Epstein-Barr virus reactivation by persistent apical periodontal pathogens.

AIM: To assess whether Epstein-Barr virus (EBV) reactivation is triggered by persistent apical periodontitis-related microbes using in vitro and ex vivo methodologies. METHODOLOGY: Surgically removed human periapical granulomas (n = 50) and healthy gingival tissues (n = 10) were analysed to determine the presence of EBV and seven persistent apical periodontitis-related microbes. In addition, real-time polymerase chain reaction was used to detect the mRNA expression of BZLF-1, an immediate-early gene of EBV. Expression of latent membrane protein (LMP)-1 and ZEBRA, an early lytic protein of EBV encoded by BZLF-1, was also examined using triple-colour immunofluorescence staining. n-Butyric acid produced by the microbes was quantified, and luciferase assays were performed in association with bacterial lysates. In addition, Daudi cells were cultured with bacterial lysates, and the expression levels of BZLF-1 mRNA and ZEBRA protein were determined. RESULTS: EBV DNA and BZLF-1 mRNA were detected in 47 out of 50 periapical granulomas, but not in healthy gingival tissues. The EBV DNA copy number and the number of Fusobacterium nucleatum were significantly positively correlated with BZLF-1 expression in periapical granulomas. The number of Prevotella intermedia was slightly correlated with BZLF-1 expression; however, the other microbes were not. CD79a-positive B cells in periapical granulomas, but not those in healthy gingival tissues, expressed both LMP-1 and ZEBRA. n-Butyric acid production was the highest in F. nucleatum and the lowest in P. intermedia. Enterococcus faecalis, Candida albicans and the other tested microbes did not produce n-butyric acid. An F. nucleatum lysate exhibited significantly increased BZLF-1-luciferase activity in the same manner of commercial butyric acid, whereas P. intermedia did not. F. nucleatum also induced the expression of BZLF-1 mRNA and ZEBRA protein by Daudi cells, indicating that EBV reactivation was induced. CONCLUSION: Among the persistent apical periodontitis-related bacteria that were tested, F. nucleatum most strongly reactivated latent EBV, whereas E. faecalis and C. albicans as well as the other microbes did not.

A potential role for the silent information regulator 2 homologue 1 (SIRT1) in periapical periodontitis.

AIM: To investigate the role played by silent information regulator 2 homologue 1 (SIRT1) during angiogenesis of periapical periodontitis. METHODOLOGY: Periapical granulomas were subjected to dual-colour immunofluorescence imaging and real-time polymerase chain reactions assaying the expression levels of SIRT1, vascular endothelial growth factor (VEGF) and VE-cadherin. The association between Ki-67 and SIRT1 expression was also examined. Human umbilical vein endothelial cells (HUVECs) were treated with a combination of lipopolysaccharide and resveratrol (a SIRT1 activator) or sirtinol (a SIRT1 inhibitor); and the levels of mRNAs encoding SIRT1, VEGF and VE-cadherin were determined. HUVEC tube formation was assayed in the presence of resveratrol or sirtinol. The Mann-Whitney U-test or the Tukey-Kramer test was used for statistical analysis. RESULTS: Ki-67-expressing cells, including endothelial cells, lay adjacent to SIRT1-expressing cells in periapical granulomas. In addition, SIRT1-expressing cells were detected adjacent to VEGF-expressing cells and VEGF- or VE-cadherin-expressing endothelial cells. SIRT1, VEGF and VE-cadherin mRNA expression levels in periapical granulomas were significantly higher (P = 0.0054, 0.0090 and 0.0090, respectively) than those in healthy gingival tissues. HUVECs treated with resveratrol exhibited significantly higher expression of mRNAs encoding SIRT1, VEGF and VE-cadherin (P = 0.0019, 0.00005 and 0.0045, respectively) compared with controls, but sirtinol inhibited such expression. Resveratrol caused HUVECs to form tube-like structures, whilst sirtinol inhibited this process. CONCLUSIONS: These findings suggest that SIRT1 may stimulate angiogenesis in periapical granulomas by triggering the proliferation of endothelial cells and inducing VEGF and VE-cadherin expression.

