Figure 1.
Schematic of the L. pneumophila type IV secretion system (T4SS).
The 27 proteins of the T4SS are shown at their predicted or experimentally determined subcellular locations in the outer membrane (OM), periplasm, inner membrane (IM) and cytoplasm. Dot proteins are labeled with the last letter of their name. Icm proteins are designated in the same manner, but are prefaced with an ‘i’.
Figure 2.
Interaction between DotF and substrates is not mediated by the secretion signal sequence.
(A) Visualization of the two-hybrid interaction on MacConkey indicator media. Shown is a portion of a plate containing a red positive control consisting of two leucine zippers fused to the T18 and T25 domains of CyaA (1) and a white negative control consisting of the T18 and T25 empty vector controls (2). The T18:DotF(29–123) fragment was assayed for interaction with T25:RalF and T25:SidG containing their signal sequences (3–4) versus fusions lacking their signal sequences (6–7) and the T25 vector control (5 and 8). (B) Quantification of the interactions in (A) by measurement of β-galactosidase activity. (C,D) The secretion signal sequence is required for translocation. CyaA:RalF (black bars), CyaA:RalF(ΔSS) (white bars), CyaA:SidG (black bars), CyaA:SidG(ΔSS) (white bars) were assayed for secretion in a wild type strain (WT) or a dotA mutant strain that is defective for secretion (T4SS−). Secretion was monitored by an ELISA assay that detects calmodulin-induced cAMP production. Assays were performed in triplicate and error bars represent the standard deviation from the mean.
Figure 3.
DotF interacts with substrates via its periplasmic and transmembrane domains.
(A) Schematic of the DotF fragments used to determine the DotF-substrate interaction domain. The outer membrane (OM), periplasm (P), inner membrane (IM) and cytoplasm (C) are shown. DotF was divided into three domains: cytoplasmic (blue ovals), transmembrane (black rectangle), and periplasmic (red rectangles). Full-length DotF (1) and the DotF fragments (2–8) are shown and their amino acid range is provided beneath the schematic. (B) The DotF fragments (2–8) were tested for interaction with RalF (black bars), SidG (gray bars) or the vector (white bars) by measurement of β-galactosidase activity. Assays were performed in triplicate and error bars represent the standard deviation from the mean.
Figure 4.
Strains lacking DotF retain the ability to partially grow intracellularly and secrete Dot/Icm substrates.
(A) ΔdotF exhibits a partial growth phenotype in A. castellanii. A wild type strain (open squares), a secretion incompetent strain (open triangles), a strain lacking dotF (filled diamonds) and a ΔdotF complemented strain (filled circles) were used to infect A. castellanii. Intracellular growth was monitored by CFU determination at the time of infection and 20, 32 and 42 hours post infection. Infections were performed in triplicate and fold growth was determined by dividing the CFUs at a given time point by the CFUs at time zero. Error bars represent the standard deviation from the mean. (B) A ΔdotF mutant strain is able to secrete Dot/Icm substrates. Full-length RalF and SidG fusions to CyaA were assayed for secretion in a wild type strain (WT, black bars) or a strain lacking dotF (ΔdotF, gray bars). Data was generated in the same secretion experiment show in Figure 2.