Assay of the GAG dependency of binding by C. caviae OmcB and C. trachomatis OmcB variants. Schematic representation of the adhesion of various yeast transformants to HEp-2 cells. Yeast cells expressing Aga2, Aga2–OmcB-C.pn., Aga2–OmcB-C.ca. GPIC, Aga2–OmcB-C.tr. L1, Aga2–OmcB-C.tr. E, Aga2–OmcB-C.tr. L1 P₆₆L or Aga2–OmcB-C.tr. E L₆₆P were pre-incubated with 500 μg ml⁻¹ soluble heparin prior to the adhesion assay. (n = 1000 HEp-2 cells, No. experiments = 4, P < 0.0001). w/o, without; C.pn., C. pneumoniae; C.ca., C. caviae; C.tr., C. trachomatis.

A. Photomicrographs of HEp-2 cells infected with C. pneumoniae at moi 1 in the presence of 1.2 μg ml⁻¹ cycloheximide for 48 h. Cells were fixed with methanol to visualize intra- and extrachlamydial antigens by staining with anti-DnaK, anti-OmpA and anti-OmcB antibodies. Fixation with formaldehyde and permeabilization with 0.5% Triton X-100 was used to visualize extrachlamydial antigens (Birkelund et al., 1990). Bar 10 μm.
B. Left panel: Photomicrograph of purified, non-fixed C. pneumoniae EBs stained with CFSE. Right panels: Equal amounts of purified, non-fixed C. pneumoniae EBs were used for microimmunofluorescence (MIF) analysis. Chlamydial antigens were detected using anti-DnaK, anti-OmpA and anti-OmcB antibodies. Bar 1 μm.

A. Affinity-purified, recombinant wild-type OmcB and OmcB_Δ45−78 proteins (black arrows) were resolved by SDS-PAGE and stained with Coomassie. Bands of higher molecular weight represent contaminants, and bands of lower molecular weight are degradation products. Positions of size markers (98, 64 and 50 kDa) are indicated.
B. Equal amounts of beads coated with OmcB or OmcB_Δ45−78 protein (black arrows) were separated by SDS-PAGE and probed with an anti-His antibody.
C. Adhesion of coated latex beads to HEp-2 cells. Latex beads (1 × 10⁶) coated with 100 μg of BSA, OmcB or OmcB_Δ45−78 were incubated with 1 × 10⁵ HEp-2 cells and the numbers of beads associated with HEp-2 cells were determined by microscopy. (n = 1000 HEp-2 cells, No. experiments = 4, P < 0.0001).

A. Top: Schematic representations of the wild-type OmcB protein from C. pneumoniae, and the deletion derivatives used here, with basic amino acids (black bars), signal peptide sequence (black box) and heparin-binding motifs (grey box) indicated. Bottom left: Western blot analysis of protein extracts from strains expressing Aga2–OmcB₁₋₂₈₂, Aga2–OmcB_275−556 and Aga2–OmcB_Δ45−78. After digestion with α-mannosidase, cell extracts were resolved by SDS-PAGE and probed with an anti-His antibody. Bottom right: Adhesion of yeast cells expressing Aga2, Aga2–OmcB₁₋₂₈₂, Aga2–OmcB_275−556 or Aga2–OmcB_Δ45−78 to HEp-2 cells. (n = 1000 HEp-2 cells, No. experiments = 4, P < 0.0001).
B. Schematic representation of the amino acid changes generated in the OmcB variant Aga2–OmcB₁₋₁₀₀. The heparin-binding motifs I and II are indicated. These variants show reduced adherence to HEp-2 cells (bottom diagram) (n = 1000 HEp-2 cells, No. experiments = 4, P < 0.0001).

A. Left: Aga2- or Aga2–OmcB-expressing yeast cells were incubated with 500 μg ml⁻¹ of the soluble GAGs heparin, chondroitin-6-sulphate (C-6-S), dermatan sulphate (DS) or chondroitin-4-sulphate (C-6-S) prior to the adhesion assay and the numbers of yeast cells adhering to human cells were determined by microscopy. Yeast cells treated with PBS served as controls (n = 1000 HEp-2 cells, No. experiments = 4, P < 0.0001). Right: Photomicrographs of GAG-treated Aga2–OmcB-expressing yeast cells adhering to human cells. Bar 10 μm.
B. Effects of treatment with heparinase on the ability of HEp-2 cells to bind Aga2-, Aga2–Inv- and Aga2–OmcB-expressing yeast cells. HEp-2 cells were incubated with PBS or 50 U heparinase I prior to the adhesion experiment. The photomicrographs show adhesion of Aga2–OmcB-expressing yeast cells to HEp-2 cells.
C. Efficiency of association of Aga2- or Aga2–OmcB-expressing yeast cells with wild-type (CHO-K1) or GAG-deficient (pgsA-745 and pgsD-677) cell lines is plotted diagrammatically and illustrated in the photomicrographs. Bar 10 μm (n = 1000 cells, No. experiments = 4, P < 0.01).

