Figure 1. Motility assay of the His-FliI(Δ1-7) mutant on soft tryptone agar plates. A fliH-fliI null strain, MMHI001 (ΔfliH-fliI), and a fliI null strain, MKM30 (ΔfliI) were transformed with pTrc99A (V), pMM1702 (His-FliI) and pBGtI(Δ1-7) (His-FliI(Δ1-7)). Plates were incubated for 7 h at 30 °C.

Figure 2. Hydrodynamic properties of His-FliI(Δ1-7). (a) Analysis of purified His-FliI(Δ1-7) by gel-filtration chromatography with a Superdex 200 HR10/30 column followed by in-line multi-angle light-scattering to determine the molecular mass. The elution profile of His-FliI(Δ1-7) is shown as a continuous line. The clustered points represent light-scattering data converted to molecular mass. (b) Protein concentration profile of FliI(Δ1-7) at sedimentation equilibrium by analytical ultracentrifugation. Open circles are data points and the continuous line is the curve fit by the monomer model.

Figure 3. Far-UV CD spectra of His-FliI monomer (black line) and His-FliI(Δ1-7) (grey line). Measurements were carried out at room temperature in 10 mM phosphate buffer (pH 8.0), in a quartz cell with a path-length of 1 mm. The difference spectrum (broken line) was obtained by subtracting the spectrum of His-FliI from that of His-FliI(Δ1-7).

Figure 4. Dependence of the ATPase activity of His-FliI (filled circle and black line) and His-FliI(Δ1-7) (filled square and grey line) on protein concentration. The specific ATPase activity was measured at different concentrations of protein in the presence of 4 mM ATP by using the malachite green assay. The activity is expressed as nmol of Pi released per min per μg of FliI(Δ1-7). These data are the average of three independent experiments. Vertical bars on plotted data are standard errors.

Figure 5. Subcellular location of His-FliI and His-FliI(Δ1-7). (a) His-FliI; (b) His-FliI(Δ1-7). The soluble and membrane fractions of MKM30 (ΔfliI) producing His-FliI or His-FliI(Δ1-7) were prepared by sonication and ultracentrifugation. Then, the membrane fraction was treated with either 1.2% Sarkosyl or 1 M NaCl, separated by ultracentrifugation into the pellet (P) and supernatant (S) fractions, and immunoblotted with polyclonal anti-FliI antibody. His-FliI and His-FliI(Δ1-7) are indicated by arrows. The position of the 43 kDa molecular mass marker is shown at the left.

Figure 6. Relative ATPase activity of His-FliI(Δ1-7) in the absence and in the presence of E. coli acidic phospholipids (PL) at three different concentrations of protein. Each ATPase activity was normalized to that of His-FliI(Δ1-7) at a protein concentration of 57 μg ml⁻¹ in the absence of acidic phospholipids. These data are the average of three independent experiments.

Figure 7. Electron micrographs showing the oligomerization ability of FliI and FliI(Δ1-7). (a) His-FliI, (b) His-FliI(Δ1-7). Purified FliI, preincubated with 5 mM MgCl₂, 5 mM ADP, 5 mM AlCl₃, 5 mM NaF and 100 mg ml⁻¹ of acidic phospholipids, was stained with 0.5% (w/v) uranyl acetate and visualized by electron microscopy. Electron micrographs were recorded at a magnification of 50, 000×. The scale bar represents 200 Å.

Figure 8. A model for FliI oligomerization and energy coupling of ATP hydrolysis in the flagellar protein export process. The six integral membrane protein components, including FlhA and FlhB, are postulated to be located within the putative central pore of the MS ring. FlhA and FlhB both have a large cytoplasmic domain that projects into the cytoplasmic space within the C ring. FlhA and FlhB are depicted as the platform for mounting the ATPase FliI and may be together with its regulator FliH. FliI forms a complex with the FliH dimer in the cytoplasm and keeps its ATPase activity low. The chaperone–substrate complex and the (FliH)₂FliI complex bind together and bind to FliN in the C ring through FliH and wait until the FlhA-FlhB platform becomes available for docking, where FliI oligomerization is induced to fully activate the ATPase activity to drive the translocation of export substrates into the central channel of the basal body.