Novel human plasma proteins, IHRP (acute phase protein) and PHBP (serine protease), which bind to glycosaminoglycans.

A novel human plasma protein was found in the eluate from the dextran sulfate column, which was used for the treatment of the patients with hypercholesteremia to reduce plasma low density lipoprotein. The results of sequence analysis revealed that this protein was a homologue of heavy chains of inter-alpha-trypsin inhibitor (ITI) family, and it was termed IHRP (ITI family heavy chain related protein). IHRP was identified as an acute-phase protein in animals, and slightly increased concentrations in human plasmas were observed in the patients with inflammatory disorders. IHRP bound to actin and inhibited its polymerization, and IHRP suppressed the phagocytosis and chemotaxis of polymorphonuclear cells. These results suggest that IHRP may function as an anti-inflammatory protein. Plasma hyaluronan binding protein (PHBP) is a novel serine protease which was also found in human plasma. It is consisted of three epidermal growth factor domains, one kringle domain and one serine protease domain from its amino terminus. The amino acid sequence of PHBP is homologous to that of hepatocyte growth factor activator. Purified 75-kDa single chain pro-form of PHBP was auto-activated (auto-cleaved) to 50-kDa heavy chain and 25-kDa light chain, both of them are bridged by a disulfied bond. PHBP digested alpha-chain and beta-chain of fibrinogen to prevent coagulation and cleaved single chain urokinase type plasminogen activator (scuPA) to the active hetero dimer form (tcuPA). The auto-activation of PHBP was accelerated in the presence of dextran sulfate or phosphatidylethanolamine as well as factor XII of the coagulation system. C1 inhibitor of the complement system was identified as the main inhibitor of PHBP in human plasma. Partial hepatectomy and administration of carbon tetrachloride or galactosamine caused the conversion of pro-PHBP to the active form in mouse but administrations of turpentine and mercury chloride did not, suggesting the hepatic injury specific activation of PHBP. These results indicate that PHBP participates not only in the fibrinolytic system but also in the degradation cascade of extracellular matrix (ECM), i.e., PHBP activates scuPA to tcuPA, tcuPA activates matrix metalloproteases (MMPs) and activated MMPs degrade ECM for the tissue remodeling after hepatic injury.

Changes in the distribution pattern of gelatin-binding protein of 28 kDa (adiponectin) in myocardial remodelling after ischaemic injury.

AIMS: Gelatin-binding protein of 28 kDa (GBP28) is a collagen-like plasma protein having a binding capacity with collagens. We investigated GBP28 role on myocardial remodelling as well as the diagnostic significance of GBP28 immunostaining in myocardial infarction. METHODS AND RESULTS: Myocardial tissues obtained from 47 autopsied hearts with infarction were immunostained with antibodies against GBP28, fibronectin, type III and IV collagens, and prolyl 4-hydroxylase. GBP28 was distributed in interstitium of infarcted lesions at an early stage. GBP28 was linearly found both along the border with vital myocardium and at the periphery of surviving cardiomyocyte around the lesion at granulative stage. However, it was not observed in the scar. GBP28 distribution patterns were similar to those of fibronectin in infarcted lesions and those of type IV collagen at the periphery of cardiomyocyte. Type III collagen was not recognized in the early-stage lesion but increased along with scar maturation. Prolyl 4-hydroxylase was found in surviving cardiomyocytes around the lesion during all stages and in interstitial cells appeared in granulation tissue. CONCLUSION: GBP28 plays a role as a scaffold of newly formed collagen in myocardial remodelling after ischaemic injury, and GBP28 immunostaining may assist in a diagnosis of healing stage.

Hepatic injury-specific conversion of mouse plasma hyaluronan binding protein to the active hetero-dimer form.

Plasma hyaluronan binding protein (PHBP) is produced only in liver and kidney in mouse. The induction of PHBP mRNA and the conversion of pro PHBP to the active hetero-dimer form were studied after CCl4, D-galactosamine, HgCl2 or turpentine administration and after partial hepatectomy. The results indicated that the administrations of CCl4 and D-galactosamine, which caused hepatic failure, and the partial hepatectomy enhanced the conversion of pro PHBP to the active two-chain form in the plasma. On the other hand, HgCl2 which injured kidney and turpentine which led to inflammation were not involved in the activation of PHBP. The weak induction and suppression of PHBP mRNA were observed in the liver at 3 h and 12 h, respectively, after the CCl4 administration. However, HgCl2 and turpentine did not influence the amount of PHBP mRNA. These results suggested the hepatic injury-specific activation of PHBP in plasma. PHBP may act as an early factor in the cascade for the tissue remodeling in liver following hepatic injury, i.e., PHBP activates urokinase, urokinase activates matrix metalloproteinases (MMPs) and MMPs degrade extracellular matrix for liver regeneration.

