Determinants of FtsZ C-terminal linker-dependent regulation of cell wall metabolism in Caulobacter crescentus

Bacterial cell division requires assembly of a multi-protein machinery or “divisome” that remodels the cell envelope to cause constriction. The cytoskeletal protein FtsZ forms a ring-like scaffold for the divisome at the incipient division site. FtsZ has three major regions – a conserved, polymerizing GTPase domain; a C-terminal conserved (CTC) peptide required for binding membrane-anchoring proteins; and a C-terminal linker (CTL) of poor length and sequence conservation. We previously demonstrated that, in Caulobacter crescentus, the CTL regulates FtsZ polymerization in vitro and cell wall metabolism in vivo. To understand the mechanism of CTL-dependent regulation of cell wall metabolism, here we investigated the impact of the CTL on Z-ring structure in cells and employed genetics to identify molecular determinants of the dominant lethal effects of ΔCTL. Deleting the CTL specifically resulted in formation of dense, asymmetric, non-ring FtsZ assemblies in vivo. Moreover, we observed that production of an FtsZ variant with the GTPase domain of Escherichia coli FtsZ fused to the CTC of C. crescentus FtsZ phenocopied the effects of C. crescentus ΔCTL, suggesting the CTC mediates signaling to cell wall metabolism. Finally, whereas overproduction of ZapA, FzlC, or FtsEX had slight protective effects against ΔCTL, depletion of FtsA partially suppressed the effects of ΔCTL. From these results, we propose that the cell wall misregulation downstream of ΔCTL results from its aberrant assembly properties and is propagated through the interaction between the CTC of FtsZ and FtsA. Our study provides mechanistic insights into CTL-dependent regulation of cell wall enzymes downstream of FtsZ. Importance Bacterial cell division is essential and requires the recruitment and regulation of a complex network of proteins needed to initiate and guide constriction and cytokinesis. FtsZ serves as a master regulator for this process, and its function is highly dependent on both its self-assembly into a canonical “Z-ring” and interaction with protein binding partners, which results in the activation of enzymes that remodel the cell wall to drive constriction. Using mutants of FtsZ and its binding partners, we have established the role of its C-terminal linker domain in regulating Z-ring organization, as well as the requirement for its C-terminal conserved peptide and interaction with the membrane-anchoring protein FtsA for regulating cell wall remodeling for constriction.

activity of the divisome-associated cell wall enzymes. Collectively these data indicate that Z-ring 73 assembly properties are directly relevant to the regulation of local cell wall remodeling. 74 However, the pathways downstream of Z-ring assembly that regulate cell wall enzymes are 75 largely unknown. 76 FtsZ has three regions: (i) a conserved GTPase domain, (ii) a C-terminal linker (CTL), and (iii) a 77 conserved C-terminal peptide (CTC) (Figure S1) (7). The GTPase domain is structurally similar 78 to eukaryotic tubulin (8-9) and is sufficient for polymerization on binding GTP (4,10). 79 Mutations in the GTPase domain affect Z-ring dynamics, organization, and regulation of cell 80 wall synthetic enzymes, at least in some bacteria (5-6). The CTC is composed of a conserved α- are tolerated to some extent, but significant changes to CTL sequence impact protein stability 91 and, therefore, cell division (4). Complete deletion of the CTL causes dominant lethal defects in 92 Z-ring assembly and cell lysis, at least in C. crescentus and B. subtilis (4,11). Identifying the 93 contributions of the CTL to FtsZ function is essential to understanding the communication 94 between Z-ring structure and cell wall enzyme activities.
We previously showed that the expression of FtsZ lacking its CTL ("ΔCTL", wherein the 96 GTPase domain is fused directly to the CTC) in the α-proteobacterium C. crescentus causes 97 misregulation of cell wall enzymes resulting in the formation of spherical envelope bulges at the 98 sites of ΔCTL assembly and rapid cell lysis (4). Using a fluorescent fusion to ZapA, a protein 99 that binds FtsZ, we found that FtsZ superstructure was affected in ΔCTL: ∆CTL formed large, 100 amorphous assemblies instead of focused rings (4). FtsZ with a minimal CTL of 14 amino acids 101 (L14) exhibited WT-like Z-ring shape and did not lead to bulging and lysis. In vitro, ∆CTL 102 polymerizes into straight multi-filament bundles that are significantly longer than the curved 103 protofilaments observed for WT FtsZ or L14 by electron microscopy (4,10). Moreover, ∆CTL 104 exhibits lower GTP hydrolysis rates, reduced polymer turnover, and increased protofilament 105 lateral interactions compared to WT FtsZ in vitro (4,10,14). These effects result in the 106 formation of stable networks of ∆CTL protofilaments on membranes, in contrast to small 107 dynamic clusters formed by WT FtsZ, when observed on supported lipid bilayers by total 108 internal reflection fluorescence microscopy (14). Unlike the CTL, the CTC does not significantly 109 contribute to polymer structure or dynamics for C. crescentus FtsZ -polymer structure, observed 110 by EM, or GTP hydrolysis rates of FtsZ lacking its CTC (∆CTC) are comparable to those of WT 111 FtsZ (10). In vivo, ∆CTC forms Z-rings similar to WT but is incapable of cytokinesis (4 suggest that the CTL is required for proper Z-ring assembly, and the interaction between the 124 CTC and FtsA is required for CTL-dependent signaling from Z-ring structure and/or dynamics to 125 the regulation of cell wall enzymes in cells.

