Uncovering the Roles and Evolved Sequence Grammar of Hypervariable Intrinsically Disordered Proteins in Bacterial Cell Division
Date of Award
Doctor of Philosophy (PhD)
Across all domains of life, a defining hallmark of the onset of cell division is the formation of a cytokinetic ring at the center of the cell. Cell division is a tightly controlled process that involves various regulatory factors that modulate the assembly of the cytokinetic ring. In rod-shaped bacteria, the ring is termed the Z-ring after the protein FtsZ, which is foundational to ring formation and is the bacterial homolog of tubulin. Like tubulin, FtsZ is an assembling GTPase, where GTP binding promotes the cooperative assembly into FtsZ polymers that laterally associate to form bundles. While the GTPase domain drives FtsZ polymerization, the formation of these higher-order structures requires domains outside the folded core. FtsZ has a bristled architecture, where a disordered tail, called the C-terminal tail (CTT), flanks the folded domain. The essentiality of the CTT was established through deletion experiments; however, the exact role that the CTT plays within the context of FtsZ function remained unclear. Here, we establish that the CTT, containing an intra-and intermolecular interaction motif (CTP) and a disordered linker (CTL), has a sticker-and-spacer architecture, where the CTL modulates the interactions of the CTP. We find that the modules of the CTT not only influence FtsZ assembly but also impact the catalytic efficiency of the GTPase domain. These findings add to recent findings that implicate disordered regions tethered to enzymes in auto-regulatory activities. The findings summarized above were obtained by focusing our investigations on the CTT of the FtsZ protein from B. subtilis (Bs-FtsZ). Is the stickers-and-spacers model applicable to understanding the functions of CTTs from other bacterial FtsZs? We analyzed the sequences of 1208 orthologous FtsZs, and the results show that while the CTPs and the core domains are reasonably well conserved, the CTLs are hypervariable across orthologs. The results of the sequence analysis have several implications: It might reflect a form of convergent evolution whereby different CTL sequences are interoperable with one other because different sequences serve the functionality of being spacers. Alternatively, the variation could be an example of divergent evolution, whereby changes to the CTLs engender different functionalities in different bacteria. Answering these questions will require methods to identify common sequence patterns across orthologous CTLs, and this cannot be achieved using traditional multiple sequence alignment approaches. Accordingly, we introduce computational methods that enable the quantitative analysis of conserved / distinct sequence-ensemble relationships across a family of IDRs. Additionally, we introduce a new method to uncover cryptic sequence patterns that define disordered regions as random versus non-random. These methods are shown to be applicable for high-throughput analysis of CTLs derived from different FtsZs. They are also effective in uncovering sequence patterns that are cryptic but conserved in intrinsically disordered regions (IDRs) from other bacterial proteins. Given the role of sequence-ensemble relationships and non-random motifs in IDP/IDR function, we hypothesized that these features might influence function and, therefore, might be encoded for within the amino acid sequence of the FtsZ CTL. This implies that designed CTL sequence variants that result in significant changes to these sequence features and to the ability of the CTL to function as a spacer could perturb function. To test this hypothesis, we developed scrambled sequence variants of the B. subtilis FtsZ CTL using the patterning of oppositely charged residues as a design parameter. Leveraging new methodologies, we found that the designed variants caused changes to the sequence-ensemble relationships, the non-random sequence patterning, and / or the spacer properties. Each variant was tested for complementary functions to wild type in vitro and in vivo. Indeed, deviations from wild type features had phenotypic impacts and / or influenced FtsZ assembly and activity, showing that the CTL is not a random disordered sequence but instead has specifically encoded sequence features that dictate function. As the global need to combat antibiotic-resistant infections continues to mount, studies that further understand the functions that IDRs contribute to essential bacterial processes such as cell division can be leveraged to create next-generation antibiotics.
Rohit V. Pappu
Petra A. Levin, Alex S. Holehouse, Timothy M. Lohman, Andrea Soranno,