Presenilin-1 and -2 are molecular targets for gamma-secretase inhibitors.

Presenilins are integral membrane protein involved in the production of amyloid beta-protein. Mutations of the presenilin-1 and -2 gene are associated with familial Alzheimer's disease and are thought to alter gamma-secretase cleavage of the beta-amyloid precursor protein, leading to increased production of longer and more amyloidogenic forms of A beta, the 4-kDa beta-peptide. Here, we show that radiolabeled gamma-secretase inhibitors bind to mammalian cell membranes, and a benzophenone analog specifically photocross-links three major membrane polypeptides. A positive correlation is observed among these compounds for inhibition of cellular A beta formation, inhibition of membrane binding and cross-linking. Immunological techniques establish N- and C-terminal fragments of presenilin-1 as specifically cross-linked polypeptides. Furthermore, binding of gamma-secretase inhibitors to embryonic membranes derived from presenilin-1 knockout embryos is reduced in a gene dose-dependent manner. In addition, C-terminal fragments of presenilin-2 are specifically cross-linked. Taken together, these results indicate that potent and selective gamma-secretase inhibitors block A beta formation by binding to presenilin-1 and -2.

The five cysteine residues located in the active site region of bovine aspartyl (asparaginyl) beta-hydroxylase are not essential for catalysis.

In previous chemical modification studies on bovine aspartyl (asparaginyl) beta-hydroxylase, cysteines were implicated as critical catalytic residues. Using site-directed mutagenesis, the five cysteine residues located in a highly conserved region of the enzyme identified as the active site were individually mutated to alanine. Substitutions at cysteine 637, 644, 656, 681, and 696 resulted in active mutant enzymes indicating that these residues are not required for catalysis.

Overexpression of human aspartyl(asparaginyl)beta-hydroxylase in hepatocellular carcinoma and cholangiocarcinoma.

To characterize genes that become upregulated with malignant transformation of human hepatocytes, a library of monoclonal antibodies was produced against the FOCUS hepatocellular carcinoma cell line. Antibody FB-50 reacted with an antigen that was highly expressed in 4 of 10 primary hepatocellular carcinomas, in all 20 cholangiocarcinomas we studied, and in a variety of transformed cell lines. This antigen was also highly expressed in neoplastic epithelial cells of breast and colon carcinomas in contrast to its low level of expression in normal hepatocytes and in non-neoplastic epithelial cells. Among the normal adult tissues studied, high levels were observed only in proliferating trophoblastic cells of the placenta and in adrenal glands. A 636-bp partial cDNA, isolated from a gamma GT11 expression library generated with HepG2 human hepatoblastoma cells, and a complete cDNA, generated by reverse transcriptase-PCR, identified the antigen as the human form of aspartyl(asparaginyl)beta-hydroxylase. This enzyme catalyzes posttranslational hydroxylation of beta carbons of specific aspartyl and asparaginyl residues in EGF-like domains of certain proteins. Analyses of extracts prepared from several human tumor cell lines compared to their normal tissue counterparts indicate that the increase in hydroxylase, approximately 10-fold, is controlled at the level of transcription and the protein is expressed in an enzymatically active form. In similar analyses, comparing hepatocellular carcinomas to adjacent uninvolved liver from five patients, enzymatic activity was much higher in the tumor tissue from the four patients whose immunoblots revealed increased hydroxylase protein in the malignant tissue. EGF repeats in the extracellular domain of Notch or its homologs contain the consensus sequence for hydroxylation. Deletion mutants lacking this domain are gain-of-function mutants, suggesting that the domain modulates signal transduction by the cytoplasmic domain. While the function imparted by beta hydroxylation is unknown, our studies raise the possibility that beta hydroxylation is regulated in proteins like the mammalian Notch homologs, whose cytoplasmic domains have been shown to be oncogenic.

The subcellular distribution of eukaryotic translation initiation factor, eIF-5A, in cultured cells.

