rev Microbiol https://doi.org/10.1038/s41579-023-00942-x. 609 610 4. Gueiros-Filho FJ, Losick R. 2002. A widely conserved bacterial cell division protein that 611 promotes assembly of the tubulin-like protein FtsZ. Genes Dev 16:2544–2556. 612 613 5. Hamoen LW, Meile J-C, De Jong W, Noirot P, Errington J. 2006. SepF, a novel FtsZ-614 interacting protein required for a late step in cell division. Mol Microbiol 59:989–999. 615 616 6. Duman R, Ishikawa S, Celik I, Strahl H, Ogasawara N, Troc P, Löwe J, Hamoen LW. 617 2013. Structural and genetic analyses reveal the protein SepF as a new membrane 618 anchor for the Z ring. Proceedings of the National Academy of Sciences 110:E4601–619 E4610. 620 621 7. Ishikawa S, Kawai Y, Hiramatsu K, Kuwano M, Ogasawara N. 2006. A new FtsZ-622 interacting protein, YlmF, complements the activity of FtsA during progression of cell 623 division in Bacillus subtilis. Mol Microbiol 60:1364–1380. 624 625 8. Levin PA, Kurtser IG, Grossman AD. 1999. Identification and characterization of a 626 negative regulator of FtsZ ring formation in Bacillus subtilis. Proceedings of the National 627 Academy of Sciences 96:9642–9647. 628 629 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 20 9. Bhattacharya D, King A, McKnight L, Horigian P, Eswara PJ. 2025. GpsB interacts with 630 FtsZ in multiple species and may serve as an accessory Z-ring anchor. Mol Biol Cell 631 36:ar10. 632 633 10. Feucht A, Lucet I, Yudkin MD, Errington J. 2001. Cytological and biochemical 634 characterization of the FtsA cell division protein of Bacillus subtilis. Mol Microbiol 40:115–635 125. 636 637 11. Gulsoy IC, Saaki TN V, Wenzel M, Syvertsson S, Morimoto T, Siersma TK, Hamoen LW. 638 2025. Minimization of the Bacillus subtilis divisome suggests FtsZ and SepF can form an 639 active Z-ring, and reveals the amino acid transporter BraB as a new cell division 640 influencing factor. PLoS Genet 21:e1011567-. 641 642 12. Beall B, Lutkenhaus J. 1992. Impaired cell division and sporulation of a Bacillus subtilis 643 strain with the ftsA gene deleted. J Bacteriol 174:2398–2403. 644 645 13. Jensen SO, Thompson LS, Harry EJ. 2005. Cell Division in Bacillus subtilis: FtsZ and 646 FtsA Association Is Z-Ring Independent, and FtsA Is Required for Efficient Midcell Z-Ring 647 Assembly. J Bacteriol 187:6536–6544. 648 649 14. Feucht A, Lucet I, Yudkin MD, Errington J. 2001. Cytological and biochemical 650 characterization of the FtsA cell division protein of Bacillus subtilis. Mol Microbiol 40:115–651 125. 652 653 15. Claessen D, Emmins R, Hamoen LW, Daniel RA, Errington J, Edwards DH. 2008. Control 654 of the cell elongation–division cycle by shuttling of PBP1 protein in Bacillus subtilis. Mol 655 Microbiol 68:1029–1046. 656 657 16. Tavares JR, de Souza RF, Silva Meira GL, Gueiros-Filho FJ. 2008. Cytological 658 Characterization of YpsB, a Novel Component of the Bacillus subtilis Divisome. J 659 Bacteriol 190:7096–7107. 660 661 17. Daniel RA, Noirot-Gros M-F, Noirot P , Errington J. 2006. Multiple Interactions between the 662 Transmembrane Division Proteins of Bacillus subtilis and the Role of FtsL Instability in 663 Divisome Assembly. J Bacteriol 188:7396–7404. 664 665 18. Daniel RA, Harry EJ, Katis VL, Wake RG, Errington J. 1998. Characterization of the 666 essential cell division gene ftsL (yllD ) of Bacillus subtilis and its role in the assembly of 667 the division apparatus. Mol Microbiol 29:593–604. 668 669 19. Sievers J, Errington J. 2000. Analysis of the Essential Cell Division Gene ftsL of Bacillus 670 subtilis by Mutagenesis and Heterologous Complementation. J Bacteriol 182:5572–5579. 671 672 20. Levin PA, Losick R. 1994. Characterization of a cell division gene from Bacillus subtilis 673 that is required for vegetative and sporulation septum formation. J Bacteriol 176:1451–674 1459. 675 676 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 21 21. Harry EJ, Wake RG. 1997. The membrane-bound cell division protein DivIB is localized to 677 the division site in Bacillus subtilis. Mol Microbiol 25:275–283. 678 679 22. Gamba P, Veening J-W, Saunders NJ, Hamoen LW, Daniel RA. 2009. Two-Step 680 Assembly Dynamics of the Bacillus subtilis Divisome. J Bacteriol 191:4186–4194. 681 682 23. Taguchi A, Welsh MA, Marmont LS, Lee W, Sjodt M, Kruse AC, Kahne D, Bernhardt TG, 683 Walker S. 2019. FtsW is a peptidoglycan polymerase that is functional only in complex 684 with its cognate penicillin-binding protein. Nat Microbiol 4:587–594. 685 686 24. Meeske AJ, Riley EP, Robins WP, Uehara T, Mekalanos JJ, Kahne D, Walker S, Kruse 687 AC, Bernhardt TG, Rudner DZ. 2016. SEDS proteins are a widespread family of bacterial 688 cell wall polymerases. Nature 537:634–638. 689 690 25. Reichmann NT, Tavares AC, Saraiva BM, Jousselin A, Reed P, Pereira AR, Monteiro JM, 691 Sobral RG, VanNieuwenhze MS, Fernandes F, Pinho MG. 2019. SEDS–bPBP pairs 692 direct lateral and septal peptidoglycan synthesis in Staphylococcus aureus. Nat Microbiol 693 4:1368–1377. 