'Molecular sandwiches' as a basis for structural and functional similarities for interferons, MSH, ACTH, LHRH, myelin basic protein, and albumins

Sequential similarities between the tryptophan peptide of myelin basic protein (residues 111-121), luteinizing hormone releasing hormone, melanotropin, adrenocorticotropin (residues 1-13), human leukocyte interferon (residues 28-40), and various segments of human and bovine serum albumin and hen ovalbumin are presented. It is suggested that these structural similarities may explain observations concerning common functional characteristics such as serotonin modulation, immunological activity with the adjuvant muramyl dipeptide, immunological cross-reactivity, and the possible MSH-ACTH-like activity of a pepsin-derived peptide of interferon.     

Isolation and properties of a liver mitochondrial precursor protein to aspartate aminotransferase expressed in Escherichia coli

The precursor to rat liver mitochondrial aspartate aminotransferase has been expressed in Escherichia coli JM105 using the pKK233-2 expression vector. This mammalian natural precursor has been isolated as a soluble dimeric protein. The amino-terminal sequence and the amino acid composition of the isolated protein correspond to those predicted from the inserted cDNA (Mattingly, J. R., Jr., Rodriguez-Berrocal, F. J., Gordon, J., Iriarte, A., and Martinez-Carrion, M. (1987) Biochem. Biophys. Res. Commun. 149, 859-865). The isolated precursor contains bound pyridoxal phosphate and shows catalytic activity with a specific activity equal to that of the mature form of the enzyme. This precursor can also be processed by mitochondria into a form with the sodium dodecyl sulfate-polyacrylamide gel electrophoresis mobility of mature enzyme. The isolation of this precursor as a stable and catalytically active entity indicates that the presequence peptide does not necessarily interfere with much of the folding and basic structural properties of the mature protein component.     

Characterization and conservation of the inner E2 core domain structure of branched-chain alpha-keto acid dehydrogenase complex from bovine liver. Construction of a cDNA encoding the entire transacylase (E2b) precursor

A cDNA clone encoding the entire transacylase (E2b) precursor of the bovine branched-chain alpha-keto acid dehydrogenase complex has been constructed from two overlapping incomplete cDNA clones which were isolated from a lambda ZAP library prepared from bovine liver poly(A)+ RNA. Nucleotide sequencing indicates that this bovine E2b cDNA insert (bE2-11) is 2701 base pairs in length with an open reading frame of 1446 base pairs. The bE2-11 cDNA insert encodes a leader peptide of 61 residues and a mature E2b polypeptide of 421 amino acid residues with a calculated monomeric molecular mass of 46,518 daltons. The molecular mass of the native E2b component isolated from bovine liver is 1,110,000 daltons as determined by sedimentation equilibrium. This value establishes the 24-subunit octahedral model for the quaternary structure of bovine E2b. The amino-terminal sequences of two tryptic fragments (A and B) of the E2b protein have been determined. Fragment A comprises residues 175 to 421 of the E2b protein and is the inner E2 core domain which contains the transacylase active site. Fragment B, produced by further tryptic cleavage of fragment, comprises residues 205 to 421, but does not have transacylase activity. Both fragments A and B confer the highly assembled 24-mer structure. The primary structure of the inner E2 core domain of bovine E2b (fragment A) is very similar to those of three other E2 proteins (human E2p, Escherichia coli E2p, and E. coli E2k). These similarities suggest that these E2 proteins are structurally and evolutionarily related.     

