Journal: bioRxiv

Article Title: Pal stabilises the bacterial outer membrane during constriction by a mobilisation-and-capture mechanism

doi: 10.1101/790931

Figure Lengend Snippet: a , ITC data of E. coli TolA C-terminal domain (cell concentration, 150 μM) dissolved in 50 mM Hepes, 50 mM NaCl, pH 7.5, into which was injected a 12-residue TolB peptide (TolB - ; syringe concentration, 2 mM) at 25°C, using a iTC200 instrument (Microcal/GE Healthcare). From three independent measurements, average values for ΔH −3.79 (±1.71) kcal.mol -1 , ΔS +5.91 (±6.84) cal.K -1 mol -1 , N = 0.56 (±0.23) and K d = 44 (±23.5) μM were obtained. These data are comparable to those reported previously by for intact TolB binding TolA under the same conditions with the exception that here TolB binding is enthalpically driven. These experiments demonstrate that the entire TolA-binding region is contained within the N-terminus of TolB and hence can be used in the form of an isolated peptide. As detailed below, we were unable to determine the structure for the E. coli complex and so switched to the equivalent complex from P. aeruginosa (binding was too weak to determine the K d by ITC for this complex). b , Binding of TolB - to the C-terminal domain of N-labelled TolA (residues 224-347) 0.6 mM from P. aeruginosa followed by changes in H, N-HSQC spectra in 20 mM phosphate buffer, 100 mM NaCl at pH 7.6. At 95% saturation, >80% of all TolA peaks were perturbed by TolB - binding. Every peak was in slow chemical exchange hence the ratio of the bound and unbound peak volumes corresponds to the ratio of bound/free TolA for each concentration of TolB. The inset shows a portion of a ZZ-exchange HSQC spectrum of a mixture of bound (68%) and free (32%) TolA. The peaks for Ser271 and Gly316 in the bound (b) and free (f) states are labeled; the additional peaks observed arise from exchange between the bound and free states of TolA and confirm the assignment of the bound and free peaks to the same amino acid residue. The titration curves for Ser271, Ser306, Gly316 and Leu347 in TolA were fitted by non-linear regression in SigmaPlot to obtain the K d . Average Kd from these fits was 220 μM ± 30 μM (n=4 residues), which is over five-fold weaker binding than the E. coli complex. c , Fluorescence anisotropy binding curves for P. aeruginosa TolB - binding the C-terminal domain of TolA. Increasing concentrations of TolA were titrated into 1 µM TolB - peptide labelled at its C-terminus (via an additional lysine residue) with fluorescein isothiocyanate (FITC) in 50 mM Tris 100 mM NaCl pH 7.0 at 25°C, λ ex 495 nm and λ em 519 nm. Fluorescence polarization was recorded in a Fluoromax-4 spectrofluorimeter (Horiba JobinYvon). Binding curves were fitted using non-linear regression in SigmaPlot. Closed symbols, wild-type TolA (residues 224-347), K d = 250 µM ±10 µM (n = 3), which is in reasonable agreement with the NMR titration data shown in b . Open symbols, TolA (residues 224-344) lacking its short C-terminal α-helix, K d = 86 µM ±3 µM (n = 3). Deletion of the TolA helix improves TolB binding three-fold.

Article Snippet: Data analysis was performed in SigmaPlot 12.0, using a non-linear regression dynamic curve fit using the quadratic equation: where: f = concentration of TolA-TolB, A T = Total concentration of TolA, B T = Total concentration of TolB, and k = K d . Experiments were carried out using either a CLARIOstar plate reader or a Fluoromax-4 spectrofluorimeter (Horiba JobinYvon).

Techniques: Concentration Assay, Injection, Binding Assay, Isolation, Labeling, Titration, Fluorescence