p120, a p120-related protein (p100), and the cadherin/catenin complex.

Cadherins and catenins play an important role in cell-cell adhesion. Two of the catenins, beta and gamma, are members of a group of proteins that contains a repeating amino acid motif originally described for the Drosophila segment polarity gene armadillo. Another member of this group is a 120-kD protein termed p120, originally identified as a substrate of the tyrosine kinase pp60src. In this paper, we show that endothelial and epithelial cells express p120 and p100, a 100-kD, p120-related protein. Peptide sequencing of p100 establishes it as highly related to p120. p120 and p100 both appear associated with the cadherin/catenin complex, but independent p120/catenin and p100/catenin complexes can be isolated. This association is shown by coimmunoprecipitation of cadherins and catenins with an anti-p120/p100 antibody, and of p120/p100 with cadherin or catenin antibodies. Immunocytochemical analysis with a p120-specific antibody reveals junctional colocalization of p120 and beta-catenin in epithelial cells. Catenins and p120/p100 also colocalize in endothelial and epithelial cells in culture and in tissue sections. The cellular content of p120/p100 and beta-catenin is similar in MDCK cells, but only approximately 20% of the p120/p100 pool associates with the cadherin/catenin complex. Our data provide further evidence for interactions among the different arm proteins and suggest that p120/p100 may participate in regulating the function of cadherins and, thereby, other processes influenced by cell-cell adhesion.

Cleavage of amyloid beta peptide during constitutive processing of its precursor.

The amyloid beta peptide (A beta P) is a small fragment of the much larger, broadly distributed amyloid precursor protein (APP). Abundant A beta P deposition in the brains of patients with Alzheimer's disease suggests that altered APP processing may represent a key pathogenic event. Direct protein structural analyses showed that constitutive processing in human embryonic kidney 293 cells cleaves APP in the interior of the A beta P, thus preventing A beta P deposition. A deficiency of this processing event may ultimately prove to be the etiological event in Alzheimer's disease that gives rise to senile plaque formation.

Evidence for excitoprotective and intraneuronal calcium-regulating roles for secreted forms of the beta-amyloid precursor protein.

The beta-amyloid precursor protein (beta APP) is a membrane-spanning glycoprotein that is the source of the beta-amyloid peptide (beta AP) which accumulates as senile plaques in the brains of patients with Alzheimer's disease. beta APP is normally processed such that a cleavage occurs within the beta AP, liberating secreted forms of beta APP (APPss) from the cell. The neuronal functions of these forms are unknown. We now report that APPss have a potent neuroprotective action in cultured rat hippocampal and septal neurons and in human cortical neurons. APPs695 and APPs751 protected neurons against hypoglycemic damage, and the neuroprotection was abolished by antibodies to a specific region common to both APPs695 and APPs751. APPss caused a rapid and prolonged reduction in [Ca2+]i and prevented the rise in [Ca2+]i that normally mediated hypoglycemic damage. APPss also protected neurons against glutamate neurotoxicity, effectively raising the excitotoxic threshold. APPss may normally play excitoprotective and neuromodulatory roles. Alternative processing of APPss in Alzheimer's disease may contribute to neuronal degeneration by compromising the normal function of APPss and by promoting the deposition of beta AP.

Complementary deoxyribonucleic acid (cDNA) cloning and DNA sequence analysis of rat ovarian inhibins.

Two forms of inhibin (A and B), gonadal polypeptide hormones that selectively suppress the secretion of FSH from the anterior pituitary, have been characterized from the porcine and human species, each being composed of a common alpha-chain and one of two distinct, but homologous beta-chains, i.e. alpha beta A and alpha beta B. Using cDNAs encoding the porcine inhibin subunits we have cloned and sequenced the cDNAs encoding the alpha, beta A, and beta B chains of rat ovarian inhibin. Northern analyses of rat testicular RNA with rat ovarian cDNA probes show the presence of mRNAs encoding alpha and beta B chains, but no detectable mRNA encoding the beta A chain under our experimental conditions. This suggests that there may be specific and distinct physiological roles for inhibins A and B. In addition, if there is no extratesticular source of beta A mRNA, then the male rat may be devoid of the stimulators of the secretion of FSH, i.e. activin (beta A beta B) and homoactivin A (beta A beta A), which are derived from the beta subunits of the two inhibins.