694 695 26. Goehring NW, Gonzalez MD, Beckwith J. 2006. Premature targeting of cell division 696 proteins to midcell reveals hierarchies of protein interactions involved in divisome 697 assembly. Mol Microbiol 61:33–45. 698 699 27. Goehring NW, Gueiros-Filho F, Beckwith J. 2005. Premature targeting of a cell division 700 protein to midcell allows dissection of divisome assembly in Escherichia coli. Genes Dev 701 19:127–137. 702 703 28. Gonzalez MD, Akbay EA, Boyd D, Beckwith J. 2010. Multiple Interaction Domains in FtsL, 704 a Protein Component of the Widely Conserved Bacterial FtsLBQ Cell Division Complex. J 705 Bacteriol 192:2757–2768. 706 707 29. Gonzalez MD, Beckwith J. 2009. Divisome under Construction: Distinct Domains of the 708 Small Membrane Protein FtsB Are Necessary for Interaction with Multiple Cell Division 709 Proteins. J Bacteriol 191:2815–2825. 710 711 30. Marmont LS, Bernhardt TG. 2020. A conserved subcomplex within the bacterial 712 cytokinetic ring activates cell wall synthesis by the FtsW-FtsI synthase. Proceedings of 713 the National Academy of Sciences 117:23879–23885. 714 715 31. Liu B, Persons L, Lee L, de Boer PAJ. 2015. Roles for both FtsA and the FtsBLQ 716 subcomplex in FtsN-stimulated cell constriction in Escherichia coli. Mol Microbiol 95:945–717 970. 718 719 32. Käshammer L, van den Ent F, Jeffery M, Jean NL, Hale VL, Löwe J. 2023. Cryo-EM 720 structure of the bacterial divisome core complex and antibiotic target FtsWIQBL. Nat 721 Microbiol 8:1149–1159. 722 723 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 22 33. Yang L, Chen Y, Chang S, Shen C, Wang X, Zhang C, Zhang Z, Ding B-S, Su Z, Dong H, 724 Tang X. 2024. Structural insights into the activation of the divisome complex FtsWIQLB. 725 Cell Discov 10:2. 726 727 34. Tsang M-J, Bernhardt TG. 2015. A role for the FtsQLB complex in cytokinetic ring 728 activation revealed by an ftsL allele that accelerates division. Mol Microbiol 95:925–944. 729 730 35. Park K-T, Pichoff S, Du S, Lutkenhaus J. 2021. FtsA acts through FtsW to promote cell 731 wall synthesis during cell division in Escherichia coli. Proceedings of the National 732 Academy of Sciences 118:e2107210118. 733 734 36. Park K-T, Du S, Lutkenhaus J. 2020. Essential Role for FtsL in Activation of Septal 735 Peptidoglycan Synthesis. mBio 11:10.1128/mbio.03012-20. 736 737 37. Kobayashi K, Ehrlich SD, Albertini A, Amati G, Andersen KK, Arnaud M, Asai K, Ashikaga 738 S, Aymerich S, Bessieres P, Boland F, Brignell SC, Bron S, Bunai K, Chapuis J, 739 Christiansen LC, Danchin A, Débarbouillé M, Dervyn E, Deuerling E, Devine K, Devine 740 SK, Dreesen O, Errington J, Fillinger S, Foster SJ, Fujita Y, Galizzi A, Gardan R, 741 Eschevins C, Fukushima T, Haga K, Harwood CR, Hecker M, Hosoya D, Hullo MF, 742 Kakeshita H, Karamata D, Kasahara Y, Kawamura F, Koga K, Koski P, Kuwana R, 743 Imamura D, Ishimaru M, Ishikawa S, Ishio I, Le Coq D, Masson A, Mauël C, Meima R, 744 Mellado RP, Moir A, Moriya S, Nagakawa E, Nanamiya H, Nakai S, Nygaard P, Ogura M, 745 Ohanan T, O’Reilly M, O’Rourke M, Pragai Z, Pooley HM, Rapoport G, Rawlins JP, Rivas 746 LA, Rivolta C, Sadaie A, Sadaie Y, Sarvas M, Sato T, Saxild HH, Scanlan E, Schumann 747 W, Seegers JFML, Sekiguchi J, Sekowska A, Séror SJ, Simon M, Stragier P, Studer R, 748 Takamatsu H, Tanaka T, Takeuchi M, Thomaides HB, Vagner V, van Dijl JM, Watabe K, 749 Wipat A, Yamamoto H, Yamamoto M, Yamamoto Y, Yamane K, Yata K, Yoshida K, 750 Yoshikawa H, Zuber U, Ogasawara N. 2003. Essential Bacillus subtilis genes. 751 Proceedings of the National Academy of Sciences 100:4678–4683. 752 753 38. Beall B, Lutkenhaus J. 1989. Nucleotide sequence and insertional inactivation of a 754 Bacillus subtilis gene that affects cell division, sporulation, and temperature sensitivity. J 755 Bacteriol 171:6821–6834. 756 757 39. Megrian D, Taib N, Jaffe AL, Banfield JF, Gribaldo S. 2022. Ancient origin and constrained 758 evolution of the division and cell wall gene cluster in Bacteria. Nat Microbiol 7:2114–2127. 759 760 40. Yang X, McQuillen R, Lyu Z, Phillips-Mason P, De La Cruz A, McCausland JW, Liang H, 761 DeMeester KE, Santiago CC, Grimes CL, de Boer P, Xiao J. 2021. A two-track model for 762 the spatiotemporal coordination of bacterial septal cell wall synthesis revealed by single-763 molecule imaging of FtsW. Nat Microbiol 6:584–593. 764 765 41. Shrestha S, Taib N, Gribaldo S, Shen A. 2023. Diversification of division mechanisms in 766 endospore-forming bacteria revealed by analyses of peptidoglycan synthesis in 767 Clostridioides difficile. Nat Commun 14:7975. 768 769 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 23 42. Straume D, Piechowiak KW, Kjos M, Håvarstein LS. 2021. Class A PBPs: It is time to 770 rethink traditional paradigms. Mol Microbiol 116:41–52. 771 772 43. Pazos M, Vollmer W. 2021. Regulation and function of class A Penicillin-binding proteins. 773 Curr Opin Microbiol 60:80–87. 