Activation of SEDS-PBP cell wall synthases by an essential regulator of bacterial division

Bacterial growth and division require insertion of new peptidoglycan (PG) into the existing cell wall by PG synthase enzymes. Emerging evidence suggests that many PG synthases require activation to function, however it is unclear how activation of division-specific PG synthases occurs. The FtsZ cytoskeleton has been implicated as a regulator of PG synthesis during division, but the mechanisms through which it acts are unknown. Here we show that FzlA, an essential regulator of constriction in Caulobacter crescentus, links FtsZ to PG synthesis to promote division. We find that hyperactive mutants of the PG synthases FtsW and FtsI specifically render fzlA, but not other division genes, non-essential. However, FzlA is still required to maintain proper constriction rate and efficiency in a hyperactive PG synthase background. Intriguingly, loss of fzlA in the presence of hyperactivated FtsWI causes cells to rotate about the division plane during constriction and sensitizes cells to cell wall-specific antibiotics. We demonstrate that FzlA-dependent signaling to division-specific PG synthesis is conserved in another α-proteobacterium, Agrobacterium tumefaciens. These data establish that FzlA links FtsZ to cell wall remodeling, serving both to activate and spatially orient PG synthesis during division. Overall, our findings support the paradigm that activation of SEDS-PBP PG synthases is a broadly conserved requirement for bacterial morphogenesis.


Abstract 21
Bacterial growth and division require insertion of new peptidoglycan (PG) into the 22 existing cell wall by PG synthase enzymes. Emerging evidence suggests that many PG 23 synthases require activation to function, however it is unclear how activation of division-24 specific PG synthases occurs. The FtsZ cytoskeleton has been implicated as a regulator of 25 PG synthesis during division, but the mechanisms through which it acts are unknown. 26 Here we show that FzlA, an essential regulator of constriction in Caulobacter crescentus, 27 links FtsZ to PG synthesis to promote division. We find that hyperactive mutants of the 28 PG synthases FtsW and FtsI specifically render fzlA, but not other division genes, non-29 essential. However, FzlA is still required to maintain proper constriction rate and 30 efficiency in a hyperactive PG synthase background. Intriguingly, loss of fzlA in the 31 presence of hyperactivated FtsWI causes cells to rotate about the division plane during 32 constriction and sensitizes cells to cell wall-specific antibiotics. We demonstrate that 33 FzlA-dependent signaling to division-specific PG synthesis is conserved in another α-34 proteobacterium, Agrobacterium tumefaciens. These data establish that FzlA links FtsZ to 35 cell wall remodeling, serving both to activate and spatially orient PG synthesis during 36 division. Overall, our findings support the paradigm that activation of SEDS-PBP PG 37 synthases is a broadly conserved requirement for bacterial morphogenesis. Bacterial division is driven by the insertion of new cell wall material at midcell in a 43 tightly regulated manner, allowing for determination of cell shape and maintenance of 44 envelope integrity 1,2 . The cell wall is made of peptidoglycan (PG), a meshwork consisting 45 of glycan strands crosslinked by peptide stems 3,4 . PG synthesis requires the coordination 46 of glycan polymerization and peptide crosslinking by either coupled monofunctional 47 glycosyltransferases (GTases) and transpeptidases (TPases), or bifunctional enzymes that 48 contain both activities, with these proteins being more generally referred to as PG 49 synthases 1 . 50 Monofunctional PG synthase pairs have been implicated as the primary synthetic 51 enzymes of the elongation (elongasome) and division (divisome) machineries. A 52 paradigm has been proposed whereby a shape, elongation, division, and sporulation 53 (SEDS) family GTase is functionally coupled to a penicillin binding protein (PBP) 54 TPase, which together facilitate cell wall synthesis [5][6][7] . Through characterization of the 55 elongation-specific PG synthases RodA and PBP2 in Escherichia coli, it has been 56 postulated that SEDS-PBP enzymes require activation to function 5 . Specifically, 57 mutations in RodA or PBP2 that increase GTase activity in vitro and PG synthesis in 58 cells render other components of the elongasome non-essential, arguing that their normal 59 function is to activate the RodA-PBP2 complex 5 . Intriguingly, analogous mutations in the 60 division-specific SEDS-PBP enzymes, FtsW and FtsI, allow cells to constrict faster than 61 normal 8 , suggesting that these mutations promote formation of an activated PG synthase 62 complex 5,9 . However, it is unclear precisely how SEDS-PBP activation normally occurs 63 during division. 64 ftsW* ΔfzlA cells are longer than ftsW**I* ΔfzlA cells, we conclude that ftsW**I* 106 suppresses loss of fzlA better than the single mutant. 107 We also observed that ftsW**I* suppresses length, width, and fitness defects 108 associated with slowly constricting fzlA point mutants fzlA NH2 and fzlA NH3 (Fig. S2, Fig.  109 S1B), further indicating that hyperactivated ftsWI are dominant to, and likely downstream 110 of fzlA. To determine the contribution of the FtsZ-FzlA interaction to activation of FtsWI, 111 we assessed cell morphology, fitness, and cell length of ftsW**I* strains containing FzlA 112 mutants with decreasing affinity for FtsZ 22 (Fig. S3, Fig. S1C; FzlA > FzlA NH2 = 113 FzlA NH3 > FzlA NH1 ; FzlA NB2 , FzlA NB1 = no binding). We found that decreased affinity of 114 FzlA towards FtsZ correlated with an increase in cell length (Fig. S3E), indicating that 115 high-affinity binding to FtsZ is required for FzlA to signal to FtsWI. 116 117

