Akihito Ishigami

Abnormal accumulation of citrullinated proteins catalyzed by peptidylarginine deiminase in hippocampal extracts from patients with Alzheimer's disease

Citrullinated proteins are the products of a posttranslational process in which arginine residues undergo modification into citrulline residues when catalyzed by peptidylarginine deiminases (PADs) in a calcium ion‐dependent manner. In our previous report, PAD2 expressed mainly in the rat cerebrum became activated early in the neurodegenerative process. To elucidate the involvement of protein citrullination in human neuronal degeneration, we examined whether citrullinated proteins are produced during Alzheimers disease (AD). By Western blot analysis with antimodified citrulline antibody, citrullinated proteins of varied molecular weights were detected in hippocampal tissues from patients with AD but not normal humans. Two of the citrullinated proteins were identified as vimentin and glial fibrillary acidic protein (GFAP) by using two‐dimensional gel electrophoresis and MALDI‐TOF mass spectrometry. Interestingly, PAD2 was detected in hippocampal extracts from AD and normal brains, but the amount of PAD2 in the AD tissue was markedly greater. Histochemical analysis revealed citrullinated proteins throughout the hippocampus, especially in the dentate gyrus and stratum radiatum of CA1 and CA2 areas. However, no citrullinated proteins were detected in the normal hippocampus. PAD2 immunoreactivity was also ubiquitous throughout both the AD and the normal hippocampal areas. PAD2 enrichment coincided well with citrullinated protein positivity. Double immunofluorescence staining revealed that citrullinated protein‐ and PAD2‐positive cells also coincided with GFAP‐positive cells, but not all GFAP‐positive cells were positive for PAD2. As with GFAP, which is an astrocyte‐specific marker protein, PAD2 is distributed mainly in astrocytes. These collective results, the abnormal accumulation of citrullinated proteins and abnormal activation of PAD2 in hippocampi of patients with AD, strongly suggest that PAD has an important role in the onset and progression of AD and that citrullinated proteins may become a useful marker for human neurodegenerative diseases.

Senescence marker protein-30 knockout mouse liver is highly susceptible to tumor necrosis factor-α- and fas-mediated apoptosis

Senescence marker protein-30 (SMP30) is a calcium-binding protein that decreases in an androgen-independent manner with aging. To elucidate the physiological role of this protein, we introduced a null mutation of the SMP30 gene into the germ line of mice. Despite the complete lack of SMP30 (SMP30-/-), these mutant mice were indistinguishable from their wild-type (SMP30+/+) littermates in terms of development and fertilization capability. We then investigated the tissue susceptibility for apoptosis induced by cytokine using primary cultured hepatocytes, because SMP30 could rescue cells from cell death caused by calcium influx, using a calcium ionophore as previously described. SMP30-/- hepatocytes were found to be more susceptible to apoptosis induced by tumor necrosis factor-alpha (TNF-alpha) plus actinomycin D (ActD) than SMP30+/+ hepatocytes. In addition, the TNF-alpha/ActD-induced caspase-8 activity in SMP30-/- hepatocytes was twofold greater than that in SMP30+/+ hepatocytes. In contrast, no significant difference was observed in the TNF-alpha/ActD-induced nuclear factor-kappa B activation of SMP30+/+ versus SMP30-/- hepatocytes, indicating that SMP30 is not related to TNF-alpha/ActD-induced nuclear factor-kappa B activation itself. Moreover, deletion of the SMP30 gene enhanced liver injury after treatment in vivo with anti-Fas antibody and the SMP30+/- mice showed intermediate susceptibility to Fas-induced apoptosis. Collectively, these results demonstrate that SMP30 acts to protect cells from apoptosis.

Hydrogen-rich pure water prevents superoxide formation in brain slices of vitamin C-depleted SMP30/GNL knockout mice.

Hydrogen is an established anti-oxidant that prevents acute oxidative stress. To clarify the mechanism of hydrogens effect in the brain, we administered hydrogen-rich pure water (H(2)) to senescence marker protein-30 (SMP30)/gluconolactonase (GNL) knockout (KO) mice, which cannot synthesize vitamin C (VC), also a well-known anti-oxidant. These KO mice were divided into three groups; recipients of H(2), VC, or pure water (H(2)O), administered for 33 days. VC levels in H(2) and H(2)O groups were <6% of those in the VC group. Subsequently, superoxide formation during hypoxia-reoxygenation treatment of brain slices from these groups was estimated by a real-time biography imaging system, which models living brain tissues, with Lucigenin used as chemiluminescence probe for superoxide. A significant 27.2% less superoxide formed in the H(2) group subjected to ischemia-reperfusion than in the H(2)O group. Thus hydrogen-rich pure water acts as an anti-oxidant in the brain slices and prevents superoxide formation.