To gain insight into the role of the eukaryotic translation initiation factor, eIF-5A, we investigated the subcellular distribution of this protein in several cultured cell types and at different stages of the cell cycle using a highly potent monospecific polyclonal antibody to eIF-5A. Studies using indirect immunofluorescence and confocal microscopy in conjunction with subcellular fractionation demonstrate that eIF-5A is primarily localized in the cytoplasm of cells. This cytoplasmic location of eIF-5A is not significantly altered in different stages of the cell cycle and the subcellular distribution pattern of eIF-5A is not changed by viral oncogene transformation. Cell fractionation experiments identified two populations of eIF-5A in the cytoplasm, a soluble fraction and a fraction bound to internal membranes. By double immunofluorescence staining with an antibody against calnexin, a resident protein of the endoplasmic reticulum (ER), we demonstrate that the membrane-bound fraction of eIF-5A colocalizes with the ER and not with the cytoskeleton. Expression of Rev, a regulatory protein of human immunodeficiency virus type 1 (HIV-1), does not alter the subcellular distribution of endogenous eIF-5A in these cells. eIF-5A is detected in all tissues and cells examined including extracts prepared from Xenopus oocytes. Our results indicate that eIF-5A is a ubiquitous cytoplasmic protein and suggest that a site of eIF-5A function is likely to be in association with the ER.

Effects of N1-guanyl-1,7-diaminoheptane, an inhibitor of deoxyhypusine synthase, on the growth of tumorigenic cell lines in culture.

N1-guanyl-1,7-diaminoheptane (GC7) is a potent inhibitor of deoxyhypusine synthase (DHS), the enzyme that catalyzes the first step in the hypusination of eukaryotic translation initiation factor 5A (eIF-5A). Since eIF-5A is the only known cellular substrate for DHS and GC7 has been reported to block the proliferation of CHO cells, it has been suggested that DHS may be a novel target for anti-cancer therapy. In the present study we investigated the antiproliferative effect of GC7 on several tumorigenic cell lines under various growth conditions. We found that this compound inhibits the proliferation of H9 cells in suspension culture and the growth of HeLa cells and v-src-transformed NIH3T3 cells under both anchorage-dependent and anchorage-independent conditions. Moreover, studies with NIH3T3 cells and v-src-transformed NIH3T3 cells show that GC7 inhibits the growth of both cell lines in monolayer culture with similar potency and could not reverse the transformed phenotype. In addition, the v-src-transformed cells grown under both anchorage-dependent and anchorage-independent conditions showed similar sensitivity toward GC7. These data indicate that GC7 acts as a general antiproliferative agent and does not appear to preferentially target tumorigenic cell types. Cell cycle analysis show that GC7 reduces the CHO-K1 cell population in the G1-phase of the cell cycle by 42% and increases the number of cells in the S-phase by 44%. This cell cycle distribution profile strikingly resembles the distribution of cells treated with puromycin. This result supports the hypothesis that the synthesis of a subset of proteins important for the S-phase progression of CHO-K1 cells might be dependent upon hypusinated eIF-5A. Thus the antiproliferative effect of GC7 appears to be related to its interference with the progression of cell cycle, which also provides a possible explanation for the lack of selectivity of GC7 between nontransformed and transformed cell types tested in this study.

Transglutaminase inhibition by 2-[(2-oxopropyl)thio]imidazolium derivatives: mechanism of factor XIIIa inactivation.

