Potential role for Streptococcus gordonii-derived hydrogen peroxide in heme acquisition by Porphyromonas gingivalis.

Streptococcus gordonii, an accessory pathogen and early colonizer of plaque, co-aggregates with many oral species including Porphyromonas gingivalis. It causes α-hemolysis on blood agar, a process mediated by H2 O2 and thought to involve concomitant oxidation of hemoglobin (Hb). Porphyromonas gingivalis has a growth requirement for heme, which is acquired mainly from Hb. The paradigm for Hb heme acquisition involves the initial oxidation of oxyhemoglobin (oxyHb) to methemoglobin (metHb), followed by heme release and extraction through the actions of K-gingipain protease and/or the HmuY hemophore-like protein. The ability of S. gordonii to mediate Hb oxidation may potentially aid heme capture during co-aggregation with P. gingivalis. Hemoglobin derived from zones of S. gordonii α-hemolysis was found to be metHb. Generation of metHb from oxyHb by S. gordonii cells was inhibited by catalase, and correlated with levels of cellular H2 O2 production. Generation of metHb by S. gordonii occurred through the higher Hb oxidation state of ferrylhemoglobin. Heme complexation by the P. gingivalis HmuY was employed as a measure of the ease of heme capture from metHb. HmuY was able to extract iron(III)protoporphyrin IX from metHb derived from zones of S. gordonii α-hemolysis and from metHb generated by the action of S. gordonii cells on isolated oxyHb. The rate of HmuY-Fe(III)heme complex formation from S. gordonii-mediated metHb was greater than from an equivalent concentration of auto-oxidized metHb. It is concluded that S. gordonii may potentially aid heme acquisition by P. gingivalis by facilitating metHb formation in the presence of oxyHb.

Heme acquisition mechanisms of Porphyromonas gingivalis - strategies used in a polymicrobial community in a heme-limited host environment.

Porphyromonas gingivalis, a main etiologic agent and key pathogen responsible for initiation and progression of chronic periodontitis requires heme as a source of iron and protoporphyrin IX for its survival and the ability to establish an infection. Porphyromonas gingivalis is able to accumulate a defensive cell-surface heme-containing pigment in the form of µ-oxo bisheme. The main sources of heme for P. gingivalis in vivo are hemoproteins present in saliva, gingival crevicular fluid, and erythrocytes. To acquire heme, P. gingivalis uses several mechanisms. Among them, the best characterized are those employing hemagglutinins, hemolysins, and gingipains (Kgp, RgpA, RgpB), TonB-dependent outer-membrane receptors (HmuR, HusB, IhtA), and hemophore-like proteins (HmuY, HusA). Proteins involved in intracellular heme transport, storage, and processing are less well characterized (e.g. PgDps). Importantly, P. gingivalis may also use the heme acquisition systems of other bacteria to fulfill its own heme requirements. Porphyromonas gingivalis displays a novel paradigm for heme acquisition from hemoglobin, whereby the Fe(II)-containing oxyhemoglobin molecule must first be oxidized to methemoglobin to facilitate heme release. This process not only involves P. gingivalis arginine- and lysine-specific gingipains, but other proteases (e.g. interpain A from Prevotella intermedia) or pyocyanin produced by Pseudomonas aeruginosa. Porphyromonas gingivalis is then able to fully proteolyze the more susceptible methemoglobin substrate to release free heme or to wrest heme from it directly through the use of the HmuY hemophore.

Evidence of mutualism between two periodontal pathogens: co-operative haem acquisition by the HmuY haemophore of Porphyromonas gingivalis and the cysteine protease interpain A (InpA) of Prevotella intermedia.