Structural characterization of follistatin: a novel follicle-stimulating hormone release-inhibiting polypeptide from the gonad.

Follistatin, a novel, single chain, glycosylated polypeptide bearing no homology with previously characterized inhibins but exhibiting potent and specific pituitary FSH-release inhibition has been structurally characterized by protein microsequencing, cDNA cloning, and DNA sequencing. Two populations of clones differing in their 3'-untranslated sequences were found to encode a 344 amino acid precursor protein and an identical but carboxyl terminal truncated 317 amino acid precursor, respectively. Additionally, one clone, FS18, contained two introns and probably resulted from reverse transcription of heterogeneous nuclear RNA during cDNA library construction. Follistatin is unusually cysteine-rich, containing 36 cysteines in the mature coding sequence of 315 amino acids and an extremely acidic carboxyl terminal region, FS(292-304), comprised of Glu-Asp-Thr-Glu-Glu-Glu-Glu-Glu-Asp-Glu-Asp-Gln-Asp which probably resides outside a tightly cross-linked protein sphere. The heparin-binding ability of follistatin can probably be ascribed to the basic region specified by FS(75-86), Lys-Lys-Cys-Arg-Met-Asn-Lys-Lys-Asn-Lys. Overall, follistatin is organized into three homologous domains, FS(66-135), FS(139-210), and FS(216-287) containing 70, 72, and 72 amino acids, respectively, which show a 52% homology among themselves and a 57% homology with the 56 amino acid human pancreatic secretory trypsin inhibitor protein when aligned for maximum homology.

Isolation and characterization of epidermal growth factor from human milk.

Epidermal growth factor (EGF) has been purified from human milk. The purification was monitored with a human placental membrane radioreceptor assay using murine salivary epidermal growth factor I (mEGF I) as a competitive ligand and was achieved exclusively by the use of reverse-phase liquid chromatography (RPLC). The sequential use of preparative, semipreparative and analytical RPLC on an octylsilica support with solvent systems of different solute selectivity such as pyridine formate, triethylammonnium phosphate or perfluorocarbonic acids in the presence of n-propanol or acetonitrile allowed purification to homogeneity with 5 consecutive runs. The molecular mass, amino acid composition and NH2-terminal sequence of human EGF were determined. Gas-phase microsequencing of residues 1-17 revealed the following sequence: Asn-Ser-Asp-Ser-Glu-X-Pro-Leu-Ser-His-Asp-Gly-Tyr-X-Leu-X-Asp which is identical with the NH2-terminof urogastrone from human urine. The purified polypeptide competes with mEGF for the placental membrane receptor with a ki of 1 ng. Furthermore, it stimulates the anchorage-dependent as well as -independent proliferation of human and rat indicator cells with half-maximal stimulation at 1 and 2.5 ng/ml, respectively. Although human epidermal growth factor has been unequivocally identified in human milk and -for the first time-shown to be identical with urogastrone from human urine, the high-resolution techniques employed have also revealed the presence of EGF-related molecules which await further characterization. It is possible that EGF and the EGF-related growth factors possess important regulatory functions in normal growth of the human breast during pregnancy and lactation as well as in abnormal growth during mammary tumor formation and progression.

Exact cleavage site of Alzheimer amyloid precursor in neuronal PC-12 cells.

We have identified the secretory cleavage site in the Alzheimer amyloid precursor (APP) in a non-transfected neuronal cell line, using cyanogen bromide digests of APP purified from medium conditioned by PC-12 cells which were differentiated to a neuronal phenotype. The results obtained are most consistent with proteolysis of the Lys16-Leu17 bond in the beta amyloid peptide, followed by partial removal of Lys16 by a basic carboxypeptidase.

Axonal sprouting induced in the sciatic nerve by the amyloid precursor protein (APP) and other antiproteases.