The physiologic role of several transglutaminases could be more precisely defined with the development of specific inhibitors for these enzymes. In addition, specific plasma transglutaminase (fXIIIa) inhibitors may have therapeutic utility in the treatment of thrombosis. For these purposes, the inactivation of fXIIIa and human erythrocyte transglutaminase (HET) by 2-[(2-oxopropyl)thio]imidazolium derivatives, which comprise a novel class of transglutaminase inactivators, was studied. As a specific example, 1,3,4,5-tetramethyl-2-[(2-oxopropyl)thio]imidazolium chloride (III) inactivated fXIIIa with an apparent second-order rate constant (specificity constant of inactivation) of 6.3 x 10(4) M-1 s-1, corresponding to a rate 4 x 10(7) times greater than its reaction rate with glutathione (GSH). The mechanism of fXIIIa inactivation by this class of compounds was investigated utilizing two [14C]-isotopic regioisomers of 1,3-dimethyl-2-[(2-oxopropyl)thio]imidazolium iodide (II). Structural analyses demonstrated that acetonylation of the active site cysteinyl residue of fXIIIa occurred along with the stoichiometric release of the complementary fragment of the inactivator as the corresponding thione. Kinetic analysis of the inactivation of fXIIIa by nonquarternary analogs of II and III indicated the formation of a reversible complex between the inactivator and fXIIIa prior to irreversible modification of the enzyme. At 1 mM, III displayed no detectable levels of inhibition or inactivation with several serine proteases and thiol reagent-sensitive enzymes. 2-[(2-Oxopropyl)thio]imidazolium derivatives and the related molecule 2-(1-acetonylthio)-5-methylthiazolo-[2,3]-1,3,4-thiadiazo lium perchlorate (I), when present at the time of clot formation at 1-10 microM, enhanced the rates of tissue plasminogen activator catalyzed clot lysis in vitro. These inactivators prevented the fXIIIa-catalyzed covalent incorporation of alpha 2-antiplasmin into the alpha chain of fibrin and the formation of high molecular weight fibrin alpha chain polymers, providing the basis for the observed enhancements in clot lysis rates.

Identification and characterization of endothelin converting activity from EAHY 926 cells: evidence for the physiologically relevant human enzyme.

A neutral proteolytic activity that converts human Big endothelin-1 (Big Et-1) to endothelin-1 has been identified from a human endothelial hybrid cell line, EAHY 926. This enzyme is an integral membrane protein and cofractionates with other enzymes typically found in the plasma membrane. The activity has been solubilized with nonionic detergents and purified 1000-fold by a combination of lectin affinity chromatography, ion-exchange chromatography, and chromatography on red-dye agarose. The partially purified activity is a metalloenzyme based upon its sensitivity to chelating agents, competitive inhibition by phosphoramidon, and reconstitution with ZnCl2 or CoCl2 following EDTA inactivation. The enzyme appears to be unique, however, as it is not inhibited by specific inhibitors of known metalloproteases. It correctly processes Big Et-1 to Et-1 and the complementary C-terminal fragment with a sharp pH optimum near 7.0. Both the Km for Big Et-1 and the Ki for phosphoramidon are pH-dependent, with values of 5-7 and 3.5 microM, respectively, at pH 7.0. The enzyme also cleaves Big Et-2 with a Km of 27.9 microM and a Vmax one-third that for Big Et-1 but has no appreciable activity toward Big Et-3. An s20,w of 9.5 S was determined by sucrose density ultracentrifugation in H2O and D2O. When combined with a Stokes radius of 56 A determined by gel filtration, the enzyme had a calculated apparent molecular weight of 250,000. Conditions have been established to renature the activity after sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Under nonreducing conditions activity was detected in a protein band at 280 kDa, in agreement with the aforementioned molecular weight determination. From these results, a kcat/Km of 1 x 10(6) M-1 s-1 was estimated for the purified enzyme with Big Et-1 as a substrate, which is a reasonable value for a protease acting upon its physiologic substrate. Several criteria indicate that the activity isolated from EAHY cells is the physiologically relevant endothelial-derived endothelin converting enzyme. On the basis of our results, this enzyme is present in low abundance in endothelial cells and at least a 100,000-fold purification will be required to obtain a homogeneous preparation. However, because EAHY cells can be grown in large numbers, they can supply the quantities of enzyme required both for biochemical studies and for the development of specific inhibitors.