Internal quadruplication in the structure of human interstitial retinol-binding protein deduced from its cloned cDNA

Interstitial retinol-binding protein (IRBP) is a glycoprotein that shuttles retinoids between the retina and pigment epithelium and is secreted by the photoreceptor cells of the vertebrate eye. Human retina cDNA libraries in lambda gt10 were screened with a previously isolated human IRBP probe (H.4 IRBP), yielding five overlapping cDNA clones generating a 4223-base sequence. A 17-kilobase pair clone (HGL.3) isolated by screening a human genomic library in EMBL3 with H.4 IRBP yielded a 2.5-kilobase pair SstI fragment that overlapped the 5' end of the cDNA sequence by 329 nucleotide residues. An open reading frame encoded the N-terminal sequence of human IRBP and predicted a protein consisting of 1262 amino acids with a molecular mass of 136,600. Two putative N-linked glycosylation sites were identified. The translated sequence suggests that there is a 16-amino acid presumptive signal peptide rich in hydrophobic residues and with a high alpha-helix probability preceding the N terminus of the mature protein. The amino acid sequence of human IRBP could be aligned with 87% identity with the amino acid sequences of 31 peptides (605 residues) purified from a tryptic digest of bovine IRBP. The protein sequence of human IRBP contains four duplicated segments (302-310 residues in length) with 33-38% identity. From the degree of identity between the bovine and human sequences, it is possible that IRBP evolved by several gene duplications that occurred 600-800 million years ago, before the emergence of the vertebrates.     

On the reactivities of the tryptophan residues of human alpha-lactalbumin to 2-hydroxy-5-nitrobenzyl bromide.

The reaction of human alpha-lactalbumin with the tryptophan reagent 2-hydroxy-5-nitrobenzyl bromide has been studied. This protein has 3 tryptophan residues (Trp-60, Trp-104 and Trp-118) all of which are accessible to the reagent at pH 2.7 or 7. Trp-60 of human alpha-lactalbumin is much more reactive than Trp-60 of bovine alpha-lactalbumin (Barman, T. E. (1972) Biochim. Biophys. Acta 257, 297-313). As with bovine alpha-lactalbumin, at pH 2.7, 2-hydroxy-5-nitrobenzyl bromide is specific for tryptophan but at pH 7 His-32 also reacts. When treated with the tryptophan reagent, both alpha-lactalbumins lose their specifier protein activities in the lactose synthase (UDPgalactose:D-glucose 4-beta-galactosyltransferase, EC 2.4.1.22) reaction.     

Sequence of an intestinal cDNA encoding human motilin precursor

A cDNA clone encoding the human motilin precursor was isolated from an intestinal library using synthetic oligonucleotide probes. The predicted amino acid sequence indicates that the motilin precursor consists of 115 amino acids and includes a 25-residue N-terminal signal peptide followed by the 22-amino-acid motilin sequence and a long, 68-residue C-terminal peptide. The amino acid sequence of human motilin predicted from the cDNA sequence is identical to its porcine counterpart, which has been determined by protein sequencing. Proteolytic processing of promotilin to motilin occurs at the sequence, Lys-Lys, this being the first reported instance of processing occurring at a pair of Lys residues. In other precursors it occurs at Lys-Arg, Arg-Arg, Arg, or very rarely Lys.     

cDNA cloning, overproduction and characterization of rat adrenodoxin reductase.

We isolated a full-length cDNA clone for rat adrenodoxin reductase (AdR). The precursor of rat AdR was predicted to consist of 34 amino-terminal residues of extrapeptide for transport into mitochondria and the following 460 residues of the mature peptide region. The deduced amino acid sequence was 70.8 and 61.8% homologous to those of bovine and human AdRs in the extrapeptide region, respectively, and 88.5% homologous to both the sequences of bovine and human AdRs in the mature peptide region. The predicted mature form of rat AdR was directly expressed in Escherichia coli, using cDNA, and was purified with a yield of 32 mg/l of culture. The purified recombinant rat AdR showed an absorption spectrum characteristic of a flavoprotein with peaks at 270, 378 and 450 nm and shoulders at 280, 425 and 474 nm. The extinction coefficient was estimated to be 10.9 mM(-1) cm(-1) at 450 nm. The absorbance ratio at 270 nm/450 nm was 7.1. From the θ(208) value in the circular dichroism spectrum, the alpha-helix content in the rat AdR was calculated to be 30%. In NADPH-cytochrome c reductase activity reconstituted with adrenodoxin (Ad), the apparent K(m) value of rat AdR for NADPH was 0.32 microM, a value significantly lower than that of bovine AdR (1.4 microM). The rat AdR showed a higher affinity to the heterologous redox partner (bovine Ad, K(m)=9.3 nM) than to the native partner (rat Ad, K(m)=16.7 nM), whereas the affinity of bovine AdR was slightly higher to the native partner (bovine Ad, K(m)=37.1 nM) than to the heterologous partner (rat Ad, K(m)=46.8 nM). The K(m) values showed a reverse correlation to the difference of pI values between the redox partners. These results indicate that AdR binds to Ad mainly by ionic interaction.     