774 775 44. Cho H, Wivagg CN, Kapoor M, Barry Z, Rohs PDA, Suh H, Marto JA, Garner EC, 776 Bernhardt TG. 2016. Bacterial cell wall biogenesis is mediated by SEDS and PBP 777 polymerase families functioning semi-autonomously. Nat Microbiol 1:16172. 778 779 45. Shlosman I, Vettiger A, Bernhardt TG, Kruse AC, Loparo JJ. 2025. The hit-and-run of cell 780 wall synthesis: LpoB transiently binds and activates PBP1b through a conserved 781 allosteric switch. Nat Commun 16:6723. 782 783 46. Paradis-Bleau C, Markovski M, Uehara T, Lupoli TJ, Walker S, Kahne DE, Bernhardt TG. 784 2010. Lipoprotein Cofactors Located in the Outer Membrane Activate Bacterial Cell Wall 785 Polymerases. Cell 143:1110–1120. 786 787 47. Egan AJF, Jean NL, Koumoutsi A, Bougault CM, Biboy J, Sassine J, Solovyova AS, 788 Breukink E, Typas A, Vollmer W, Simorre J-P. 2014. Outer-membrane lipoprotein LpoB 789 spans the periplasm to stimulate the peptidoglycan synthase PBP1B. Proceedings of the 790 National Academy of Sciences 111:8197–8202. 791 792 48. Vigouroux A, Cordier B, Aristov A, Alvarez L, Özbaykal G, Chaze T, Oldewurtel ER, 793 Matondo M, Cava F, Bikard D, van Teeffelen S. 2020. Class-A penicillin binding proteins 794 do not contribute to cell shape but repair cell-wall defects. Elife 9:e51998. 795 796 49. Straume D, Piechowiak KW, Olsen S, Stamsås GA, Berg KH, Kjos M, Heggenhougen 797 MV, Alcorlo M, Hermoso JA, Håvarstein LS. 2020. Class A PBPs have a distinct and 798 unique role in the construction of the pneumococcal cell wall. Proceedings of the National 799 Academy of Sciences 117:6129–6138. 800 801 50. Navarro PP, Vettiger A, Ananda VY, Llopis PM, Allolio C, Bernhardt TG, Chao LH. 2022. 802 Cell wall synthesis and remodelling dynamics determine division site architecture and cell 803 shape in Escherichia coli. Nat Microbiol 7:1621–1634. 804 805 51. Navarro PP, Vettiger A, Hajdu R, Ananda VY, López-Tavares A, Schmid EW, Walter JC, 806 Loose M, Chao LH, Bernhardt TG. 2025. The aPBP-type cell wall synthase PBP1b plays 807 a specialized role in fortifying the Escherichia coli division site against osmotic rupture. 808 bioRxiv 2025.04.02.646830. 809 810 52. Brown PJB, de Pedro MA, Kysela DT, Van der Henst C, Kim J, De Bolle X, Fuqua C, 811 Brun Y V. 2012. Polar growth in the Alphaproteobacterial order Rhizobiales. Proceedings 812 of the National Academy of Sciences 109:1697–1701. 813 814 53. Williams MA, Aliashkevich A, Krol E, Kuru E, Bouchier JM, Rittichier J, Brun Y V, 815 VanNieuwenhze MS, Becker A, Cava F, Brown PJB. 2021. Unipolar Peptidoglycan 816 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 24 Synthesis in the Rhizobiales Requires an Essential Class A Penicillin-Binding Protein. 817 mBio 12:10.1128/mbio.02346-21. 818 819 54. Arora D, Chawla Y, Malakar B, Singh A, Nandicoori VK. 2018. The transpeptidase PbpA 820 and noncanonical transglycosylase RodA of Mycobacterium tuberculosis play important 821 roles in regulating bacterial cell lengths. Journal of Biological Chemistry 293:6497–6516. 822 823 55. Sieger B, Schubert K, Donovan C, Bramkamp M. 2013. The lipid II flippase RodA 824 determines morphology and growth in Corynebacterium glutamicum. Mol Microbiol 825 90:966–982. 826 827 56. Sher JW, Lim HC, Bernhardt TG. 2021. Polar Growth in Corynebacterium glutamicum 828 Has a Flexible Cell Wall Synthase Requirement. mBio 12:10.1128/mbio.00682-21. 829 830 57. Melzer ES, Kado T, García-Heredia A, Gupta Kuldeepkumar R, Meniche X, Morita YS, 831 Sassetti CM, Rego EH, Siegrist M Sloan. 2022. Cell Wall Damage Reveals Spatial 832 Flexibility in Peptidoglycan Synthesis and a Nonredundant Role for RodA in 833 Mycobacteria. J Bacteriol 204:e00540-21. 834 835 58. Cuenot E, Garcia-Garcia T, Douche T, Gorgette O, Courtin P, Denis-Quanquin S, Hoys S, 836 Tremblay YD, Matondo M, Chapot-Chartier M-P, Janoir C, Dupuy B, Candela T, Martin-837 Verstraete I. 2019. The Ser/Thr Kinase PrkC Participates in Cell Wall Homeostasis and 838 Antimicrobial Resistance in Clostridium difficile. Infect Immun 87:10.1128/iai.00005-19. 839 840 59. Garcia-Garcia T, Douché T, Giai Gianetto Q, Poncet S, El Omrani N, Smits WK, Cuenot 841 E, Matondo M, Martin-Verstraete I. 2022. In-Depth Characterization of the Clostridioides 842 difficile Phosphoproteome to Identify Ser/Thr Kinase Substrates. Molecular & Cellular 843 Proteomics 21. 844 845 60. Buckley AM, Jukes C, Candlish D, Irvine JJ, Spencer J, Fagan RP , Roe AJ, Christie JM, 846 Fairweather NF, Douce GR. 2016. Lighting up Clostridium difficile: reporting gene 847 expression using fluorescent Lov domains. Sci Rep 6:23463. 848 849 61. Ransom EM, Ellermeier CD, Weiss DS. 2015. Use of mCherry red fluorescent protein for 850 studies of protein localization and gene expression in Clostridium difficile. Appl Environ 851 Microbiol 81:1652–1660. 852 853 62. Harrison GA, Shen A. 2025. Molecular dissection of Class A PBP function uncovers novel 854 features of the non-canonical Clostridioides difficile divisome complex. PLoS Genet 855 21:e1011746-. 