Antitumor activity of DL-threo-beta-fluoroasparagine against human leukemia cells in culture and L1210 cells in DBA mice.

DL-threo-beta-Fluoroasparagine (DL-threo-beta-F-Asn) inhibited the growth of murine leukemia L1210 cells and three human leukemia cell lines in culture. Fifty % inhibiting dose values ranged between 30 and 50 microM DL-threo-beta-F-Asn. L1210 cells were not sensitive to DL-erythro-beta-fluoroasparagine, DL-threo-beta-fluoroaspartic acid, or DL-erythro-beta-fluoroaspartic acid at 300 microM, the highest dose studied. The antileukemia activity of DL-threo-beta-F-Asn was studied in further detail using the L1210 model system. Inhibition of growth in culture was prevented by L-asparagine but not by D-asparagine. Inhibition of growth of L1210 cells incubated for 40 hr in the presence of 300 microM DL-threo-beta-F-Asn was reversed after DL-threo-beta-F-Asn removal. Treatment for longer periods of time resulted in cell lysis. DL-threo-beta-F-Asn at doses of 250 mg/kg increased life span in mice bearing L1210 tumors by 60%. These results demonstrate the chemotherapeutic potential of the amino acid analogue DL-threo-beta-F-Asn.

Liver fibrosis (cardiac cirrhosis) five years after modified Fontan operation for tricuspid atresia.

A 15-year-old girl was found to have severe liver fibrosis on liver biopsy at the time of cholecystectomy, 5 1/2 years following a modified Fontan procedure (right atrial-right ventricular conduit) for tricuspid atresia. Postoperative right atrial pressures were consistently elevated above 13 mm Hg and this, in part, may have been due to progressive mild conduit stenosis. Because of increasing symptoms, the patient underwent successful revision of the conduit at the age of 15 years. It is suggested that sustained systemic venous hypertension caused the striking morphologic changes in the liver and that this serious complication may significantly affect the long-term prognosis of patients surviving the Fontan procedure.

Inhibition of asparagine-linked glycosylation of pro-opiomelanocortin in mouse pituitary cells by DL-threo-beta-fluoroasparagine.

The actions of DL-threo-beta-fluoroasparagine (DL-beta-F-Asn) on the glycosylation of proteins were examined in AtT-20/D16v cells which synthesize several forms of the glycoprotein prohormone, pro-opiomelanocortin (POMC). Treatment with threo-beta-F-Asn(5-10 mM) resulted in: 1) a reduction in the amount of the more highly glycosylated form of POMC (Mr = 32,000) relative to the less glycosylated form (Mr = 29,000) and 2) the appearance of a new species of POMC (Mr = 27,000). 35S]Methionine-labeled tryptic peptides prepared from 27,000 POMC were identical to those from 29,000 and 32,000 POMC; however, 27,000 POMC was found to contain 10% as much [3H]glucosamine relative to [35S]methionine as the 32,000 molecule. Furthermore, 27,000 POMC comigrated with a previously characterized unglycosylated form of this prohormone produced by treatment of cells with tunicamycin. These findings indicate that treatment of cells with threo-beta-F-Asn results in the production of a species of POMC which contains little or no carbohydrate. The effects of beta-F-Asn were specific for the threo diastereomer, were reversible by equimolar concentrations of Asn, but not Asp, and were dose-dependent. Evidence that threo-beta-F-Asn can replace Asn in proteins was obtained by showing that an identified Asn-containing tryptic peptide from threo-beta-F-Asn-treated cells displayed an altered mobility during electrophoresis consistent with threo-beta-F-Asn substitution within this peptide. We conclude that threo-beta-F-Asn can inhibit the glycosylation of proteins in intact cells and that this effect is due to its ability to replace Asn at glycosylation sites.

DL-threo-beta-fluoroasparagine inhibits asparagine-linked glycosylation in cell-free lysates.

