Human fatty acid synthase: properties and molecular cloning.

Fatty acid synthase (FAS; EC 2.3.1.85) was purified to near homogeneity from a human hepatoma cell line, HepG2. The HepG2 FAS has a specific activity of 600 nmol of NADPH oxidized per min per mg, which is about half that of chicken liver FAS. All the partial activities of human FAS are comparable to those of other animal FASs, except for the beta-ketoacyl synthase, whose significantly lower activity is attributable to the low 4'-phosphopantetheine content of HepG2 FAS. We cloned the human brain FAS cDNA. The cDNA sequence has an open reading frame of 7512 bp that encodes 2504 amino acids (M(r), 272,516). The amino acid sequence of the human FAS has 79% and 63% identity, respectively, with the sequences of the rat and chicken enzymes. Northern analysis revealed that human FAS mRNA was about 9.3 kb in size and that its level varied among human tissues, with brain, lung, and liver tissues showing prominent expression. The nucleotide sequence of a segment of the HepG2 FAS cDNA (bases 2327-3964) was identical to that of the cDNA from normal human liver and brain tissues, except for a 53-bp sequence (bases 3892-3944) that does not alter the reading frame. This altered sequence is also present in HepG2 genomic DNA. The origin and significance of this sequence variance in the HepG2 FAS gene are unclear, but the variance apparently does not contribute to the lower activity of HepG2 FAS.

Eukaryotic methionyl aminopeptidases: two classes of cobalt-dependent enzymes.

Using partial amino acid sequence data derived from porcine methionyl aminopeptidase (MetAP; methionine aminopeptidase, peptidase M; EC 3.4.11.18), a full-length clone of the homologous human enzyme has been obtained. The cDNA sequence contains 2569 nt with a single open reading frame corresponding to a protein of 478 amino acids. The C-terminal portion representing the catalytic domain shows limited identity with MetAP sequences from various prokaryotes and yeast, while the N terminus is rich in charged amino acids, including extended strings of basic and acidic residues. These highly polar stretches likely result in the spuriously high observed molecular mass (67 kDa). This cDNA sequence is highly similar to a rat protein, termed p67, which was identified as an inhibitor of phosphorylation of initiation factor eIF2 alpha and was previously predicted to be a metallopeptidase based on limited sequence homology. Model building established that human MetAP (p67) could be readily accommodated into the Escherichia coli MetAP structure and that the Co2+ ligands were fully preserved. However, human MetAP was found to be much more similar to a yeast open reading frame that differed markedly from the previously reported yeast MetAP. A similar partial sequence from Methanothermus fervidus suggests that this p67-like sequence is also found in prokaryotes. These findings suggest that there are two cobalt-dependent MetAP families, presently composed of the prokaryote and yeast sequences (and represented by the E. coli structure) (type I), on the one hand, and by human MetAP, the yeast open reading frame, and the partial prokaryotic sequence (type II), on the other.

Autoantibodies to calpastatin (an endogenous inhibitor for calcium-dependent neutral protease, calpain) in systemic rheumatic diseases.

We identified an autoantibody that reacts with calpastatin [an inhibitor protein of the calcium-dependent neutral protease calpain (EC 3.4.22.17)]. In early immunoblot studies, sera from patients with rheumatoid arthritis (RA) recognized unidentified 60-, 45-, and 75-kDa proteins in HeLa cell extracts. To identify these autoantigens, we used patient sera to clone cDNAs from a lambda gt11 expression library. We isolated clones of four genes that expressed fusion proteins recognized by RA sera. The 1.2-kb cDNA insert (termed RA-6) appeared to encode a polypeptide corresponding to the 60-kDa antigen from HeLa cells, since antibodies bound to the RA-6 fusion protein also reacted with a 60-kDa HeLa protein. The deduced amino acid sequence of the RA-6 cDNA was completely identical with the C-terminal 178 amino acids of human calpastatin except for one amino acid substitution. Patient sera that reacted with the RA-6 also bound pig muscle calpastatin, and a monoclonal antibody to human calpastatin recognized the RA-6 fusion protein, confirming the identity of RA-6 with calpastatin. Moreover, the purified RA-6 fusion protein inhibited the proteolytic activity of calpain, and IgG from a serum containing anti-calpastatin antibodies blocked the calpastatin activity of the RA-6 fusion protein. Immunoblots of the RA-6 product detected autoantibodies to calpastatin in 57% of RA patients; this incidence was significantly higher than that observed in other systemic rheumatic diseases, including systemic lupus erythematosus (27%), polymyositis/dermatomyositis (24%), systemic sclerosis (38%), and overlap syndrome (29%). Thus, anti-calpastatin antibodies are present most frequently in patients with RA and may participate in pathogenic mechanisms of rheumatic diseases.

