Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 290 (32), 19597-613

Structural Hot Spots Determine Functional Diversity of the Candida Glabrata Epithelial Adhesin Family

Affiliations

Structural Hot Spots Determine Functional Diversity of the Candida Glabrata Epithelial Adhesin Family

Rike Diderrich et al. J Biol Chem.

Abstract

For host colonization, the human fungal pathogen Candida glabrata is known to utilize a large family of highly related surface-exposed cell wall proteins, the lectin-like epithelial adhesins (Epas). To reveal the structure-function relationships within the entire Epa family, we have performed a large scale functional analysis of the adhesion (A) domains of 17 Epa paralogs in combination with three-dimensional structural studies of selected members with cognate ligands. Our study shows that most EpaA domains exert lectin-like functions and together recognize a wide variety of glycans with terminal galactosides for conferring epithelial cell adhesion. We further identify several conserved and variable structural features within the diverse Epa ligand binding pockets, which affect affinity and specificity. These features rationalize why mere phylogenetic relationships within the Epa family are weak indicators for functional classification and explain how Epa-like adhesins have evolved in C. glabrata and related fungal species.

Keywords: Candida glabrata; PA14 domain; cell adhesion; cell surface; epithelial adhesin; host-pathogen interaction; lectin; structure-function; yeast.

Figures

FIGURE 1.
FIGURE 1.
Glycan binding profiles. Binding profiles of 17 EpaA domains from C. glabrata strain CBS138 (ATCC 2001) were obtained by using the CFG mammalian glycan array Version 5.1 and fluorescently labeled EpaA proteins purified from E. coli at a concentration of 200 μg/ml. Fluorescence units measured for the different EpaA domains and the 610 different glycans present on the array are shown as black lines. Glycans with unambiguous terminal disaccharide types, which were best bound by the different EpaA domains, are indicated with numbers according to their position on the CFG array. Profiles are depicted according to their classification (roman numbers) obtained by the hierarchical clustering analysis presented in Fig. 2.
FIGURE 2.
FIGURE 2.
Glycan profile-based relationship of the Epa family. A, hierarchical clustering analysis. The 17 Epa paralogs from C. glabrata strain CBS138 were clustered according to the relative in vitro binding strength of their A domains purified from E. coli toward a total of 610 different glycans present on the CFG array Version 5.1 using Cluster 3.0 and Java TreeView software. Relationships among the 17 Epa adhesins as given by their binding profiles are shown by the tree on the left; glycans are clustered as shown on the top according to the similarity of their recognition by different Epa paralogs. Relative binding strength is shown in color with values ranging from 0 (yellow) to 100 (dark blue). Classification is indicated by roman numbers and dotted lines. B, comparison of relative binding strength of best bound glycan ligands. The left part shows the terminal residues of all glycan structures with unambiguous terminal disaccharide types best bound by the individual EpaA domains. Numbers label the different polyglycan types according to their position on the CFG array. The right side shows the relative binding strength of these glycans to all EpaA domains. CFG array numbers are indicated. Relative binding is shown in color as in A.
FIGURE 3.
FIGURE 3.
Ligand binding profiles of EpaA domains. Pie charts show the glycan types, for which the CFG array binding signals obtained by the different EpaA domains exceeded either 50 or 20%, respectively, of the signal of the best binder and for which terminal disaccharide types are unambiguous. The different glycan types are color-coded according the terminal (primary) carbohydrate moiety and the linkage type to the following (second) carbohydrate unit as indicated on the right side. The areas shown correspond to the logarithm of the fluorescence signal over noise for the best binder as exemplified by the two circles at the lower right. Functional classes are indicated by roman numbers.
FIGURE 4.
FIGURE 4.
