The Pseudomonas aeruginosa Tse4 toxin assembles ion-selective and voltage-sensitive ion channels to couple membrane depolarisation with K+ efflux

Abstract

Pseudomonas aeruginosa employs the Type VI secretion system (T6SS) to outcompete other bacteria in its environment. Among the effectors secreted by the T6SS of P. aeruginosa PAO1, Tse4 is known for its potent antibacterial activity. This study elucidates the molecular function of Tse4, which promotes cell depolarization in competing bacteria. Our results show that Tse4 spontaneously incorporates into lipid monolayers and forms multi-ionic channels in planar bilayers, with either ohmic conduction or diode-like rectifying currents and a preference for cations over anions. These observations allow us to propose a model of action whereby Tse4 channels couple cell depolarization with K+ efflux. These insights into Tse4's pore-forming activity enhance our understanding of bacterial competition and exemplify a finely tuned antibacterial strategy, coupling its ability to cause membrane depolarization with potassium efflux that synergises with other T6SS effectors. These results highlight the sophistication of Pseudomonas aeruginosa's competitive arsenal.

Fig 1. Heterologous expression of a Tse5^ΔCT-Tse4 chimera.

A. The top panel shows a schematic representation of Tse5 and Tse5^ΔCT-Tse4 constructs. Tse4 replaces the Tse5-CT fragment. The Tse5^ΔCT-Tse4 construct contains a double point mutation (K47G-P48A) that inhibits cleavage between residues G47 and A48. The bottom panel shows a predicted 3D structure of Tse5 and Tse5^ΔCT-Tse4 constructs. The construct is designed so that Tse4 is encapsulated inside the Tse5 shell/cocoon structure. B. SEC-SAXS analysis of Tse5 and Tse5^ΔCT-Tse4 proteins. The left panel shows the normalised SEC signal profile for both proteins, indicating they have very similar elution properties. The right panel plot represents the SAXS data for the two proteins. C. SDS-PAGE of purified wild-type Tse5 (left) showing that WT Tse5 auto-cleaves, resulting in three fragments, Tse5-Shell, Tse5-CT and Tse5-NT. On the right is the SDS-PAGE of Tse5^ΔCT-Tse4 and Tse4 following its separation from the Tse5^ΔCT fragment.

Fig 2. Tse4 inserts spontaneously into model membranes to assemble narrow membrane pores.

A. Representative Langmuir-Blodgett balance data showing the lateral pressure increase on lipid monolayers assembled from E. coli polar lipid extract (Avanti Polar lipids) after the addition of Tse4 at time 0. Initial lateral pressures (Π₀) in mN m⁻¹ for representative experiments are indicated next to each curve. B. Plot of lateral pressure increases (ΔΠ) as a function of initial lateral pressure (Π₀) for Tse4, Tse5^ΔCT-Tse4, and Tse5^ΔCT (n = 10). A maximal insertion pressure (Π_c) of 31.08 (Tse4), 26.53 (Tse5^ΔCT-Tse4), and 27.17 mN m⁻¹ (Tse5^ΔCT) has been determined by extrapolating the fitted curve to ΔΠ = 0. The dotted line indicates the threshold value of lateral pressure consistent with unstressed biological membranes. The equations obtained from the linear regression analysis are y = −0.4148x + 12.89 (R² = 0.94), y = −0.5832x + 15.47 (R² = 0.98),y = −0.6141x + 16.68 (R² = 0.96) for Tse4, Tse5^ΔCT-Tse4, and Tse5^ΔCT, respectively. C. Representative current traces showing small current jumps recorded at a constant voltage of 100 mV in symmetrical 150 mM KCl. D. Frequency distribution analysis of 644 independent current jumps obtained at 100 mV was conducted using a bin width of 0.2 pS. The resulting dataset, comprising 170 data points, was fitted to a Gaussian function, indicating a peak in the distribution at 19.6 ± 2.7 pS.

Fig 3. Tse4 forms a variety of cation-selective pores with mild cation specificity and diverse voltage sensitivity.

A. Histograms illustrating the reversal potentials (RP) for Tse4-induced channels using E. coli lipid mixture in KCl (orange, n = 43), NaCl (blue, n = 11) or LiCl (green, n = 15). The frequency distribution analysis was performed with a bin width of 2 mV. Rectifying pores in KCl display the most negative RP. B. Box and whiskers plot displaying the P₊/P_- ratios in KCl, NaCl and LiCl derived from the experiments presented in panel A. Here and elsewhere, the boundary of the box closest to zero indicates the 25^th percentile, a line within the box marks the median, and the boundary of the box farthest from zero indicates the 75^th percentile. Whiskers (error bars) above and below the box indicate the 90^th and 10^th percentiles. Rectifying pores in KCl exhibit the maximum P₊/P_- values. C. RP measured for Tse4 in E. coli lipids and compared with Tse5 and OmpF in KCl (circles), NaCl (triangles) or LiCl (squares) plotted versus the corresponding diffusion potential (V_diff) for each salt. For Tse5, results are shown for charged (solid symbols) and neutral (open symbols) lipids. RP for Tse4 in KCl only includes ohmic channels, as there are no rectifying currents in other salts. For Tse4, data are means of n = 37 (KCl), 11 (NaCl) and 15 (LiCl); for Tse5 in charged lipid n = 15 (KCl), 13 (NaCl) and 10 (LiCl); for Tse5 in neutral lipid n = 34 (KCl), 18 (NaCl) and 13 (LiCl); and for OmpF n = 8 (KCl), 6 (NaCl) and 7 (LiCl). n indicates independent channels. D. Box and whiskers plot displaying the P₊/P_- ratios measured in Tse4 in KCl using different lipid compositions, as indicated. Significance was tested using one-way analysis of variance (ANOVA) followed by a Holm-Sidak test for pair-wise comparison. The differences between groups were all non-significant (p > 0.05) except for the difference between E. coli and PE/E. coli (p = 0.03), indicated with an asterisk. E. Example current-voltage curves depicting the electrical behaviour of ohmic (filled circles) and rectifying (open circles) pores induced by Tse4 in KCl and E. coli lipid mixture. The solid lines represent a fit with an empirical equation used to categorize I-V curves as ohmic or rectifying (see Methods for details). F. Scatter plot of the RP as a function of the conductance measured in KCl (filled circles correspond to ohmic currents while open circles indicate rectifying currents), NaCl, or LiCl, derived from the experiments presented in panels A and B. In all panels, experiments were performed in a 250/50 mM salt gradient.

Fig 4. A conceptual model for Tse4-induced changes in cell membrane potential.

Tse4 forms ion-selective pores in bacterial membranes, exhibiting mild specificity for Na⁺ and Li⁺ over K ⁺ . These pores facilitate Na⁺ influx (A) and K⁺ efflux (B), driven by concentration gradients and membrane potential. Tse4-induced K⁺ efflux is enhanced under depolarised conditions due to rectifying pore conduction, maintaining cytoplasmic electroneutrality without affecting pH. This model integrates observations of K⁺ efflux, membrane depolarisation, and unchanged pH, highlighting Tse4’s role in disrupting membrane potential and sensitising bacteria to other T6SS effectors.