856 857 63. Alberts ME, Kurtz MP , Müh U, Bernardi JP, Bollinger KW, Dobrila HA, Duncan L, Laster 858 HM, Orea AJ, Pannullo AG, Rivera-Rosado JG, Torres F V, Ellermeier CD, Weiss DS. 859 2025. Analysis of essential genes in Clostridioides difficile by CRISPRi and Tn-seq. J 860 Bacteriol 0:e00220-25. 861 862 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 25 64. Ransom EM, Williams KB, Weiss DS, Ellermeier CD. 2014. Identification and 863 characterization of a gene cluster required for proper rod shape, cell division, and 864 pathogenesis in Clostridium difficile. J Bacteriol 196:2290–2300. 865 866 65. Kirk JA, Fagan RP. 2016. Heat shock increases conjugation efficiency in Clostridium 867 difficile. Anaerobe 42:1–5. 868 869 66. Dembek M, Barquist L, Boinett CJ, Cain AK, Mayho M, Lawley TD, Fairweather NF, 870 Fagan RP. 2015. High-Throughput Analysis of Gene Essentiality and Sporulation in 871 Clostridium difficile. mBio 6:10.1128/mbio.02383-14. 872 873 67. Müh U, Pannullo AG, Weiss DS, Ellermeier CD. 2019. A Xylose-Inducible Expression 874 System and a CRISPR Interference Plasmid for Targeted Knockdown of Gene 875 Expression in Clostridioides difficile. J Bacteriol 201:e00711-18. 876 877 68. Fujita J, Kasai K, Hibino K, Kagoshima G, Kamimura N, Tobita S, Kato Y, Uehara R, 878 Namba K, Uchihashi T, Matsumura H. 2025. Structural basis for the interaction between 879 the bacterial cell division proteins FtsZ and ZapA. Nat Commun 16:5985. 880 881 69. Small E, Marrington R, Rodger A, Scott DJ, Sloan K, Roper D, Dafforn TR, Addinall SG. 882 2007. FtsZ Polymer-bundling by the Escherichia coli ZapA Orthologue, YgfE, Involves a 883 Conformational Change in Bound GTP . J Mol Biol 369:210–221. 884 885 70. Low HH, Moncrieffe MC, Löwe J. 2004. The Crystal Structure of ZapA and its Modulation 886 of FtsZ Polymerisation. J Mol Biol 341:839–852. 887 888 71. Surdova K, Gamba P, Claessen D, Siersma T, Jonker MJ, Errington J, Hamoen LW. 2013. 889 The Conserved DNA-Binding Protein WhiA Is Involved in Cell Division in Bacillus subtilis. 890 J Bacteriol 195:5450–5460. 891 892 72. Johnson JE, Lackner LL, Hale CA, de Boer PAJ. 2004. ZipA Is Required for Targeting of 893 DMinC/DicB, but Not DMinC/MinD, Complexes to Septal Ring Assemblies in Escherichia 894 coli. J Bacteriol 186:2418–2429. 895 896 73. Galli E, Gerdes K. 2010. Spatial resolution of two bacterial cell division proteins: ZapA 897 recruits ZapB to the inner face of the Z-ring. Mol Microbiol 76:1514–1526. 898 899 74. Durand-Heredia JM, Yu HH, De Carlo S, Lesser CF, Janakiraman A. 2011. Identification 900 and Characterization of ZapC, a Stabilizer of the FtsZ Ring in Escherichia coli. J Bacteriol 901 193:1405–1413. 902 903 75. Hale CA, Shiomi D, Liu B, Bernhardt TG, Margolin W, Niki H, de Boer PAJ. 2011. 904 Identification of Escherichia coli ZapC (YcbW) as a Component of the Division Apparatus 905 That Binds and Bundles FtsZ Polymers. J Bacteriol 193:1393–1404. 906 907 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 26 76. Krupka M, Sobrinos-Sanguino M, Jiménez M, Rivas G, Margolin W. 2018. Escherichia 908 coli ZipA Organizes FtsZ Polymers into Dynamic Ring-Like Protofilament Structures. 909 mBio 9:10.1128/mbio.01008-18. 910 911 77. Gong H, Yan D, Cui Y, Li Y, Yang J, Yang W, Zhan R, Wan Q, Wang X, He H, Chen X, 912 Lutkenhaus J, Yang X, Du S. 2024. The divisome is a self-enhancing machine in 913 Escherichia coli and Caulobacter crescentus. Nat Commun 15:8198. 914 915 78. Gao T, Li T, Zhu J, Zheng L, Chen M, Liu W, Yang K, Zhang T, Yuan F, Liu Z, Guo R, Li C, 916 Wu Q, Tian Y, Zhou R, Zhou D. 2025. Identification of SepF in Streptococcus suis 917 involving cell division. BMC Microbiol 25:179. 918 919 79. Pende N, Sogues A, Megrian D, Sartori-Rupp A, England P, Palabikyan H, Rittmann SK-920 MR, Graña M, Wehenkel AM, Alzari PM, Gribaldo S. 2021. SepF is the FtsZ anchor in 921 archaea, with features of an ancestral cell division system. Nat Commun 12:3214. 922 923 80. Gupta S, Banerjee SK, Chatterjee A, Sharma AK, Kundu M, Basu J. 2015. Essential 924 protein SepF of mycobacteria interacts with FtsZ and MurG to regulate cell growth and 925 division. Microbiology (N Y) 161:1627–1638. 926 927 81. Sogues A, Martinez M, Gaday Q, Ben Assaya M, Graña M, Voegele A, VanNieuwenhze 928 M, England P, Haouz A, Chenal A, Trépout S, Duran R, Wehenkel AM, Alzari PM. 2020. 929 Essential dynamic interdependence of FtsZ and SepF for Z-ring and septum formation in 930 Corynebacterium glutamicum. Nat Commun 11:1641. 931 932 82. Dey S, Zhou H-X. 2022. Membrane Tethering of SepF, a Membrane Anchor for the 933 Mycobacterium tuberculosis Z-ring. J Mol Biol 434:167817. 934 935 83. Nußbaum P, Gerstner M, Dingethal M, Erb C, Albers S-V. 2021. The archaeal protein 936 SepF is essential for cell division in Haloferax volcanii. Nat Commun 12:3469. 937 938 84. Ricard M, Hirota Y. 1973. Process of Cellular Division in Escherichia coli: Physiological 939 Study on Thermosensitive Mutants Defective in Cell Division. J Bacteriol 116:314–322. 940 941 85. Nagai K, Tamura G. 1972. Mutant of Escherichia coli with Thermosensitive Protein in the 942 Process of Cellular Division. J Bacteriol 112:959–966. 