Sequence of membrane-associated thyroid hormone binding protein from bovine liver: its identity with protein disulphide isomerase

The complementary DNAs of the bovine liver membrane-associated 3,5,3'-triiodo-L-thyronine binding protein with 55 k-dalton (T3BP) were cloned and the nucleotide sequences were determined. Monospecific antibodies against T3BP were used to screen a bovine liver cDNA library in lambda gtll. We analyzed the sequences of two cloned T3BP cDNAs. The clones encoded a polypeptide of 510 amino acid residues, including a signal peptide of 20 amino acid. Northern blot analysis of bovine and human RNA showed that the mRNAs encoding T3BP are 2.7 kilobase in length. Amino acid sequence of N-terminal and other three peptides isolated from purified T3BP were found in the predicted amino acid sequence from the cDNA sequence. The predicted amino acid sequence is closely homologous (93%) with that of rat protein disulphide isomerase (EC 5.3.4.1), which catalyzes the isomerization of the protein disulphide bonds and has been shown to play an important role in post-translational regulation. The results suggest that T3BP and protein disulphide isomerase should be the same protein.     

Adrenomedullin (11-26): a novel endogenous hypertensive peptide isolated from bovine adrenal medulla.

Adrenomedullin (AM) is a potent hypotensive peptide originally isolated from pheochromocytoma tissue. Both the ring structure and the C-terminal amide structure of AM are essential for its hypotensive activity. We have developed an RIA which recognizes the ring structure of human AM. Using this RIA, we have characterized the molecular form of AM in bovine adrenal medulla. Gel filtration chromatography revealed that three major peaks of immunoreactive AM existed in the adrenal medulla. The peptide corresponding to Mr 1500 Da was further purified to homogeneity. The peptide was determined to be AM (11-26) which has one intramolecular disulfide bond. Amino acid sequences of bovine AM and its precursor were deduced from the analyses of cDNA encoding bovine AM precursor. The synthetic AM (11-26) produced dose-dependent strong pressor responses in unanesthetized rats in vivo. The hypertensive activity lasted about one minute, and a dose dependent increase in heart rate was also observed. The present data indicate that AM (11-26) is a major component of immunoreactive AM in bovine adrenal medulla and shows pressor activity.     

Photo-induced riboflavin binding to the tryptophan residues of bovine and human serum albumins

Summary The photobinding between riboflavin and the Trp residues from human and bovine serum albumins at two pH-dependent protein conformations was studied. At pH 7.0 both proteins showed photo-adduct formation with hyperbolic kinetics. In the bovine serum albumin this is attributed to the different locations of the two Trp residues. In the case of the human serum albumin, which has only one Trp residue, this behaviour may be related to different molecular conformations of the protein, as is also manifest in the iodide quenching experiments. At pH 3.5, the kinetics of the photo-adduct formation were found to be slower and showed a monophasic behaviour. These results are due to the conformational change of these proteins at acidic pH; the Trp residues of both proteins being now located in a more hydrophobic environment. When bovine serum albumin was anaerobically irradiated at pH 7.0 in the presence of¹⁴C-riboflavin and then cleaved by CNBr, two peptides were obtained, containing the Trp-134 and Trp-212 residues, respectively. The incorporation of¹⁴C-riboflavin in these samples was significantly higher at the level of the peptide containing the Trp-134 residue. Furthermore, it was demonstrated, that the energy transfer from enzymatically generated triplet acetone to riboflavin can also promote the binding of this vitamin to the Trp residues of human and bovine serum albumins.     