Porphyromonas endodontalis reactivates latent Epstein-Barr virus.

AIM: To determine whether Porphyromonas endodontalis can reactivate latent Epstein-Barr virus (EBV). METHODOLOGY: The concentrations of short-chain fatty acids (SCFAs) in P. endodontalis culture supernatants were determined using high-performance liquid chromatography. A promoter region of BamHI fragment Z leftward open reading frame 1 (BZLF-1), which is a transcription factor that controls the EBV lytic cycle, was cloned into luciferase expression vectors. Then, the luciferase assay was performed using P. endodontalis culture supernatants. Histone acetylation using Daudi cells treated with P. endodontalis culture supernatants was examined using Western blotting. BZLF-1 mRNA and BamHI fragment Z EB replication activator (ZEBRA) protein were also detected quantitatively using real-time polymerase chain reaction (PCR) and Western blotting. Surgically removed periapical granulomas were examined to detect P. endodontalis, EBV DNA, and BZLF-1 mRNA expression using quantitative real-time PCR. Statistical analysis using Steel tests was performed. RESULTS: The concentrations of n-butyric acid in P. endodontalis culture supernatants were significantly higher than those of other SCFAs (P = 0.0173). Using B-95-8-221 Luc cells treated with P. endodontalis culture supernatants, the luciferase assay demonstrated that P. endodontalis induced BZLF-1 expression. Hyperacetylation of histones was also observed with the culture supernatants. BZLF-1 mRNA and ZEBRA protein were expressed by Daudi cells in a dose-dependent manner after the treatment with P. endodontalis culture supernatants. P. endodontalis and BZLF-1 in periapical granulomas were also detected. The expression levels of BZLF-1 mRNA were similar to the numbers of P. endodontalis cells in each specimen. CONCLUSIONS: n-butyric acid produced by P. endodontalis reactivated latent EBV.

Nitric oxide attenuates vascular endothelial cadherin-mediated vascular integrity in human chronic inflammation.

In this study, we examined the role of nitric oxide (NO) in controlling vascular integrity mediated by vascular endothelial (VE)-cadherin in chronic inflammation. Periapical granulomas were analysed for the expression of inducible NO synthase (iNOS) and VE-cadherin, and more iNOS expression than VE-cadherin was shown. Human umbilical vein endothelial cells (HUVECs) were stimulated with proinflammatory cytokines and lipopolysaccharide extracted from Porphyromonas gingivalis and it induced iNOS expression, whereas it reduced VE-cadherin expression, compared with negative controls. On the other hand, pre-incubation with 1400W, an iNOS-specific inhibitor, markedly reduced iNOS expression in stimulated HUVECs and restored VE-cadherin expression to its control level, suggesting that vascular integrity was modulated in conjunction with the reduction of NO. Immunocytochemistry confirmed the functional role of NO in cultured HUVEC monolayers with or without 1400W. These data are consistent with a hypothesis suggesting that NO could attenuate VE-cadherin-mediated vascular integrity in human chronic inflammation.

Confocal immunolocalization of VE-cadherin- and CXC chemokine-expressing endothelial cells in periapical granulomas.

