Porphyromonas gingivalis (P. gingivalis) is a gram-negative, anaerobic bacterium recognized as a major player in progression of periodontal disease. P. gingivalis survives in the oral cavity while being exposed to dynamic environmental conditions such as pH, temperature, nutrient availability and host immune responses such as oxygen tension and nitrosative stress. Survival and pathogenesis of P. gingivalis in the oral cavity require mechanisms to regulate gene expression in response to the extracellular signals. Little is known about the regulatory mechanisms of P. gingivalis in the oral cavity, so it is important to investigate and characterize these regulatory mechanisms. Adaptation to environmental cues using riboregulation is a significant mechanism for post-transcriptional regulation in bacteria. Using bioinformatics, we have identified a putative RNA-binding protein in P. gingivalis: RBP. Bioinformatic studies have led to the selection of HUβ and HUα nucleoid associated proteins as controls for RNA binding. I hypothesize that the candidate proteins RBP, HUβ and HUα bind RNA in P. gingivalis. The first aim is to show that RBP, HUβ and HUα bind RNA. Using electrophoretic mobility shift assays with IRE RNA and synthesized RNA motifs, I have confirmed that the proteins do bind RNA. The second aim is to isolate and sequence the P. gingivalis RNA that bind to RBP, HUβ and HUα. I have isolated the RNAs that bound the proteins and determined identity of the RNA using high throughput sequencing. Finally, I have identified an antibody that specifically binds RBP to use for in vivo immunoprecipitation of RNA-protein complexes from P. gingivalis. In conclusion RBP, HUβ and HUα are novel RNA binding proteins in P. gingivalis, and further investigation of these proteins is necessary to understand the mechanisms of gene regulation in P. gingivalis.

Porphyromonas gingivalis, a gram-negative anaerobic bacterium, is a major etiological agent in the initiation and progression of severe forms of periodontal disease. Oral bacteria like P. gingivalis are subject to continually changing conditions as a consequence of host eating, oral hygiene patterns and subgingival temperatures. As such survival requires an adaptive response to environmental cues, but little is known about the mechanism by which P. gingivalis controls co- and post-transcriptional regulation of RNA levels and potentially protein expression. RNA-binding proteins (RBPs) are evolutionarily conserved across species and are involved in such regulatory mechanisms. However, P. gingivalis currently has no identified RBP. Recently, PG0627 has become an ideal candidate for a putative RBP due to its sequence homology to RBPs across various species. By characterizing PG0627, we can gain better insight into the function of this hypothetical protein and determine if it indeed behaves like an RNA-binding protein. A host of studies were done on a PG0627-deficient P. gingivalis mutant, V3139, in order to determine the biological role of the protein encoded by the gene. Our bioinformatics analysis indicated that PG0627 had sequence homology to several RNA recognition motifs or RBPs. Furthermore, our PG0627-deficient mutant, when compared to W83, exhibited decreased cell-associated iron content, decreased total interactions and invasions with eukaryotic cells, and decreased protease activity. Conversely, our PG0627-deficient mutant displayed slightly increased growth in the presence of nitrosative stress, and in hemin-depleted conditions. In conclusion, our results support that PG0627 is a valid candidate for an RNA-binding protein in P. gingivalis.

The facultative anaerobe, Escherichia coli and the obligate anaerobe, Porphyromonas gingivalis are two bacteria that reside in our body. Although they reside in separate environments, they are both subject to hydrogen peroxide stress and have mechanisms to regulate the stress. OxyR is the primary transcriptional regulator/sensor of oxidative stress response caused by hydrogen peroxide. OxyR in P. gingivalis is not well-characterized compared to OxyR in E. coli. We sought to characterize and compare the two forms of OxyR in order to gain a better understanding of the protein. We determined the oligomeric state of both proteins: primarily a tetramer for E. coli and primarily a tetramer for P. gingivalis OxyR.. We demonstrated DNA binding with E. coli OxyR, indicating purification of the functional form of E. coli OxyR.Through pulldown assays we discovered potential novel binding targets, mobB for E. coli OxyR and PG1209 for P. gingivalis OxyR. Many of the other targets corresponded to intergenic regions within genes, which may pertain to small RNAs or small proteins. These results show that OxyR in E. coli and P. gingivalis has novel function and properties indicating an expanded role in addition to the well-characterized oxidative stress response.

