A bacterial sense of touch: T4P retraction motor as a means of surface sensing by Pseudomonas aeruginosa PA14

Abstract

Most microbial cells found in nature exist in matrix-covered, surface-attached communities known as biofilms. This mode of growth is initiated by the ability of the microbe to sense a surface on which to grow. The opportunistic pathogen Pseudomonas aeruginosa (Pa) PA14 utilizes a single polar flagellum and type 4 pili (T4P) to sense surfaces. For Pa, T4P-dependent "twitching" motility is characterized by effectively pulling the cell across a surface through a complex process of cooperative binding, pulling, and unbinding. T4P retraction is powered by hexameric ATPases. Pa cells that have engaged a surface increase production of the second messenger cyclic AMP (cAMP) over multiple generations via the Pil-Chp system. This rise in cAMP allows cells and their progeny to become better adapted for surface attachment and activates virulence pathways through the cAMP-binding transcription factor Vfr. While many studies have focused on mechanisms of T4P twitching and regulation of T4P production and function by the Pil-Chp system, the mechanism by which Pa senses and relays a surface-engagement signal to the cell is still an open question. Here we review the current state of the surface sensing literature for Pa, with a focus on T4P, and propose an integrated model of surface sensing whereby the retraction motor PilT senses and relays the signal to the Pil-Chp system via PilJ to drive cAMP production and adaptation to a surface lifestyle.

Keywords: Pil-Chp; PilT; Pseudomonas aeruginosa; biofilm; motor; surface sensing; type 4 pili.

Fig 1

Diagram of the biofilm cycle of Pseudomonas aeruginosa PA14. Biofilm formation by Pseudomonas aeruginosa PA14 (green) begins with free-swimming planktonic cells (1) making initial surface attachment through its single polar flagellum (2). After initial surface contact, bacteria can either continue to explore the surface using their motility appendages or return to the planktonic population in a process known as reversible attachment (3). As cells continue to explore the surface using T4P, they become better surface adapted in a process that is mediated by cAMP level (blue triangle). Surface contact also stimulates c-di-GMP production (green triangle). Both second messengers oscillate at the single-cell level over multiple generations. How the levels of these two messengers are related to each other remains an area of active research. Once cells are surface adapted, they commit to the biofilm lifestyle and become irreversibly attached (4). As biofilm cells continue to grow on a surface, they produce matrix components necessary for the formation of a mature biofilm (5). When conditions become unfavorable for the biofilm, cells can either passively or actively disperse into the planktonic population (6). cAMP, cyclic AMP; c-di-GMP, cyclic di-guanosine monophosphate.

Fig 2

New proposed model for surface sensing by P. aeruginosa. (A) After surface binding by extended T4P, retraction of pili begins with the retraction motor PilT binding PilC. The accessory retraction motor PilU binds PilT and aids in depolymerization of the pilus fiber via conformational changes in PilC. However, if PilU is absent during retraction, we hypothesize that PilT alone is unable to exert sufficient force to power retraction, and the PilT motor will stall, potentially entering a conformational state due to improper ATP binding and/or hydrolysis and/or ADP release. We believe that PilT in this stalled conformation binds PilJ of the Pil-Chp complex to transduce the surface signal from T4P to Pil-Chp (B). After PilJ activation, the signal is transmitted to the kinase ChpA, which leads to the phosphorylation of several response regulators including PilG. PilG, along with FimV and FimL, then activates the adenylate cyclase CyaB, leading to an increase in cAMP (C). All depicted proteins are products of pil genes unless otherwise stated. Note that not all known interactions among pili proteins can be shown in this figure. Abbreviations: IM, inner membrane; OM, outer membrane; PG, peptidoglycan layer.

Fig 3

Predicting the interface between the C-terminal face of the PilT hexamer and the PilC hexamer. (A) An illustration of the T4P of P. aeruginosa with the motor proteins highlighted. (B) A linear representation of the PilT protein with known motifs labeled and the beginning and ending aa of each motif indicated. This depiction is based on a previous publication (121). (C) A representative image of how two PilT monomers oligomerize. (D) The AlphaFold-Multimer program was used to predict the structure of PilT hexamer (blue) bound to the PilC trimer (green) with the model giving the best score shown. The residues comprising the PilT-PilJ interface are highlighted in green. (E) AlphaFold-Multimer was used to predict the structure of PilT hexamer bound to the hexamer PilU with the model giving the best score shown. The Pa PA14 amino acid sequences for each protein were used to build the model. The pLDDT scores have been mapped onto the structure in Fig. S1. (F) AlphaFold-Multimer was used to predict the structure of the PilT hexamer bound to the hexamer PilU as well as the PilC trimer with the model giving the best score shown. The PA14 amino acid sequences for each protein were used to build the model. The pLDDT scores have been mapped onto the structure in Fig. S1.

Fig 4

Predicting the interface between the N-terminal face of the PilT hexamer and the PilU or PilC hexamers. (A) The C-terminal face of the PilT hexamer with PilC removed. Each PilT monomer is in a different shade of blue, and the conserved AIRNLIRE motif is yellow. A red box encircles the monomers with conservation information mapped on the surface as shown in panel B. (B) A BLAST search was performed on taxid 286 for the genus Pseudomonas using the PA14 PilT sequence as a search query. Homologs were identified, and genomes were pooled by whether a homolog of PilU could also be identified (center) or not (right). An MSA was then generated using Clustal Omega, and the resulting MSA (see Fig. S2) was used with CONSURF to calculate and map conservation scores for each residue on the center monomer of the trimer to visualize the conserved areas of the protein, as previously described (138–142). PDB: 3JVV chain B was used to map the conservation information onto each residue. The legend shows the extent of sequence identity for each residue for the central monomer. The AIRNLIRE motif (A288-E295) in the left and right monomers is highlighted in yellow. A yellow ellipse encompasses the AIRNLIRE motif on each monomer of PilT. (C) N-terminal face of the PilT hexamer with PilU removed. Each PilT monomer is in a different shade of blue, and the potential PilT-PilJ binding interface (H44, K45, H48, and N87) is in green. A red box includes the monomers with conservation information mapped on the surface of the central monomer as shown in panel D. (D) A BLAST search was performed on taxid 286 for the genus Pseudomonas using PA14 pilT as a search query. Homologs were identified, and genomes were pooled by whether a homolog of PilU could also be identified (center) or not (right). An MSA was then generated using Clustal Omega (Fig. S3 and S4) with CONSURF to calculate and map conservation scores for each residue on the center monomer of the trimer to visualize the conserved areas of the protein as previously described (138–142). PDB: 3JVV chain B was then used to map the conservation information onto each residue. The legend shows the extent of sequence identity for each residue. In green are the residues that are predicted to mediate binding with PilJ on the left and right monomers. A green ellipse encompasses the predicted PilT-PilJ interface on each PilT monomer. MSA, multiple sequence alignment.