Immunolocalization of apolipoproteins in aortic atherosclerosis in American youths and young adults: findings from the PDAY study.

The immunohistochemical distribution of apolipoproteins in the abdominal aortas of 142 men, 15-34 years of age, collected in a cooperative multicenter study group (Pathobiological Determinants of Atherosclerosis in Youth) was examined in relationship to serum VLDL+LDL+HDL cholesterol levels. ApoB deposits were limited to the intima of specimens with intimal fibro cellular thickening or atherosclerotic lesions. Apo A-I, E and J were observed in both the intima and media of the aortas with intimal lesions. The pattern of apoJ distribution was similar to that of apoA-I and E. The distribution patterns of these apolipoproteins in these young adults were very similar to those in adults and old men seen in an earlier study. The extent of apolipoprotein distribution in the intima and media increased with age and the stage of atherosclerosis development, but was not correlated significantly with serum VLDL+LDL or HDL cholesterol levels. The infiltration of lipoprotein particles into the aortic wall seems to be more strongly associated with the progression of intimal lesions rather than with serum cholesterol levels.

Change in expression of GBP28/adiponectin in carbon tetrachloride-administrated mouse liver.

GBP28 (gelatin-binding protein of 28 kDa)/adiponectin is an adipocyte-producing plasma protein proposed to interact with the extracellular matrix. To examine the production of GBP28 in non-adipose tissues, we herein analyzed its expression and localization in mouse livers before and after CCl(4) treatment. In immunohistochemical analyses, the boundary of hepatocytes provided positive signals for GBP28 after 3-6 h and their cytoplasm was intensely stained after 18 h of CCl(4) treatment. Quantitative RT-PCR and in situ hybridization revealed that GBP28 mRNA expression was markedly elevated in CCl(4)-treated mouse livers. These results suggest that the circulating GBP28 binds the extracellular matrices of hepatocytes during the initial stage of CCl(4)-induced hepatic injury and the damaged hepatocytes themselves started to produce GBP28 thereafter. The induced expression of GBP28 was also observed in human hepatoma HepG2 cells after treatment with IL-6. Thus, GBP28 is also produced by the liver, where it undergoes tissue damage-induced transcriptional regulation.

Proteolytic activation and inactivation of the serine protease activity of plasma hyaluronan binding protein.

We prepared anti-plasma hyaluronan binding protein (PHBP) mouse monoclonal antibodies and studied the fragmentation profile of PHBP with them. PHBP is present in human plasma as a single polypeptide chain (70 kDa). During the purification, PHBP partially fragmentated into the 50-kDa N-terminal fragment and the 27-kDa C-terminal fragment. After the incubation of the purified PHBP, the 70-kDa precursor form was completely cleaved to the 50- and 27-kDa fragments, followed by the 50-kDa to the 26-kDa, and the 27-kDa to the 17-kDa plus the 8-kDa fragments, respectively. Because the purified PHBP contained no other detectable proteins and PHBP has a typical serine protease domain, we concluded that the fragmentation of PHBP was caused by own serine protease activity. PHBP cleaved the C-terminal side of Arg in the peptide effectively and that of Lys weakly. The results of the pre-incubation experiments of PHBP suggested that the single-chain form of PHBP is a precursor, the two-subunit structure is an active form and the three- or four-chain structure is an inactive form of a serine protease.

Characterization of equine microsatellites and microsatellite-linked repetitive elements (eMLREs) by efficient cloning and genotyping methods.

