Threonine-rich repeats increase fibronectin binding in the Candida albicans adhesin Als5p

Abstract

Commensal and pathogenic states of Candida albicans depend on cell surface-expressed adhesins, including those of the Als family. Mature Als proteins consist of a 300-residue N-terminal region predicted to have an immunoglobulin (Ig)-like fold, a 104-residue conserved Thr-rich region (T), a central domain of a variable number of tandem repeats (TR) of a 36-residue Thr-rich sequence, and a heavily glycosylated C-terminal Ser/Thr-rich stalk region, also of variable length (N. K. Gaur and S. A. Klotz, Infect. Immun. 65: 5289-5294, 1997). Domain deletions in ALS5 were expressed in Saccharomyces cerevisiae to excrete soluble protein and for surface display. Far UV circular dichroism indicated that soluble Ig-T showed a single negative peak at 212 nm, consistent with previous data indicating that this region has high beta-sheet content with very little alpha-helix. A truncation of Als5p with six tandem repeats (Ig-T-TR(6)) gave spectra with additional negative ellipticity at 200 nm and, at 227 to 240 nm, spectra characteristic of a structure with a similar fraction of beta-sheet but with additional structural elements as well. Soluble Als5p Ig-T and Ig-T-TR(6) fragments bound to fibronectin in vitro, but the inclusion of the TR region substantially increased affinity. Cellular adhesion assays with S. cerevisiae showed that the Ig-T domain mediated adherence to fibronectin and that TR repeats greatly increased cell-to-cell aggregation. Thus, the TR region of Als5p modulated the structure of the Ig-T region, augmented cell adhesion activity through increased binding to mammalian ligands, and simultaneously promoted fungal cell-cell interactions.

FIG. 1.

Expressed versions of Als5p. The regions of Als5p are marked: S, signal sequence (residues 1 to 17); Ig, tandem Ig-like domains (residues 18 to 327); T, conserved Thr-rich region (residues 328 to 431); TR, tandem Thr-rich repeats (residues 432 to 664); Stalk, Ser-Thr-rich glycosylated stalk; GPI, glycosylphosphatidylinositol addition signal; V, V5 epitope tag; H, His₆ tag. For the protein names, the N-terminal-expressed residues are numbered, and “W” denotes that residues 650 to 1419 (stalk and GPI) are also expressed. The results of immunofluorescence assays (IFA) for each cell-bound construct next to each surface-bound construct are shown. The primary antibodies were anti-Als1p for all constructs except Als5p^{1-17,432-662-W}, which was labeled with anti-V5.

FIG. 2.

Expression and secretion of soluble forms of Als5p. Equal aliquots of His-TRAP elution fractions were spotted on nitrocellulose and probed with peroxidase-conjugated anti-His_6. White vertical arrows denote antigen-negative fractions. Horizontal arrowheads denote positive controls (black arrowhead, purified Als5p^1-664 or His₆-labeled α-agglutinin [36]) and negative controls (white arrowhead, BSA).

FIG. 3.

Purified soluble proteins. Purified Als5p^1-431 (lane 1) and Als5p^1-664 (lane 4) were electrophoresed on a 4 to 20% gel and stained with Coomassie blue. The apparent molecular masses for standard proteins (lane 3) are on the left. Lane 2 is empty.

FIG. 4.

Far UV circular dichroism spectra for Als5p^1-431 and Als5p^1-664. Multiple spectra were averaged, smoothed, and baseline corrected for 0.4-mg/ml solutions of each protein, as described in Materials and Methods. The buffer was 20 mM sodium phosphate, pH 7.0.

FIG. 5.

Dot blots with horseradish peroxidase-conjugated concanavalin A. Aliquots of His-TRAP elution fractions were spotted on nitrocellulose and probed with peroxidase-conjugated concanavalin A. White vertical arrows denote antibody-negative fractions. Horizontal arrowheads denote positive controls (black arrowhead, purified His₆-labeled α-agglutinin [36]) and negative controls (white arrowhead, BSA).

FIG. 6.

Binding of Als5p^1-431 and Als5p^1-664 to FN-coated microtiter wells. Wells were coated at the indicated concentration of FN (determined as protein added), and Als5p binding was assayed as described in Materials and Methods. The results for three independent experiments with different preparations of Als5p are shown. Error bars are ranges for duplicate samples in each experiment. (Inset) Expanded scale showing binding for Als5p^1-431; the filled diamond indicates the binding in wells coated with BSA alone.

FIG. 7.

Adhesion to FN-coated beads and yeast cell aggregation. S. cerevisiae cells expressing Als5p (A and B), Als5p^1-431-W (C to F), Als5p^{1-17,432-664-W} (G), or no Als (H) were incubated with FN-coated magnetic beads (orange-brown), and the aggregates were separated and photographed (9). Panels B and F show immunofluorescence with anti-Als1p to demonstrate surface expression of Als5p in the aggregates. For scale comparison, the magnetic beads have a diameter of 4.5 μm.

FIG. 8.

Adhesion to FN-coated beads and aggregation of S. cerevisiae cells expressing Als5p with different numbers of tandem repeats. Binding in the bead aggregation assay was quantified for Als5p containing six repeats (Als5p), four repeats (Als5p^1-575-W), two repeats (Als5p^1-503-W), or no tandem repeats (Als5p^1-431-W). Error bars show standard deviations for quadruplicate samples. Surface expression levels were similar for all constructs (Fig. 1).