Porphyromonas gingivalis é um bacilo anaeróbio Gram negativo envolvido com o início e progressão de doenças periodontais. É considerado um colonizador tardio ou secundário do biofilme oral capaz de aderir a células de Streptococcus gordonii, um colonizador inicial do biofilme dental. Este estudo se propôs a comparar a expressão de genes de amostras de P. gingivalis isoladas de diferentes condições periodontais usando Differential Display (DD) Reverse Transcription PCR, durante a formação de biofilme misto com S. gordonii. O perfil de expressão gênica de células em biofilme (saúde e periodontite) foi comparado, assim como com células planctônicas pareadas. Bandas diferencialmente expressas nas diferentes condições foram clonadas e seqüenciadas, seguida de identificação dos genes em bancos de dados. A confirmação da expressão diferencial dos genes detectados foi realizada através de PCR em Tempo Real. Desses genes, alguns se relacionam com fatores de virulência, outros com obtenção de energia além de genes relacionados com proteínas de membrana e de transporte. / Porphyromonas gingivalis is a Gram negative anaerobic rod involved with the beginning and progression of periodontal diseases P. gingivalis is a late or secondary colonizer of oral biofilms and adhere to cells of Streptococcus gordonii, an early colonizer of dental plaque. This study aim to compare the gene expression of P. gingivalis strains isolated from different periodontal conditions using Differential Display (DD) Reverse Transcription PCR, in a mixed biofilm with S. gordonii. The gene expression profile was determined with biofilm cells (health and disease) and also with their planktonic samples partners. Bands differentially expressed were cloned, sequenciated and analyzed in gene data bank. Differential expression was confirmed by Real Time PCR. Some of the confirmed genes are related to virulence factors, energy metabolism and transport and membrane proteins.

Publicación a texto completo no autorizada por el autor / La realización del estudio tiene como objetivo el aislamiento de la bacteria Porphyromonas gingivalis, de pacientes con cuadros de periodontitis de la clínica estomatológica, perteneciente a la Facultad de Odontología de la Universidad Nacional Mayor de San Marcos, para luego enfrentarlas con la oleorresina de Copaifera reticulata “Copaiba”. Las muestras para la obtención del microorganismo, son tomadas con conos de papel N° 30 ó 40, colocados dentro del surco gingival a profundidad, por un tiempo de 60 segundos, luego se llevan al medio de transporte BHI (infusión cerebro corazón), diluidos en diferentes concentraciones y sembradas en el medio de agar sangre suplementado, incubándose en condiciones de anaerobiosis, a 37 °C durante 7 a 14 días. Para su identificación preliminar, se realizan pruebas de catalasa, oxidasa, y medios diferenciales como SIM, Urea, TSI (Tripe azúcar hierro) y Citrato. Para la identificación definitiva de la bacteria purificada, se realiza la prueba automatizada de Api 20 Anaerobios. El enfrentamiento se realiza por el test de difusión en agar con disco, para lo cual se prepara diez concentraciones distintas de la oleorresina, siendo el diluyente dimetilsulfoxido. Así mismo se prepara una suspensión equivalente al patrón 1 de Mc Farland de P. gingivalis, para ser sembrada, en un medio de agar sangre suplementado, luego se colocan los discos equidistantemente y se incuba a 37 °C durante siete a diez días en anaerobiosis. Los resultados de las mediciones de los halos de inhibición, dan una media en la Concentración Mínima Inhibitoria (CMI) de 3,4345 %. Se concluye que la oleorresina de copaiba, es un posible fitoproducto, que complementaria el tratamiento odontológico. / Tesis

Histatins, a group of proteins produced by human salivary glands, have a variety of innate immune functions including the ability to: kill oral microorganisms, neutralize toxins, inactivate protease/collagenase activities, inhibit co-aggregation of oral bacteria, and inhibit lipopolysaccharide mediated activities. Hemagglutinin B (HagB), a virulence factor of the periodontal pathogen Porphyromonas gingivalis, induces a robust cytokine and chemokine response in human myeloid dendritic cells. In this study, I hypothesize that histatin 5 can attenuate a HagB-induced chemokine response. Objectives: To characterize an expanded cytokine and chemokine response induced in human myeloid dendritic cells by HagB, and to determine if prior incubation of HagB with histatin 5 attenuates these responses. Methods: In my first experiment, 0.040 M HagB was mixed with dilutions of histatin 5 and histatin 8 (Sigma, 0.04 to 40.0 M), incubated at 37C for 30 minutes, and added to 2 x 104 human myeloid dendritic cells (Lonza, Walkersville, MD). At 24 hours, culture media was removed, and 6 cytokines and chemokines (pg/ml) were determined in cell-free supernatants (Millipore, Billerica, MA) using the Luminex 100 IS instrument (Luminex, Austin, TX). In my second experiment, 0.040 M HagB was mixed with 40.0 M histatin 5 only (e.g., 1:1000), incubated at 37C for 30 minutes, and added to 2 x 104 human myeloid