Purification, characterization, and sequence analysis of a potential virulence factor from Porphyromonas gingivalis, peptidylarginine deiminase

Abstract

The initiation and progression of adult-onset periodontitis has been associated with infection of the gingival sulcus by Porphyromonas gingivalis. This organism utilizes a multitude of virulence factors to evade host defenses as it establishes itself as one of the predominant pathogens in periodontal pockets. A feature common to many other oral pathogens is the production of ammonia due to its protective effect during acidic cleansing cycles in the mouth. Additionally, ammonia production by P. gingivalis has been proposed as a virulence factor due to its negative effects on neutrophil function. In this study, we describe the first purification of a peptidylarginine deiminase (PAD) from a prokaryote. PAD exhibits biochemical characteristics and properties that suggest that it may be a virulence agent. PAD deiminates the guanidino group of carboxyl-terminal arginine residues on a variety of peptides, including the vasoregulatory peptide-hormone bradykinin, to yield ammonia and a citrulline residue. The soluble protein has an apparent mass of 46 kDa, while the DNA sequence predicts a full-length protein of 61.7 kDa. PAD is optimally active at 55 degrees C, stable at low pH, and shows the greatest activity above pH 9.0. Interestingly, in the presence of stabilizing factors, PAD is resistant to limited proteolysis and retains significant activity after short-term boiling. We propose that PAD, acting in concert with arginine-specific proteinases from P. gingivalis, promotes the growth of the pathogen in the periodontal pocket, initially by enhancing its survivability and then by assisting the organism in its circumvention of host humoral defenses.

FIG. 1

Distribution of PAD in various strains of P. gingivalis. The production of citrulline from BAEE was determined in cellular (open bar), supernatant (solid bar), and vesicle (hatched bar) fractions from HG66, ATCC 53978 (W50), and ATCC 33277 strains of P. gingivalis.

FIG. 2

Anion-exchange chromatography of 50-kDa peak from P. gingivalis. The active fractions from the 50-kDa range were pooled, dialyzed, and concentrated prior to being loaded onto a DE-52 anion-exchange column (1.5 by 21 cm; 20 ml/h) previously equilibrated with 20 mM Bis-Tris-HCl–1 μM FMN–1 mM CaCl₂, pH 6.8. A linear gradient of 0 to 500 mM NaCl in buffer B was applied over 3 column volumes. The active enzyme is found primarily in the first peak (about 150 mM NaCl). ——, A₂₈₀; ●, deiminase activity; □, amidolytic activity.

FIG. 3

Hydroxyapatite adsorption chromatography of P. gingivalis deiminase activity. The citrulline-producing fractions from anion-exchange chromatography were pooled and applied to a ceramic hydroxyapatite column in sodium phosphate buffer (10 mM, pH 6.5). A linear multistep gradient (arrows [left to right], 0 to 12, 12 to 15, 15 to 37, 37 to 100%) with 2 column volumes of eluent (0.5 M NaH₂PO₄, pH 6.5) for each step was applied to the column, and fractions were assayed. ——, A₂₈₀; ●, deiminase activity; □, amidolytic activity.

FIG. 4

SDS-PAGE of P. gingivalis PAD at various stages of purification. Samples were loaded for equal activity. Lane A, culture supernatant; lane B, acetone precipitate; lane C, Sephadex G-150 pooled activity; lane D, DE52 pooled activity; lane E, hydroxyapatite pooled activity (1 μg); lane F, purified PAD 5 times overloaded (5 μg). Molecular mass markers (phosphorylase b, 97 kDa; bovine serum albumin, 68 kDa; ovalbumin, 43 kDa; carbonic anhydrase, 30 kDa; and α-lactalbumin, 14 kDa) are on the left.

FIG. 5

Measurement of PAD activity at various pH values. PAD activity was measured at various pH values by the universal phosphate buffer system. PAD was preincubated with the 50 mM phosphate-citrate buffer with 1 mM DTT and 1 μM FMN prior to the addition of substrate.

FIG. 6

PAD sequence. The underlined amino acid sequence was determined by sequencing the amino terminus of the isolated protein.

FIG. 7

Effect of electron donors and acceptors on PAD activity. The enzyme was dialyzed overnight with low levels of cofactor (50 mM Tris-HCl, 1 μM FMN, pH 8.0) and without any cofactor present (50 mM Tris-HCl, pH 8.0) as noted. Pretreatment of the dialyzed enzyme with higher levels of cofactors FMN (25 μM), FAD (25 μM), hemin (500 μM), or dilute whole blood (1:5,000; 50 mM Tris-HCl, 118 mM NaCl, 1 μM FMN, pH 8.0) was performed prior to the enzymatic assay (37°C; 10 min). Samples were then assayed for citrulline production in triplicate as described in Materials and Methods. The error bars indicate standard deviations.

FIG. 8

Enhanced thermal stability of PAD with and without flavin nucleotides. Purified PAD was boiled with (⧫) and without (▴) flavin nucleotides (FMN-FAD, 25 μM) or incubated at 37°C (●) in substrateless assay buffer (50 mM Tris-HCl, 25 μM FMN, pH 8.0). Aliquots were removed at various time points and assayed for the production of citrulline as described in Materials and Methods.