Interaction between the type 4 pili machinery and a diguanylate cyclase fine-tune c-di-GMP levels during early biofilm formation

Abstract

To initiate biofilm formation, it is critical for bacteria to sense a surface and respond precisely to activate downstream components of the biofilm program. Type 4 pili (T4P) and increasing levels of c-di-GMP have been shown to be important for surface sensing and biofilm formation, respectively; however, mechanisms important in modulating the levels of this dinucleotide molecule to define a precise output response are unknown. Here, using macroscopic bulk assays and single-cell tracking analyses of Pseudomonas aeruginosa, we uncover a role of the T4P alignment complex protein, PilO, in modulating the activity of the diguanylate cyclase (DGC) SadC. Two-hybrid and bimolecular fluorescence complementation assays, combined with genetic studies, are consistent with a model whereby PilO interacts with SadC and that the PilO-SadC interaction inhibits SadC's activity, resulting in decreased biofilm formation and increased motility. Using single-cell tracking, we monitor both the mean c-di-GMP and the variance of this dinucleotide in individual cells. Mutations that increase PilO-SadC interaction modestly, but significantly, decrease both the average and variance in c-di-GMP levels on a cell-by-cell basis, while mutants that disrupt PilO-SadC interaction increase the mean and variance of c-di-GMP levels. This work is consistent with a model wherein P. aeruginosa uses a component of the T4P scaffold to fine-tune the levels of this dinucleotide signal during surface commitment. Finally, given our previous findings linking SadC to the flagellar machinery, we propose that this DGC acts as a bridge to integrate T4P and flagellar-derived input signals during initial surface engagement.

Keywords: Pseudomonas aeruginosa; alignment complex; bacterial biofilms; c-di-GMP; surface sensing.

Fig. 1.

Diagram of the T4P machinery and the DGC SadC. (A) Shown is a diagram of the T4P machinery illustrating interaction between SadC and PilO, a component of the T4P alignment complex. PG, peptidoglycan. The secretin PilQ and the platform protein PilC are shown in light blue while the pilus fiber, which consists of the major pilin protein PilA (pink) and the minor pilins PilVWX (green, blue and magenta), are also shown. The T4P protein PilY1 consisting of a von Willebrand A (vWA) domain and a C-terminal domain are shown at the tip of the pilus fiber. The extension and retraction ATPases, PilB (magenta), and PilT (green) are also shown. The PilMNOP proteins, which comprise the alignment complex and span the cytoplasm to the OM, are shown in yellow. PilP is present in the periplasm, while PilN and PilO, which are structurally similar, span the periplasm and extend across the IM to the cytoplasm. PilM is localized to the cytoplasm and interacts with PilN. (B) PilO interacts with SadC in the BACTH assay. Images of spots of cotransformations with the indicated proteins fused to the C terminus of the T25 or T18 domains of adenylate cyclase following incubation at 30 °C for 40 h on X-gal–containing agar supplemented with the appropriate antibiotics. Empty vectors (EV) are the negative controls in this and subsequent experiments. (C) β-Galactosidase activity in Miller units for interactions shown in B. (D) Images from the BACTH analysis for SadC cotransformed with PilO, PilN, or PilN-O_TM chimera. PilN-O_TM is a chimeric protein of PilN with its TMD replaced with that of PilO. Images of spots of cotransformations with the indicated proteins fused to the C terminus of the T25 or T18 domains of adenylate cyclase following incubation at 30 °C for 40 h on X-gal–containing agar supplemented with the appropriate antibiotics, then incubated at 4 °C for an additional 3 d to allow for further color development. Note the difference in incubation times in this panel compared to B. (E) Quantification of β-galactosidase activity of cotransformation from D shown in Miller units. For C and E, β-galactosidase activity was quantified from cells scraped from transformation plates supplemented with antibiotics; the data are from four biological replicates. Error bars show SEM and statistical significance was determined using one-way ANOVA and Dunnett’s multiple comparison post hoc test. ***P ≤ 0.0001, ****P ≤ 0.0001. (F) Representative DIC and YFP images for PilO–SadC interaction shown by BiFC analysis. The N terminus of PilO and SadC proteins were fused to the C-terminal (YC) and N-terminal (YN) portions of the YFP, respectively. Note the fluorescence background in the vector. (G) DIC (Upper) and fluorescent (Lower) images from BiFC assay shown for PilO with either empty vector, WT SadC, and SadC-T83A protein variant. The vector is included as the negative control. (H) Quantification of mean fluorescence intensity per cell. Dashed lines on violin plots represent the median and solid lines represent the first and third quartiles. Data points are the mean fluorescence intensity per cell from at least six fields. Data are from three independent experiments performed on different days. P value from a Mann–Whitney U test. *P ≤ 0.05, ****P ≤ 0.0001.

Fig. 2.

