Force Sensitivity in Saccharomyces cerevisiae Flocculins

Abstract

Many fungal adhesins have short, β-aggregation-prone sequences that play important functional roles, and in the Candida albicans adhesin Als5p, these sequences cluster the adhesins after exposure to shear force. Here, we report that Saccharomyces cerevisiae flocculins Flo11p and Flo1p have similar β-aggregation-prone sequences and are similarly stimulated by shear force, despite being nonhomologous. Shear from vortex mixing induced the formation of small flocs in cells expressing either adhesin. After the addition of Ca(2+), yeast cells from vortex-sheared populations showed greatly enhanced flocculation and displayed more pronounced thioflavin-bright surface nanodomains. At high concentrations, amyloidophilic dyes inhibited Flo1p- and Flo11p-mediated agar invasion and the shear-induced increase in flocculation. Consistent with these results, atomic force microscopy of Flo11p showed successive force-distance peaks characteristic of sequentially unfolding tandem repeat domains, like Flo1p and Als5p. Flo11p-expressing cells bound together through homophilic interactions with adhesion forces of up to 700 pN and rupture lengths of up to 600 nm. These results are consistent with the potentiation of yeast flocculation by shear-induced formation of high-avidity domains of clustered adhesins at the cell surface, similar to the activation of Candida albicans adhesin Als5p. Thus, yeast adhesins from three independent gene families use similar force-dependent interactions to drive cell adhesion. IMPORTANCE The Saccharomyces cerevisiae flocculins mediate the formation of cellular aggregates and biofilm-like mats, useful in clearing yeast from fermentations. An important property of fungal adhesion proteins, including flocculins, is the ability to form catch bonds, i.e., bonds that strengthen under tension. This strengthening is based, at least in part, on increased avidity of binding due to clustering of adhesins in cell surface nanodomains. This clustering depends on amyloid-like β-aggregation of short amino acid sequences in the adhesins. In Candida albicans adhesin Als5, shear stress from vortex mixing can unfold part of the protein to expose aggregation-prone sequences, and then adhesins aggregate into nanodomains. We therefore tested whether shear stress from mixing can increase flocculation activity by potentiating similar protein remodeling and aggregation in the flocculins. The results demonstrate the applicability of the Als adhesin model and provide a rational framework for the enhancement or inhibition of flocculation in industrial applications.

Keywords: biofilms; functional amyloid; fungal adhesins; glycoproteins.

FIG 1

Primary-structure analyses of Flo1p and Flo11p. Hydrophobic-cluster analyses highlight domain structure and patterns of repeats (15). Secretion signal sequences are boxed in blue, C-terminal GPI addition signals are boxed in green, and the blue line denotes the position of GPI signal cleavage and anchorage to cell wall glucan (13). Central repeat regions are in unshaded boxes. Cys residues are black arrowheads, with disulfides marked where they have been mapped in the N-terminal domains (12). Potential N-glycosylation sites are marked with purple hexagons, and sequences with a β-aggregation potential of >30% in TANGO are marked with blue triangles (21).

FIG 2

Effects of vortex mixing on the extent and rate of flocculation of S. cerevisiae var. diastaticus cells expressing Flo11p. (A) Cells were left untouched or vortex mixed and then visualized in the absence of Ca²⁺. Scale bars = 20 µm. (B) Flocculation assays after the addition of 330 µM Ca²⁺. Cells assayed: Δflo11 mutant not mixed (black circles, top) or vortex mixed for 5 min (inverted triangles, top); FLO11 cells not mixed (dark gray boxes) or vortex mixed in the absence of inhibitors (light-shaded diamonds) or in the presence of 200 µM ThS (dark triangles, top) or 500 µM CR (gray circles). Each error bar shows the standard error of the mean of three or more values. (C) Inhibition by ThS (200 µM) and CR (500 µM) (n = 3). Flocculation rates were calculated between 40 and 80 s. *, P < 0.05; **, P < 0.001 (Student’s t test). (D) Effect of the duration of vortex mixing on the flocculation rate. Each error bar shows the standard error of the mean of three values.

FIG 3

Effects of vortex mixing on the flocculation of S. cerevisiae cells expressing Flo1p. (A) Cells were left untouched or vortex mixed and then visualized in the absence of Ca²⁺. Scale bars = 20 µm. (B) Flocculation assays after the addition of 1.0 mM Ca²⁺. Cells were not mixed (●) or vortex mixed for 5 min (▽). (C) Flocculation rates in the absence or presence of the inhibitors ThS (200 µM) and CR (0.5 mM) (n = 3). *, P < 0.01 (Student’s t test). (D) Effect of duration of vortex mixing on activation of flocculation.

FIG 4

Agar invasion assays. Cells were grown on YPD plates for 2 weeks with or without amyloidophilic dyes and then washed and imaged from the bottom of the petri plate. Concentrations: CR, 30 µM; ThS, 200 µM. Scale bars represent 5 mm.

FIG 5

ThT fluorescence on the surface of Flo11p- and Flo1p-expressing S. cerevisiae. Shown are fluorescence confocal micrographs of cells under quiescent and vortex-mixed conditions taken right after vortex mixing as described in Materials and Methods and then stained with 500 nM ThT. Scale bars represent 5 µm. The boxed cells are further enlarged in the insets.

FIG 6

Single-molecule imaging of V5-Flo11p proteins on yeast cells. (A) AFM deflection image recorded in buffer, showing an S. cerevisiae cell expressing V5-tagged Flo11p proteins trapped in a porous membrane for AFM analysis. (B) Single-molecule detection of V5-tagged Flo11p proteins with anti-V5 antibody-coated AFM tips. (C to F) Adhesion force maps (1 by 1 µm) (C, E) and adhesion force histograms (n = 1,024 curves) (D, F), with representative force curves, obtained in buffer between anti-V5 tips and three different yeast cells expressing V5-tagged Flo11p (C, D) or no flocculin (E, F) (different cultures and different tips). Pixels are shaded according to the rupture force necessary to disrupt binding, with no binding depicted as black and strong interactions shown as white.

FIG 7

Forces in Flo11p-mediated cell-cell adhesion. (A) Principle of single-cell force spectroscopy. (Left) Single yeast cells were attached to a polydopamine-coated tipless cantilever, and curves of force between cell probes and yeast aggregates were acquired. (Right) Cartoon of homotypic interactions of the flocculins. Shown are adhesion force histograms (B, D, F, H, J, L, N, P) and rupture length histograms (C, E, G, I, K, M, O, Q) with representative force curves obtained by recording multiple force-distance curves for Flo11p-expressing cells in acetate buffer containing 200 µM CaCl₂ (B, C) or after the addition of 10 mM EDTA (D, E) and further addition of 400 µM CaCl₂ (F, G). (H, I) Force data obtained with Flo11p-expressing cells and yeast cells expressing no flocculin (EV). (L, M) Effects of methyl alpha-d-mannopyranoside at 40 mg/ml added to the buffer in the experiments shown in panels J and K. For those shown in panels P and Q, 200 µM ThS was added to the cells shown in panels N and O. Red, blue, and green represent results from three different cell pairs from independent cultures (n = >400 curves for each cell). All curves were obtained at 20°C by a constant approach at a retraction speed of 1 µm ⋅ s⁻¹ with a 1-s contact time.