128
The CTL of FtsZ impacts Z-ring superstructure in vivo 129 We previously observed that the CTL affects the higher order assembly of FtsZ polymers in vitro 130 (10, 14). Z-ring structure also appears to be regulated in a CTL-dependent manner in cells when 131 imaged using a fluorescently-labeled FtsZ binding protein (ZapA-Venus) (4). To test if the 132 differences in ZapA-Venus structures observed previously reliably reflect differences in Z-ring 133 organization, we directly visualized N-terminal monomeric-NeonGreen (mNG) fluorescent 134 fusions to FtsZ or ΔCTL using epifluorescence microscopy. We found that mNG-ΔCTL 135 produced from a xylose-inducible promoter (P xylX ) in the presence or absence of WT FtsZ caused 136 filamentation, local envelope bulges, and rapid cell lysis, indicating that mNG-ΔCTL is 137 dominant lethal, similar to untagged ΔCTL ( Figures 1A, S2, S5). This allowed us to compare 138 structures formed by FtsZ or ΔCTL in vivo to identify CTL-dependent differences in the Z-ring 139 that might correlate with bulging and lysis. In our analysis, we also included mNG fusions to (i) 140 L14 -an FtsZ CTL variant with a truncated 14 amino acid CTL that is incapable of cytokinesis 141 but does not cause bulging and lysis, and (ii) HnCTL -an FtsZ variant with the CcCTL sequence 142 replaced with the CTL from Hyphomonas neptunium FtsZ that causes inefficient cytokinesis 143 (elongation and slower doubling time), as controls (4) ( Figure S1). 144 We expressed mNG fusions to FtsZ or CTL variants using the xylose-inducible P xylX promoter 145 while simultaneously depleting WT FtsZ using strains wherein the only copy of ftsZ is under the 146 control of the vanillate-inducible P vanA promoter. mNG-FtsZ formed ring-like structures that 147 appear as a band or two closely spaced foci aligned along the short axis or a single focus per 148 dividing cell after 1 hour of induction ( Figure 1A). At longer induction times, mNG-FtsZ Z-ring 149 structure was maintained despite cell filamentation due to depletion of WT FtsZ ( Figure S2). Unlike WT FtsZ, within 1 hour of induction, mNG-∆CTL formed one or more wider and less 151 ring-like structures ( Figure 1A). These structures increased in size and intensity over time 152 ( Figure S2). Frequently, mNG-∆CTL structures appeared to be asymmetrically distributed along 153 the short axis of the cell ( Figure 1A, S2). mNG-L14 assembled into apparently less dense, 154 diffuse structures after 1 hour of induction ( Figure 1A) which became more diffuse and scattered 155 at longer induction times ( Figure S2). mNG-HnCTL structures appeared predominantly as faint 156 rings or foci or more dispersed structures similar to mNG-L14, and did not change significantly 157 with longer induction or cell filamentation ( Figure 1A, S2). 158 We quantitatively analyzed Z-ring intensities and structures using MicrobeJ (15) and Oufti (16).