In vivo binding of EpaA domains. A, expression of different EpaA domains using a heterologous S. cerevisiae expression system. S. cerevisiae strain BY4741 carrying either of the plasmids B2445 (control), BHUM2157 (no A), BHUM1983 (Epa1A), BHUM1985 (Epa2A), BHUM2150 (Epa3A), BHUM2018 (Epa6A), BHUM1988 (Epa7A), BHUM2051 (Epa8A), BHUM2020 (Epa9A), BHUM2152 (Epa11A), BHUM2153 (Epa12A), BHUM1987 (Epa13A), BHUM2154 (Epa15A), BHUM2155 (Epa19A), BHUM2156 (Epa20A), BHUM1993 (Epa21A) or BHUM1988 (Epa23A) was grown to logarithmic phase before EpaA domains were visualized at the cell surface by immunofluorescence microscopy using anti-HA primary and Cy3-conjugated secondary antibodies. Specific signals were quantified, and the values obtained are shown in percentage relative to the strain expressing the carrier domain alone (no A) set to 100. Scale bar corresponds to 10 μm. B, epithelial cell adhesion conferred by 15 EpaA domains. Adhesion was determined using S. cerevisiae strains shown in A or a C. glabrata strain (triple-auxotrophic derivative of CBS138) after 2 h of incubation with a monolayer of Caco-2 cells. Adhesion of S. cerevisiae strains expressing no adhesin (control) or only the carrier domain (no A) is shown in black. For S. cerevisiae strains, EpaA-specific adhesion is indicated by the white part of bars that results from subtracting the value obtained by the strain expressing the carrier domain alone (no A) from the total adhesion. Error bars indicate standard deviation obtained by 10 independent experiments. The functional classification is shown at the bottom. DIC, differential interference contrast.
FIGURE 5.
FIGURE 5.
In vitro binding of selected EpaA domains. A, dissociation constants (KD values) for binding of Epa1A, Epa6A, and Epa7A to five different galactosides (α1–3-galactobiose = Galα1–3Gal; β1–3-galactobiose = Galβ1–3Gal; T-antigen = Galβ1–3GalNAc; lacto-N-biose = Galβ1–3GlcNAc; and N-acetyl-d-lactosamine = Galβ1–4GlcNAc) were determined by fluorescence titration analysis and are shown in regular letters, including values for standard deviations. Binding values obtained for Epa1A, Epa6A, and Epa7A and the five galactosides on glycan arrays (corresponding array numbers are indicated in parentheses) are shown below KD values in italic letters. B, discrimination ratios with respect to different linkage types (β1–4/β1–3 and β/α) and with regard to the nature of the sugar moiety attached to galactose (Gal/GalNAc and GalNAc/GlcNAc) were calculated from the corresponding association constants (1/KD) obtained by fluorescence titration (ft) analysis or from the glycan array fluorescence signals (array) shown in A.
FIGURE 6.
FIGURE 6.
Overall structural features of Epa6A and Epa1A. A, overall fold of Epa6A (blue) shows a β-sandwich comparable with that of Epa1A (gray). The complex structures of Epa6A and Epa1A show a bound T-antigen ligand (Galβ1–3GalNAc, yellow) complexed by a calcium ion (orange) via a DcisD motif that is highly conserved in fungal adhesins. B, selectivity and affinity in ligand binding is achieved by two calcium binding loops (CBL1 and CBL2; brown) in combination with three flexible loops, L1 to L3 (green). L1 and L2 are connected by a disulfide bridge via Cys-78 and Cys-119, which shields the binding pocket from surrounding solvent. L3 contains a tryptophan residue (green) that is highly conserved in Epa adhesins and is essential for ligand binding.
FIGURE 7.
FIGURE 7.
Ligand discrimination by Epa6A and Epa1A binding pockets. A, binding pocket of Epa6A (blue) in complex with the T-antigen (yellow) and residues conferring binding affinity and specificity. A calcium ion (orange) is complexed via a DcisD motif and Asn-225. Interaction between Epa6A and the galactose moiety is accomplished by residues in CBL2 and Trp-198 of loop L3. B, binding mode of T-antigen to the pocket of Epa1A (gray) is similar to that of T-antigen to Epa6A, but the interaction pattern is different. The SigmaA-weighted 2FobsFcalc electron density highlights the orientation of the carbohydrate (contoured at 2.0 σ). C–F, structures of the Epa6A binding pocket in complex with four different galactosides. Although the orientation of the terminal galactose moiety is fixed in all four complexes, the binding pose of the second hexose is more variable and depends on its type and the linkage to galactose.
FIGURE 8.
FIGURE 8.