Overexpression of DR-nm23, a protein encoded by a member of the nm23 gene family, inhibits granulocyte differentiation and induces apoptosis in 32Dc13 myeloid cells.

Chronic myelogenous leukemia evolves in two clinically distinct stages: a chronic and a blast crisis phase. The molecular changes associated with chronic phase to blast crisis transition are largely unknown. We have identified a cDNA clone, DR-nm23, differentially expressed in a blast-crisis cDNA library, which has approximately 70% sequence similarity to the putative metastatic suppressor genes, nm23-H1 and nm23-H2. The deduced amino acid sequence similarity to the proteins encoded by these two latter genes is approximately 65% and includes domains and amino acid residues (the leucine zipper-like and the RGD domain, a serine and a histidine residue in the NH2- and in the COOH-terminal portion of the protein, respectively) postulated to be important for nm23 function. DR-nm23 mRNA is preferentially expressed at early stages of myeloid differentiation of highly purified CD34+ cells. Its constitutive expression in the myeloid precursor 32Dc13 cell line, which is growth-factor dependent for both proliferation and differentiation, results in inhibition of granulocytic differentiation induced by granulocyte colony-stimulating factor and causes apoptotic cell death. These results are consistent with a role for DR-nm23 in normal hematopoiesis and raise the possibility that its overexpression contributes to differentiation arrest, a feature of blastic transformation in chronic myelogenous leukemia.

Characterization of a voltage-gated K+ channel beta subunit expressed in human heart.

Voltage-gated K+ channels are important modulators of the cardiac action potential. However, the correlation of endogenous myocyte currents with K+ channels cloned from human heart is complicated by the possibility that heterotetrameric alpha-subunit combinations and function-altering beta subunits exist in native tissue. Therefore, a variety of subunit interactions may generate cardiac K+ channel diversity. We report here the cloning of a voltage-gated K+ channel beta subunit, hKv beta 3, from adult human left ventricle that shows 84% and 74% amino acid sequence identity with the previously cloned rat Kv beta 1 and Kv beta 2 subunits, respectively. Together these three Kv beta subunits share > 82% identity in the carboxyl-terminal 329 aa and show low identity in the amino-terminal 79 aa. RNA analysis indicated that hKv beta 3 message is 2-fold more abundant in human ventricle than in atrium and is expressed in both healthy and diseased human hearts. Coinjection of hKv beta 3 with a human cardiac delayed rectifier, hKv1.5, in Xenopus oocytes increased inactivation, induced an 18-mV hyperpolarizing shift in the activation curve, and slowed deactivation (tau = 8.0 msec vs. 35.4 msec at -50 mV). hKv beta 3 was localized to human chromosome 3 by using a human/rodent cell hybrid mapping panel. These data confirm the presence of functionally important K+ channel beta subunits in human heart and indicate that beta-subunit composition must be accounted for when comparing cloned channels with endogenous cardiac currents.

Periodicity of polar and nonpolar amino acids is the major determinant of secondary structure in self-assembling oligomeric peptides.

The tendency of a polypeptide chain to form alpha-helical or beta-strand secondary structure depends upon local and nonlocal effects. Local effects reflect the intrinsic propensities of the amino acid residues for particular secondary structures, while nonlocal effects reflect the positioning of the individual residues in the context of the entire amino acid sequence. In particular, the periodicity of polar and nonpolar residues specifies whether a given sequence is consistent with amphiphilic alpha-helices or beta-strands. The importance of intrinsic propensities was compared to that of polar/nonpolar periodicity by a direct competition. Synthetic peptides were designed using residues with intrinsic propensities that favored one or the other type of secondary structure. The polar/nonpolar periodicities of the peptides were designed either to be consistent with the secondary structure favored by the intrinsic propensities of the component residues or in other cases to oppose these intrinsic propensities. Characterization of the synthetic peptides demonstrated that in all cases the observed secondary structure correlates with the periodicity of the peptide sequence--even when this secondary structure differs from that predicted from the intrinsic propensities of the component amino acids. The observed secondary structures are concentration dependent, indicating that oligomerization of the amphiphilic peptides is responsible for the observed secondary structures. Thus, for self-assembling oligomeric peptides, the polar/nonpolar periodicity can overwhelm the intrinsic propensities of the amino acid residues and serves as the major determinant of peptide secondary structure.