Immunocytochemical localization of peptidylarginine deiminase in human eosinophils and neutrophils

Peptidylarginine deiminase, registered as PAD V in the DDBJ/GenBank/EMBL data banks, is expressed in HL‐60 cells differentiated into granulocytes or monocytes. We analyzed PAD activities in density‐fractionated human peripheral blood cell fractions. PAD activity with similar substrate specificity to that of PAD V was found in the eosinophil and neutrophil fractions, which showed single bands comigrating with authentic PAD V on immunoblotting with an anti‐PAD V antibody. Both the biochemical and immunoblotting analyses showed marked enrichment of PAD V in the eosinophil fraction. Its immunoreactivity appeared to localize in eosinophilic granules at high density and in myeloperoxidase‐negative cytoplasmic granules of neutrophils at low density, as determined by confocal laser‐scanning microscopy. Possible roles of PAD V in myeloid differentiation and granulocyte function are discussed. In addition, we present evidence for the presence of PAD(s) that are antigenically different from PAD V in monocytes and lymphocytes.

cDNA cloning, gene organization and expression analysis of human peptidylarginine deiminase type I

Peptidylarginine deiminases (PADs) catalyse a post-translational modification of proteins through the conversion of arginine residues into citrullines. The existence of four isoforms of PAD (types I, II, III and IV) encoded by four different genes, which are distinct in their substrate specificities and tissue-specific expression, was reported in rodents. In the present study, starting from epidermis polyadenylated RNA, we cloned by reverse transcriptase-PCR a full-length cDNA encoding human PAD type I. The cDNA was 2711 bp in length and encoded a 663-amino-acid sequence. The predicted protein shares 75% identity with the rat PAD type I sequence, but displays only 50-57% identity with the three other known human isoforms. We have described the organization of the human PAD type I gene on chromosome 1p36. A recombinant PAD type I was produced in Escherichia coli and shown to be enzymically active. Human PAD type I mRNAs were detected by reverse transcriptase-PCR not only in the epidermis, but also in various organs, including prostate, testis, placenta, spleen and thymus. In human epidermis extracts analysed by Western blotting, PAD type I was detected as a 70 kDa polypeptide, in agreement with its predicted molecular mass. As shown by immunohistochemistry, the enzyme was expressed in all the living layers of human epidermis, with the labelling being increased in the granular layer. This is the first description of the human PAD type I gene and the first demonstration of its expression in epidermis.

Human peptidylarginine deiminase type II : molecular cloning, gene organization, and expression in human skin

Peptidylarginine deiminases (PADs) are posttranslational modification enzymes that convert protein arginine to citrulline residues in a calcium ion-dependent manner. Rodents have four isoforms of PAD (types I, II, III, and IV), each of which is distinct in substrate and tissue specificity. In fact, the only tissue in which all four PAD mRNAs have been detected is the epidermis. In this study, we found PAD activity in HSC-1 human cutaneous squamous carcinoma cells in vitro, and this activity increased during cultivation. Using a homology-based strategy, we cloned a full-length cDNA encoding human PAD type II. The cDNA was 2348 bp long and encoded a 665-amino-acid sequence with a predicted molecular mass of 75 kDa. The predicted protein shared 93% identity with the rat and mouse PAD type II sequence. Alignment of the amino acid sequences from both species revealed notable conservation in the C-terminal region, suggesting the presence of a functional region such as an enzyme catalytic site and/or a calcium-binding domain. Gene organization analysis established that human PAD type II on chromosome 1p35.2-p35.21 spanned more than 50 kb and contained 16 exons and 15 introns. A recombinant PAD protein subsequently produced in Escherichia coli proved to be enzymatically active, with substrate specificities similar to those of the rat PAD type II. In an immunohistochemical study of human skin, the type II enzyme was expressed by all the living epidermal layers, suggesting that PAD type II is functionally important during terminal differentiation of epidermal keratinocytes.

