Fig 1.
Fractionation of bacterial cell lysates.
P. gingivalis TDC60 lysates were fractionated by anion-exchange chromatography after ammonium sulfate precipitation (A). Fractions 30 to 50 (10 μl each) were subjected to SDS-PAGE, followed by CBB staining (B) and western blotting using anti-FimA, anti-Mfa1, and anti-PGN_1808 antisera (C). We demonstrated reproducibility by repeating the fractionation experiment three times (data not shown). Numbers in left indicate molecular weights (kDa).
Fig 2.
Electrophoretic analyses of PGN_1808 polymerization.
Fractionated products of PGN_1808 (1808), FimA and Mfa1 in SDS-containing buffer were heated at 37 (1), 60 (2), 80 (3) and 100 (4)°C for 10 min, and then applied to SDS-PAGE. The SDS-PAGE gel was stained with CBB (A), and western blotting was performed with anti-PGN_1808, anti-FimA, and anti-Mfa1 antisera (B). Fractionated PGN_1808 was applied to blue native-PAGE followed by western blotting (C). Samples containing 5, 2 and 20 μg of proteins were applied to the experiments in panels A, B, and C, respectively. Numbers in left indicate molecular weights (kDa). Note that anti-Mfa1 antiserum weakly reacts to ladder-like band at high molecular weight (Mfa1 polymer) when compared to its monomer, resulting in the differential intensity of bands between CBB and western blot analyses.
Fig 3.
Deduced amino acid sequence of PGN_1808.
The LipoP program predicted that the N-terminal 20 amino acids were a signal peptide (underlined). N-terminal sequencing of the 50-kDa PGN_1808 monomer band in the SDS-PAGE analysis (Fig 2A) showed that the N-terminal amino acid was Gly-56, indicating that the N-terminal 55 amino acids were removed for processing (highlighted in gray).
Fig 4.
Transmission electron micrographs.
Whole cells of the P. gingivalis mutant that was deficient in both FimA and Mfa1 fimbriae (A), whole cells of the PGN_1808-overexpressing mutant (B), and fractionated PGN_1808 (C) were negatively stained, then observed by TEM. Filamentous structures (2‒3 nm × 200‒400 nm) were observed on cell surfaces of the PGN_1808-overexpressing mutant (indicated by arrows) and in the PGN_1808 fraction. Scale bars indicate 200 nm.
Fig 5.
Structure and polymerization models.
PGN_1808 without the signal sequence (the N-terminal 20 amino acids) was submitted to structure homology-modeling (A). The color (orange to blue) indicates the QMEAN score, which indicates the reliability of the model (low to high quality, respectively) [23]. The N-terminal sequence (from the 21st to 55th amino acid), surrounded by the black dotted line, was removed for processing (see text). PGN_1808 is predicted to polymerize by donor-strand complementation and cleft-mediated anchorage (B). White letter N denotes the N-terminal region.