Getting Closer to Decrypting the Phase Transitions of Bacterial Biomolecules

Abstract

Liquid-liquid phase separation (LLPS) of biomolecules has emerged as a new paradigm in cell biology, and the process is one proposed mechanism for the formation of membraneless organelles (MLOs). Bacterial cells have only recently drawn strong interest in terms of studies on both liquid-to-liquid and liquid-to-solid phase transitions. It seems that these processes drive the formation of prokaryotic cellular condensates that resemble eukaryotic MLOs. In this review, we present an overview of the key microbial biomolecules that undergo LLPS, as well as the formation and organization of biomacromolecular condensates within the intracellular space. We also discuss the current challenges in investigating bacterial biomacromolecular condensates. Additionally, we highlight a summary of recent knowledge about the participation of bacterial biomolecules in a phase transition and provide some new in silico analyses that can be helpful for further investigations.

Keywords: bacterial cells; biomacromolecular condensates; liquid–liquid phase separation; membraneless organelles; multivalent interactions; phase transitions.

Figure 1

Proposed phase transitions in bacterial cells. Bacterial biomacromolecular assemblies can be present in optimal growth conditions (e.g., bacterial microcompartments). The condensates can fuse upon contact, and the components can be exchanged between condensates. Sudden changes in the environment (e.g., stress) can lead to phase transitions to form more solid-like forms. The formation of such condensates may provide an efficient means for protecting genetic material. For more details, see the main text.

Figure 2

The structural organization of bacterial proteins with propensity for LLPS. (A) Architecture of the ABC transporter Rv1747. Rv1747 is composed of a cytoplasmic regulatory module containing the forkhead-associated (FHA-1 and FHA-2) domains joined by an ID linker, cytoplasmic nucleotide-binding domain (NBD), and a helical transmembrane domain (TMD) through which substrate is transported. (B) Domain organization of FtsZ protein. FtsZ contains 10 unstructured residues at the N-terminus, a conserved globular core domain containing the GTPase active site, a flexible variable linker of approximately 50 residues, a conserved C-terminal tail (CTT), and a C-terminal variable region of 4 residues (CTV). (C) Schematic representation of PopZ structure. PopZ comprises two conserved and mostly α-helical domains: N- and C-terminal (shown in red) and proline-glutamate rich domain (PED) located between them. (D) Domain organization of SSB protein. SSB protein molecule contains an N-terminal well-folded domain that is responsible for DNA binding (DBD), LCR, and C-terminal protein–protein interaction region (C-term). Disordered, flexible linkers are shown as blue lines. These regions are often supposed to drive LLPS.

Figure 3

In silico analysis of LLPS-related proteins. Summary graph of the pLLPS values obtained with FuzDrop for SSB (P0AGE0), RNase E (A0A0H3CAR6), Rv1747^1–310 (O65934), CsoS2 (O85041), CcmM35 (Q03513-2), PopZ (Q9A8N4), DivJ (Q03228), FtsZ (P0A9A6), SlmA (P0A9A6), McdB (Q8GJM6), and NusA (P0AFF6). The gray bars show pLLPS values of proteins for which the in silico data were consistent with the experimental data, the blue bars show pLLPS values for proteins for which the in silico data were found to be inconsistent with the experimental data. The red line is the threshold of 0.6 pLLPS values.

Figure 4

In silico analysis of C. crescentus RNase E and degradosome proteins interacting with it. Summary graph of pLLPS values obtained with FuzDrop for RNase E (A0A0H3CAR6), MetK (A0A0H3C5U2), DbpA (A0A0H3C896), RhlE (A0A0H3C5T6), RhlB (A0A0H3C8I9), PNPase (B8GWz0), RNase D (A0A0H3CAA2), OdhA (A0A0H3C4B5), S1 (A0A0H3CCW5), Aconitase (A0A0H3CE29), NudC (A0A0H3C5J2), and acetoacetyl-CoA reductase (A0A0H3C629). Gray bars show pLLPS values for potential scaffold protein sequences (RNase E in yellow box—result confirmed experimentally [111]), green—for clients, blue—for non-LLPS-related proteins. The red line is the threshold of 0.6 pLLPS value.