943 944 86. Hirota Y, Ryter A, Jacob F. 1968. Thermosensitive Mutants of E. coli Affected in the 945 Processes of DNA Synthesis and Cellular Division. Cold Spring Harb Symp Quant Biol 946 33:677–693. 947 948 87. Van De Putte P, Van Dillewijn J, Rörsch A. 1964. The selection of mutants of Escherichia 949 coli with impaired cell division at elevated temperature. Mutation Research/Fundamental 950 and Molecular Mechanisms of Mutagenesis 1:121–128. 951 952 88. Fuchs M, Lamm-Schmidt V, Sulzer J, Ponath F, Jenniches L, Kirk JA, Fagan RP, Barquist 953 L, Vogel J, Faber F. 2021. An RNA-centric global view of Clostridioides difficile reveals 954 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 27 broad activity of Hfq in a clinically important gram-positive bacterium. Proceedings of the 955 National Academy of Sciences 118:e2103579118. 956 957 89. White ML, Eswara PJ. 2021. ylm Has More than a (Z Anchor) Ring to It! J Bacteriol 958 203:10.1128/jb.00460-20. 959 960 90. Murphy BT, Wiepen JJ, He H, Pramanik AS, Peters JM, Stevenson B, Zückert WR. 2023. 961 Inducible CRISPRi-Based Operon Silencing and Selective in Trans Gene 962 Complementation in Borrelia burgdorferi. J Bacteriol 205:e00468-22. 963 964 91. Li S, Poulton NC, Chang JS, Azadian ZA, DeJesus MA, Ruecker N, Zimmerman MD, 965 Eckartt KA, Bosch B, Engelhart CA, Sullivan DF, Gengenbacher M, Dartois VA, 966 Schnappinger D, Rock JM. 2022. CRISPRi chemical genetics and comparative genomics 967 identify genes mediating drug potency in Mycobacterium tuberculosis. Nat Microbiol 968 7:766–779. 969 970 92. Fagan RP, Fairweather NF. 2011. Clostridium difficile has two parallel and essential Sec 971 secretion systems. Journal of Biological Chemistry 286:27483–27493. 972 973 93. Fleurie A, Manuse S, Zhao C, Campo N, Cluzel C, Lavergne J-P, Freton C, Combet C, 974 Guiral S, Soufi B, Macek B, Kuru E, VanNieuwenhze MS, Brun Y V, Di Guilmi A-M, 975 Claverys J-P, Galinier A, Grangeasse C. 2014. Interplay of the Serine/Threonine-Kinase 976 StkP and the Paralogs DivIVA and GpsB in Pneumococcal Cell Elongation and Division. 977 PLoS Genet 10:e1004275-. 978 979 94. Trouve J, Zapun A, Bellard L, Juillot D, Pelletier A, Freton C, Baudoin M, Carballido-980 Lopez R, Campo N, Wong Y-S, Grangeasse C, Morlot C. 2024. DivIVA controls the 981 dynamics of septum splitting and cell elongation in Streptococcus pneumoniae. mBio 982 15:e01311-24. 983 984 95. Fadda D, Santona A, D’Ulisse V, Ghelardini P, Ennas MG, Whalen MB, Massidda O. 985 2007. Streptococcus pneumoniae DivIVA: Localization and Interactions in a MinCD-Free 986 Context. J Bacteriol 189:1288–1298. 987 988 96. Jiang Q, Li B, Zhang L, Li T, Hu Q, Li H, Zou W, Hu Z, Huang Q, Zhou R. 2023. DivIVA 989 Interacts with the Cell Wall Hydrolase MltG To Regulate Peptidoglycan Synthesis in 990 Streptococcus suis. Microbiol Spectr 11:e04750-22. 991 992 97. Halbedel S, Hahn B, Daniel RA, Flieger A. 2012. DivIVA affects secretion of virulence-993 related autolysins in Listeria monocytogenes. Mol Microbiol 83:821–839. 994 995 98. Schäper S, Yau HCL, Krol E, Skotnicka D, Heimerl T, Gray J, Kaever V, Søgaard-996 Andersen L, Vollmer W, Becker A. 2018. Seven-transmembrane receptor protein RgsP 997 and cell wall-binding protein RgsM promote unipolar growth in Rhizobiales. PLoS Genet 998 14:e1007594-. 999 1000 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 28 99. Ma X, Ehrhardt DW, Margolin W. 1996. Colocalization of cell division proteins FtsZ and 1001 FtsA to cytoskeletal structures in living Escherichia coli cells by using 1002 green fluorescent protein. Proceedings of the National Academy of Sciences 93:12998–1003 13003. 1004 1005 100. Goley ED, Yeh Y-C, Hong S-H, Fero MJ, Abeliuk E, McAdams HH, Shapiro L. 2011. 1006 Assembly of the Caulobacter cell division machine. Mol Microbiol 80:1680–1698. 1007 1008 101. Sun Q, Margolin W. 1998. FtsZ Dynamics during the Division Cycle of Live Escherichia 1009 coli Cells. J Bacteriol 180:2050–2056. 1010 1011 102. Stricker J, Maddox P, Salmon ED, Erickson HP. 2002. Rapid assembly dynamics of the 1012 Escherichia coli FtsZ-ring demonstrated by fluorescence recovery after photobleaching. 1013 Proceedings of the National Academy of Sciences 99:3171–3175. 1014 1015 103. Scheffers D-J, Jones LJF, Errington J. 2004. Several distinct localization patterns for 1016 penicillin-binding proteins in Bacillus subtilis. Mol Microbiol 51:749–764. 1017 1018 104. Cha JH, Stewart GC. 1997. The divIVA minicell locus of Bacillus subtilis. J Bacteriol 1019 179:1671–1683. 1020 1021 105. Ramamurthi KS, Losick R. 2009. Negative membrane curvature as a cue for subcellular 1022 localization of a bacterial protein. Proceedings of the National Academy of Sciences 1023 106:13541–13545. 1024 1025 106. Lenarcic R, Halbedel S, Visser L, Shaw M, Wu LJ, Errington J, Marenduzzo D, Hamoen 1026 LW. 2009. Localisation of DivIVA by targeting to negatively curved membranes. EMBO J 1027 28:2272–2282. 