Ikuro Maruyama

Aquaporin 3 (AQP3) is the predominant water channel protein in human keratinocytes and acts as an inflammatory mediator in some lesions. A chronic, inflammatory process of periodontitis is related with a dramatic change of surrounding fluid homeostasis to plasma extravasation. The exact pattern of aquaporin (AQP) water channel expression and its mechanism in periodontal disease is still unknown. We describe herein an up‐regulated AQP3 expression in the epithelial lesion with chronic periodontitis and its functional role. The levels of AQP3 expression in inflamed gingival epithelial tissues were significantly higher than those of healthy subjects. Consistent with these results, AQP3 expression (i.e., levels of mRNA and protein) in cultured rat primary gingival epithelial cells and the human gingival epithelial cell line Ca9‐22 were strongly increased in response to TNF‐α treatment through the 55 kDa TNF‐α receptor (TNFR I). In this context, small interfering RNA‐ (siRNA)‐mediated “aqp‐3gene silencing,” which could reduce AQP3 expression by more than 65%, significantly attenuated selected proinflammatory events of ICAM‐1 expression induced by TNF‐α in Ca9‐22. A sixfold increase in leukocyte adherence to TNF‐α‐stimulated epithelial cells was demonstrated by an adherence assay (P < 0.001) and pretreatment with AQP3 siRNA and anti‐ICAM‐1 antibody reduced leukocyte retention by 85% (P < 0.001). Our study indicates for the first time a novel important mode in the regulation of the inflammatory response through TNF‐α/TNFR I ligation at the site of epithelial lesions by specialized membrane channel AQP3 and ICAM‐1 protein, which is closely implicated in the development of periodontitis mechanisms. J. Cell. Physiol. 217: 338–349, 2008.

Abstract:  High mobility group box‐1 (HMGB1) protein, primarily from the nucleus, is released into the extracellular milieu either passively by necrotic or damaged cells, or actively by secretion from monocytes/macrophages. Extracellular HMGB1 acts as a potent inflammatory stimulator by promoting cytokine (for example, tumor necrosis factor‐α) production, and also has pro‐coagulant activity. The signaling pathway initiated by receptor for advanced glycation end‐product (RAGE), which is the HMGB1 receptor, also induces complement activation. Recent studies have implicated HMGB1 in acute cardiac allograft rejection, and have identified infiltrating T cells and other damaged cells as its main sources. HMGB1 blockade using the anti‐HMGB1 antibody HMGB1 box‐A (amino‐terminal region) and soluble RAGE rescues mice from acute rejection. We therefore studied the release of HMGB1 in co‐cultures of porcine aortic endothelial cells (PAEC) and human leukocytes. Human T cells, but not B cells, monocytes or neutrophils, stimulated significant HMGB1 release in culture with PAEC; this activity required cell–cell contact and was dose‐dependent, as determined by Western blotting. The released HMGB1 originated from both cell types, as immunofluorescent microscopy showed that it was present in the cytosol of PAEC in contact with T cells, and had disappeared from the T‐cell nuclei. These results demonstrate that direct interactions between PAEC and T cells might be a key factor in triggering HMGB1 release, which suggests that HMGB1 is associated with graft rejection in the early phase.

We reported earlier that hydroxyapatite (HA) formed on/in agarose gels (HA/agarose) produced by alternate soaking process is a bone‐filling material possessing osteoconductive and hemostatic effects. This process could allow us to make bone‐like apatite that was formed on/in organic polymer hydrogel matrices. Here, we investigated the mechanism of hemostasis induced by HA/agarose and found that HA/agarose, but not agarose or HA powder, significantly shortened activated partial thromboplastin time (APTT). While HA/agarose did not show significant platelet aggregation, it markedly enhanced adenosine diphosphate (ADP)‐induced platelet aggregation. Moreover, Western blot analysis revealed selective adsorption of vitronectin onto HA/agarose. We also observed marked differences between HA powder and HA/agarose in their XRD patterns. The crystallinity of HA powder was much higher compared to that of HA/agarose. Furthermore, 50–100 nm of tube‐form aggregations was observed in HA powder on the other hand 100–200 nm of particles was observed in HA/agarose by SEM observation. Thus 100–200 nm of low crystallized particles on the surface structure of HA/agarose may play an important role in hemostasis. Our results demonstrated a crucial role of HA/agarose in the mechanism of hemostasis and suggested a potential role for HA/agarose as a bone‐grafting material. © 2006 Wiley Periodicals, Inc. J Biomed Mater Res Part B: Appl Biomater, 2006