Isolation and characterization of the bovine corticotropin/beta-lipotropin precursor gene

The entire bovine corticotropin/beta-lipotropin precursor gene has been isolated as a set of overlapping genomic DNA fragments which extend over a length of approximately 17000 base pairs. Restriction mapping of the cloned DNA fragments and nucleotide sequence analysis of the whole mRNA-coding segments and their surrounding regions have established that the corticotropin/beta-lipotropin precursor gene is approximately 7300-base-pairs long and contains two intervening sequences; one with an approximate length of 4000 base pairs is located within the segment encoding the 5'-untranslated region of the mRNA, and the other with an approximate length of 220 base pairs interrupts the protein-coding sequence near the signal peptide region. Sequence analysis of more than 200 base pairs preceding the proximal end of the corticotropin/beta-lipotropin precursor gene has revealed a 'Hogness box' and a variant of the model sequence d(G-G-TC-C-A-A-T-C-T) as well as palindrome structures as observed in other eukaryotic genes. Furthermore, some sequence similarities in the 5'-flanking region are found between the corticotropin/beta-lipotropin precursor gene and the mouse alpha-globin and beta-globin genes, all of which are negatively regulated by glucocorticoids. At least four homologous repetitive sequences are distributed at 3000-5000-base-pair distances in the corticotropin/beta-lipotropin precursor gene region; two such sequences are located in the 5'-flanking region, and one within each intervening sequence. Blot hybridization analysis of bovine pituitary nuclear RNA has indicated that the entire corticotropin/beta-lipotropin precursor gene is transcribed into a primary hnRNA product, which is then spliced to form the mature mRNA.     

cDNA cloning of halocidin and a new antimicrobial peptide derived from the N-terminus of Ci-META4

Halocidin is an antimicrobial peptide, which is isolated from hemocytes from the tunicate, Halocynthia aurantium. In this study, we cloned the full-length cDNA of halocidin from pharyngeal tissue, using a combination of RT-PCR and 5′-RACE-PCR. The observed cDNA structure indicated that halocidin is synthesized as a 10.37 kDa prepropeptide. Based on the cDNA structure and the known amino acid sequence of the mature peptide, it was concluded that the precursor of halocidin contains a 21-residue signal peptide, followed by the 18 residues of the mature peptide, and a 56-residue anionic C-terminal extension, which is removed later on in the process. The signal sequence of halocidin exhibited a high degree of similarity with the corresponding portion of the Ci-META4 protein, which had been previously discovered in the coelomic cells of another tunicate, Ciona intestinalis, and is considered to play a role in metamorphosis. However, in several respects, the cDNA structure of Ci-META4 suggested that it might constitute a precursor for an antimicrobial peptide. Thus, we prepared a synthetic peptide, which was comprised of 19 N-terminal amino acid residues in the predicted mature region of Ci-META4, and tested it with regard to its antimicrobial activity. As a result, we confirmed that the synthetic peptide exhibited potent antimicrobial activity against Gram (+) and (−) bacteria, while evidencing no hemolytic activity toward human erythrocytes.     

Human lysosomal alpha-glucosidase. Characterization of the catalytic site.

The substrate analogue conduritol B epoxide (CBE) is demonstrated to be an active site-directed inhibitor of human lysosomal alpha-glucosidase. A competitive mode of inhibition is obtained with glycogen as natural and 4-methylumbelliferyl-alpha-D-glucopyranoside as artificial substrate. The inactivation of the enzyme is time and concentration dependent and results in the covalent binding of CBE. Catalytic activity is required for binding to occur. CBE-labeled peptides containing the catalytic residue of lysosomal alpha-glucosidase were isolated and identified by microsequencing and amino acid analysis. The peptides appeared to originate from a protein domain which is highly conserved among alpha-amylases, maltase, glucoamylases, and transglucanosylases. Based on the sequence similarity and the mechanism of CBE binding, Asp-518 is predicted to be the essential carboxylate in the active site of lysosomal alpha-glucosidase. The functional importance of Asp-518 and other residues around the catalytic site was studied by expression of in vitro mutagenized alpha-glucosidase cDNA in transiently transfected COS cells. Substitution of Asp-513 by Glu-513 is shown to interfere with the posttranslational modification and the intracellular transport of the alpha-glucosidase precursor. The residues Trp-516 and Asp-518 are demonstrated to be critical for catalytic function.     