FzlA plays a specific and unique role in activating FtsWI 118
To assess the specificity of the fzlA-ftsWI genetic interaction and potentially 119 identify additional components of this pathway, we performed comparative transposon 120 sequencing (Tn-Seq) on WT and ftsW**I* strains. Surprisingly, fzlA was the only 121 essential gene to become non-essential in the ftsW**I* background, with few insertions 122 in WT but plentiful insertions in ftsW**I* cells (Fig. 1F,G, Supplementary Table 1). 123 All other known essential division genes, e.g. ftsZ (Fig. 1G), had few transposon 124 insertions in either background. These data indicate that fzlA is specific and unique in its 125 essential role upstream of ftsWI. We suspect that other essential division proteins 126 participate in this pathway as well, but that they play additional essential functions in 127 divisome assembly or activity. 128

FzlA contributes to efficient division in a hyperactive PG synthase background 129
Given that cells lacking fzlA in the hyperactive PG synthase backgrounds were elongated, 130 we assessed constriction rate and division efficiency in these strains in more detail. 131 Specifically, we performed time-lapse microscopy on ftsW**I* and ftsW* cells ± fzlA and 132 tracked division in cells using MicrobeJ 22,23 ( Fig. 2A, Supplementary Video 1). 133 Consistent with previous findings 8 , ftsW**I* and ftsW* cells constrict more quickly than 134 WT (Fig. 2B). Intriguingly, the hyperactive PG synthase strains lacking fzlA constricted 135 significantly more slowly than the corresponding strain with fzlA present, with 136 constriction rates cut nearly in half (Fig. 2B). This suggests that hyperactivated FtsWI are 137 not sufficient for efficient division and underscores the importance of FzlA in dictating 138 constriction rate. As with cell length and fitness, ftsW**I* acted as a better suppressor to 139 fzlA deletion, allowing for a faster constriction rate than did ftsW* (Fig. 2B). 140 To ensure that changes in constriction rate were not due to global differences in 141 PG synthesis, we determined elongation rates across strains (Fig. 2C), which enabled 142 calculation of the ratio of constriction to elongation rate (Fig. 2D). We saw the same 143 trend as for constriction rate itself, with ftsWI** and ftsW* mutant strains having higher 144 ratios of constriction to elongation and loss of fzlA giving lower ratios (Fig. 2D). 145 Interestingly, elongation rate was inversely correlated with constriction rate in all mutant 146 strains (Fig. 2C, Fig. 2D), perhaps reflecting competition between the elongasome and 147 divisome for PG precursor substrate 24 . Altogether, these data support the conclusion that 148 alterations to the ftsZ-fzlA-ftsWI pathway specifically affect constriction, with FzlA 149 increasing the constriction rate in both WT and hyperactive PG synthase mutant 150