Role of CBL2 in conferring ligand discrimination by Epa1A and Epa6A. A, structural basis for α-/β-discrimination by Epa1A and Epa6A. Shown is a superposition of the structure of Epa1A (gray) with the structure of the Epa6A·α-galactobiose complex (blue). Higher selectivity of Epa6A toward α-linked galactosides is achieved by the sterically preferred Asp residue at CBL2 position II (Asp-227) in comparison with the sterically more demanding Glu residue at the corresponding position of Epa1A (Glu-227). In addition, a hydrogen bond is observed between α-galactobiose and the CBL2 position III (Asn-228) of Epa6A, which is absent in the Epa1A that carries a Tyr at this position (Tyr-228). B, ligand discrimination by Epa6A. Comparison of Galβ1–3GlcNAc (dark blue) and Galβ1–4GlcNAc (light blue) bound to Epa6A reveals divergent ligand-binding patterns. Whereas the backbone of residue Cys-119 of Epa6A interacts with the 6-OH group of the secondary sugar in the case of Galβ1–3GlcNAc, the same residue of Epa6A contacts the N-acetyl group of the GlcNAc moiety of Galβ1–4GlcNAc. C, ligand discrimination by Epa6A. Comparison of Galβ1–3GalNAc (yellow) with Galβ1–4Glc (light blue) bound to Epa6A reveals highly deviating binding patterns. The secondary carbohydrate is rotated by 90° and different amino acid residues of Epa6A are used for interaction.
FIGURE 9.
FIGURE 9.
Phylogeny and structural conservation of C. glabrata EpaA domains. A, conservation and variability of individual residues of Epa family A domains. Shown is a structure-based sequence alignment of the 17 Epa paralogs from C. glabrata CBS138 as analyzed in this study. The alignment was generated using a local copy of the T-Coffee software implemented with 3DCoffee (43, 52) in combination with the three-dimensional structure of Epa1A (14) and was further processed using the ConSurf server (41, 53). The degree of conservation/variability of individual residues is color-coded according to the ConSurf-server, and yellow letters show minor reliability. Positions of the loops L1, L2, and L3, which form the outer binding pocket, and the calcium-binding loops CBL1 and CBL2, which constitute the inner binding pocket, are indicated above the sequences. Indicated below the sequences are positions I–IV of CBL2, the positions of the DcisD (DD) motif, and the Asn residue of CBL2 that form the DD-N structural motif, as well as positions of the Trp residue of loop L3 and the Arg residue of CBL2 (position I), which constitute the W-R corner of the binding pocket. Arrows indicate three further residues (70, 106, 122) discussed in the text. B, phylogenetic tree of EpaA domains. The tree was created with the MEGA6 software using 500 bootstrap replications (44) and is based on the structure-guided multiple sequence alignment described in A. Bootstrapping values above 70% are indicated. Functional classification obtained by glycan array analysis is indicated by white (class I), black (class II), or gray (class III) circles. Sequence motifs of CBL2 positions II–IV are shown in turquoise. A bar refers to phylogenetic distances. C, conservation of surface properties of EpaA domains. A structural model of Epa1A is shown on the right, which depicts conserved and variable surface residues. The degree of conservation/variability is color-coded and was obtained by using the ConSurf server (41, 53) and the multiple sequence alignment shown in A. The ligand binding pocket is presented on the left and shows highly conserved residues, including the DcisD motif of CBL1 and an asparagine of CBL2, which both confer coordination of the Ca2+ ion, as well as a tryptophan residue of loop L3 and an arginine residue at position I of CBL2, which form a specific corner of the inner binding pocket. In contrast, residues at positions II–IV of CBL2 are highly variable.
FIGURE 10.
FIGURE 10.
Phylogenetic analysis of Epa family members and related PA14 domain-containing adhesins. A structure-guided alignment of the Epa family and Epa-related orthologs from C. glabrata, C. bracarensis, C. nivariensis, N. delphensis, N. bacillisporus, and C. castellii was carried out together with other PA14 domain-harboring adhesins (Pwp proteins from C. glabrata and Flo adhesins from S. cerevisiae) using appropriate sequences (19, 40) and the three-dimensional structure of Epa1A (14) as described for Fig. 9A. Molecular evolutionary genetic analysis was then performed and visualized using the MEGA software (44). PA14-like A domains in the tree are colored with respect to the following structural motifs: no DD-N (gray), only DD-N (black), DD-N and W-R (purple), or DD-N and W-I (pink). For A domains containing W-R or W-I motifs, the residues at CBL2 positions II–IV are shown in turquoise. Details on sequence motifs are described in Fig. 9A and in the text.

Similar articles

See all similar articles

Cited by 13 articles

See all "Cited by" articles

Publication types

MeSH terms

LinkOut - more resources

Feedback