AIM: To determine whether endothelial cells (ECs) in periapical granulomas can express vascular endothelial (VE)-cadherin, CXCL8 and CXCL10 by examining with two-colour confocal laser scanning microscope. METHODOLOGY: Periapical lesions were surgically removed from patients with chronic periapical periodontitis (n = 20), and the paraffin-embedded sections were prepared after being fixed with cold acetone. The 7-mum-thick sections were stained with haematoxylin-eosin and then examined pathologically using a light microscope. The lesions diagnosed as periapical granulomas (17 specimens) were analysed further using immunofluorescence and antibodies specific for human VE-cadherin, CXCL8, and CXCL10. The slides were carefully examined using a confocal laser scanning microscope. The numbers of positive ECs were counted, and the comparison between VE-cadherin-positive ECs and CXCL8 or CXCL10 was assessed statistically using one-way ANOVA followed by a Student-Newman-Keuls test. RESULTS: The expression of CXCL8 and CXCL10 by ECs was detected in 60.4 +/- 13.4 and 67.2 +/- 13.9%, respectively. However, the percentage of VE-cadherin-expressing ECs was 40.4 +/- 10.5%, which was significantly lower (P < 0.01) than CXCL8 and CXCL10-expressing ECs. Two-colour immunofluorescence staining revealed that ECs co-expressed VE-cadherin and CXCL8 (37.4 +/- 14.1%) or CXCL10 (39.1 +/- 13.8%). CONCLUSIONS: VE-cadherin expression in ECs was lower than CXCL8 and CXCL10, suggesting that inflamed ECs in periapical granulomas could increase vascular permeability and that leukocyte chemotaxis mediated by ECs might occur. These findings may suggest the possibility that ECs could play a pivotal role in cell recruitment in periapical granulomas.

Cytokine profiles of T-lymphocytes from gingival tissues with pathological pocketing.

Periodontal disease is an infection in which destruction occurs at sites remote from the infection, resulting in pathological pocketing. Intervening between the infection and the destruction is a dense mononuclear inflammatory infiltrate. It has been suggested that this infiltrate might have characteristics and the destructive potential of Th1-type T lymphocytes. To ascertain the nature of the infiltrates we investigated the expression of mRNA for IL-2, IL-5, and IFN-gamma by gingival mononuclear cells (GMC) from healthy (n = 8) or adult periodontitis (AP) patients (n = 25) by using cytokine-specific reverse-transcription/polymerase-chain-reaction (RT-PCR). GMC, as obtained from patients' tissues, expressed IL-2, IFN-gamma, or IL-5 mRNA. Significantly higher proportions of GMC from AP patients expressed IL-2 and IFN-gamma mRNA than did those from healthy subjects. IFN-gamma was the most consistent cytokine message detected. In other experiments, gingival T-lymphocytes (n = 12) and CD4+ and CD8+ gingival T-lymphocytes (n = 16) were isolated from gingival tissues removed surgically from AP patients. AP gingival T-lymphocytes expressed mRNA for IL-2, IFN-gamma, or IL-6 prior to stimulation. After stimulation with Con A, the cells significantly up-regulated IL-5 and IL-6 message expression. Both CD4+ and CD8+ gingival T-lymphocytes expressed IFN-gamma, IL-5, and some IL-2. This cumulative cytokine profile observed in these experiments is consistent with the predominance of Th1-type cells in pathological tissues and with Th2-type cells, which can also be present, being up-regulated under appropriate stimulation. Importantly, CD4+ and CD8+ lymphocytes were shown to express T1- and T2-type cytokine message, emphasizing the potential for CD8+ T-lymphocytes to participate in periodontal disease pathology.

Production of human-inducible nitric oxide synthase in radicular cysts.

To examine if nitric oxide (NO) is produced in radicular cysts, NO synthase (NOS) production was analyzed. Periapical tissues were removed from patients at the time of endodontic surgery. Frozen tissue sections were histologically evaluated with hematoxylin-eosin staining. Production of human-inducible NOS (iNOS) in apical cysts was then immunohistochemically examined. Immunoreactive human iNOS was widely distributed in epithelial cells, endothelial cells, fibroblasts, macrophages, or polymorphonuclear leukocytes. Remarkably, iNOS-positive cells were significantly present around blood vessels, and cells residing apart from the blood vessels showed weak or no iNOS production, suggesting that only cells around blood vessels could be stimulated for iNOS synthesis. These data demonstrated the possibility that several, but not all, cells could be stimulated to synthesize iNOS in inflamed tissues. In the presence of iNOS, NO can be produced spontaneously in periapical lesions and may play a crucial role in the regulation of chronic infection.

