[Proteasome degradation of protein C and plasmin inhibitor mutants: molecular mechanism of congenital protein deficiency].

In many inherited disorders, protein deficiency is one of the major aetiologies, but the molecular and cellular mechanisms remain unclear. We investigated the intracellular degradation of mutant proteins, using naturally occurring PC and PI mutants that lead to congenital deficiencies. We have shown that proteasomes are very important for the degradation of PC and PI mutants, irrespective of the presence or absence of N-glycosylation moieties. Furthermore, mannose trimming after glucose removal is very important for initiation of the degradation. Inhibition of glucose trimming of the mutant proteins accelerated degradation by the proteasomes, and initiation of the degradation occurs after mannose trimming of the middle chain of N-linked glycosylation by mannosidase I. The binding of molecular chaperons influenced by the presence of N-glycosylation moieties may affect the efficient degradation of the mutant proteins. Cotransfection of endoplasmic reticulum (ER) degradation enhancing alpha-mannosidase like protein (EDEM) accelerated the degradation of N-glycosylated PC. The mutant PC or PI molecules were ubiquitin-independently degraded by proteasomes. Autophagy does not appear to contribute to the degradation of PC and PI mutants. These findings might help to elucidate the molecular mechanisms and potential treatments of congenital deficiencies of proteins in a system of coagulation and fibrinolysis.

Proteasome degradation of protein C and plasmin inhibitor mutants.

Protein C (PC) deficiency and plasmin inhibitor (PI) deficiency are inherited thrombotic and haemorrhagic disorders. We investigated the intracellular degradation of mutant proteins, using naturally occurring PC and PI mutants that lead to congenital deficiencies. To examine the necessity of N-linked glycosylation for the proteasomal degradation of PC and PI, PC178 and PC331 mutants treated with tunicamycin and N-glycosylation-lacking mutants, PC92Stop and PI-America were pulse chased. The analysis revealed that the speed of degradation of the tunicamycin-treated PC mutants, PC92Stop and PI-America lacking glycosylation, was slower than that of N-glycosylated mutants. Immunoprecipitation and immunoblot analysis showed that PC178 and PC331 mutants were associated with molecular chaperones, Bip, GRP94, and calreticulin. PI-America was associated with only Bip. Although degradation of mutants was mediated by proteasomes, no association with ubiquitin was detected. Cotransfection of endoplasmic reticulum (ER) degradation enhancing alpha-mannosidase-like protein (EDEM) accelerated the degradation of N-glycosylated PC. In the absence of autophagy using Atg5-deficient cell lines, the degradation of the PC331 mutant was mildly accelerated but that of PC178, PI-America and PI-Okinawa mutants was not influenced. While the degradation of the PC and PI mutants was facilitated by N-glycosylation moieties, they were ubiquitin-independently degraded by proteasomes, irrespective of the presence or absence of N-glycosylation. Molecular chaperone binding was influenced by the presence of N-glycosylation moieties. When the misfolded or truncated mutant proteins are functionally active, proteasome inhibitors such as bortezomib may have therapeutic potential for treatment of protein deficiencies.

Gradually glycosylated protein C mutants (Arg178Gln and Cys331Arg) are degraded by proteasome after mannose trimming.

Proteins that fail to attain their correct three-dimensional structure are retained in the endoplasmic reticulum (ER) and eventually degraded within the cells. We investigated the degradation of mutant proteins, using naturally occurring protein C (PC) mutants (Arg178Gln and Cys331Arg) which lead to congenital deficiencies. Chinese hamster ovary (CHO) cells were transfected with normal or mutant expression vectors. The introduction of mutation at Asn329 of an unusual sequence Asn-X-Cys for N-linked glycosylation revealed that the mutation at Cys331, which may preclude a formation of disulfide bond with Cys345, resulted in no addition of N-linked oligosaccharides at Asn329. PC mutants with 4 glycosylation sites were gradually glycosylated in the ER, and the fourth glycosylation site is less accessible for glycosylation as reported for PC in plasma. The half lives of PC178 and PC331 mutants were about 5 and 4 h, respectively. PC mutants were degraded, but the degradation was inhibited by inhibitors for proteasome. Mannose trimming of N-linked oligosaccharides after glucose removal targeted PC mutants for degradation by proteasomes. And also the inhibition of glucose trimming immediately led to mannose trimming, resulting in the accelerated degradation of PC mutants. These degradations were inhibited by mannosidase I inhibitor, kifunensine. These results indicate that the initiation of mannose trimming by mannosidase I leads to the proteasomemediated degradation of glucose-trimmed or untrimmed PC mutants.