Mutations in SadC’s TMD modulate interaction with PilO and impact c-di-GMP–associated behaviors in P. aeruginosa. (A) Images from the BACTH analysis for cotransformations with plasmids expressing PilO and either WT SadC, SadC-T83A, or SadC-L172Q proteins. EV, empty vector. (B) Quantification of β-galactosidase activity in Miller units for interactions in A. Details of experiments and analysis are provided in the legend of Fig. 1. (C) Predicted structure of four of the six N-terminal TMD (amino acids 1 to 187) of SadC. The structure was generated using the prediction software Phyre (21). TMD3 (red), TMD4 (blue), TMD5 (green), and TMD6 (magenta) are shown. Residues T83 and L172 located on TMD3 and TMD6, respectively, are highlighted in yellow. (D) Quantification of global c-di-GMP levels for WT and sadC variants. (E) Representative swarm images. (F) Quantification of pixel intensity (PI) of swarm area for images shown in E. Error bars in B, D, and F are SEM and statistical significance was determined by one-way ANOVA and a Dunnett’s post hoc test, *P ≤ 0.01, **P ≤ 0.001, ***P ≤ 0.0001, and ****P ≤ 0.00001; ns, not significant. (G) Representative blot for normalized SadC-3xFLAG protein levels. The band at ∼30 kDa is a nonspecific, cross-reacting band with the anti-FLAG antibody and serves as an additional loading control. (H) Quantification of normalized SadC-FLAG protein levels of WT and mutants relative to the cross-reacting band. Data are from three biological replicates. Dots with the same color represent the same biological replicate; different colors indicate different biological replicates. Error bars are SEM and statistical significance was determined by one-way ANOVA and a Dunnett’s post hoc test. *P ≤ 0.05; ns, not significant.

Fig. 3.

A conserved Sm-xxx-Sm motif in the TMD of PilO is important for interaction with SadC. (A) β-Galactosidase activity from the BACTH assay for SadC cotransformed with PilO or the PilO-VxxxL mutant protein. EV, empty vector. Dots shown on graph for β-galactosidase assay represent data points from five biological replicates. Error bars are SEM, and statistical significance was determined by a one-way ANOVA and a Dunnett’s post hoc test. *P ≤ 0.05, **P ≤ 0.01. (B) Surface representation of the structures for SadC, the PilO-TMD, and MotC determined using Phyre. The helices of the TMD of SadC colored as in Fig. 2 and the SadC-L94P mutation that disrupts interaction with MotC is shown in green, while SadC mutations (T83A and L172Q) that modulate interaction with PilO are shown in yellow. The PilO-TMD with the alanine residues of the Sm-xxx-Sm motif are shown in cyan. Rotation of SadC (180°) is shown to better view TMD3 with L172 and L94 residues. The location of SadC-L94P and SadC-T83A on the same face of the α-helix suggests that SadC does not simultaneously interact with both PilO and MotC.

Fig. 4.

Single-cell tracking reveals the correlation between PilO–SadC interaction strength and intracellular c-di-GMP. Scatter plots of the mean and variance of c-di-GMP of WT, sadC, and pilO variants during early biofilm formation. GFP intensity was determined on a cell-by-cell basis for strains carrying the P_cdrA-gfp construct, a reporter of c-di-GMP levels. Each data point represents one tracked cell through an entire division cycle. The mean and variance characterize the average and spread in c-di-GMP level during the division cycle, respectively. Ellipses are generated from fitting a multivariate Gaussian distribution to the data points, and each ellipse encloses roughly 1 SD of this distribution from the center of mass (centroid) of the points. The size of the ellipse roughly correlates to the range of c-di-GMP levels that the strain can exhibit. The high interaction strength mutants PilO-VxxxL and SadC-T83A (A and B) have the lowest range of the mean and variance of c-di-GMP levels in comparison to WT (C) and the other mutants, while the low interaction strength mutant, SadC-L172Q, has the higher average mean and a wider range of c-di-GMP levels (E). WT sits in the middle of these two scenarios, where its interaction strength, mean, and c-di-GMP variance levels fall between the extremes of these mutants. (F) The null mutant, ΔsadC, shows a pattern similar to the noninteracting, SadC-L172Q mutant variant. A summary plot of all strains using the fitted ellipses in shown in D. The number of data points per strain and the number of cells used to generate the GFP intensities for the mean and variance c-di-GMP calculations for each time point and strain ellipse area overlap between strains, calculated as intersection over union of the ellipse areas, is are summarized in SI Appendix, Table S1. The individual axes of the data are shown as violin plots in SI Appendix, Fig. S9, with associated P values of multiple comparisons tests summarized in SI Appendix, Tables S2 and S3.

Fig. 5.

Proposed model. Proposed model for the role of the PilO–SadC interaction during transition from planktonic to surface-associated or a biofilm state. In a planktonic state, PilO–SadC interaction inhibits SadC’s activity, which results in decreased biofilm formation and increased motility. During surface association, PilO–SadC interaction is disrupted relieving repression of SadC activity; in turn, SadC is stimulated through interaction with MotC. The SadC–MotC interaction results is in increased c-di-GMP levels, which promotes biofilm formation and inhibits motility.