A chimeric light-regulated amino acid transport system allows the isolation of blue light regulator (blr) mutants of Neurospora crassa.

We have developed a system for the isolation of Neurospora crassa mutants that shows altered responses to blue light. To this end we have used the light-regulated promoter of the albino-3 gene fused to the neutral amino acid permease gene mtr. The product of the mtr gene is required for the uptake of neutral aliphatic and aromatic amino acids, as well as toxic analogs such as p-flurophenylalanine or 4-methyltryptophan. mtr trp-2-carrying cells were transformed with the al-3 promoter-mtr wild-type gene (al-3p-mtr+) to obtain a strain with a light-regulated tryptophan uptake. This strain is sensitive to p-fluorophenylalanine when grown under illumination and resistant when grown in the dark. UV mutagenesis of the al-3p-mtr(+)-carrying strain allowed us to isolate two mutant strains, BLR-1 and BLR-2 (blue light regulator), that are light-resistant to p-fluorophenylalanine and have lost the ability to grow on tryptophan. These two strains have a pale-orange phenotype and show down-regulation of all the photoregulated genes tested (al-3, al-1, con-8, and con-10). Mutations in the BLR strains are not allelic with white collar 1 or white collar 2, regulatory genes that are also involved in the response to blue light.

Age-related learning deficits in transgenic mice expressing the 751-amino acid isoform of human beta-amyloid precursor protein.

The beta-amyloid precursor protein (beta-APP), from which the beta-A4 peptide is derived, is considered to be central to the pathogenesis of Alzheimer disease (AD). Transgenic mice expressing the 751-amino acid isoform of human beta-APP (beta-APP751) have been shown to develop early AD-like histopathology with diffuse deposits of beta-A4 and aberrant tau protein expression in the brain, particularly in the hippocampus, cortex, and amygdala. We now report that beta-APP751 transgenic mice exhibit age-dependent deficits in spatial learning in a water-maze task and in spontaneous alternation in a Y maze. These deficits were mild or absent in 6-month-old transgenic mice but were severe in 12-month-old transgenic mice compared to age-matched wild-type control mice. No other behavioral abnormalities were observed. These mice therefore model the progressive learning and memory impairment that is a cardinal feature of AD. These results provide evidence for a relationship between abnormal expression of beta-APP and cognitive impairments.

Genetic and biochemical dissection of protein linkages in the cadherin-catenin complex.

The cadherin-catenin complex is important for mediating homotypic, calcium-dependent cell-cell interactions in diverse tissue types. Although proteins of this complex have been identified, little is known about their interactions. Using a genetic assay in yeast and an in vitro protein-binding assay, we demonstrate that beta-catenin is the linker protein between E-cadherin and alpha-catenin and that E-cadherin does not bind directly to alpha-catenin. We show that a 25-amino acid sequence in the cytoplasmic domain of E-cadherin and the amino-terminal domain of alpha-catenin are independent binding sites for beta-catenin. In addition to beta-catenin and plakoglobin, another member of the armadillo family, p120 binds to E-cadherin. However, unlike beta-catenin, p120 does not bind alpha-catenin in vitro, although a complex of p120 and endogenous alpha-catenin could be immunoprecipitated from cell extracts. In vitro protein-binding assays using recombinant E-cadherin cytoplasmic domain and alpha-catenin revealed two catenin pools in cell lysates: an approximately 1000- to approximately 2000-kDa complex bound to E-cadherin and an approximately 220-kDa pool that did not contain E-cadherin. Only beta-catenin in the approximately 220-kDa pool bound exogenous E-cadherin. Delineation of these molecular linkages and the demonstration of separate pools of catenins in different cell lines provide a foundation for examining regulatory mechanisms involved in the assembly and function of the cadherin-catenin complex.

Replication factor encoded by a putative oncogene, set, associated with myeloid leukemogenesis.