dendritic cells. At 0, 1, 2, 4, 8, 16, and 24 hours post-inoculation, culture media was removed, and 26 cytokines and chemokines (pg/ml) were determined in cell-free supernatants. Results: In both experiments, human myeloid dendritic cells incubated with HagB produced Th1, Th2, and Th17 cytokines (IL-2, IL-12(p70), IFN-, IL-3, IL-4, IL-5, , IL-15, IL-17); pro-inflammatory cytokines (IL-1, IL-1, IL-6, TNF-, IL-12(p40); anti-inflammatory v cytokines (IL-10, IL-13, IFN2); chemokines (CXCL8/IL-8, CXCL10/IP-10, CCL2/MCP-1, CCL3/MIP-1, MIP-1b, CCL11/eotaxin); and colony stimulating factors (IL-7, G-CSF, GM-CSF). Histatin 5 significantly attenuated (p < 0.05) the IL-8 response induced by HagB at 8 - 16 hours and to a lesser extent, the IL-6, GM-CSF, MCP-1, MIP-1α, MIP-1β, and TNF-α response. Conclusion: Histatin 5 is an important salivary component capable of attenuating an IL-8 response. Together with human beta defensin 3, another peptide previously shown to attenuate pro-inflammatory cytokines, histatin 5 may help control and contain oral infection and inflammation by down regulating IL-8 chemotactic response.

Periodontal disease is a chronic inflammation of the gingiva and periodontium that leads to progressive destruction and irreversible damage to the supportive structures of the teeth. It affects nearly half of the United States population and is a particular risk factor in adults older than 65 years of age. Oral microorganisms assemble in plaque as a polymicrobial biofilm and Porphyromonas gingivalis, an important secondary colonizer in oral biofilms, has been implicated in periodontal disease. Although the protective functions of various salivary molecules such as antimicrobial proteins have been delineated, lipids present in saliva and on the oral mucosa have been largely ignored and there is growing evidence that the role of lipids in innate immunity is more important than previously realized. In fact, recent studies suggest that sphingoid bases and fatty acids, which exhibit potent broad spectrum antimicrobial activity against a variety of bacteria and fungi, are likely important innate immune molecules involved in the defense against oral bacterial and fungal infections. However little is known about their spectrum of activity or mechanisms of action. In addition, the effects of these lipids that are endogenous to the oral cavity have not been explored against oral bacteria. In this study I hypothesized that oral mucosal and salivary lipids exhibit dose-dependent antimicrobial activity against P. gingivalis and alter cell morphology and metabolic events. To test this hypothesis, I first examined the effects of two fatty acids: sapienic acid and lauric acid, and three sphingoid bases: sphingosine, dihydrosphingosine, and phytosphingosine, against a variety of gram-positive and gram-negative bacteria including P. gingivalis. Using broth microdilution assays to determine minimum inhibitory and minimum bactericidal concentrations, I show that antimicrobial activity against bacteria is dose-dependent, lipid specific, and microorganism specific. Kill kinetics were also variable across each bacteria-lipid combination. Upon examination of select bacteria-lipid combinations via scanning and transmission electron microscopy, different morphologies were evident across all treatments, demonstrating differential activity of each lipid for a particular bacterium as well as for each bacterium across different lipids. In addition, all sphingoid bases and fatty acids were taken up and retained in association with P. gingivalis cells and could be extracted along with bacterial lipids and separated using thin layer chromatography. Using a combination of two-dimensional in-gel electrophoresis and Western blots followed by mass spectroscopy and n-terminus degradation sequencing, I show that sapienic-acid treatment induces a unique stress response in P. gingivalis, as evidenced by the ability of P. gingivalis to upregulate a set of proteins involved in fatty acid biosynthesis metabolism and energy production, protein processing, cell adhesion, and virulence. Finally, utilizing flow cytometry and confocal microscopy, I assessed the effects of oral antimicrobial lipids against a representative host cell and describe oral lipid concentrations that are both antimicrobial to P. gingivalis cells and non-cytotoxic to the representative host cells tested. Combined, these data strongly suggest that sphingoid bases and fatty acids found within the saliva and on oral mucosa likely do contribute to the innate antimicrobial activity of saliva, mucosal surfaces, and skin and this dose-dependent activity is both lipid specific and bacteria specific. This information adds to current knowledge of the innate functions of endogenous lipids in the oral cavity. With bacterial resistance to current antibiotics increasing, the exploration of new antimicrobial agents is important and these lipid treatments may be beneficial for prophylactic treatments or therapeutic intervention of infection by supplementing the natural immune function of endogenous lipids on skin and other mucosal membranes.