Co-production of vascular endothelial cadherin and inducible nitric oxide synthase by endothelial cells in periapical granuloma.

AIM: To clarify the mechanisms of inflammatory cell migration in human periapical granulomas by examining vascular endothelial (VE) cadherin and inducible nitric oxide synthase (iNOS)-producing cells. METHODOLOGY: Periapical tissues were obtained from patients during endodontic surgery and were divided into two portions. After fixing the tissues with acetone or 4% paraformaldehyde in phosphate-buffered saline, 5-microm-thick paraffin or cryostat sections were prepared, respectively. The paraffin sections of the inflamed tissues were evaluated histologically with haematoxylin-eosin stains. Cryostat sections of the tissue, diagnosed as periapical granulomas, were then examined by either immunohistochemistry using anti-human VE-cadherin or iNOS antibodies (Abs) for the characterization of infiltrating cells. In addition, co-localization of VE-cadherin and iNOS production was also analysed by two-colour immunofluorescence image analysis. RESULTS: Endothelial cells were strongly stained with iNOS Abs. Macrophages, lymphocytes, polymorphonuclear leucocytes and fibroblasts also exhibited iNOS production. These iNOS-positive cells accumulated around the blood vessels. On the other hand, VE-cadherin production was exhibited in only endothelial cells. Two-colour immunofluorescence image analysis using VE-cadherin and iNOS Abs demonstrated that iNOS-producing endothelial cells also showed VE-cadherin production. CONCLUSIONS: Vascular endothelial-cadherin produced by endothelial cells could be regulated by iNOS-producing cells in periapical granulomas and might play a pivotal role in vascular permeability.

Mitogen-activated protein kinase/extracellular signal-regulated protein kinase activation of cultured human dental pulp cells by low-power gallium-aluminium-arsenic laser irradiation.

AIM: To examine whether low-power laser irradiation (LPLI) promotes cellular proliferation of human dental pulp-derived fibroblast-like cells (dental pulp cells). METHODOLOGY: Dental pulp cells were obtained by primary culture of human dental pulp tissues from extracted third molar teeth. The phosphorylation of the mitogen-activated protein kinase (MAPK) family after LPLI of these cells was investigated by Western blotting. By using a specific MAPK/ERK kinase (MEK) inhibitor (PD098059), the specific effect of LPLI on the MAPK pathway was also investigated by Western blotting as described above. The incorporation of [3H]thymidine into the cells after LPLI was determined, and statistical analysis was performed by Wilcoxon signed-ranks test. RESULTS: Extracellular signal-regulated protein kinase (ERK) 1/2 was phosphorylated between 5 and 30 min after LPLI. Moreover, PD098059 inhibited LPLI-mediated ERK1/2 activation. LPLI did not affect p38 MAPK or c-Jun N-terminal kinase (JNK) phosphorylation. But LPLI did not stimulate [3H]thymidine incorporation into these cells. CONCLUSIONS: These results indicated that LPLI activated MAPK/ERK, a signal for proliferation, differentiation and survival, but did not activate the stress signals p38 MAPK and JNK in human dental pulp cells.

Effect of mineral trioxide aggregate on proliferation of cultured human dental pulp cells.