SMP30 deficiency causes increased oxidative stress in brain

Senescence marker protein 30 (SMP30), an important aging marker molecule, has been identified functionally as a calcium regulatory protein. Recent evidence showed its new assumed role as an effective anti-oxidative property. However, the role of SMP30 in the brain has not been explored. To delineate its role in the brain, we utilized SMP30 knock-out (SMP30 KO) mice in the current study. We focused on the oxidative status of the brain by examining selected oxidative markers in brains of SMP30 KO mice. Results showed that the generation of reactive species (RS) and NADPH oxidase activities were significantly elevated in SMP30 deficient brain. The increased oxidative status in these mice was further confirmed by increased oxidatively modified proteins such as dityrosine formation and carbonylation in the cortex of SMP30 KO mice. Moreover, SMP30 deficient brain showed the increased Mac-1 protein and myeloperoxidase (MPO) activity in the brain, supporting the putative anti-oxidative action of SMP30. Interestingly, the activities of major antioxidant enzymes, superoxide dismutase, catalase and reduced glutathione peroxidase in the brain were not affected by SMP30 depletion. Our results documented that brain SMP30 has a protective action against oxidative damage, without influencing antioxidant enzyme status.

Vitamin C depletion increases superoxide generation in brains of SMP30/GNL knockout mice.

Vitamin C (VC) has a strong antioxidant function evident as its ability to scavenge superoxide radicals in vitro. We verified that this property actually exists in vivo by using a real-time imaging system in which Lucigenin is the chemiluminescent probe for detecting superoxide in senescence marker protein-30 (SMP30)/gluconolactonase (GNL) knockout (KO) mice, which cannot synthesize VC in vivo. SMP30/GNL KO mice were given 1.5 g/L VC [VC(+)] for 2, 4, or 8 weeks or denied VC [VC(-)]. At 4 and 8 weeks, VC levels in brains from VC(-) KO mice were <6% of that in VC(+) KO mice. Accordingly, superoxide-dependent chemiluminescence levels determined by ischemia-reperfusion at the 4- and 8 weeks test intervals were 3.0-fold and 2.1-fold higher, respectively, in VC(-) KO mice than in VC(+) KO mice. However, total superoxide dismutase activity and protein levels were not altered. Thus, VC depletion specifically increased superoxide generation in a model of the living brain.

Protein deimination in the rat brain after kainate administration : citrulline-containing proteins as a novel marker of neurodegeneration

Peptidylarginine deiminases (PADs) are a group of enzymes that convert protein arginine residues to citrulline residues in a Ca2+-dependent manner. In the central nervous system, PAD type II localizes in glial cells, but its biological role is little understood. We examined the timing and region dependence of protein deimination in the rat cerebrum after a systemic injection of kainic acid (KA). Citrulline-containing proteins were consistently found in neurodegenerating regions. Western blot analyses showed deimination of numerous proteins in a broad-molecular-weight range. By immunocytochemical scrutiny, deiminated protein-positive astrocytes were found at 2 h after KA administration, and they increased in number until the 6 h. Furthermore, shrunken neurons became deiminated protein-positive at 12-24 h. These data suggest that PAD type II becomes activated in regions undergoing neurodegeneration and functions to deiminate various proteins. Therefore, citrulline-containing proteins seem to be a useful marker of acute neurodegeneration.

Senescence marker protein-30 knockout mouse as a novel murine model of senile lung

Senescence marker protein‐30 (SMP30) was originally identified as a novel protein of which expression decreases in an androgen‐independent manner with aging in the rat liver and functions to protect cells from apoptosis. By reverse transcription–polymerase chain reaction analysis, SMP30 mRNA transcripts were found in the mouse lung, liver, kidney, testis and cerebrum. We examined SMP30 expression in the mouse liver, kidney and lung during aging and a distinct temporal profile of SMP30 expression was found in each tissue; the SMP30 mRNA level peaked at 1–3 months of age and decreased thereafter in the liver (the highest at 1 month of age followed by a rapid decline and consistently low thereafter in the kidney), and peaked at 12 months of age in the lung. To investigate the physiological role of SMP30 in the lung, immunohistochemical studies of wild‐type (SMP30Y/+) mice and histopathological examinations of SMP30 knockout (SMP30Y/–) mice were performed. Immunoreactivity against anti‐SMP30 antibody was mainly detected in bronchial epithelial cells and strongly detected at 6–12 months of age. Morphometric analysis was performed to measure the mean linear intercept and destructive index, and found peripheral airspace enlargement without alveolar destruction in SMP30Y/– mice at 1, 3 and 6 months of age compared with the SMP30Y/+ mice. Our results strongly suggest that SMP30Y/– mice could be a novel model for a senile lung and further examinations of SMP30Y/– mice may offer clues to elucidate the mechanisms of the development of pulmonary diseases in the elderly.