Objectives: This investigation adapted the LIVE/DEAD� Baclight[TM] bacterial viability stain for the quantitative determination of bacterial cell viability of the aerotolerant anaerobes Porphyromonas gingivalis ATCC 33277 and Prevotella intermedia ATCC 25611. The Live/Dead stain was used to determine the influence of haem availability on the resistance of P. gingivalis and P. intermedia to the reactive oxygen species (ROS) superoxide anion and hydrogen peroxide and compare the sensitivities between the haem-requiring periodontal bacteria to ROS. Neutrophils use oxidative and non-oxidative killing mechanisms. During phagocytosis, neutrophils kill bacteria via a respiratory burst, producing ROS. P. gingivalis and P. intermedia are oxygen-tolerant gram-negative bacteria found in the gingival crevice. These bacteria express superoxide dismutase (SOD) activity, which extends some protection against superoxide radicals. Methods: Initially, experiments were performed to validate the reliability and accuracy of the fluorogenic Live/Dead stain using Escherichia coli ATCC 10798 (K-12), followed by experiments using P. gingivalis. The Live/Dead stain distinguishes viable:non-viable proportions of bacteria using mixtures of green (SYTO 9) and red (propidium iodide) fluorescent nucleic acid stains respectively. Bacterial cell viability was assessed with fluorescence microscopy and subsequently quantitative measurement using a fluorescence microplate reader (BMG Fluorostar plus Optima). P. gingivalis and P. intermedia colonies were subcultured from frozen cultures, in Tryptic soy broth (TSB) (Difco) and incubated anaerobically for approximately five days. They were further subcultured in pre-reduced TSB, supplemented with menadione 0.5[mu]g/ml (TSB-M) and either 5 [mu]g/ml haemin (Haem 5), 50 [mu]g/ml haemin (Haem 50) or without supplemental haemin (Haem 0). Cultures were grown anaerobically at 37�C to early stationary phase (approximately 48 hours). For experimental purposes, bacteria were harvested, washed and resuspended in 10 mM Tris-buffered saline (pH 7.5) containing peptone (TBS-P) (0.1 mg/ml), with a final adjustment to OD₅₄₀ [approximately equals] 2.0 (which corresponds to 1 x 10⁹ bacteria/ml). Bacterial suspensions were diluted ([approximately equals] 10⁸/ml) into TBS-P containing the fluorogenic viability stain (BacLight, Molecular Probes). Either pyrogallol (0.02 - 2 mM) or hydrogen peroxide (0.01 - 100 mM) was added (except to control tubes); tubes were vortexed for ten seconds and incubated at 37�C. Viability was monitored fluorimetrically for three hours. Results: For both P. gingivalis and P. intermedia, a pyrogallol concentration of 0.2 mM resulted in 80 to 90% cell death; and a hydrogen peroxide concentration of 10 mM killed approximately 80 to 90% of cells. Irrespective of the haem status, no significant difference was determined between the overall maximum rate of killing of P. gingivalis and P. intermedia, in their response to either superoxide or hydrogen peroxide; with the exception that the P. intermedia Haem 0 group was significantly less susceptible to hydrogen peroxide than the P. gingivalis Haem 0 group. For the majority of the experiments, there was no significant difference between final bacterial cell viability in the Haem 0 and Haem 5 cells for both species, after 3 hours exposure to various concentrations of ROS. However, the Haem 50 cells showed a significant increased susceptibility (albeit, a small difference) to both hydrogen peroxide and superoxide. Conclusions: The Live/Dead bacterial viability stain provided a valuable method to monitor "real-time" killing, avoiding the difficulties associated with culture-based methods for assessing viability. Haem availability had no clear influence on the resistance to ROS of either P. gingivalis or P. intermedia Haem 0 and Haem 5 cells. The Haem 50 cells showed a very slight increase in susceptibility to hydrogen peroxide and superoxide. Although P. intermedia may be isolated in significant numbers from healthy gingivae, as well as from periodontally diseased sites, it was no more resistant to ROS than was P. gingivalis, which is associated with periodontal lesions and difficult to cultivate from relatively healthy (more oxygenated) sites. This suggests that resistance to ROS does not contribute to the ecological distinction between these two species. The finding that haem availability did not influence sensitivity implies that these bacteria do not accumulate haem for the purpose of protection from ROS.