We performed efficient cloning and genotyping methods for isolation of a large number of polymorphic microsatellites. The methods contain the time-efficient cloning method of constructing microsatellite-enriched libraries and the economic genotyping method of fluorescent labeling of PCR products. Eighty novel equine microsatellites cloned were efficiently isolated from the enrichment library and analyzed for genotype polymorphism. Of these, 72 microsatellites were analyzed with a good resolution. The average heterozygosity of all loci was 0.52, and the number of alleles ranged from one to 9 with an average of 4.5 alleles. The other eight loci showed multiple bands of PCR products, suggesting the occurrence of microsatellites in a repetitive element, in which the number of microsatellite repeats varies among different members of the repetitive element. We found five homologous groups at flanking regions in comparison with the flanking regions of microsatellites from DNA databases. One of them showed homology to equine repetitive element-2. In the other four homologous groups, the two groups were named equine microsatellite-linked repetitive element-1 (eMLRE-1) and equine microsatellite-linked repetitive element-2 (eMLRE-2) as novel equine repetitive elements identified from equine genome. These data should help the analysis of equine DNA sequences and the design of equine genome markers.

Quantitative measurement of the novel human plasma protein, IHRP, by sandwich ELISA.

Inter-alpha-trypsin inhibitor family heavy chain-related protein (IHRP), which has a sequence similarity to the heavy chains of the inter-alpha-trypsin inhibitor (ITI) family, is a novel glycoprotein found in human plasma. We prepared two clones (1A4 and 6E11) of anti-IHRP mouse monoclonal antibody. Both of them recognized the C-terminal 35-kDa fragment which was produced by plasma kallikrein-digestion of IHRP. We developed a sandwich ELISA for measurement of the plasma IHRP concentration with the monoclonal antibody coated microtiter plate and the anti-N-terminal 57-kDa fragment of IHRP rabbit polyclonal antibody (anti-GP57). We found that the average concentration of IHRP in the plasma of healthy donors was 101.3+/-31.8 microg/ml (average+/-S.D.). The IHRP concentration in the plasma of patients with inflammatory disorders was slightly increased (137.5+/-40.2 microg/ml: average+/-S.D.). Together with the previous data indicating the induction of the porcine mRNA, which is thought to be the species counterpart of human IHRP mRNA, in the liver after resuscitation from cardiogenic shock, we propose that IHRP is a member of the acute-phase protein family.

Regulation mechanism of the serine protease activity of plasma hyaluronan binding protein.

The inhibitor for the serine protease activity of plasma hyaluronan binding protein (PHBP) was purified from human plasma by polyethylene glycol (PEG) fractionation, diethylaminoethyl (DEAE)-Sephacel ion-exchange chromatography, Phenyl Toyopearl 650M hydrophobic chromatography, Bio Gel A-0.5 m gel-filtration and hydroxyapatite chromatography. The serine protease activity of PHBP was measured with Boc-Phe-Ser-Arg-methylcoumarine amide (MCA) as the synthetic substrate of PHBP. The results of the amino acid sequence analyses of the purified PHBP inhibitor indicated that it was C1 inhibitor of the serpin family. C1 inhibitor formed a complex with PHBP, suggesting that it is the actual inhibitor of PHBP in human plasma. On the other hand, dextran sulfate and phosphatidylethanolamine enhanced the auto-fragmentation and the serine protease activity of pro-PHBP, but kaolin did not. These results suggested that the serine protease activity of PHBP was regulated in a similar manner to that of factor XII of the coagulation system.

Characterization of mouse GBP28 and its induction by exposure to cold.

OBJECTIVE: To investigate whether the expression of the novel adipose tissue-specific protein GBP28 in adipose tissue and serum are altered in mice under a variety of conditions. DESIGN: Mice were fed a high-fat diet for 4 weeks, fasted for 48 h or exposed at 4 degrees C. SUBJECTS: C57BL/6J mouse, male, 4--6 weeks old. MEASUREMENTS: GBP28 mRNA, GBP28 protein, blood glucose, insulin and fad pad weight of the mice. RESULTS: We first confirmed that the mouse has GBP28 and its characteristics are the same as human GBP28. Serum concentration and mRNA levels of GBP28 significantly increased in the mice exposed to cold. CONCLUSION: GBP28 may play a role in homeostasis, regulating body temperature and basal metabolic rate in response to changing environmental conditions. International Journal of Obesity (2001) 25, 75-83

Identification of the substrates for plasma hyaluronan binding protein.

Plasma hyaluronan biding protein (PHBP) is a novel serine protease, which has an amino acid sequence homology to that of hepatocyte growth factor activator (HGFA), and has a similar domain structure to that of urinary plasminogen activator (u-PA), found in human plasma. We searched the PHBP substrate in human plasma by measuring the digested protein bands on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The results showed that fibrinogen and fibronectin were the major substrates of PHBP. PHBP cleaved the alpha-chain at multiple sites and the beta-chain between lysine53 and lysine54 but not the gamma-chain of fibrinogen. Therefore, PHBP did not initiate the formation of the fibrin clot and did not cause the fibrinolysis directly. PHBP did not cleave (activate) prothrombin and plasminogen, but it converted the inactive single chain urinary plasminogen activator to the active two chain form.