Haem (iron protoporphyrin IX) is both an essential growth factor and a virulence regulator of the periodontal pathogens Porphyromonas gingivalis and Prevotella intermedia, which acquire it through the proteolytic degradation of haemoglobin and other haem-carrying plasma proteins. The haem-binding lipoprotein HmuY haemophore and the gingipain proteases of P. gingivalis form a unique synthrophic system responsible for capture of haem from haemoglobin and methaemalbumin. In this system, methaemoglobin is formed from oxyhaemoglobin by the activities of gingipain proteases and serves as a facile substrate from which HmuY can capture haem. This study examined the possibility of cooperation between HmuY and the cysteine protease interpain A (InpA) of Pr. intermedia in the haem acquisition process. Using UV-visible spectroscopy and polyacrylamide gel electrophoresis, HmuY was demonstrated to be resistant to proteolysis and so able to cooperate with InpA to extract haem from haemoglobin, which was proteolytically converted to methaemoglobin by the protease. Spectroscopic pH titrations showed that both the iron(II) and iron(III) protoporphyrin IX-HmuY complexes were stable over the pH range 4-10, demonstrating that the haemophore could function over a range of pH that may be encountered in the dental plaque biofilm. This is the first demonstration of a bacterial haemophore working in conjunction with a protease from another bacterial species to acquire haem from haemoglobin and may represent mutualism between P. gingivalis and Pr. intermedia co-inhabiting the periodontal pocket.

The HA2 haemagglutinin domain of the lysine-specific gingipain (Kgp) of Porphyromonas gingivalis promotes micro-oxo bishaem formation from monomeric iron(III) protoporphyrin IX.

The lysine- and arginine-specific gingipains (Kgp, and RgpA and RgpB) are the major proteinases produced by the black-pigmented periodontopathogen Porphyromonas gingivalis. They play a role in degrading host proteins, including haemoglobin, from which is formed the mu-oxo bishaem complex of iron(III) protoporphyrin IX, [Fe(III)PPIX]2O, the major haem component of the black pigment. Kgp and RgpA bind haem and haemoglobin via the haemagglutinin-adhesin 2 (HA2) domain, but the role of this domain in the formation of mu-oxo bishaem-containing pigment is not known. UV-visible spectroscopy was used to examine the interaction of iron(III) protoporphyrin IX monomers [Fe(III)PPIX.OH] with recombinant HA2 and purified HRgpA, Kgp and RgpB gingipains. The HA2 domain reacted with Fe(III)PPIX.OH to form mu-oxo bishaem, the presence of which was confirmed by Fourier transform infrared spectroscopy. Both HRgpA and Kgp, but not RgpB, also mediated mu-oxo bishaem formation and aggregation. It is concluded that the Arg- and Lys-gingipains with HA2 haemagglutinin domains may play a crucial role in haem-pigment formation by converting Fe(III)PPIX.OH monomers into [Fe(III)PPIX]2O and promoting their aggregation.

Detection of heme-binding proteins in epidemic strains of Burkholderia cepacia.

A panel of 30 previously characterized strains representing five genomovars from the Burkholderia cepacia complex (E. Mahenthiralingam, T. Coenye, J. W. Chung, D. P. Speert, J. R. W. Govan, P. Taylor, and P. Vandamme, J. Clin. Microbiol. 38:910--913, 2000) were examined for their iron protoporphyrin IX-binding ability. These included B. cepacia genomovars I and III and B. stabilis (formerly B. cepacia genomovar IV), B. multivorans (formerly B. cepacia genomovar II), and B. vietnamiensis (formerly B. cepacia genomovar V). Cells were exposed to micro-oxo bisheme of iron protoporphyrin IX (micro-oxo dimers) and examined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis under nonreducing, nondenaturing conditions for the presence of heme-binding proteins using tetramethylbenzidine-H(2)O(2) staining. Seven of the 30 strains, each belonging to B. cepacia genomovar III and designated epidemic (in possessing the B. cepacia epidemic strain marker), expressed a 96- to 100-kDa heme-binding protein which was located in the outer membrane. The heme-binding protein of B. cepacia genomovar III epidemic strain C5424 bound iron(III) protoporphyrin IX in both the monomeric and micro-oxo bisheme forms. Cells of all strains grown on Columbia agar bound iron protoporphyrin IX in the micro-oxo bisheme (dimeric) form. There were no statistical differences between the five genomovars, or those possessing the heme-binding protein, in their micro-oxo bisheme-binding ability. Possession of the outer membrane heme-binding protein may be a pathogenicity trait in enabling the bacterium to withstand oxidative stresses in inflammatory exudates in the lung and may aid identification of invasive epidemic strains of B. cepacia.