Protease inhibition is the mechanism by which some trophic factors promote the extension of neurites. In the rat sciatic nerve, we assessed the ability to induce sprouts of the APP isoform that embodies the Kunitz antiprotease domain and other antiproteases. With the electron microscope, axonal sprouts were found when antiproteases were supplied but not after administration of inactive substances. We conclude that axons have a drive to sprout which can be released by the unbalance of an extracellular protease-antiprotease system. We propose that this system is involved in the pathogenesis of Alzheimer's disease.

Polypeptide microsequence analysis with the commercially available gas-phase sequencer.

An evaluation of the commercially available gas-phase sequencer with regard to its sensitivity, ease of operation, and reliability is presented. Techniques for the preparation of ultra-pure reagents and solvents for this new technology are reviewed, as is a highly sensitive reverse-phase liquid chromatography program for the identification and quantitation of phenylthiohydantoin amino acid residues from the new sequenator.

On the subunit stoichiometry of the F1-ATPase and the sites in it that react specifically with p-fluorosulfonylbenzoyl-5'-adenosine.

During the inactivation of the nucleotide-free F1-ATPase at pH 7.0, by p-fluorosulfonyl[14C]benzoyl-5'-adenosine ([14C]FSBA) in the presence of 20% glycerol, about 4.5 g atoms of 14C are incorporated/350,000 g of enzyme. Isolation of the subunits has shown: (a) over 90% of the incorporated label is associated with the alpha and beta subunits; (b) the amount of label incorporated into the alpha subunit is about 0.5 g atoms/mol which is nonspecifically associated with a number of tyrosine and lysine residues; (c) the amount of radioactivity incorporated into the beta subunit is about 0.9 g atoms/mol which correlates with the degree of inactivation of the enzyme and resides on a single tyrosine residue; (d) up to 2.2 mol of alpha subunit have been isolated from each mole of inactivated enzyme; and (e) about 2 mol of beta subunit have been isolated from each mole of inactivated enzyme. These results account for the incorporation of 4.5 g atoms of 14C which are incorporated/mol of ATPase during inactivation if there are three copies each of the alpha and beta subunit present in the enzyme. It has also been shown that 4-chloro-7-nitrobenzofurazan (NBD-Cl) and FSBA react with different tyrosine residues when they inactivate the ATPase. In addition, it has been shown that the ATPase inactivated with FSBA retains the capacity to bind up to 2.2 mol of [14C]ADP/350,000 g of enzyme.

Identification of a tyrosine residue at a nucleotide binding site in the beta subunit of the mitochondrial ATPase with p-fluorosulfonyl[14C]-benzoyl-5'-adenosine.

The mitochondrial F1-ATPase is irreversibly inactivated by the adenine nucleotide analogue, p-fluorosulfonylbenzoyl-5'-adenosine. This inactivation is partly prevented by the presence of bound adenine nucleotides. Inactivations of the ATPase with p-fluorosulfonyl[14C]benzoyl-5'-adenosine were most efficiently accomplished with the nucleotide-free enzyme at pH 7.0, in a buffer containing 20% glycerol. Under these conditions, 4.2 g atoms of 14C are incorporated per 350,000 g of enzyme when the ATPase is inactivated by 90% by its reaction with 2 mM p-fluorosulfonyl[14C]benzoyl-5'-adenosine. Isolation of the component polypeptide chains of the labeled ATPase showed that all of the radioactivity was associated with the two largest subunits. The isolated alpha subunit contained 0.45 g atom of 14C/mol and the isolated beta subunit contained 0.88 g atom of 14C/mol. Hence, the inactivation can be correlated with the incorporation of 14C into the beta subunit. This suggests that the hydrolytic site of the enzyme resides on this subunit. The majority of the radioactivity in a tryptic digest of labeled beta subunit is contained ina tryptic peptide that has the following amino acid sequence: Ile-Met-Asp-Pro-Asn-Ile-Val-Gly-Ser-Glu-His-Tyr-Asp-Val-Ala-Arg, where Tyr is the radioactive derivative of the tyrosine residue that was sulfonylated during the inactivation.

Identification of a lysine residue at a nucleotide binding site in the firefly luciferase with p-fluorosulfonyl[14C]benzoyl-5'-adenosine.