Clusterin is up-regulated in glomerular mesangial cells in complement-mediated injury.

BACKGROUND: Clusterin is a soluble complement regulatory protein that binds to C5b-7 and inhibits generation of membrane attack complex, C5b-9. Glomerular deposition of clusterin has been observed in human and experimental membranous nephropathy in association with C5b-9 and immune deposits. However, it is controversial as to whether clusterin observed in glomeruli is synthesized by the resident glomerular cells or is derived from the circulation. We examined whether clusterin is expressed by resident glomerular cells exposed to complement-mediated injury. METHODS: In vitro, cultured mesangial cells were exposed to antithymocyte serum immunoglobulin G and 5% normal rat serum as a complement source. In vivo, we induced anti-Thy1 nephritis in rats and examined the kidneys on days 8 and 29. RESULTS: We observed increased expression of clusterin in cultured rat glomerular mesangial cells stimulated by sublytic complement attack. We also demonstrated that in comparison with control rats, both a marked increase in clusterin mRNA in the glomeruli and marked deposition of clusterin protein in the mesangial area occurred in the OX-7-treated rats on day 8 in association with C5b-9 deposition and on day 29. CONCLUSION: Clusterin was induced in glomerular mesangial cells during the course of immune-mediated injuries. This up-regulation of clusterin may play a critical role in protecting mesangial cells from complement attack.

Population study and validation of paternity testing for Thoroughbred horses by 15 microsatellite loci.

Microsatellite 15 TKY System was characterized for parentage verification of horse registry. The Microsatellite 15 TKY System was constructed by using 15 microsatellites, TKY279, TKY287, TKY294, TKY297, TKY301, TKY312, TKY321, TKY325, TKY333, TKY337, TKY341, TKY343, TKY344, TKY374, and TKY394, to provide stringent PCR-based microsatellite typing specifically optimized for multicolor fluorescence detection. The Microsatellite 15 TKY System showed good resolutions for 250 unrelated Thoroughbred horses, and the probability of exclusion (PE) at each microsatellite ranged from 0.437 to 0.621, resulting in a total PE value of 99.998% for Thoroughbred horses. These results indicated that the Microsatellite 15 TKY System is useful for paternity testing of Thoroughbred horses. A paternity testing case for a Thoroughbred horse family, in which candidate sires had close relations, was analyzed using the Microsatellite 15 TKY System. In this case, the Microsatellite 15 TKY System excluded paternity of a false sire. We concluded that the Microsatellite 15 TKY System can give sufficient and reliable information for paternity testing.

The novel acute phase protein, IHRP, inhibits actin polymerization and phagocytosis of polymorphonuclear cells.

OBJECTIVE AND DESIGN: In the present study, the involvement of the binding of IHRP (inter-alpha-trypsin inhibitor family heavy chain-related protein) and actin in phagocyte activity was investigated. MATERIALS AND METHODS: The actin polymerization and the phagocytic activity of the polymorphonuclear (PMN) cells were studied in the presence of IHRP. RESULTS: IHRP inhibited the polymerization of actin and the phagocytic activity of the PMN cells. CONCLUSION: 1) IHRP may bind to actin released from the damaged cells and suppress its toxic action by preventing the formation of actin fibril. 2) IHRP may bind to cell surface actin on PMN cells and inhibit their phagocytic activities. 3) From these results, IHRP may act as an anti-inflammatory protein.

Regulation of gelatin-binding protein 28 (GBP28) gene expression by C/EBP.

We have previously reported the isolation of human gelatin-binding protein 28 (GBP28) gene which is specifically expressed in adipose tissue. The transcriptional activity of the flanking region of the GBP28 gene was examined by the transient transfection of promoter-luciferase reporter constructs into 3T3 adipocytes and electrophoretic mobility shift assay. This revealed the existence of a protein which binds to the 5'-flanking region of the GBP28 gene in nuclear extracts from human adipose tissue, but not in nuclear extracts from mouse liver. The C/EBP sites contained in this region are thought to take part in the regulation of GBP28 gene expression.