Temperature elevation regulates iron protoporphyrin IX and hemoglobin binding by Porphyromonas gingivalis.

Porphyromonas gingivalis, an obligate anerobe with a growth requirement for iron protoporphyrin IX (FePPIX), is exposed to increased temperatures in the inflamed periodontal pocket. In this study, P. gingivalis was grown in a chemostat at 37 degrees C (control), 39 degrees C, and 41 degrees C, and examined for hemagglutinating (HA) activity, hemoglobin binding and degrading activity, and iron protoporphyrin IX binding. HA activity decreased in cells as the growth temperature increased. Binding of mu-oxo bishaem (dimeric haem), and Fe(II)- and Fe(III)-monomeric forms was increased in 39 degrees C-grown cells but decreased in 41 degrees C-grown cells compared with controls. Cellular hemoglobin binding and degradation decreased with increased growth temperature. The decrease in cellular hemagglutination and hemoglobin degradation occurring with increased growth temperature would limit the potential overproduction of toxic monomeric haem molecules. The increased binding of mu-oxo bishaem and monomeric forms of FePPIX at 39 degrees C may reflect a defense strategy against reactive oxidants and a mechanism of dampening down the inflammatory response to maintain an ecological balance.

The periodontal pathogen Porphyromonas gingivalis harnesses the chemistry of the mu-oxo bishaem of iron protoporphyrin IX to protect against hydrogen peroxide.

The major haem component in the black pigment of Porphyromonas gingivalis is the mu-oxo bishaem of iron protoporphyrin IX and formation and cell-surface binding of this haem species is proposed as an extracellular buffer against reactive oxidants [Smalley, J.W. et al. (1998) Biochem. J. 331, 681-685]. P. gingivalis cells grown in the presence of the mu-oxo bishaem were protected against H(2)O(2) compared to control cells grown without it. When added to the growth medium, soluble mu-oxo bishaem inactivated H(2)O(2) and supported cell growth. Cells carrying a surface layer of mu-oxo bishaem were less susceptible to peroxidation by H(2)O(2). Cell-surface haems were slowly destroyed during reaction with H(2)O(2). Binding of mu-oxo bishaem by P. gingivalis may aid survival during neutrophil attack through inactivation of hydrogen peroxide.

Iron protoporphyrin IX-albumin complexing increases the capacity and avidity of its binding to the periodontopathogen Porphyromonas gingivalis.

Cells of Porphyromonas gingivalis strains W50 and WPH35 bound albumin and haemalbumin complexes (with 2:1 and 1:1 molar ratios of protein to iron protoporphyrin IX) in a concentration-dependent manner. The binding capacity for both haemalbumins was greater than for albumin. Scatchard analysis of binding to strain W50 revealed monophasic binding for albumin with an association constant (Ka) approximately 10(5)/M. Binding of the haemalbumin complexes was biphasic. The Kas of the lower-affinity binding phases were similar to that for albumin, whilst those for the higher-affinity binding were approximately 20-30-fold greater. It is concluded that both the capacity and avidity for albumin binding to P. gingivalis are increased following haemalbumin complex formation. This phenomenon would enable cells to discriminate between albumin and haem-bearing albumin molecules as a potential source of haem. Such binding behaviour may confer a nutritional and ecological advantage in the periodontal pocket or gingival sulcus under conditions of haem limitation.