1028 1029 107. Makroczyová J, Jamroškovič J, Krascsenitsová E, Labajová N, Barák I. 2016. Oscillating 1030 behavior of Clostridium difficile Min proteins in Bacillus subtilis. Microbiologyopen 5:387–1031 401. 1032 1033 108. Valenčíková R, Krascsenitsová E, Labajová N, Makroczyová J, Barák I. 2018. Clostridial 1034 DivIVA and MinD interact in the absence of MinJ. Anaerobe 50:22–31. 1035 1036 109. Wang L, Lutkenhaus J. 1998. FtsK is an essential cell division protein that is localized to 1037 the septum and induced as part of the SOS response. Mol Microbiol 29:731–740. 1038 1039 110. Yu X, Tran AH, Sun Q, Margolin W. 1998. Localization of Cell Division Protein FtsK to the 1040 Escherichia coli Septum and Identification of a Potential N-Terminal Targeting Domain. J 1041 Bacteriol 180:1296–1304. 1042 1043 111. Begg KJ, Dewar SJ, Donachie WD. 1995. A new Escherichia coli cell division gene, ftsK. 1044 J Bacteriol 177:6211–6222. 1045 1046 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 29 112. Capiaux H, Lesterlin C, Pérals K, Louarn JM, Cornet F. 2002. A dual role for the FtsK 1047 protein in Escherichia coli chromosome segregation. The EMBO Reports 3:532–536. 1048 1049 113. Wang L, Lutkenhaus J. 1998. FtsK is an essential cell division protein that is localized to 1050 the septum and induced as part of the SOS response. Mol Microbiol 29:731–740. 1051 1052 114. Bigot S, Sivanathan V, Possoz C, Barre F-X, Cornet F. 2007. FtsK, a literate chromosome 1053 segregation machine. Mol Microbiol 64:1434–1441. 1054 1055 115. Naha A, Haeusser DP, Margolin W. 2023. Anchors: A way for FtsZ filaments to stay 1056 membrane bound. Mol Microbiol 120:525–538. 1057 1058 116. Martinez M, Petit J, Leyva A, Sogues A, Megrian D, Rodriguez A, Gaday Q, Ben Assaya 1059 M, Portela MM, Haouz A, Ducret A, Grangeasse C, Alzari PM, Durán R, Wehenkel AM. 1060 2023. Eukaryotic-like gephyrin and cognate membrane receptor coordinate 1061 corynebacterial cell division and polar elongation. Nat Microbiol 1062 https://doi.org/10.1038/s41564-023-01473-0. 1063 1064 117. Zhao S, Makarova KS, Zheng W, Zhan L, Wan Q, Liu Y, Gong H, Krupovic M, Lutkenhaus 1065 J, Chen X, Koonin E V, Du S. 2024. Widespread photosynthesis reaction centre barrel 1066 proteins are necessary for haloarchaeal cell division. Nat Microbiol 9:712–726. 1067 1068 118. Nußbaum P, Kureisaite-Ciziene D, Bellini D, van der Does C, Kojic M, Taib N, Yeates A, 1069 Tourte M, Gribaldo S, Loose M, Löwe J, Albers S-V. 2024. Proteins containing 1070 photosynthetic reaction centre domains modulate FtsZ-based archaeal cell division. Nat 1071 Microbiol 9:698–711. 1072 1073 119. Liao Y, Shinde VD, Hu D, Xu Z, Söderström B, Michie KA, Duggin IG. 2025. Cell division 1074 protein CdpA organises and anchors the midcell ring in haloarchaea. Nat Commun 1075 16:5076. 1076 1077 120. Fadda D, Pischedda C, Caldara F, Whalen MB, Anderluzzi D, Domenici E, Massidda O. 1078 2003. Characterization of divIVA and Other Genes Located in the Chromosomal Region 1079 Downstream of the dcw Cluster in Streptococcus pneumoniae. J Bacteriol 185:6209–1080 6214. 1081 1082 121. Shrestha S, Harrison GA, Dressler JM, McNellis ME, Shen A. 2025. Penicillin-binding 1083 proteins exhibit functional redundancy during asymmetric cell division in Clostridioides 1084 difficile. J Bacteriol 207:e00503-25. 1085 1086 122. Osawa M, Erickson HP. 2006. FtsZ from Divergent Foreign Bacteria Can Function for Cell 1087 Division in Escherichia coli. J Bacteriol 188:7132–7140. 1088 1089 123. Moore DA, Whatley ZN, Joshi CP, Osawa M, Erickson HP. 2016. Probing for Binding 1090 Regions of the FtsZ Protein Surface through Site-Directed Insertions: Discovery of Fully 1091 Functional FtsZ-Fluorescent Proteins. J Bacteriol 199:10.1128/jb.00553-16. 1092 1093 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 30 124. Jacq M, Adam V, Bourgeois D, Moriscot C, Di Guilmi A-M, Vernet T, Morlot C. 2015. 1094 Remodeling of the Z-Ring Nanostructure during the Streptococcus pneumoniae Cell 1095 Cycle Revealed by Photoactivated Localization Microscopy. mBio 6:10.1128/mbio.01108-1096 15. 1097 1098 125. Ng YK, Ehsaan M, Philip S, Collery MM, Janoir C, Collignon A, Cartman ST, Minton NP. 1099 2013. Expanding the repertoire of gene tools for precise manipulation of the Clostridium 1100 difficile genome: allelic exchange using pyrE alleles. PLoS One 8:e56051-. 1101 1102 126. Karasawa T, Ikoma S, Yamakawa K, Nakamura S. 1995. A defined growth medium for 1103 Clostridium difficile. Microbiology (N Y) 141:371–375. 1104 1105 127. Bouillaut L, McBride SM, Sorg JA. 2011. Genetic manipulation of Clostridium difficile. 1106 Curr Protoc Microbiol 20:9A.2.1-9A.2.17. 1107 1108 128. Stylianidou S, Brennan C, Nissen SB, Kuwada NJ, Wiggins PA. 2016. SuperSegger: 1109 robust image segmentation, analysis and lineage tracking of bacterial cells. Mol Microbiol 1110 102:690–700. 1111 1112 129. Ducret A, Quardokus EM, Brun Y V. 2016. MicrobeJ, a tool for high throughput bacterial 1113 cell detection and quantitative analysis. Nat Microbiol 1:16077. 