A single thymine nucleotide deletion responsible for congenital deficiency of plasmin inhibitor.

Plasma plasmin inhibitor (PI) is a physiological inhibitor of plasmin-mediated fibrinolysis and constitutes a hemostatic component in blood plasma; hence its deficiency results in a severe hemorrhagic diathesis. We have carried out molecular analysis of American family members with congenital PI deficiency, and detected a single thymine deletion at nucleotide position 332 in exon 5. The deletion was found in both alleles of the homozygotes and in one allele of the heterozygotes, and the patterns of restriction fragment length polymorphism created by the mutation in the family members were compatible with their phenotypes. The deletion caused a frameshift leading to an alteration and shortening of the deduced amino acid sequence. The amino acid sequence consists of the first 83 amino acids of the N-terminal sequence of the normal PI and additional new amino acids, resulting in a mutant composed of 94 amino acids in contrast to 464 amino acids of the normal PI. In transient expression analysis, the mutant PI whose molecular size was compatible with the predicted amino acid sequence was detected in the lysates of the cells transfected with the mutated PI expression vector. The mutant PI was retained and underwent progressive degradation within the cells, and was minimally excreted into the media. These data indicate that this mutation is the cause of PI deficiency in this pedigree.

Induction of tissue factor expression in human monocytic cells by protease inhibitors through activating activator protein-1 (AP-1) with phosphorylation of Jun-N-terminal kinase and p38.

Tissue factor (TF) is expressed rapidly by human monocytes exposed to a variety of agonists such as lipopolysaccharide (LPS) or tumor necrosis factor-alpha. Activation of both activator protein-1 (AP-1; c-Jun/c-Fos) and nuclear factor-kappaB (NF-kappaB) pathways is necessary for maximal induction of the TF gene. It has been demonstrated that activation of both AP-1 and NF-kappaB is correlated with the degradation of both phosphorylated c-Jun and inhibitor kappaB (IkappaB) by proteasome. The present study was designed to investigate whether various protease inhibitors, including proteasome inhibitors, affect TF expression in monocytic cells. Protease inhibitors, 3,4-dichloroisocoumarin (DCI), N-tosyl-l-phenylalanine chloromethyl ketone (TPCK), and N-acetyl-Leu-Leu-norleucinal (ALLN) induced TF activity in monocytic cells in a dose- and time-dependent manner at the level of the transcription of the TF gene, which was mediated through inducing phosphorylation of both Jun-N-terminal kinase and p38. The early growth response-1 (Egr-1) pathway was not affected. The NF-kappaB pathway was not activated; rather it was inhibited. These results were distinct from the findings previously reported for LPS-stimulated cells. The present study demonstrated that some protease inhibitors might act as stress and induced TF expression with direct phosphorylation of JNK and p38, followed by phosphorylation and activation of AP-1 in monocytic cells. This evidence may help elucidate further regulatory mechanisms of TF induction, and might have physiological significance for the clinically challenged use of proteasome inhibitors. In addition to phosphorylation of JNK and p38, an unknown signal pathway needs to be clarified for TF induction.

Non-viral in vivo thrombomodulin gene transfer prevents early loss of thromboresistance of grafted veins.