Proteolysis and utilization of albumin by enrichment cultures of subgingival microbiota.

Subgingival dental plaque consists mainly of microorganisms that derive their energy from amino acid fermentation. Their nutrient requirements are met by the subgingival proteolytic system, which includes proteases from microorganism and inflammatory cells, and substrate proteins from sulcus exudate, including albumin. To determine the selective effect of individual proteins on microbiota, we used albumin as the main substrate for growth. Eight subgingval plaque samples from untreated periodontal pockets of patients with adult periodontitis were inoculated in peptone yeast medium with bovine albumin (9 g/l). After three subculture steps, cell yields of the enrichment cultures at the medium with 0, 1.25, 2.5, 5, 10, and 20 g/l albumin were determined. Proteolytic activity (U/absorbance at 550 nm) of the enrichment cultures and different isolates derived from the cultures was estimated by the degradation of resorufin-labeled casein. It was observed that the yield of the mixed culture was albumin limited, and the proteolytic activities of the cultures in albumin broth were higher than in control (peptone broth). Among the isolates from the enrichment cultures, Peptostreptococcus micros, Prevotella melaninogenica, Prevotella buccae and Prevotella bivia demonstrated proteolysis. The frequent occurrence of Streptococcus gordonii and Streptococcus anginosus in the albumin cultures is explained by their ability to utilize arginine as an energy source for growth. Albumin in the medium was partly degraded by pure cultures but completely consumed in enrichment cultures, indicating synergy of bacterial proteinases. It is concluded that the subgingival microbiota possesses proteolytic activity and may use albumin as a substrate for their growth. Enrichment cultures on albumin may serve as a relatively simple in vitro model to evaluate the effects of proteinase inhibitors.

The periodontopathogen Porphyromonas gingivalis binds iron protoporphyrin IX in the mu-oxo dimeric form: an oxidative buffer and possible pathogenic mechanism.

Mössbauer spectroscopy was used to re-evaluate iron protoporphyrin IX, FePPIX, binding and the chemical nature of the black iron porphyrin pigment of Porphyromonas gingivalis. We demonstrate that FePPIX is bound to the cell in the mu-oxo dimeric form, [Fe(III)PPIX]2O, and that the iron porphyrin pigment is also composed of this material. P. gingivalis also assimilated monomeric Fe(II)- and Fe(III)PPIX into mu-oxo dimers in vitro. Scatchard analysis revealed a greater binding maximum of cells for mu-oxo dimers than for monomeric Fe(III)-or Fe(II)PPIX, although the relative affinity constant for the dimers was lower. Formation of [Fe(III)PPIX]2O via reactions of Fe(II)PPIX with oxygen, and its toxic derivatives, would serve as an oxidative buffer and permit P. gingivalis and other black-pigmenting anaerobes to engender and maintain a local anaerobic environment. Tying up of free oxygen species with iron protoporphyrin IX would also reduce and limit Fe(II)PPIX-mediated oxygen-radical cell damage. More importantly, formation of a cell-surface mu-oxo dimer layer may function as a protective barrier against assault by reactive oxidants generated by neutrophils. Selective interference with these mechanisms would offer the possibility of attenuating the pathogenicity of P. gingivalis and other iron protoporphyrin IX-binding pathogens whose virulence is regulated by this reactive molecule.

Hemin regulation of hemoglobin binding by Porphyromonas gingivalis.