Patients with inflammatory or infectious conditions such as periodontitis, peri-implantitis, osteomyelitis, rheumatoid arthritis, septic arthritis and loosened joint prosthesis display varying severity of destruction in the adjacent bone tissue. Bone loss in inflammatory diseases is considered a consequence of cytokine induced RANKL and subsequent enhanced osteoclast formation. Hence, osteotropic cytokines and their receptors have been suggested to be important for the pathogenesis of inflammation-induced osteolysis. It is, here, suggested that bacterial components, so called “pathogen associated molecular patterns=PAMPs”, may also be involved. Varieties of cells express receptors for PAMPs, including Toll-like receptors (TLRs) which are the first line of defence in the innate immune system. LPS (lipopolysaccharide), fimbria and lipoproteins from pathogenic bacteria such as P. gingivalis, S. aureus are ligands for TLR2 and flagellin from pathogenic flagellated bacteria like S. typhimurium is a ligand for TLR5.   Since the susceptibility to, or the severity of inflammation-associated bone diseases are likely related to differences in the tissue response, and the mechanisms by which PAMPs interact with bone cells are not fully understood, we aimed to elucidate the importance of different TLRs for inflammation induced bone loss by conducting in vitro and in vivo investigations. Activation of TLR2 and TLR5 in organ cultured mouse parietal bones increased bone resorption in a time- and concentration-dependent manner by a process inhibited by OPG and bisphosphonate, showing the crucial role of RANKL-induced osteoclast formation. In addition, the number of osteoclasts, expression of osteoclastic genes and osteoclastogenic transcription factors were increased. In the bones and in osteoblasts isolated from the bones, TLR2 agonists increased the expression of RANKL without affecting OPG, while TLR5 activation resulted in enhanced RANKL and decreased OPG. Activation of both TLR2 and TLR5 stimulated the expression in both bones and osteoblasts of prostaglandins and pro-inflammatory cytokines, known to stimulate RANKL. By blocking the cytokines and prostaglandin, we showed that TLR2 and TLR5 induced bone resorption and RANKL expression are independent of these molecules. Activation of TLR2, but not TLR5, in mouse bone marrow macrophage cultures inhibited RANKL-induced osteoclast formation, an effect not observed in committed pre-osteoclasts. Local administration in vivo of TLR2 and TLR5 agonists on the top of mouse skull bones enhanced local and systemic osteoclast formation and bone resorption. Using knockout mice, we showed that the effects by LPS from P. gingivalis (used as TLR2 agonist) and flagellins (used as TLR5 agonists) are explicit for TLR2 and TLR5 ex vivo and in vivo, respectively. These data show that stimulation of TLR2 and TLR5 results in bone resorption in vitro and in vivo mediated by increased RANKL in osteoblasts and thus may be one mechanism for developing inflammatory bone loss. Interestingly, histological analyses of skull bones of mice treated locally with TLR2 and TLR5 agonists revealed that the bones not only reacted with locally increased osteoclastogenesis (osteoclast formation), but also with locally increased new bone formation. This was observed on both periosteal and endosteal sides of the bones, as well as in the bone marrow compartment. The formation of new bone was seen close to osteoclasts in some parts, but also in other areas, distant from these cells. The response was associated with active, cuboidal osteoblasts, extensive cell proliferation and increased expression of genes coding for bone matrix proteins and osteoblastic transcription factors. In conclusion, activation of TLR2 and TLR5 in osteoblasts results in bone loss associated with enhanced osteoclast formation and bone resorption, as well as with increased osteoblast differentiation and new bone formation, indicating that inflammation causes bone modeling. The data provide an explanation why LPS from P. gingivalis and flagellin from flagella-expressing bacteria can stimulate bone loss. Since TLR2 and TLR5 can be activated not only by bacterial components, but also by endogenous ligands produced in inflammatory processes, the data also contribute to the understanding of inflammation induced bone loss in autoimmune diseases. Hopefully, these findings will contribute to the development of treatment strategies for inflammation induced bone loss.