1114 1115 1116 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 31 FIGURES: 1117 1118 1119 Fig 1: SepF is required for the synthesis of division septa. (A) Organization of the operon containing 1120 sepF. The location targeted by CRISPRi sgRNAs are indicated. Both sepF and divIVA sgRNAs target the 1121 non-template strand. (B) A C. difficile strain was generated harboring a plasmid -encoded xylose-inducible 1122 CRISPRi sepF-KD cassette in addition to a CRISPRi -resistant sepFim complementation cassette 1123 integrated in the genome under control of an aTc -inducible Ptet promoter. Culturing in 2.5% xylose results 1124 in induction of the sepF-KD cassette, and addition of 2.5 ng/mL aTc induces expression of the sepFim 1125 complementation construct. Cells were labeled with HADA to visualize sites of peptidoglycan synthesis 1126 and/or remodeling, then fixed for microscopy. Yellow arrows indicate division septa. Scale bars = 5 μm. 1127 (C) A C. difficile strain harboring a plasmid -encoded xylose-inducible CRISPRi divIVA-KD cassette was 1128 cultured in the presence of 2.5% xylose, then labeled with HADA and fixed for microscopy. Yellow arrows 1129 indicate division septa. Scale bars = 5 μm. (D) C. difficile harboring a plasmid -encoded xylose-inducible 1130 sepF-KD or divIVA-KD CRISPRi cassette were cultured in the absence ( -) or presence (+) of 2.5% xylose 1131 for approximately 6 hr ( ~8 doublings) and western blotting was performed for DivIVA with GDH as a 1132 loading control. Data are representative of at least two independent experiments. 1133 1134 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 32 1135 1136 1137 Fig 2: Localization profile of essential C. difficile divisome proteins. (A-B) C. difficile strains were 1138 made harboring expression cassettes to produce mScarlet -I3 (mScI3) fusion proteins under control of an 1139 aTc-inducible Ptet promoter from an ectopic site in the genome. To induce production of the fusion protein, 1140 logarithmically growing cells were cultured in the presence of the following concentrations of aTc for 1 hr: 1141 0.5 ng/mL aTc for FtsZ -mScI3 and mScI3 -ZapA, 1 ng/mL aTc for SepF -mScI3, and 2.5 ng/mL aTc for 1142 mScI3-PBP1. Cells were labeled with HADA for 10 min to visualize sites of peptidoglycan synthesis 1143 and/or remodeling and then fixed for fluorescence microscopy. (A) Merged images containing phase and 1144 mScI3 signal (top) or phase and HADA signal (bottom) are shown, and yellow arrows point to examples of 1145 divisome protein foci. Scale bars = 5 μm. (B) Demographs were generated using MicrobeJ to visualize the 1146 medial axis fluorescence profile of ≥500 cells. Cells are ordered by length, and the mScI3 or HADA signal 1147 along the distance of the cell relative to mid -cell is shown in magenta and cyan, respectively. Data in A -B 1148 are representative of at least three independent experiments. 1149 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 33 1150 1151 Fig 3: Assessing the functionality of fluorescent protein fusions to essential divisome proteins. 1152 (A-E) Plasmid -encoded xylose -inducible CRISPRi -KD cassettes that either (A) are non -targeting 1153 (negative control), or target (B) ftsZ, (C) zapA, (D) sepF, or ( E) pbp1, were introduced into C. difficile 1154 strains carrying chromosomally -encoded complementation constructs that are “immune” to CRISPRi 1155 targeting. These CRISPRi -immune constructs were expressed under control of an aTc -inducible P tet 1156 promoter. The indicated strains of C. difficile were cultured for ~6 hr in the presence of 2.5% xylose, to 1157 induce the CRISPRi-KD cassette, and aTc (ftsZ/negative control = 1 ng/mL; zapA = 50 ng/mL; sepF = 2.5 1158 ng/mL; pbp1 = 2 ng/mL), to induce expression of the CRISPRi -immune complementation construct. The 1159 cultures were then labeled with HADA to visualize sites of peptidoglycan synthesis and/or remodeling and 1160 fixed for microscopy. Merged images containing phase/HADA (blue) or phase/mScI3 (magenta) are 1161 shown. The HADA signal is normalized across all images, but the mScI3 signal was scaled such that the 1162 localization for each individual fusion protein was more easily detected. The yellow arrow in C points to an 1163 example of a HADA “spiral”; these structures were only observed in zapA-KD/mScI3-zapAim cells. More 1164 examples of these structures can be found in Fig S3. All data are representative of at least two 1165 independent experiments. Scale bars = 5 μm. 1166 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 34 1167 1168 Fig 4: Order of assembly of essential divisome proteins. (A-B) Constructs encoding tagged divisome 1169 proteins were each expressed from an ectopic locus under the control of an aTc -inducible Ptet promoter; 1170 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 35 xylose-inducible CRISPRi-KD constructs are encoded on a plasmid. Cells were cultured with 2.5% xylose 1171 for 6 hr and pulsed with aTc (FtsZ-mScI3 = 0.5 ng/mL; mScI3-ZapA = 0.5 