Hemoglobin binding to chemostat-grown hemin-excess and hemin-limited cells of Porphyromonas gingivalis W50, and to cells of the avirulent, beige-pigmenting variant W50/BE1, was quantified. Hemin-excess W50 bound more hemoglobin than hemin-limited W50, mirroring the hemin-binding ability of these cells [Microb Ecol Health Dis 7:9-15, 1994]. In contrast to hemin, hemoglobin binding was not enhanced by sodium dithionite. The hemoglobin-binding capacity of hemin-excess W50/BE1 was below that of hemin-limited W50 and only observed under oxidizing conditions. Scatchard analysis revealed similar affinity constants for hemin-excess and hemin-limited W50, and confirmed a lower binding maximum for the latter. Hemin-excess W50/BE1 displayed cooperative binding of hemoglobin. These differences in binding were reflected in the binding of a horse radish peroxidase-conjugated hemoglobin (HHRPO) in a dot-blot assay. However, neither the 32-kDa hemin-binding protein, nor its 19-kDa heat-modified form, from either hemin-limited W50 or hemin-excess W50/BE1, bound this conjugate. These data indicate that hemoglobin binding by P. gingivalis is hemin-regulated and occurs via a mechanism different from hemin binding.

Albumin and hemalbumin degradation by Porphyromonas gingivalis.

Degradation of bovine albumin and hemalbumin by Porphyromonas gingivalis W50 cells under non-reducing conditions at 37 degrees C was examined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and densitometry. Albumin and hemalbumins with heme:protein molar ratios of 1:1, 4:1 and 8:1 were degraded, yielding protease-resistant 55.6-kDa peptides. Cells of strains WPH 35, 11834 and Bg 381 also produced a similar digestion pattern. N-terminal sequencing of the 55.6-kDa albumin digestion fragment revealed two peptides with the sequences 82glu-thr-tyr-gly-asp-met-ala and 95gln-pro-glu-arg-asn-glu-cys, indicating cleavage in the N-terminal hinge region. Tosyllysylchloromethylketone and N-ethylmaleimide were the most effective in inhibiting breakdown of albumin and hemalbumin with a 1:1 heme:protein ratio. Initial degradation rates of albumin and all hemalbumins were similar, but the total amount of hemalbumins degraded over 7.5 h decreased with increased ratio of bound hemin. The specific proteolysis of hemalbumin may enable P. gingivalis to release hemin from a region of the molecule where heme binding is least avid.

Haemin binding as a factor in the virulence of Porphyromonas gingivalis.

Haemin (iron protoporphyrin IX) is an essential growth factor for the periodontal pathogen. Porphyromonas gingivalis. Iron protoporphyrin IX (IPP IX) binding to the avirulent P. gingivalis beige variant (W50/BE1) and the black-pigmenting parent wild-type strain W50 was quantified. W50/BE1 grown in a chemostat under haemin excess-bound IPP IX under both oxidising and reducing conditions but with both lower capacity and avidity than either the haemin-limited- and haemin-excess-grown parent strain W50. Rosenthal plots for W50/BE1 indicated cooperative binding. W50/BE1 cells expressed a 32 kDa outer membrane haemin-binding protein when grown under conditions of haemin excess, and this strain might serve as a useful source from which to isolate this protein. The reduced IPP IX binding ability of W50/BE1 may be the rate-limiting factor for haem uptake and explain the reduced virulence and slower rate of pigmentation of this strain.

Kinetics of Congo-red binding by haemin-limited and haemin-excess cells of Porphyromonas gingivalis W50.

The binding of Congo red to P. gingivalis W50 grown in a chemostat under haemin-limitation and haemin-excess was quantified. Congo red bound to both haemin-excess and haemin-limited cells with similar capacity and affinity. Binding of Congo red was greater than for ferri- (haemin) or ferroprotoporphyrin IX (haem), and was not influenced by redox potential at low added ligand concentrations. Both haemin-limited and haemin-excess cells showed positive co-operativity towards Congo red binding. Pre-exposure of haemin-limited and haemin-excess cells to sub-saturating concentrations of ferriprotoporphyrin IX did not affect Congo red binding, whereas pre-exposure of haemin-excess cells to ferroprotoporphyrin IX increased binding. Iron protoporphyrin IX binding was enhanced after exposure of both haemin-excess and haemin-limited cells to Congo red, especially under reducing conditions. These results confirm that Congo red binding cannot be used as an indirect measure of haemin binding, nor can Congo red be used to inhibit haemin binding to P. gingivalis.

