Dissecting the contribution of Staphylococcus aureus α-phenol-soluble modulins to biofilm amyloid structure

Abstract

The opportunistic pathogen Staphylococcus aureus is recognized as one of the most frequent causes of biofilm-associated infections. The recently discovered phenol soluble modulins (PSMs) are small α-helical amphipathic peptides that act as the main molecular effectors of staphylococcal biofilm maturation, promoting the formation of an extracellular fibril structure with amyloid-like properties. Here, we combine computational, biophysical and in cell analysis to address the specific contribution of individual PSMs to biofilm structure. We demonstrate that despite their highly similar sequence and structure, contrary to what it was previously thought, not all PSMs participate in amyloid fibril formation. A balance of hydrophobic/hydrophilic forces and helical propensity seems to define the aggregation propensity of PSMs and control their assembly and function. This knowledge would allow to target specifically the amyloid properties of these peptides. In this way, we show that Epigallocatechin-3-gallate (EGCG), the principal polyphenol in green tea, prevents the assembly of amyloidogenic PSMs and disentangles their preformed amyloid fibrils.

Figure 1. Phenol soluble modulins alignment and in vitro aggregation assays.

(a) Sequences of PSMs are aligned and coloured using ClustalX. (b) Aggregation kinetics were monitored by following the change in relative Th-T fluorescence during 25 days. Error bars indicate ± SE (n = 3). (c) Congo red (CR) binding was registered at 6 days and 25 days. Free Congo red is represented by a discontinuous black line.

Figure 2. Conformational conversion of PSMs followed by far-UV CD.

(a–e) Evolution of the far-UV CD spectra of PSM peptides during the first 18 days of incubation; (a–e) panels correspond to α-PSM1, α-PSM2, α-PSM3, α-PSM4 and δ-toxin, respectively. (f) Far-UV spectra of PSM peptides after 25 days.

Figure 3. Morphological and structural properties of PSM aggregates.

(a) TEM micrographs of the aggregates formed by PSMs at different time points. The scale bar represents 0.5 μm. (b) α-PSM1 and α-PSM4 fibrils (25 days) are shown at different magnification (the scale bar represents 100 nm) and characterized by ATR-FTIR. Absorbance spectra of the amide I region (thick line) showing the component bands (thin lines). The sum of individual spectral components after Fourier self-deconvolution closely matches the experimental data.

Figure 4. Seeding and cross seeding test.

(a) PSM seeding and cross-seeding Th-T emission spectra recorded after 6 days of incubation are reported. In all the cases, Th-T free is represented by a black dotted line. (b) TEM micrographs of α-PSM1 and α-PSM4 in absence (α-PSM control) and in presence of α-PSM preformed fibrils are shown. The scale bar represents 200 nm.

Figure 5. Disaggregation kinetics of α-PSM1 and α-PSM4 fibrils by EGCg.

(a) Chemical structure of EGCg. (b) Disaggregation of α-PSM1 (grey) and α-PSM4 (black) mature fibrils were followed by Th-T emission fluorescence at different EGCg concentration; error bars indicate ± SE (n = 2). (c) TEM micrographs of α-PSM1 and α-PSM4 preformed fibrils are displayed in the upper and bottom panels respectively, in presence of 0, 1, 5, 10 and 20 μM of EGCg (from left to right). The scale bar represents 0.5 μm.

Figure 6. EGCg inhibition effect upon α-PSM1 and α-PSM4 aggregation kinetics.

Th-T kinetics of α-PSM1 (a) and α-PSM4 (b) peptide solutions in presence of EGCg were monitored at different time points. Error bars indicate ± SE (n = 2). Inhibitory effect of EGCg upon α-PSM1 (c) and α-PSM4 (d) fibril formation at final point was also tested by TEM at 0, 20, 100 and 200 μM EGCg (from left to right). The scale bar represents 200 nm.

Figure 7. Characterization of α-PSM-GFP fusion protein expression in E. coli.

(a) GFP fluorescence emission spectra of entire cells expressing α-PSM1 (red), α-PSM2 (yellow), α-PSM3 (blue) and α-PSM4 (green) fusions were collected. (b) α-PSM-GFP containing cells were observed using fluorescence microscopy (right panels) and phase-contrast microscopy (left panels); the scale bar represents 20 μm. (c) Total (T), insoluble (I) and soluble (S) cell fractions of α-PSM fusions were detected by western blotting.

Figure 8. Impact of α-PSMs on cell viability.

(a) α-PSM fusion cells growth in presence (black) and in absence (grey) of IPTG was analyzed measuring OD600 after 8 h of expression. Error bars indicate ± SE (n = 3). (b) Cell density was determined by calculating the dry weight of the α-PSM fusion cells. Viability of E. coli expressing α-PSMs was evaluated by (c) IP staining and (d) CFU counts. Error bars indicate ± SE (n = 3). Bars labelled as 1, 2, 3 and 4 correspond to cells expressing α-PSM1, α-PSM2, α-PSM3 α-PSM4 fusions to GFP, respectively.