ng/mL; SepF-mScI3 = 1 ng/mL; 1172 mScI3-PBP1 = 2.5 ng/mL). for the last hour of the treatment before labeling with HADA and fixing cells for 1173 microscopy. (A) Control C. difficile strains producing the indicated divisome protein -mScarlet-I3 (mScI3) 1174 fusion and harboring a negative control CRISPRi cassette with non -targeting sgRNA were visualized by 1175 microscopy to ensure that the CRISPRi cassette itself does not impact the protein localization . Yellow 1176 arrows point to divisome protein foci. (B) C. difficile cells producing divisome protein -mScI3 fusions upon 1177 the indicated CRISPRi ftsZ-KD, zapA-KD, sepF-KD, and pbp1-KD cassettes were visualized by 1178 microscopy. Yellow arrows point to divisome protein foci. The HADA signal is scaled equally across all 1179 images, but the mScI3 signal was scaled independently for optimal protein localization The data in A -B 1180 are representative of at least two independent experiments. Scale bars = 5 μm. (C) The order of 1181 assembly of the essential divisome proteins is depicted. The dotted line indicates that while SepF can 1182 form foci without ZapA, the foci appear irregular and diffuse, suggesting that ZapA is partially required for 1183 proper SepF localization. 1184 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 36 1185 1186 Fig 5: Localization profile of non-essential divisome proteins. (A-B) Constructs encoding mScarlet-I3 1187 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 37 (mScI3) fusion proteins under the control of an aTc- inducible P tet promoter were expressed from an 1188 ectopic site in the genome. To induce production of the fusion protein, logarithmically growing cells were 1189 cultured in the presence of aTc for 1 hr (mScI3-MldA = 1 ng/mL; mScI3-MldB = 0.5 ng/mL; mScI3-MldC = 1190 0.5 ng/mL; DivIVA -mScI3 = 0.5 ng/mL; mScI3 -FtsK = 5 ng/mL; mScI3 -PBP3 = 5 ng/mL). Cells were 1191 labeled with HADA for 10 min to visualize sites of peptidoglycan synthesis and/or remodeling and then 1192 fixed for fluorescence microscopy. (A) Merged images containing phase and mScI3 signal (top) or phase 1193 and HADA signal (bottom) are shown, and yellow arrows point to examples of divisome protein foci. Scale 1194 bars = 5 μm. (B) Demographs were generated using MicrobeJ to visualize the medial axis fluorescence 1195 profile of ≥500 cells. Cells are ordered by length, and the mScI3 or HADA signal along the distance of the 1196 cell relative to mid -cell is shown in magenta and cyan, respectively. Data in A- B are representative of at 1197 least three independent experiments. 1198 1199 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 38 1200 1201 Fig 6: Dependence of non -essential divisome proteins on FtsZ, SepF, and PBP1 for localization. 1202 (A) Constructs encoding fluorescent protein fusions to divisome proteins were expressed from an ectopic 1203 locus under the control of an aTc -inducible P tet promoter during xylose -inducible CRISPRi- KD of the 1204 indicated genes. The CRISPRi -KD constructs were expressed from a plasmid. Cells were cultured with 1205 2.5% xylose for 6 hr and pulsed with aTc for the last hour of the treatment (mScI3 -MldA = 1 ng/mL; 1206 mScI3-MldB = 0.5 ng/mL; mScI3 -MldC = 0.5 ng/mL; DivIVA -mScI3 = 0.5 ng/mL; mScI3 -FtsK = 5 ng/mL; 1207 mScI3-PBP3 = 5 ng/mL) before labeling with HADA and fixing cells for microscopy . Divisome protein foci 1208 are indicated by arrows. Cyan arrows represent divisome protein foci that are co -localized with septa in 1209 the pbp1-KD strain, which is likely due to incomplete depletion of PBP1 protein leading to a low -level of 1210 septum synthesis. White arrows represent divisome protein foci within cells that did not label efficiently 1211 with HADA, so the presence or absence of septa at the site of protein localization is unknown. Yellow 1212 arrows indicate divisome protein foci that localize in the absence of septa, specifically in cells that label 1213 well with HADA. Data are representative of at least two independent experiments. Scale bars = 5 μm. (B) 1214 The order of divisome assembly is depicted. The dotted line indicates that ZapA is only partially required 1215 for proper SepF localization. The proteins labeled with an * indicate that the protein only formed foci when 1216 septa were detectable; these septa likely form because PBP1 is only partially depleted. 1217 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint 39 1218 1219 Fig 7: Phenotypes of deletion mutants lacking non -essential divisome genes. (A-B) The indicated 1220 clean deletion strains of C. difficile were cultured to logarithmic phase and labeled with HADA. (A) yellow 1221 arrows point to division septa. (B) The ΔmldA or ΔdivIVA mutants were complemented with either an 1222 empty vector integrated into an ectopic locus “-“ or the indicated expression construct. Scale bars = 5 μm. 1223 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 19, 2026. ; https://doi.org/10.64898/2026.01.19.700290doi: bioRxiv preprint