AIM: To investigate the effect of mineral trioxide aggregate (MTA) on the proliferation of human dental pulp (HDP) cells ex-vivo. METHODOLOGY: Human dental pulp cells were cultured with MTA or calcium hydroxide-containing cement (Dycal) using culture plate inserts. Control cells were cultured with culture plate inserts only. Cell proliferation was measured for up to 14 days using a Cell Counting kit, and the concentration of calcium ions released from the tested materials was assessed using a Calcium E-test kit. To confirm that the effect of MTA was attributable to released calcium ions, cell proliferation was measured in the presence of exogenous calcium chloride as a source of calcium ions while in the absence of MTA. RESULTS: Mineral trioxide aggregate significantly stimulated cell proliferation after 12 days, whereas Dycal had no such effect. The number of calcium ions released from MTA was significantly higher than that released from Dycal. Following the addition of calcium chloride, cell proliferation increased in a dose-dependent manner after 12 days. Moreover, cell proliferation showed a similar pattern whether a given concentration of calcium ions was produced by calcium chloride or by release from MTA. CONCLUSIONS: In this ex-vivo study, the elution components such as calcium ions from MTA had higher proliferation ability of HDP cells than control and Dycal.

Behaviour of bone marrow osteoblast-like cells on mineral trioxide aggregate: morphology and expression of type I collagen and bone-related protein mRNAs.

AIM: To investigate the in vitro behaviour of rat bone marrow cells (RBM) on mineral trioxide aggregate (MTA) (ProRoot, MTA Root Canal Repair Material; Dentsply Tulsa, Tulsa, OK, USA) compared with intermediate restorative materials (IRM) (Dentsply Caulk, Milford, DE, USA). METHODOLOGY: RBM were obtained from rat femur and were primary cultured and then subcultured. Cells were then seeded on three dishes of each material, and cultured for 3 days, after which they were evaluated morphologically using scanning (SEM) and transmission (TEM) electron microscopy. Furthermore, the calcium released from hydrated material, the cell proliferation ratio and alkaline phosphatase (ALP) activity were analysed, and the expression of type I collagen and bone-related protein mRNAs were evaluated. The data were averaged and analysed via one-way analysis of variance (anova) and were then compared by the Scheffe's test. RESULTS: SEM showed that RBM attached to MTA and had a flattened appearance without nuclear protrusions and microspikes. TEM showed that the cells attached in the same manner as the control group, but gaps larger than 2 microm were frequently seen. The calcium released from hydrated MTA was about 130 ppm after 3 days of immersion in saline. The ALP activity was similar to the control group. Cell proliferation and expression of type I collagen mRNA was significantly lower, while the expression of osteopontin mRNA was significantly higher than the control group at the third day of culture. In IRM groups, a few rounded cells were observed on the material but no living cells were seen. CONCLUSIONS: MTA is a material of low toxicity which does not inhibit cell growth, but does suppress the differentiation of osteoblast-like cells.

Non-surgical root canal treatment of dens invaginatus type 2 in a maxillary lateral incisor.

AIM: To describe the clinical management of an unusual dens invaginatus type 2. SUMMARY: A case of dens invaginatus in a maxillary lateral incisor with a periapical lesion is reported. The patient presented with pain and localized swelling. Despite the complex anatomy and diagnosis of dens invaginatus, non-surgical root canal treatment was performed successfully. KEY LEARNING POINTS: * Dens invaginatus may be presented in many forms, and the aetiology of this phenomenon is not fully understood. * Due to abnormal anatomical configuration, dens invaginatus presents technical difficulties in its clinical management. * Non-surgical root canal treatment can be performed successfully.

Non-surgical root canal treatment of dens invaginatus type 2 in a maxillary lateral incisor.

AIM: To describe the clinical management of an unusual dens invaginatus type 2. SUMMARY: A case of dens invaginatus in a maxillary lateral incisor with a periapical lesion is reported. The patient presented with pain and localized swelling. Despite the complex anatomy and diagnosis of dens invaginatus, non-surgical root canal treatment was performed successfully. Key learning points Dens invaginatus may be presented in many forms, and the aetiology of this phenomenon is not fully understood. Due to abnormal anatomical configuration, dens invaginatus presents technical difficulties in its clinical management. Non-surgical root canal treatment can be performed successfully.

Human polymorphonuclear leukocytes derived from chronically inflamed tissue express inflammatory cytokines in vivo.