Objectives: This investigation adapted the LIVE/DEAD� Baclight[TM] bacterial viability stain for the quantitative determination of bacterial cell viability of the aerotolerant anaerobes Porphyromonas gingivalis ATCC 33277 and Prevotella intermedia ATCC 25611. The Live/Dead stain was used to determine the influence of haem availability on the resistance of P. gingivalis and P. intermedia to the reactive oxygen species (ROS) superoxide anion and hydrogen peroxide and compare the sensitivities between the haem-requiring periodontal bacteria to ROS. Neutrophils use oxidative and non-oxidative killing mechanisms. During phagocytosis, neutrophils kill bacteria via a respiratory burst, producing ROS. P. gingivalis and P. intermedia are oxygen-tolerant gram-negative bacteria found in the gingival crevice. These bacteria express superoxide dismutase (SOD) activity, which extends some protection against superoxide radicals. Methods: Initially, experiments were performed to validate the reliability and accuracy of the fluorogenic Live/Dead stain using Escherichia coli ATCC 10798 (K-12), followed by experiments using P. gingivalis. The Live/Dead stain distinguishes viable:non-viable proportions of bacteria using mixtures of green (SYTO 9) and red (propidium iodide) fluorescent nucleic acid stains respectively. Bacterial cell viability was assessed with fluorescence microscopy and subsequently quantitative measurement using a fluorescence microplate reader (BMG Fluorostar plus Optima). P. gingivalis and P. intermedia colonies were subcultured from frozen cultures, in Tryptic soy broth (TSB) (Difco) and incubated anaerobically for approximately five days. They were further subcultured in pre-reduced TSB, supplemented with menadione 0.5[mu]g/ml (TSB-M) and either 5 [mu]g/ml haemin (Haem 5), 50 [mu]g/ml haemin (Haem 50) or without supplemental haemin (Haem 0). Cultures were grown anaerobically at 37�C to early stationary phase (approximately 48 hours). For experimental purposes, bacteria were harvested, washed and resuspended in 10 mM Tris-buffered saline (pH 7.5) containing peptone (TBS-P) (0.1 mg/ml), with a final adjustment to OD₅₄₀ [approximately equals] 2.0 (which corresponds to 1 x 10⁹ bacteria/ml). Bacterial suspensions were diluted ([approximately equals] 10⁸/ml) into TBS-P containing the fluorogenic viability stain (BacLight, Molecular Probes). Either pyrogallol (0.02 - 2 mM) or hydrogen peroxide (0.01 - 100 mM) was added (except to control tubes); tubes were vortexed for ten seconds and incubated at 37�C. Viability was monitored fluorimetrically for three hours. Results: For both P. gingivalis and P. intermedia, a pyrogallol concentration of 0.2 mM resulted in 80 to 90% cell death; and a hydrogen peroxide concentration of 10 mM killed approximately 80 to 90% of cells. Irrespective of the haem status, no significant difference was determined between the overall maximum rate of killing of P. gingivalis and P. intermedia, in their response to either superoxide or hydrogen peroxide; with the exception that the P. intermedia Haem 0 group was significantly less susceptible to hydrogen peroxide than the P. gingivalis Haem 0 group. For the majority of the experiments, there was no significant difference between final bacterial cell viability in the Haem 0 and Haem 5 cells for both species, after 3 hours exposure to various concentrations of ROS. However, the Haem 50 cells showed a significant increased susceptibility (albeit, a small difference) to both hydrogen peroxide and superoxide. Conclusions: The Live/Dead bacterial viability stain provided a valuable method to monitor "real-time" killing, avoiding the difficulties associated with culture-based methods for assessing viability. Haem availability had no clear influence on the resistance to ROS of either P. gingivalis or P. intermedia Haem 0 and Haem 5 cells. The Haem 50 cells showed a very slight increase in susceptibility to hydrogen peroxide and superoxide. Although P. intermedia may be isolated in significant numbers from healthy gingivae, as well as from periodontally diseased sites, it was no more resistant to ROS than was P. gingivalis, which is associated with periodontal lesions and difficult to cultivate from relatively healthy (more oxygenated) sites. This suggests that resistance to ROS does not contribute to the ecological distinction between these two species. The finding that haem availability did not influence sensitivity implies that these bacteria do not accumulate haem for the purpose of protection from ROS.