DNA replication of the adenovirus genome complexed with viral core proteins is dependent on the host factor designated template activating factor I (TAF-I) in addition to factors required for replication of the naked genome. Recently, we have purified TAF-I as 39- and 41-kDa polypeptides from HeLa cells. Here we describe the cloning of two human cDNAs encoding TAF-I. Nucleotide sequence analysis revealed that the 39-kDa polypeptide corresponds to the protein encoded by the set gene, which is the part of the putative oncogene associated with acute undifferentiated leukemia when translocated to the can gene. The 41-kDa protein contains the same amino acid sequence as the 39-kDa protein except that short N-terminal regions differ in both proteins. Recombinant proteins, which were purified from extracts of Escherichia coli, expressing the proteins from cloned cDNAs, possessed TAF-I activities in the in vitro replication assay. A particular feature of TAF-I proteins is the presence of a long acidic tail in the C-terminal region, which is thought to be an essential part of the SET-CAN fusion protein. Studies with mutant TAF-I proteins devoid of this acidic region indicated that the acidic region is essential for TAF-I activity.

Proline-rich sequences that bind to Src homology 3 domains with individual specificities.

To study the binding specificity of Src homology 3 (SH3) domains, we have screened a mouse embryonic expression library for peptide fragments that interact with them. Several clones were identified that express fragments of proteins which, through proline-rich binding sites, exhibit differential binding specificity to various SH3 domains. Src-SH3-specific binding uses a sequence of 7 aa of the consensus RPLPXXP, in which the N-terminal arginine is very important. The SH3 domains of the Src-related kinases Fyn, Lyn, and Hck bind to this sequence with the same affinity as that of the Src SH3. In contrast, a quite different proline-rich sequence from the Btk protein kinase binds to the Fyn, Lyn, and Hck SH3 domains, but not to the Src SH3. Specific binding of the Abl SH3 requires a longer, more proline-rich sequence but no arginine. One clone that binds to both Src and Abl SH3 domains through a common site exhibits reversed binding orientation, in that an arginine indispensable for binding to all tested SH3 domains occurs at the C terminus. Another clone contains overlapping yet distinct Src and Abl SH3 binding sites. Binding to the SH3 domains is mediated by a common PXXP amino acid sequence motif present on all ligands, and specificity comes about from other interactions, often ones involving arginine. The rules governing in vivo usage of particular sites by particular SH3 domains are not clear, but one binding orientation may be more specific than another.

A cyclophilin from the polycentric anaerobic rumen fungus Orpinomyces sp. strain PC-2 is highly homologous to vertebrate cyclophilin B.

A cyclophilin (CyP) purified to homogeneity from the polycentric anaerobic rumen fungus Orpinomyces sp. strain PC-2 had a molecular mass of 20.5 kDa and a pI of 8.1. The protein catalyzed the isomerization of the prolyl peptide bond of N-succinyl-Ala-Ala-(cis,trans)-Pro-Phe p-nitroanilide with a kcat/Km value of 9.3 x 10(6) M-1.s-1 at 10 degrees C and pH 7.8. Cyclosporin A strongly inhibited this peptidylprolyl cis-trans isomerase activity with an IC50 of 19.6 nM. The sequence of the first 30 N-terminal amino acids of this CyP had high homology with the N-terminal sequences of other eukaryotic CyPs. By use of a DNA hybridization probe amplified by PCR with degenerate oligonucleotide primers designed based on the amino acid sequences of the N terminus of this CyP and highly conserved internal regions of other CyPs, a full-length cDNA clone was isolated. It possessed an open reading frame encoding a polypeptide of 203 amino acids with a calculated molecular weight of 21,969, containing a putative hydrophobic signal peptide sequence of 22 amino acids preceding the N terminus of the mature enzyme and a C-terminal sequence, Lys-Ala-Glu-Leu, characteristic of an endoplasmic reticulum retention signal. The Orpinomyces PC-2 CyP is a typical type B CyP. The amino acid sequence of the Orpinomyces CyP exhibits striking degrees of identity with the corresponding human (70%), bovine (69%), mouse (68%), chicken (66%), maize (61%), and yeast (54%) proteins. Phylogenetic analysis based on the CyP sequences indicated that the evolutionary origin of the Orpinomyces CyP was closely related with CyPs of animals.