In this study, we sought to determine if human polymorphonuclear leukocytes (PMNs) derived from chronically inflamed tissues can produce inflammatory cytokines in vivo. Human gingival crevicular fluid (GCF) with adult periodontitis was collected, and PMNs in GCF were examined after purified by Ficoll-Hypaque gradient method. Cytokines from peripheral blood (PB) cells stimulated with concanavalin A, LPS, or zymosan were also characterized, since GCF contains predominantly PMNs (> 95%) with a small number of lymphocytes or macrophages. Production of interleukin-1 alpha (IL-1 alpha), IL-1 beta, tumor necrosis factor-alpha (TNF alpha), and IL-6 in GCF or culture supernatants of peripheral blood cells was determined by ELISA. Significant levels of IL-1 alpha and IL-1 beta secretion were found in GCF. PB cells in culture showed prominent cytokine production from monocytes/macrophages, followed by lymphocytes. Human peripheral blood PMNs (PB-PMNs) also produced low levels of IL-1 alpha and IL-1 beta, but not TNF alpha and IL-6. These cells were also examined for cytokine mRNA expression using the reverse transcription-polymerase chain reaction analysis. Highly purified PMNs (> 99.5%) from GCF expressed mRNA for IL-1 alpha, IL-1 beta and TNF alpha, but not for IL-6. PB-PMNs in culture also showed mRNA expression for IL-1 alpha, IL-1 beta, and TNF alpha in a time- and dose-dependent manner, especially after stimulation with zymosan. Therefore, we concluded that human PMNs from inflamed tissues can produce IL-1 alpha, IL-1 beta, and TNF alpha in vivo, but not IL-6.

Expression of inflammatory cytokine genes in vivo by human alveolar bone-derived polymorphonuclear leukocytes isolated from chronically inflamed sites of bone resorption.

Alveolar bone-derived polymorphonuclear leukocytes (PMNs) were characterized for their ability to produce inflammatory cytokines such as interleukin-1 alpha (IL-1 alpha), IL-1 beta, tumor necrosis factor alpha (TNF alpha), and IL-6 in vivo. Periapical exudates (PE) were collected from periapical lesions with chronic periapical periodontitis through root canals. Cells and noncellular supernatants were then isolated by centrifugation. The concentration of cytokines present in the noncellular supernatants were determined by ELISA. High concentrations of IL-1 alpha, IL-1 beta, and IL-6 were detected in PE, however, TNF alpha was not. PE contains predominantly PMNs ( > 95% of residing cells) with a few percent of lymphocytes and/or macrophages. These alveolar bone-derived PMNs were purified by the Ficoll-Hypaque gradient method and were analyzed for cytokine mRNA expression using the cytokine-specific reverse-transcription polymerase chain reaction. Highly purified PMNs ( > 99.5%) isolated from PE expressed significant levels of mRNA for IL-alpha, IL-1 beta, and TNF alpha. IL-6 mRNA was not detected, although a high concentration of IL-6 was detected in supernatants of PE by ELISA. The IL-6 secretion in PE could be derived from macrophages, T lymphocytes, osteoblasts, or fibroblasts around periapical lesions. These data strongly suggest that human PMNs derived from alveolar bone can spontaneously produce IL-1 alpha, IL-1 beta, and TNF alpha at sites of inflammation, and probably initiate inflammation and regulate augmentation of bone resorption in vivo.

Histochemical studies of monoamine oxidase in experimental myocardial infarctions.