A and B. C. pneumoniae infection is inhibited by addition of recombinant OmcB protein or an OmcB-specific antibody. (A) HEp-2 cells (1 × 10⁶) were incubated with PBS, 200 μg of BSA, or 20 μg or 200 μg of recombinant OmcB or OmcB_Δ45−78 protein prior to challenge with purified C. pneumoniae EBs. Cells were fixed at 48 h post infection and the numbers of inclusions determined by microscopy. The number of inclusions found in the PBS control was set to 100%. (n = 20 microscopic fields, No. experiments = 4, P < 0.0001). (B) Purified C. pneumoniae EBs were incubated with PBS, pre-immune serum or anti-OmcB antibody before being exposed to HEp-2 cells. The numbers of inclusions formed was determined by microscopy 48 h post infection. (n = 20 microscopic fields, No. experiments = 4, P < 0.0001).
C. Binding of purified C. pneumoniae EBs to HEp-2 cells was detected by a colorimetric assay using a chlamydia-specific primary antibody and a HRP-coupled secondary antibody. Attachment of C. pneumoniae EBs in the presence of PBS or 200 μg of BSA served as controls, and these data were compared with samples to which heparin (500 μg ml⁻¹) or recombinant OmcB or OmcB_Δ45−78 protein (200 μg) had been added. (n = 2 wells, No. experiments = 6, P < 0.01).
D. Adhesion of viable, CFSE-stained C. pneumoniae EBs to HEp-2 or human umbilical vein endothelial cells (HUVEC) was detected by flow cytometric analysis. The same amounts of protein and heparin as in (C) were used in this experiment.

A. Schematic representation of the yeast display system. The protein of interest is presented on the yeast cell surface as a consequence of its fusion to Aga2p.
B. Western analysis of Aga2 (lane 1), Aga2–Inv (lane 2) and Aga2–OmcB (lane 3). Left: Protein extracts from strains expressing Aga2, Aga2–Inv or Aga2–OmcB were digested with α-mannosidase to remove Aga2 O-glycosylation, resolved by SDS-PAGE and probed with an anti-His antibody. Right: Coomassie-stained SDS-PAGE of the protein extracts shown in the left panel. The positions of the size markers (250, 98, 50 and 22 kDa) are indicated.
C. Detection of Aga2 fusion proteins on the yeast cell surface. Yeast cells expressing Aga2, Aga2–Inv or Aga2–OmcB were fixed and stained with anti-invasin or anti-OmcB antibodies for visualization of Aga2–Inv and Aga2–OmcB proteins respectively. Bar 1 μm.
D–G. Adhesion of yeast cells to HEp-2 cells. (D) Varying numbers of untransformed yeast cells or yeast cells expressing Aga2 or Aga2–Inv from a plasmid were incubated with 1 × 10⁵ HEp-2 cells, and the number of yeast cells associated with HEp-2 cells was determined by microscopy. (n = 1000 HEp-2 cells, No. experiments = 4, P < 0.0001). (E) Schematic representation of the numbers of yeast control cells or yeast cells expressing Aga2, Aga2–Inv and Aga2–OmcB adhering to 1000 HEp-2 cells. (n = 1000 HEp-2 cells, No. experiments = 4, P < 0.0001). (F) Photomicrographs of HEp-2 cells after incubation with yeast cells expressing the indicated proteins. Bar 10 μm. The cells indicated by the white box are shown enlarged in the rightmost panel. Bar 5 μm. (G) Adhesion of Aga2 or Aga2–OmcB-expressing yeast cells to HEp-2 cells. Yeast cells (1 × 10⁶) were treated with PBS or proteinase K (4 mg ml⁻¹) for 1 h at 37°C, washed three times and then added to HEp-2 cells (n = 1000 HEp-2 cells, No. experiments = 4, P < 0.0001).