The emergence of phase separation as an organizing principle in bacteria

Abstract

Recent investigations in bacteria suggest that membraneless organelles play a crucial role in the subcellular organization of bacterial cells. However, the biochemical functions and assembly mechanisms of these compartments have not yet been completely characterized. This article assesses the current methodologies used in the study of membraneless organelles in bacteria, highlights the limitations in determining the phase of complexes in cells that are typically an order of magnitude smaller than a eukaryotic cell, and identifies gaps in our current knowledge about the functional role of membraneless organelles in bacteria. Liquid-liquid phase separation (LLPS) is one proposed mechanism for membraneless organelle assembly. Overall, we outline the framework to evaluate LLPS in vivo in bacteria, we describe the bacterial systems with proposed LLPS activity, and we comment on the general role LLPS plays in bacteria and how it may regulate cellular function. Lastly, we provide an outlook for super-resolution microscopy and single-molecule tracking as tools to assess condensates in bacteria.

Figure 1

Proposed criteria to determine whether a cluster assembles by LLPS. (A) Condensates are spherical because of surface tension. (B) Droplets can fuse upon contact. (C) Mobility of condensate components and their exchange across the boundary are shown. Arrow length denotes the rate of diffusion of the molecule. (D) Restricted diffusion across the membraneless boundary is shown. (E) Condensate formation and size scales with component concentration; cytoplasm component concentration is buffered. To see this figure in color, go online.

Figure 2

Proposed bacterial LLPS systems and evidence for the criteria in Fig. 1. References are given in the figure. ND, not determined. ^†Overexpressed construct in a heterologous system. Not shown: The α-carboxysome CcmM analog CsoS2 can phase separate in vitro. DEAD-box RNA helicases were also briefly discussed but are not shown in the figure. To see this figure in color, go online.

Figure 3

ParB forms clusters at parS sites, which are segregated by ParA. (A) Super-resolution images of ParB clusters are given. Scale bars, 500 nm. Reprinted with permission from (30). (B) In cells expressing ParB-mEos2 and ParA fused to the SsrA degron tag, time-lapse images of a ParB cluster separating without induction of ParA degradation (top) and fusing upon induction of ParA-Ssra degradation (bottom) are given. Reprinted with permission from (26). (C) Representative ParB single-molecule tracks are shown. Reprinted with permission from (26). (D) A hybrid model for ParB condensation at parS is shown that includes aspects from the ParB nucleation and caging model (28,30), ParB CTPase activity (75), and potential interplay between ParB CTPase and ParA ATPase activities. To see this figure in color, go online.

Figure 4

LLPS may play a role in carboxysome biogenesis and organization. (A) The model for CcmM-RuBisCO nucleation is shown. Once assembled, carboxysomes may be fluidized by McdB. (B) Progression of carboxysome formation after induction of the ccm operon in ΔccmK2-ccmO S. elongatus expressing RbcL-GFP (green) is shown. White arrow denotes budding event from a procarboxysome. Scale bars, 1 μm. Reprinted with permission from (94). (C) RbcL-GFP (green) forms bar carboxysomes upon CcmL deletion (+IPTG). Scale bars, 1 μm. Reprinted with permission from (94). (D) Carboxysome positioning is governed by the McdAB system via a Brownian ratchet. Inactive carboxysomes, possibly with no McdB, have been shown to become polarly localized. (E) Microscopy images of S. elongatus McdB droplets under varying pH are given. Scale bars, 10 μm. Reprinted with permission from (24). (F) McdB droplet fusion events (yellow arrows) are shown. Scale bars, 5 μm. Reprinted with permission from (24). To see this figure in color, go online.

Figure 5

Single-molecule tracking characterizations of molecular condensates in vivo. (A) Super-resolution reconstructions of ParB clusters in E. coli (left) are class-averaged to estimate size, shape, and concentration (right). Reprinted with permission from (26). (B) Three-dimensional single-molecule tracks (time-coded connected dots) overlaid with the super-resolution images of PopZ (orange) in C. crescentus are shown. Scale bars, 200 nm. Reprinted with permission from (32). (C) Super-resolution localizations of SpmX (violet) and PopZ (green) are overlaid with an electron tomogram of the C. crescentus cell at the stalked pole. Reprinted with permission from (42). (D) Super-resolution images of RNaseE-eYFP (green) and ribosomal component L1-PAmCherry (violet) in E. coli are given. Arrows denote regions of ribosome exclusion with BR-bodies. Scale bars, 500 nm. Reprinted with permission from (18). To see this figure in color, go online.