OBJECTIVE: Immediate loss of thrombomodulin activity in the endothelium of vein grafts has been demonstrated during 90 min exposure to arterial circulation; this loss of activity is ascribed as an important cause of early thrombosis. Conventional ex vivo gene transfection after vein harvest cannot cover this acute period immediately after implantation. We have established a highly efficient non-viral gene therapy protocol utilizing modified transferrin receptor-facilitated gene transfer. Using this technique, we examined whether in vivo thrombomodulin gene therapy, directed to the endothelium of rat veins 2 days prior to grafting, may prevent thromboresistance impairment of vein grafts under simulated arterial circulation. METHODS: Abdomen of SD rat was opened and cationic liposome:transferrin:thrombomodulin gene complexes or the vector without DNAs were applied to the inferior vena cava of rats while blood flow was reduced by proximal and distal clamping. After 2 days, the transfected veins were harvested and thrombomodulin expression and thromboresistance properties determined before and after exposure to an artificial circuit. RESULTS: The trial of gene transfection using variable doses of DNAs confirmed that 7.5 microg of total DNAs was the most efficient quantity for thrombomodulin gene transfection to IVCs, although accompanying an increase of gene expression in other downstream organs. By transfection of the thrombomodulin gene in IVCs, the generation capacity of activated protein C in venous endothelium increased three-fold compared with veins treated with vector alone (P<0.01). Under simulated arterial circulation, perfusion of veins treated with vector alone decreased thrombomodulin activity to 36% of preperfused levels (P<0.01), whereas transfected grafts preserved the activity at normal vein endothelium levels even after perfusion. Consequently, the increase in endothelial thrombin activity induced by simulated arterial circulation was markedly attenuated in transfected veins (P<0.01), while immunohistochemistry confirmed the preservation of endothelial lining. CONCLUSIONS: Transferrin receptor-facilitated in vivo gene transfer to the inferior vena cava resulted in sufficient thrombomodulin gene expression immediately after graft implantation and subsequent maintenance of thromboresistance even after exposure to arterial pressure. Although further studies are needed, the present results suggest the possibility of gene therapy targeting acute phases of vein graft disease.

1,25(OH)(2)D(3) blocks TNF-induced monocytic tissue factor expression by inhibition of transcription factors AP-1 and NF-kappaB.

An essential coagulation factor, tissue factor (TF), is rapidly expressed by human monocytes when exposed to a variety of agonists, such as lipopolysaccharide or tumor necrosis factor (TNF). We previously found that 1alpha,25-dihydroxyvitamin D(3) (1,25(OH)(2)D(3)) and its potent synthetic analogs downregulate TF and upregulate thrombomodulin expression on monocytic cells, counteracting the effects of TNF at the level of transcription. The human TF gene has characteristic binding sequences for activator protein-1 (AP-1) (c-Jun/c-Fos), nuclear factor-kappaB (NF-kappaB), Sp-1, and early growth response factor-1 (Egr-1). In this study, we investigated the regulatory mechanisms by which 1,25(OH)(2)D(3) inhibits TNF-induced TF expression in human monocytic cells. 1,25(OH)(2)D(3) reduced basal and TNF-induced TF activities. Gel-shift assay and luciferase assay with the respective reporter vectors showed that 1,25(OH)(2)D(3) reduced basal and TNF-induced activities of the nuclear proteins AP-1 and NF-kappaB, but not Egr-1. 1,25(OH)(2)D(3) inhibited TNF-induced phosphorylation of c-Jun without affecting phosphorylation of the other pathways. On the other hand, 1,25(OH)(2)D(3) directly inhibited nuclear binding and activities of NF-kappaB in the nucleus without affecting phosphorylation of the NF-kappaB activation pathway. These results indicate that 1,25(OH)(2)D(3) suppresses basal and TNF-induced TF expression in monocytic cells by inhibition of AP-1 and NF-kappaB activation pathways, but not of Egr-1. Our results may help to elucidate the regulatory mechanisms of 1,25(OH)(2)D(3) in TF induction, and may have physiological significance in the clinical challenge to use potential 1,25(OH)(2)D(3) analogs in antithrombotic therapy as well as immunomodulation and antineoplastic therapy of leukemia.

Cloning and characterization of the human BAZF gene, a homologue of the BCL6 oncogene.

The BAZF gene has recently been identified as a novel homologue of the BCL6 oncogene. Here we cloned the human BAZF gene using murine BAZF as a probe. The predicted amino acid sequence was 91% identical to that of murine BAZF. The BTB/POZ and zinc finger domains were almost completely conserved between human and murine BAZF. Fluorescence in situ hybridization analysis revealed that the human BAZF gene is located on chromosome 17p13.1. Although expression of human BAZF mRNA was ubiquitously detected in human tissues, abundant expression was detected in heart and placenta. BAZF mRNA was expressed in some immature B cell lines and erythroleukemia cell lines. The expression in a human erythroleukemia cell line, HEL cells, was upregulated during megakaryocytic differentiation induced by 12-O-tetradecanoyl-phorbol-13-acetate. These expression patterns of BAZF mRNA suggest that BAZF may regulate differentiation in stages or lineages that are different from those regulated by BCL6.