backgrounds. 151
While tracking division in ftsW**I* ΔfzlA and ftsW* ΔfzlA cells to measure 152 constriction, we noticed that some cells initiated constriction at one location, then aborted 153 division at that location before successfully dividing at a second (or third or fourth) site 154 (Fig. 2E, Supplementary Video 2). We quantified the frequency of such constriction 155 failure events and found that 16.6-19.5% of the hyperactive PG synthase cells lacking 156 fzlA aborted division at one site before successfully dividing at another, compared to a 0-157 0.3% failure rate for WT or hyperactive PG synthase cells with fzlA present (Fig. 2F). 158 These data further demonstrate that ftsW**I* are not sufficient for efficient division, and 159 that fzlA is required to ensure division processivity and efficiency. 160 161 fzlA is required for maintenance of proper cell shape 162 As mentioned earlier, deletion of fzlA in the hyperactive PG synthase backgrounds 163 impacted global cell morphology, with many pre-divisional cells appearing "S-shaped". 164 In order to more carefully assess this phenotype, we imaged cells by scanning electron 165 microscopy (SEM). We saw a relatively high frequency of S-shaped ftsW**I* DfzlA cells,

166
whereas most WT or ftsW**I* cells displayed the typical "C-shaped" morphology 167 characteristic of Caulobacter (Fig. 3A). We quantified the frequency of S-shaped cells in 168 a population of dividing cells by phase contrast microscopy to assess penetrance of this 169 morphological phenotype. We extracted outlines of individual cells and performed 170 principal component analysis using Celltool to isolate variance in cell shape to features 171 referred to as shape modes 22,25 . Shape mode 3 captured the variation due to degree of S-172 versus C-shape and we set a cutoff such that cells with a standard deviation |sd| > 1 from 173 the mean for this shape mode are considered S-shaped (Fig. 3B,C). Means and medians 9 were similar for degree of S-shape across populations, with no significant difference for 175 means, and a statistically significant but numerically small difference for medians. 176 However, there was an obvious and significant difference in variance in degree of S-177 shape across populations (Fig. 3B), corresponding with a large difference in the number 178 of cells found to be S-shaped in different strains. Over a quarter (26.9%) of dividing 179 ftsW**I* DfzlA cells displayed an S-shaped morphology, compared to 2.4% of WT and 180 1.1% of ftsW**I* cells that are S-shaped (Fig. 3D). 181 To shed light on the origin of S-shape, we next asked at what point during growth 182 do ftsW**I* DfzlA cells begin to adopt this morphology. Using time-lapse microscopy,

183
we observed that ftsW**I* DfzlA cells were C-shaped at the beginning of the cell cycle 184 and began to twist or rotate about the division plane after constriction initiated. S-shape 185 only became apparent in the latter part of constriction, when daughters had rotated ~180° 186 relative to each other (Fig. 3E, Supplementary Video 3). This finding suggests that the 187 fzlA-ftsWI pathway determines geometry of PG insertion at the site of division in a 188 manner that influences global cell morphology, normally constraining cells in their 189 characteristic C-shape as constriction progresses. These results also indicate that our 190 quantification method for S-shape likely underestimates the number of twisted ftsW**I* 191 DfzlA cells, since S-shape is not obvious by phase contrast until the end of constriction,

192
and we quantified cell shape at all stages of the constriction process. and Venus-MreB localization in ftsW**I* DfzlA cells was comparable to ftsW**I* cells 199 (Fig. S4, S5). Additionally, we visualized the localization of PG synthesis using the 200 fluorescent D-amino acid HADA 27 , in order to assess whether cell twisting might be 201 induced by mislocalized PG synthesis in spite of properly localized FtsZ and MreB. 202 However, we did not detect any gross changes in HADA localization (Fig. S6). Together, 203 these findings suggest that cell twisting is likely induced by a finer scale alteration of PG 204 synthesis at the division site due to disruption of the fzlA-ftsWI pathway. 205 To determine whether global shape regulation depends on the FtsZ-FzlA 206 interaction, we assessed S-shaped cell frequency in strains containing mutants of FzlA 207 displaying decreasing affinities towards FtsZ. We observed that affinity of the mutant 208 FzlA for FtsZ was inversely correlated with the frequency of S-shaped cells (Fig. S7), 209 verifying that the FtsZ-FzlA interaction is important for maintaining proper morphology. 210

