Search

The amino acid sequence of bovine heart mitochondrial coupling factor 6 (F6) has been determined by automated Edman degradation of the whole protein and derived peptides. Preparations based on heat precipitation and ethanol extraction showed allotypic variation at three positions while material further purified by HPLC yielded only one sequence that also differed by a Phe-Thr replacement at residue 62. The mature protein contains 76 amino acids with a calculated molecular weight of 9006 and a pI of approximately equal to 5, in good agreement with experimentally measured values. The charged amino acids are mainly clustered at the termini and in one section in the middle; these three polar segments are separated by two segments relatively rich in nonpolar residues. Chou-Fasman analysis suggests three stretches of alpha-helix coinciding (or within) the high-charge-density sequences with a single beta-turn at the first polar-nonpolar junction. Comparison of the F6 sequence with those of other proteins did not reveal any homologous structures.

The amino acid sequence of porcine heart mitochondrial malate dehydrogenase (mMDH; L-malate: NAD+ oxidoreductase, EC 1.1.1.37) has been compared with the sequences of six different lactate dehydrogenases (LDH; L-lactate: NAD+ oxidoreductase, EC 1.1.1.27) and with the "x-ray" sequence of cytoplasmic malate dehydrogenase (sMDH). The main points are that (i) all three enzymes are homologous; (ii) invariant residues in the catalytic center of these enzymes include a histidine and an internally located aspartate that function as a proton relay system; (iii) numerous residues important to coenzyme binding are conserved, including several glycines and charged residues; and (iv) amino acid side chains present in the subunit interface common to the MDHs and LDHs appear to be better conserved than those in the protein interior. It is concluded that LDH, sMDH, and mMDH are derived from a common ancestral gene and probably have similar catalytic mechanisms.

The change in cell surface adhesive specificity previously shown to occur between day 7 and 8 of development in the chick optic tectum ]Gottlieb et al. (1974), Proc. Nat. Acad. Sci. USA 71, 1800-1802] can be induced in rotating cultures of tectal cells by the addition of 10(-7) M mouse submaxillary gland nerve growth factor. Insulin, proinsulin, dexamethasone, and performic-acid-oxidized nerve growth factor are individually inactive in this system, but nerve growth factor in which the three tryptophan residues/subunit have been oxidized with N-bromosuccinimide is active. Thus, the specificity of this system for nerve growth factor is different than that observed with embryonic dorsal root or sympathetic ganglia, where the oxidized tryptophan derivative is inactive in stimulating neurite production. It is possible that in this system nerve growth factor serves as an analog of another specific trophic factor, presumably structurally related to nerve growth factor, may be active at much lower concentrations.

The complete amino acid sequence of the Escherichia coli alkaline phosphatase subunit [orthophosphoric-monoester phosphohydrolase (alkaline optimum), EC 3.1.3.1, isozyme 3] has been determined. The monomer contains 449 amino acid residues in a single unglycosylated polypeptide chain having a calculated Mr of 47,029. Isozyme 1 has an additional arginine residue at the NH2 terminus that presumably results from variability in processing of precursor molecules. Sequence data were obtained from both manual and automatic Edman degradation of the tryptic and cyanogen bromide peptides, as well as other peptides derived therefrom. The two disulfide bonds were determined from analyses of the appropriate peptic peptides. This structure confirms earlier reports of the sequence surrounding the active-site serine and both the NH2- and COOH-terminal cyanogen bromide fragments. A secondary structure prediction places nearly half the residues in alpha-helical segments that have 13% and 16%, respectively, in beta-strand and beta-turn orientations.

Two classes of receptors for 125I-labeled nerve growth factor in chick embryonic dorsal root neurons have been observed. One type is associated with the plasma membrane (or microsomal fraction) and can be completely solubilized by Triton X-100. These receptors display the nonsaturable binding isotherms and curvilinear Scatchard plots previously reported for nerve growth factor receptors in whole cells. The second class of binding sites is located in the nucleus, firmly bound to chromatin. These receptors are not solubilized by detergent, show saturable binding, and yield linear Scatchard plots of the type associated with a single class of binding sites of high affinity. The presence of the two receptor types suggests a bimodal mechanism of action for nerve growth factor.

The amino-acid sequence and three-dimensional structure of a calcium-binding protein prepared from carp muscle has been determined. This protein, designated carp-muscle calcium-binding protein B, is one of three closely related parvalbumins found in this tissue. The electron density map, calculated by heavyatom substitution crystallographic methods to 2.0-Å resolution, reveals the orientation of most of the amino-acid side chains. The calcium coordination site consists of one glutamic- and three aspartic-acid carboxyl groups in a tetrahedral arrangement. The core of this spherical molecule is remarkably hydrophobic, with 8 of its 10 phenylalanine side chains packed in an approximate herringbone pattern. 52 of the 108 residues are in six α-helixes; there is no β-pleated sheet. The acetylated amino-terminal alanine appears not to be accessible to solvent. All of the heavy-atom derivatives are bound at the sole cysteine. The properties of this protein suggest a relationship to troponin A of mammalian tissue.

An argument is presented in favor of the hypothesis that, bovine carboxypeptidases A and B fit by and large, the concept of homology on the levels of both amino acid sequence and three-dimensional structure. Two peptide sequences from carboxypeptidases A and B, comprising 46 amino acid residues, display structural homology to the extent of 51 per cent. One of the two peptides contains two of the amino acid residues which have been identified by X-ray crystallography as zinc ligands in carboxypeptidase A. The same residues are conserved in carboxypeptidase B in a loop of comparable dimensions. The other homologous pair of amino acid sequences appears to occur in different regions of the two enzymes which include in carboxypeptidase B the cysteinyl residue proposed to function as the third zinc ligand but not the amino acid residue believed to be the analogous ligand in carboxypeptidase A. The uncertainties of this prediction are considered in the light of existing chemical and structural evidence.