159
To avoid potential effects of cell length on Z-ring organization, we focused on cells 3-5 µm long, proportion of fluorescence signal in the Z-ring than those expressing mNG-FtsZ ( Figure 1C).

169
The fraction of mNG-HnCTL was lower than each of the other CTL variants, suggesting a lower 170 tendency to assemble into polymers at the Z-ring, while that of mNG-L14 was similar to WT 171 FtsZ ( Figure 1C). We next measured the mean fluorescence intensity (i.e. density) of Z-rings in 172 each strain to determine whether variant Z-rings were more or less diffuse than those formed by WT FtsZ. While mNG-FtsZ and mNG-∆CTL had similar mean intensity, mNG-L14 and mNG-

174
HnCTL each formed less intense structures ( Figure 1D), consistent with their apparent 175 "dimness" in the images and our biochemical studies indicating that L14 and HnCTL do not 176 polymerize as robustly as FtsZ or CTL (4, 10).

177
To address whether protein levels are affected by CTL composition, we determined the relative concentration of protein for each of these variants. Using quantitative immunoblotting, we found 182 that indeed, mNG-∆CTL and mNG-L14 levels were ~5-fold higher than mNG-FtsZ or mNG-

183
HnCTL and that these levels increased relative to mNG-FtsZ over time ( Figure S4). mNG-L14 184 was present at higher levels than mNG-∆CTL, whereas mNG-HnCTL had levels nearly 185 equivalent to mNG-FtsZ ( Figure S4). Since all mNG fusions were expressed using identical  We previously observed that, in vitro, HnCTL and L14 form relatively few, unbundled filaments 204 while ∆CTL forms bundled, more stable filaments when compared to WT FtsZ (4, 10). We EcGTPase and CcCTC (EcGTPase-CcL14-CcCTC) did not cause constriction or cell envelope 254 defects. We confirmed by immunoblotting using antibodies against both C. crescentus and E. 255 coli FtsZ that there were no significant differences in the expression levels of these chimeras that 256 could account for the differences in phenotypes observed ( Figure S7).

The CTL of E. coli FtsZ has a modest effect on lateral interactions in vitro
While Ec∆CTL by itself was unable to cause bulging and rapid lysis, the pattern of HADA 266 localization -diffuse, asymmetric foci -suggests that Ec∆CTL (EcGTPase-EcCTC) could still 267 affect the organization of cell wall synthetic enzymes. We hypothesized that, similar to Cc∆CTL, 268 Ec∆CTL also had aberrant assembly properties that affect downstream localization or activity of  Collectively, our data thus far indicate that the CTC is required for ∆CTL-mediated signaling 307 through FtsA to misregulate cell wall metabolism, but that all other non-essential FtsZ-binding 308 proteins are dispensable for this signaling.  Figure S9). We confirmed that the effects of 331 overproducing these proteins on ∆CTL-induced bulging were not due to differences in 332 expression of ∆CTL using immunoblotting ( Figure S10).
In contrast to the effects of overproducing ZapA or FzlC, overproduction of FtsA caused an 334 exacerbation of the effects of ΔCTL production -envelope bulges were larger, less symmetric, 335 and appeared earlier in ΔCTL-producing cells overexpressing ftsA compared to those not 336 overexpressing ftsA ( Figure 5A). We conclude that of all the division proteins tested, FtsE,            Figure 5