211
The ftsZ-fzlA-ftsWI pathway contributes to resistance to PBP-targeting antibiotics 212 Because FzlA is important for regulation of PG synthesis in the context of determining 213 constriction rate and cell shape, we hypothesized that it might also contribute to 214 resistance to cell wall-targeting antibiotics. To test this, we challenged cells with 215 antibiotics targeting PG synthetic processes and assessed resulting cell fitness. ftsW**I*, 216 as has been previously shown 9 , displayed sensitivity to cephalexin (Fig. 4A), which 217 inhibits FtsI and other penicillin-binding proteins in Caulobacter 28,29 . Interestingly, 218 deletion of fzlA in the ftsW**I* background exacerbated sensitivity to cephalexin ( Fig. elongation-specific PG synthase PBP2 30,31 , whereby the minimum inhibitory 221 concentration (MIC) for ftsW**I* cells was decreased compared to WT, with deletion of 222 fzlA further lowering the MIC (Fig. 4B). We also treated cells with the β-lactam 223 ampicillin 32 and with the cell wall targeting antibiotics vancomycin, which blocks 224 transpeptidation by a distinct mechanism from β-lactams 33 , and fosfomycin, which 225 inhibits cell wall synthesis by blocking PG precursor availability 34 (Fig. S8). Neither 226 hyperactivation of ftsWI nor loss of fzlA yielded a change in MIC in the presence of any 227 of these antibiotics (Fig. S8). fzlA therefore supports robust cell wall synthesis in the 228 presence of certain PG-targeting drugs, perhaps by compensating for inactivation of 229 specific PBPs. To determine if the interaction between FtsZ and FzlA is important for 230 maintaining cell wall integrity, we assessed sensitivity to cephalexin using the panel of 231 fzlA mutants which display varying affinities towards FtsZ (Fig. S9). We found that 232 mutants with decreased FzlA affinity towards FtsZ in fact became more sensitive to 233 cephalexin (Fig. S9), demonstrating that the entire ftsZ-fzlA-ftsWI pathway is required for 234 promoting cell wall integrity during antibiotic treatment. 235 Since ftsW**I* cells are more sensitive to perturbation of other PG synthetic 236 activities even when fzlA is present, we asked if any normally non-essential division 237 genes become more important for fitness in an ftsW**I* background, as they might help 238 bolster resistance to assaults on PG synthesis. Examination of the ftsW**I* Tn-Seq data 239 indicated that pbpX (encoding a bifunctional PG synthase that localizes to midcell) 35,36 , 240 and to a lesser extent ftsX (encoding a cell separation factor) 37 and dipM (encoding an 241 envelope maintenance/cell separation factor) 37-40 , had fewer transposon insertions in an misregulated division site PG synthase activity, we suspect that FtsEX, DipM, and PbpX 244 become important for ensuring robust PG synthesis during constriction and, later, 245 efficient cell separation. Surprisingly, the normally non-essential nhaA locus, coding for a 246 putative sodium-proton antiporter 41 , was also predicted by Tn-Seq to become essential in 247 ftsW**I* cells (Fig. 1F, Fig. S10B). Disruption of nhaA in the presence of sucrose has 248 been shown to arrest division 41 , suggesting nhaA may be important for division under 249 certain conditions. It is unclear why it also becomes important upon PG synthesis mis-250 regulation, but its role in osmoregulation may contribute to its apparent synthetic lethality 251 with ftsW**I*. division arrest, and ectopic pole formation at midcell (Fig. 5, Fig. S11), reminiscent of 267 FtsW depletion in A. tumefaciens 18 . Importantly, we found that the decrease in viability 268 and morphology defects associated with depletion of FzlA were rescued by ftsW F137L 269 (Fig. 5, Fig. S11). These data indicate that FzlA's essential role in regulating division-270 specific PG synthesis is conserved in another α-proteobacterium and further highlight the 271 importance of FzlA as a key regulator of constriction and cell morphology. Here we have described a conserved PG synthesis activation pathway in which FtsZ and 275 FzlA signal through FtsWI to regulate wall synthesis during division in α-proteobacteria 276 ( Fig. 6, left panel). Specifically, the FtsZ-FzlA-FtsWI pathway determines geometry of 277 cell wall insertion at the site of division, sets the constriction rate, and promotes cell wall 278 integrity (Fig. 6, left panel). FtsW**I* can still receive input from FzlA which, in 279 combination with their intrinsic hyperactivity, leads to shorter, faster-constricting cells 280 with sensitivity to cell wall antibiotics (Fig. 6, middle panel). In the absence of fzlA, 281 ftsW**I* cells lose critical regulation of PG synthesis, leading to twisting during division, 282 slower constriction, and increased sensitivity to cell wall antibiotics. We establish FzlA 283 as a key intermediary in signaling from FtsZ to FtsWI and demonstrate that this division-284 specific SEDS-PBP pair require activation for normal division. Notably, our observations 285 indicate that FtsWI activity is regulated in multiple ways, likely including input into their 286 catalytic rates and modulation of the fine-scale geometry by which they insert new PG for 287