Firefly luciferase is 80-90% inactivated within 3 h upon incubation with the adenine nucleotide analogue p-fluorosulfonylbenzoyl-5'-adenosine (FSBA). Although 4 mol of 14C-FSBA/mol of enzyme is irreversibly bound during inactivation, only 1 mol of 14C-FSBA appears to be specifically directed to an adenine nucleotide binding site on the enzyme. The other 3 mol of 14C-FSBA is bound to the protein nonspecifically. The major radioactive peptide in a tryptic digest os labeled luciferase was isolated and shown to have the following amino acid sequence: *Lys-Gly-Glx-Asx-Ser-Lys, where *Lys is the radioactive derivative of the lysine residue that was sulfonylated during the inactivation.

Amino-terminal sequence analysis of alphavirus polypeptides.

The single late 26S mRNA of Semliki Forest virus (SFV) directs the synthesis of the four viral structural proteins, C, E3, E2, and E1, and the recently described nonstructural protein, 6K. We report here partial NH2-terminal amino acid sequences of the SFV polypeptides E3 and 6K and of p62, the precursor to E3 and E2. In addition, were have determined a partial NH2-terminal sequence of the Sindbis virus homolog of 6K, the 4.2K protein. p62 and E3 of SFV have identical NH2-terminal amino acid sequences. Comparison of the partial NH2-terminal sequences of 6K of SFV and 4.2K of Sindbis virus with the deduced amino acid sequence encoded by the 26S mRNA of each virus reveals that the genes for these peptides are located in each case between those for E2 and E1. The order of the genes on the 26S mRNA of the alphaviruses is therefore 5'-C-E3-E2-6K-E1-3'. We discuss two mechanisms by which the nascent viral glycoproteins may be inserted into the membrane of the endoplasmic reticulum.

cDNA cloning and sequencing of the protein-tyrosine kinase substrate, ezrin, reveals homology to band 4.1.

Ezrin is a component of the microvilli of intestinal epithelial cells and serves as a major cytoplasmic substrate for certain protein-tyrosine kinases. We have cloned and sequenced a human ezrin cDNA and report here the entire protein sequence derived from the nucleotide sequence of the cDNA as well as from partial direct protein sequencing. The deduced protein sequence indicates that ezrin is a highly charged protein with an overall pI of 6.1 and a calculated molecular mass of 69,000. The cDNA clone was used to survey the distribution of the ezrin transcript, and the 3.2 kb ezrin mRNA was found to be expressed in the same tissues that are known to express the protein and at the same relative levels. Highest expression was found in intestine, kidney and lung. The cDNA clone hybridized to DNAs from widely divergent organisms indicating that its sequence is highly conserved throughout evolution. The amino acid sequence of ezrin revealed a high degree of similarity within its N-terminal domain to the erythrocyte cytoskeletal protein, band 4.1 and secondary structure predictions indicate that a second region of ezrin contains a long alpha-helix, a feature also common to band 4.1. The structural similarity of ezrin to band 4.1 suggests a mechanism for the observed localization to the membrane, and a role for ezrin in modulating the association of the cortical cytoskeleton with the plasma membrane.

Isolation and characterization of an endogenous C-terminal fragment of the alpha-neo-endorphin/dynorphin precursor from bovine caudate nucleus.

Antibodies have been raised to a synthetic peptide corresponding to the C-terminal 15-amino acid residues of prodynorphin, the common precursor to the neo-endorphins and dynorphins. The amino acid sequence of the antigen was based on the sequence deduced from mRNA isolated and cloned from porcine hypothalamus (Kakidani, H., Y. Furutani, H. Takahashi, M. Noda, Y. Morimoto, T. Hirose, M. Asai, S. Inayama, S. Nakanishi, and S. Numa (1982) Nature 298: 245-248). Using a radioimmunoassay developed from these antibodies we have isolated an endogenous prodynorphin C-fragment from bovine caudate nucleus. The isolated peptide displayed characteristics on gel filtration similar to those of synthetic prodynorphin C-fragment predicted from the porcine mRNA sequence but had low cross-reactivity in the radioimmunoassay. Sequencing and amino acid analysis showed a substitution of serine for asparagine at position 6 in the porcine sequence. Dynorphin B (rimorphin), which is adjacent to prodynorphin C-fragment in the precursor, was isolated from the same extract. Amino acid analysis and elution position on a gel filtration column confirmed its structure as that previously characterized from bovine pituitary extracts. The release of prodynorphin C-fragment and the C-terminus of dynorphin B from the porcine precursor would require cleavage at a single arginine residue. However, a terminal arginine was not present on either of these prodynorphin peptides isolated from bovine caudate. The data would suggest that processing at a single arginine residue results in elimination of the arginine, a feature in common with processing at paired basic residues.