Congo red binding by Porphyromonas gingivalis is mediated by a 66 kDa outer-membrane protein.

Congo red was bound from solution by strains of Porphyromonas gingivalis including W50, HG189, HG184, NCTC 11834, Bg 381, WPH35, the slower brown pigmenting colonial variant W50/BR1, and the avirulent mutant W50/BE1, and by Porphyromonas endodontalis HG370 and Porphyromonas asaccharolytica B537. SDS-PAGE of whole cells of all species examined displayed a 66 kDa Congo-red-binding component which was also detected in the outer membranes of P. gingivalis W50 grown in the chemostat under both haemin limitation and haemin excess, and which corresponded to a Coomassie-blue-stained band of the same mobility. Pretreatment of haemin-excess batch-grown cells of P. gingivalis W50 with polymyxin B, which binds to lipid A, did not inhibit binding, whilst binding was enhanced in the presence of 2 M ammonium sulphate, suggesting the involvement of non-specific hydrophobic interactions. Binding was also reduced by pretreatment with trypsin and papain, and by 8-anilino-1-naphthalenesulphonic acid, which binds to hydrophobic amino acids. The 66 kDa binding component was sensitive to proteinase K digestion, and loss of Congo red staining of this band correlated with the quantitative reduction in Congo red binding by whole cells. These data, and our previous work, show that Congo red and iron protoporphyrin IX (haemin) are bound to different outer-membrane components, and that Congo red binding may be of little value as a marker to detect virulent strains of P. gingivalis or those expressing haemin-binding proteins.

Pathogenic mechanisms in periodontal disease.

Periodontal diseases have been considered as "infections" in which micro-organisms initiate and maintain the destructive inflammatory response. Host-mediated tissue destruction occurs via complement activation and the release of lysosomal enzymes, and connective tissue matrix metalloproteinases. Microbial enzymes may damage connective tissues directly, and, together with toxic metabolites and structural materials, are thought to disrupt the reparative activities of fibroblasts and cells of the immune defenses. The significance and relative contributions of host and microbial factors to the disease process remain unresolved. Environmental changes in the gingival sulcus and periodontal pocket and tissues, the degree of the host response and nutrient availability, concomitant with disease progression, compromise tissue metabolism and repair, and allow for enhanced or de novo expression of microbial virulence factors, such as proteases, which alter microbial pathogenicity. Proteolytic destruction of specific antibodies and complement by both viable and non-viable bacterial cells may retard phagocytic killing and removal of pathogens, thus prolonging the inflammatory response. Bacterial products may indirectly mediate tissue destruction by stimulating release of matrix metalloproteinases or by proteolytically inactivating the specific inhibitors of these enzymes.

Mucin-sulphatase activity of some oral streptococci.

Mucin-sulphatase activity, measured using a 35S-[SO4(2-)]-labelled colonic mucin substrate, was detected in whole cells of Streptococci isolated from the human oral cavity. The highest levels of sulphatase activity were found in all strains of Streptococcus salivarius, Streptococcus mitis and in half of the strains of Streptococcus mutans tested. Little or no activity was detected in 9 of the 11 Streptococcus oralis strains examined, in the 4 Streptococcus constellatus strains, and in the 3 Streptococcus anginosus isolates tested. The highest enzyme levels were obtained from the two fresh Streptococcus gordonii isolates. This is the first report of such activity in oral microorganisms. Streptococcal mucin-sulphatase may contribute to the destruction of salivary mucins and mitigate their protective functions in the oral cavity, and be a determinant in the development of dental caries.

Haemin-binding proteins of Porphyromonas gingivalis W50 grown in a chemostat under haemin-limitation.