constriction. 288
Our findings provide the foundation for further mechanistic investigation into this 289 pathway and raise a number of questions. For one, the nature and timing of the activation 290 signal(s) are still unknown: is there a signal always emanating from FtsZ-FzlA that 291 induces constriction as soon as FtsW arrives, or is constriction triggered by a discrete 292 cellular event, such as clearance of the chromosomal termini or the arrival of a sparkplug 293 factor that jumpstarts FtsWI activity? Additionally, it is unclear why hyperactivation of 294 ftsWI together or ftsW alone can rescue loss of fzlA -are TPase and GTase activities study demonstrated that FtsW is activated by FtsI 7 , so it is possible that the ftsI* mutation 297 described here in fact hyperactivates FtsW. Finally, we have no evidence that FzlA and 298 FtsWI directly interact, and suspect that other intermediary factor(s) transduce the 299 activation signal from FzlA to FtsWI. 300 Our model advances the idea that regulation of SEDS-PBP pairs for growth and 301 division is conserved at numerous levels. The finding that FzlA governs division-specific 302 PG synthesis in both Caulobacter and A. tumefaciens argues that α-proteobacteria use 303 FzlA as a conserved and dedicated FtsWI activator. FzlA is absent outside of this clade, 304 however, so we propose that other divisome components serve as FtsWI activators in 305 other organisms. More broadly, our findings expand the paradigm for PG synthesis by 306

SEDS-PBP PG synthase pairs in bacteria and provide evidence that the requirement for 307
PG synthase activation is conserved. Elongation is facilitated by the coordination of the 308 SEDS family GTase RodA and the monofunctional TPase PBP2, orthologs to FtsW and 309 FtsI, respectively 5 . The proposed model for elongation activation, as described for E. coli, 310 holds that these PG synthases are activated by another protein, MreC, forming an 311 activated complex that in turn regulates assembly and directional motion of the 312 polymerizing scaffold MreB 5 . In this system, hyperactivating mutations in RodA or PBP2 313 allow for bypass of the activator, MreC, similar to our finding that FtsW**I* can bypass 314 the activator FzlA. Our data provide experimental support for the proposal that the 315 requirement for activation of the SEDS-PBP pair of PG synthases is generally conserved 316 for elongation and division. 317 There are prominent differences between the models for elongasome and version of the divisome 45 : the elongasome contains fewer proteins than the divisome 45 , 320 and when either RodA or PBP2 is hyperactivated through mutation, all elongasome 321 components except MreB and the PG synthases are rendered dispensable 5 . This would 322 suggest that for elongation, the cell needs an activated SEDS-PBP pair and a spatial 323 regulator to orient their motion 5 . Conversely, hyperactive FtsWI in Caulobacter only 324 allows for disruption of FzlA, with the rest of the divisome remaining essential. This may 325 be because division is a more complex process than elongation, requiring invagination  For A. tumefaciens time-lapse microscopy, images were collected every ten minutes. 395

Microscopy of A. tumefaciens cells was performed with an inverted Nikon Eclipse TiE 396
with a QImaging Rolera em-c 2 1K EMCCD camera and Nikon Elements Imaging 397