Sequential histochemical studies were conducted to determine the level of monoamine oxidase (MAO) activity in the infarcted tissue of the experimental myocardial infarction in dog. MAO activity was not found in the normal heart muscle, but the activities were present in the wall of the coronary artery histochemically. Fibroblasts and collagen fibers began to take the place of the destroyed heart muscles in the infarcted area from 10 days after the coronary occlusion in dogs, and MAO activities were noted sporadically in the fibroblasts and the interstices of the collagen fibers in the infarcted area. MAO activities increased histochemically in the fibroblasts and the fiber interstices in the infarcted area 10 days or more after the coronary occlusion. These findings suggested the presence of the active collagen metabolism outside the myocardial cells in the infarcted area in the recovery stage of the experimental myocardial infarction. It was also suggested that the interstices of the collagen fibers in the myocardial wall constituted the lymphatic ducts outside the blood vessels and that the MAO activity in serum determined by the method in which tryptamine hydrochloride was used as substrate might indicate the grade of fibrosis of the myocardial tissue in the infarcted areas.

Inducible nitric oxide synthase activity by interferon-gamma-producing cells in human radicular cysts.

AIM: In this study, the interaction of interferon-gamma-(IFN-gamma) and inducible nitric oxide synthase (iNOS)-producing cells in human radicular cysts were investigated. METHODOLOGY: Inflamed periapical tissues were obtained from patients at the time of endodontic surgical treatments and were cut into two pieces. After fixing with acetone or 4% paraformaldehyde in phosphate-buffered saline, 5-m-thick paraffin and cryostat sections were prepared. The paraffin sections of the inflamed tissues were evaluated histologically with haematoxylineosin stains. The specimens diagnosed as radicular cysts were then examined by immunostaining. Immunohistochemistry for iNOS and fluoresence microscopy for IFN-gamma using the cryostat sections were performed with a mixture of affinity purified human iNOS antiserum and human IFN-gamma monoclonal antibodies. RESULTS: The results revealed that iNOS-gamma producing cells localized adjacent to IFN-gamma-producing cells. In addition, some of iNOS-producing cells exhibited immunoreactive IFN-gamma. On the other hand, epithelial cells showed significant levels of iNOS production, but not IFN-gamma. CONCLUSIONS: The data would suggest the possibility that iNOS production could be precisely controlled by autocrine or paracrine effects of IFN-gamma producing cells in radicular cysts and might play a pivotal role in periapical lesions. These findings are consistent with a hypothesis suggesting that NO inhibitors could be used through the root canals as a pharmacological treatment for periapical lesions.

Cytokine regulation on the synthesis of nitric oxide in vivo by chronically infected human polymorphonuclear leucocytes.

To determine if nitric oxide (NO) is produced by chronically infected human polymorphonuclear leucocytes (PMNs) in vivo, inflamed exudates (periapical exudates: PE) collected from periapical periodontitis patients were examined. Cell-free supernatants and cells were separated by centrifugation. Significant levels of nitrite concentrations were observed in the supernatants. The production of inducible NO synthase (iNOS) in highly purified PMNs derived from PEs was then immunocytochemically determined using rabbit anti-human iNOS antiserum. In vitro, human peripheral blood PMNs (PB-PMNs) isolated from patients were cultured with a combination of Esherichia coli-lipopolysaccharide (LPS), recombinant human interferon-gamma (rhIFN-gamma) and/or interleukin-1 beta (rhIL-1 beta). The stimulated PB-PMNs showed steady-state levels of nitrite. The stimulation of LPS, rhIFN-gamma and rhIL-1 beta showed more NO induction than that of LPS with either IFN-gamma or IL-1 beta, suggesting the synergistic effects of cytokines. Cryostat sections of surgically removed periapical tissues were also immunohistochemically examined for iNOS, IFN-gamma and IL-1 beta. Two-colour immunohistochemistry revealed the interaction of iNOS-producing PMNs and IFN-gamma- or IL-1 beta-producing mononuclear cells. On the basis of these data, we concluded that with the stimulation of inflammatory cytokines derived from mononuclear cells, PMNs can spontaneously produce NO at the site of chronic infection. The present studies are consistent with a hypothesis suggesting that PMNs could be regulated and delicately balanced to produce NO by mononuclear cell-derived cytokines in vivo. NO-producing cells may play a pivotal role in chronic inflammation.