The endogenous cannabinoid anandamide, a lipid mediator, induces various physiologic events such as vascular relaxation, inhibition of gap–junctions formation, tumor proliferation, neurologic analgesia, and apoptosis. Although increased concentration of anandamide in plasma has been implicated in pathophysiologic states including endotoxin–induced hypotension, the effects of anandamide on hepatocytes still remain unclear. In this study, we present evidence that plasma anandamide concentration is highly increased in severe hepatitis and cirrhosis patients. In addition, concentrations of anandamide within the pathophysiologic range potently induced apoptosis of hepatoma cell line (Hep G2) and primary hepatocytes, suggesting a possible link between increased anandamide level and hepatocyte damage. Anandamide–induced cell death was preceded by G0/G1 cell–cycle arrest, activation of proapoptotic signaling (i.e., p38 MAPK and JNK), and inhibition of antiapoptotic signaling (i.e., PKB/Akt) pathways. Moreover, anandamide increased susceptibility to oxidative stress–induced hepatocyte damage. In this context, methyl–β–cyclodextrin (MCD), a membrane cholesterol depletor, or mevastatin, an HMG–CoA reductase inhibitor, or N–acetyl cysteine, an antioxidant, potently inhibited the anandamide–induced proapoptotic events and cell death, whereas putative cannabinoid receptor antagonists did not exhibit an inhibitory effect on anandamide–induced cell death. Furthermore, binding assay using polymyxin beads revealed that anandamide could interact with cholesterol. In conclusion, our data suggest that cholesterol present in the cell membrane determines the fate of hepatocytes exposed to anandamide, possibly functioning as an anandamide receptor.

Anandamide is a neuroimmunoregulatory molecule that triggers apoptosis in a number of cell types including PC12 cells. Here, we investigated the molecular mechanisms underlying anandamide‐induced cell death in PC12 cells. Anandamide treatment resulted in the activation of p38 mitogen‐activated protein kinase (MAPK), c‐Jun N‐terminal kinase (JNK), and p44/42 MAPK in apoptosing cells. A selective p38 MAPK inhibitor, SB203580, or dn‐JNK, JNK1(A‐F) or SAPKβ(K‐R), blocked anandamide‐induced cell death, whereas a specific inhibitor of MEK‐1/2, U0126, had no effect, indicating that activation of p38 MAPK and JNK is critical in anandamide‐induced cell death. An important role for apoptosis signal‐regulating kinase 1 (ASK1) in this event was also demonstrated by the inhibition of p38 MAPK/JNK activation and death in cells overexpressing dn‐ASK1, ASK1 (K709M). Conversely, the constitutively active ASK1, ASK1ΔN, caused prolonged p38 MAPK/JNK activation and increased cell death. These indicate that ASK1 mediates anandamide‐induced cell death via p38 MAPK and JNK activation. Here, we also found that activation of p38 MAPK/JNK is accompanied by cytochromecrelease from the mitochondria and caspase activation (which can be inhibited by SB203580), suggesting that anandamide triggers a mitochondrial dependent apoptotic pathway. The caspase inhibitor, zVAD, and the mitochondrial pore opening inhibitor, cyclosporine A, blocked anandamide‐induced cell death but not p38 MAPK/JNK activation, suggesting that activation of these kinases may occur upstream of mitochondrial associated events.