Porphyromonas gingivalis W50 was grown in a chemostat at pH 7.3 under haemin-limitation and haemin-excess at a constant mean doubling time of 6.9 h. Outer membranes (OM) were extracted from whole cells using EDTA and compared by SDS-PAGE. Haemin-limited cells expressed novel outer membrane proteins (OMPs) of mol. mass 115, 113 and 19 kDa when samples were solubilized at 100 degrees C. A 46 kDa OMP was observed in haemin-excess cells but not in those from haemin-limited conditions. Tetramethylbenzidine (TMBZ) staining of gels, after OM solubilization at 20 degrees C, was used to detect haemin-binding proteins (HBPs). HBPs were observed only in OM from haemin-limited cells. The major HBP (mol. mass 32.4 kDa) corresponded to a similar sized Kenacid-blue-stained protein which was not observed in haemin-excess-derived OM. Haemin-limited cells and OM displayed a ladder-like series of Kenacid-blue-stained proteins. Lighter TMBZ-stained proteins of mol. mass 51, 53, 56 and 60 kDa, with mobilities corresponding to those of silver-stained LPS components, were observed in haemin-limited OM. No soluble HBPs were detected extracellularly. The greater number of HBPs expressed by cells grown under haemin-limitation may reflect an additional cell surface receptor system for haemin acquisition under low environmental levels of this essential cofactor.

Haemin-restriction influences haemin-binding, haemagglutination and protease activity of cells and extracellular membrane vesicles of Porphyromonas gingivalis W50.

Porphyromonas gingivalis strain W50 was grown in a chemostat either under haemin limitation or haemin excess at pH 7.3. Cells and the extracellular vesicle (ECV) and extracellular protein (EP) fractions were separated, quantified, and assayed for haemagglutination, protease activity and haemin binding. Under haemin-limitation, despite a reduction in cell yield, there was a 2.5-fold increase in the gravimetric yield of extracellular vesicles. Cells and vesicles from haemin-limited cultures, haemagglutinated sheep red blood cells to higher titres than their haemin-excess counterparts. Growth in haemin-excess conditions resulted in increased haemin-binding capacities of ECV, cells and EDTA-extracted outer membrane. Cells grown under haemin-excess showed a 2-fold elevation in specific activity towards the substrate N-alpha-benzoyl-L-arginine-p-nitroanilide (L-BAPNA) compared to haemin-limited cells. The specific activities against L-BAPNA for haemin-limited ECV were 3-fold greater than their haemin-excess counterparts. These vesicle activities represented 25% and 3% of the total culture protease activity under haemin limited and haemin excess conditions respectively.

Extracellular vesicle-associated and soluble trypsin-like enzyme fractions of Porphyromonas gingivalis W50.

Soluble vesicle-associated trypsin-like enzyme fractions (VSF) were prepared by sonication from extracellular vesicles (ECV) from strains W50 and W50/BE1. High-(H), intermediate-(I) and low-(L) molecular-weight VSF enzyme subfractions were identified by non-dissociative gel filtration chromatography with Mr 160, 95 and 60 kDa respectively. The chromatographic profiles of W50 VSF from 48-h and 72-h cultures were identical. W50/BE1 VSF displayed a higher ratio of the 160 to 60 kDa components. This ratio was reduced in VSF from 72-h cultures. Extracellular soluble protein (EP) trypsin profiles were similar to their respective VSF, but the 60 kDa component predominated. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) revealed a loss of soluble extracellular polypeptides with culture age. A polyclonal antiserum to EP subfraction L reacted in immunoblots with a 50 kDa peptide of subfraction L of W50. Whole EP and its subfraction H displayed a 50 kDa immunoreactive peptide but no peptides of higher molecular weight. This antiserum reacted with a similar sized peptide, and with lower-molecular-weight components in whole ECV. Gelatin substrate zymography of whole EP following non-reducing SDS-PAGE revealed a major 80 kDa protease that increased with culture age. Minor protease bands of 70 and 50 kDa were also detected.