Apolipoproteins J and E co-localise with amyloid in gelatinous drop-like and lattice type I corneal dystrophies.

AIMS: Apolipoprotein J (apoJ) and apolipoprotein E (apoE) are thought to contribute to amyloid formation in patients with Alzheimer's disease. The aim of this investigation was to discover whether or not these apolipoproteins associate with corneal amyloid in gelatinous drop-like corneal dystrophy (GDCD) and lattice corneal dystrophy type I (LCD-I). METHODS: Corneas from three eyes of three patients with GDCD and one eye of one patient with LCD-I were examined immunohistochemically using antibodies against apoJ and apoE. Two normal corneas were similarly examined. Tissue sections of brain from a patient with Alzheimer's disease were used as positive controls for the antibodies. For all negative controls, mouse IgG was used instead of the primary antibody. RESULTS: Intense apoJ and apoE immunoreactivities were found in congophilic amyloid deposits in GDCD and LCD-I. These deposits were located subepithelially in GDCD, and subepithelially and intrastromally in LCD-I. In GDCD, immunostaining of subepithelial amyloid with anti-apoJ was noticeably stronger than with anti-apoE. CONCLUSIONS: As in senile plaques in brain from a patient with Alzheimer's disease, apoJ and apoE co-localise with amyloid in corneas with GDCD and LCD-I.

Distribution and synthesis of apolipoprotein J in the atherosclerotic aorta.

The distribution of apolipoprotein (apo) J during the development of atherosclerosis in the human aorta was evaluated by immununohistochemical observation, together with the other apolipoprotein A-I, A-II, B, C-III, and E. Although apoJ was never observed in the normal aorta (ie, without any intimal lesions or intimal thickening), it was distributed not only in the intima but also in the media of aortas with diffuse, intimal thickening or atherosclerotic lesions. Double immunostaining with antibodies for apoJ and alpha-smooth muscle actin revealed apoJ deposition in smooth muscle cells (SMCs) or the aortic stroma in the vicinity of SMCs. The extent of apoJ distribution in the aortic wall increased with the degree of atherosclerosis development. In addition, the distribution pattern of apoJ was very similar to that of apoA-I and E. In situ hybridization with human apoJ cDNA demonstrated intense signals in cells scattered within the subendothelial space and medial SMCs of the aorta with advanced atherosclerosis but not in those of the normal aorta without intimal thickening. Furthermore, reverse transcriptase-polymerase chain reaction of the cultured human aortic SMCs revealed apoJ mRNA expression in these cells. The results indicate that apoJ in the aortic wall originates from not only apoJ circulated in the plasma but also apoJ produced by SMCs in the aortic wall. Considering the similarities of the distribution between apoJ and apo-A-I or E, we hypothesize that apoJ possibly has a protective role against human atherosclerosis by its involvement with cholesterol transport from the aortic wall to the liver.

Cloning of the cDNA for a mouse homologue of human PHBP: a novel hyaluronan-binding protein.

The cDNA which encodes the mouse counterpart of human plasma hyaluronan-binding protein (PHBP) was isolated and characterized. The clone contained an insert of 2153 bp, which contained the 1674-bp open reading frame coding for a polypeptide of 558 amino acid residues. The amino acid sequence of mouse PHBP predicted from the nucleotide sequence of cDNA shows reasonable homology to that of human PHBP. Like human PHBP, the amino acid sequence predicted from the nucleotide sequence of mouse PHBP cDNA exhibited significant homology to that of human hepatocyte growth factor activator (HGFA).

Isolation and characterization of the plasma hyaluronan-binding protein (PHBP) gene (HABP2).

PHBP is a novel human plasma hyaluronan-binding protein that shows significant homology in amino acid sequence to hepatocyte growth factor activator. Two overlapping clones that encode the human plasma hyaluronan-binding protein (PHBP) gene (HABP2) were isolated and characterized. The PHBP gene spans 35 kb and is composed of 13 exons from 37 to 1,394 bp in size with consensus splice sites. The gene's regulatory sequences contain putative promoter elements, but no typical TATA box. Some exons of this gene showed significant similarities to those of coagulation factor XII, tissue-type plasminogen activator, and urokinase genes in nucleotide length and in intron phasing. We also report the chromosome mapping of this gene by fluorescence in situ hybridization (FISH) using a genomic DNA fragment as a probe. The PHBP gene (HABP2) was located on chromosome 10q25-q26.