The effect of beta-fluoroasparagine on N-linked glycosylation was examined in a cell-free translation system in which glycosylation is coupled to protein synthesis. The threo-isomer markedly inhibited glycosylation at a concentration of 1 mM, and this effect was blocked by L-asparagine, indicating that glycosylation was inhibited secondary to incorporation of the asparagine analog into protein. The erythro-isomer, at similar concentrations, was not incorporated into protein and had no effect on glycosylation. threo-beta-Fluoroasparagine is highly toxic to some mammalian cells in culture. Our observations suggest that its toxicity may be due in part to the failure of the fluoroasparagine-containing protein to become glycosylated. The data suggest that this analog will be useful for examining the structural determinants for glycosylation.

DL-threo-beta-Fluoroaspartate and DL-threo-beta-fluoroasparagine: selective cytotoxic agents for mammalian cells in culture.

Absolute configuration assignments have been made for the diastereomers of DL-beta-fluoroaspartate by X-ray analysis. The cytotoxicity of these isomers against various mammalian cells was examined. DL-threo-beta-Fluoroaspartate shows selective cytotoxicity. Growth of the most sensitive cells is completely inhibited by 13 micrometers DL-threo-beta-fluoroaspartate in the presence of 100 micrometers L-aspartate, a component of the culture medium. A difference in the rate of transport of DL-beta-fluoroaspartate among the cells studied is an important factor determining cell specificity. For those cells that are sensitive to DL-beta-fluoroaspartate, the threo isomer is, in all cases, more potent than the erythro isomer. Radioactivity derived from L-threo-beta-fluoro[14C]aspartate is incorporated into proteins at a rate comparable to the rate of incorporation from L-[14C]aspartate. We synthesized DL-threo-beta-fluoroasparagine. This compound is also cytotoxic but less specific and less potent than DL-threo-beta-fluoroaspartate. However, the cell specificity can be enhanced in the presence of 1 mM L-aspartate, which can protect some cells but not others from the cytotoxic effects of DL-threo-beta-fluoroasparagine. Jensen sarcoma cells, which require asparagine, are not protected by L-aspartate. Therefore, a combination of L-aspartate and DL-threo-beta-fluroasparagine can be used to inhibit specifically the growth of asparagine-requiring tumors.

Ventricular septal defect: results after repair in infancy.

During the 19 years from 1957 through 1975, there have been 106 patients under age 2 years who have undergone surgery for repair of a large ventricular septal defect at the University of Michigan Medical Center. The majority of the patients had either severe pulmonary hypertension or intractable congestive heart failure. Eighty-three infants survived operation; there has been one late death. The greatest mortality occurred in patients under age 6 months and in those with severe pulmonary hypertension. Surviving infants showed marked symptomatic improvement and change in growth patterns. Complications included the development of complete right bundle branch blodk or left anterior hemiblock in approximately 50 percent of patients and, in one instance, complete atrioventricular block. Forty-five patients have undergone cardiac catheterization 1 to 8 years postoperatively. Although 17 were found to have residual septal defects only 3 of these had a pulmonary to systemic flow ratio of 1.5:1 or more, and reoperation was accomplished without incident in these 3 patients and in 3 others with smaller shunts. With one exception, postoperative pulmonary arterial pressures and pulmonary to systemic vascular resistance ratios were normal or near normal, thus representing a significant contrast with findings in patients operated on after age 2 years. Whereas the complications of surgery appear no greater in the infant than in the older patient, many of the benefits can be realized only with operation at the earlier age.

A different approach for the total correction of tricuspid atresia.

A new surgical procedure has been used successfully to correct tricuspid atresia in a 9-year-old girl. An external conduit containing a porcine aortic valve was positioned to lead from the right atrium to the underdeveloped right ventricle. The right ventricle was reconstructed with a large Dacron patch, thus providing a large cavity for the small hypoplastic right ventricle. The atrial and ventricular septal defects were closed. The patient made a satisfactory recovery and is doing well four months after the operation.