Characterization of a Saponaria officinalis seed ribosome-inactivating protein: immunoreactivity and sequence homologies.

A ribosome inactivating protein from Saponaria officinalis, SO-6, was purified and the N-terminus sequenced. The sequence shows extensive homology with Pokeweed antiviral protein, Pokeweed antiviral protein II, Pokeweed antiviral seed protein and dodecandrin. SDS gel electrophoresis in the Laemmli system revealed two bands of similar intensities with a smear between them, probably an artifact due to the high pI of the protein. Use of a harsher denaturing gel system resulted in one band in electrophoresis. Immune antisera was raised in rabbits against this protein and it cross reacted with other proteins (SO-5, SO-8 and SO-9) from seeds of Saponaria officinalis, but not with gelonin, Momordica charantia inhibitor and dianthin 32.

Inactivation of the bovine mitochondrial F1-ATPase with dicyclohexyl[14C]carbodiimide leads to the modification of a specific glutamic acid residue in the beta subunit.

When the bovine mitochondrial F1-ATPase is inactivated with dicyclohexyl[14C]carbodiimide and then gel-filtered, from 2 to 3 g atoms of 14C are incorporated/mol of enzyme. Prior inactivation of the enzyme by the modification of an essential tyrosine residue with 4-chloro-7-nitrobenzofurazan, a reaction that can be reversed by thiols, does not affect the irreversible inactivation of the ATPase by dicyclohexyl[14C]carbodiimide. During the large scale modification of the F1-ATPase by dicyclohexyl[14C]carbodiimide which led to 70% inactivation, 1.9 g atoms of 14C were incorporated/mol of enzyme. Isolation of the alpha, beta, and gamma subunits from this large scale inactivation revealed that the gram atoms of 14C bound per mol of each of the subunits was: alpha, 0.04; beta, 0.56; and gamma, 0.04. The majority of the radioactivity in a cyanogen bromide digest of the 14C-labeled beta subunit was isolated in a fragment that has the following amino acid sequence: Glu-Leu-Ile-Asn-Asn-Val-Ala-Lys-Ala-His-Gly-Gly-Tyr-Ser-Val-Phe-Ala-Gly-Val-Gly -Glu-Arg-Thr-Arg-Glu-Gly-Asn-Asp-Leu-Tyr-Glu*-His-Met; where Glu* represents the N gamma-glutamyl derivative of dicyclohexyl[14C]urea.

Identification of an essential glutamic acid residue in the beta subunit of the adenosine triphosphatase from the thermophilic bacterium PS3.

The TF1-ATPase from the thermophilic bacterium, PS3, is inactivated by dicyclohexylcarbodiimide (DCCD). This inactivation is stimulated by ADP and other specific nucleotides and is inhibited by Mg2+. When the inactivation is carried out with [14C]DCCD, about 2 g atoms of 14C are bound/mol of TF1 when the enzyme is nearly completely inactivated. The isolated subunits from TF1 inactivated with [14C]DCCD contain the following amounts of 14C/mol: alpha, 0.12 g atom; beta, 0.47 g atom; gamma, approximately 0.04 g atom; delta, none; and epsilon, 0.05 g atom. Fractionation of tryptic digests have shown that the 14C bound to the alpha subunit is nonspecifically associated with several peptides, and that the 14C bound to the beta subunit is associated with a single tryptic peptide with the amino acid sequence Ala-Gly-Val-Gly-Glu-Arg, where Glu represents the N-gamma-glutamyl derivative of dicyclohexyl[14C]urea.