Structure of equine corticotropin releasing factor

A 41 amino acid peptide, probably identical in structure to human corticotropin releasing factor, was isolated from 70 equine hypothalami by methanol extraction, immunoaffinity chromatography and single step of reverse phase HPLC. The amino acid sequence was determined by gas phase sequence analysis. Probable carboxyl terminal amidation was demonstrated by similar retention times for equine and human corticotropin releasing factor on reverse phase HPLC at pH 8. The likely structure of equine corticotropin releasing factor is: Ser-Glu-Glu-Pro-Pro- Ile-Ser-Leu-Asp-Leu-Thr-Phe-His-Leu-Leu-Arg-Glu-Val-Leu-Glu-Met-Ala-Arg-Ala-Glu-Gln-Leu-Ala-Gln-Gln-Ala-His-Ser-Asn- Arg-Lys-Leu-Met-Glu-Ile-Ile-NH₂. The purified peptide is equipotent with human corticotropin releasing factor in an in vitro bioassay and in a human plasma binding protein assay.     

A single gamma-carboxyglutamic acid residue in a novel cysteine-rich secretory protein without propeptide

Gamma-glutamyl carboxylase catalyzes the modification of specific glutamyl residues to gamma-carboxyglutamyl (Gla) residues in precursor proteins that possess the appropriate gamma-carboxylation recognition signal within the propeptide region. We describe the immunopurification and first biochemical characterization of an invertebrate high molecular weight Gla-containing protein with homologues in mammals. The protein, named GlaCrisp, was isolated from the venom of the marine cone snail Conus marmoreus. GlaCrisp gave intense signals in Western blot experiments employing the Gla-specific antibody M3B, and the presence of Gla was chemically confirmed by amino acid analysis after alkaline hydrolysis. Characterization of a full-length cDNA clone encoding GlaCrisp deduced a precursor containing an N-terminal signal peptide but, unlike other Gla-containing proteins, no apparent propeptide. The predicted mature protein of 265 amino acid residues showed considerable sequence similarity to the widely distributed cysteine-rich secretory protein family and closest similarity (65% identity) to the recently described substrate-specific protease Tex31. In addition, two cDNA clones encoding the precursors of two isoforms of GlaCrisp were identified. The predicted precursor isoforms differed at three amino acid positions (-6, 9, and 25). Analysis by Edman degradation and nanoelectrospray ionization mass spectrometry, before and after methyl esterfication, identified a Gla residue at amino acid position 9 in GlaCrisp. This is the first example of a Gla-containing protein without an obvious gamma-carboxylation recognition site. The results define a new class of Gla proteins and support the notion that gamma-carboxylation of glutamyl residues is phylogenetically older than blood coagulation and the vertebrate lineage.     

Characterization of human adenylate kinase 3 (AK3) cDNA and mapping of the AK3 pseudogene to an intron of the NF1 gene.

We have isolated cDNA clones for human adenylate kinase isozyme 3 (AK3) with a genomic probe from the neurofibromatosis type 1 (NF1) region. Three overlapping clones isolated from a human frontal-cortex cDNA library gave rise to a consensus sequence of 1.7 kb. The open reading frame identified in this sequence predicted a peptide of 223 residues. A database search revealed striking homology, about 58% amino acid sequence identity, between this predicted protein and bovine AK3. Human AK3 protein also showed significant homology to other members of the adenylate kinase family isolated from various species. Genomic Southern analysis suggested that multiple AK3 loci exist in the human genome, including one located in an intron of NF1 on chromosome 17. The chromosome-17 locus appears to be a processed pseudogene, since it is intronless and contains a polyadenylate tract; it nevertheless retains coding potential because the open reading frame is not impaired by any observed base substitutions.     

Molecular characterization of nuclear basic protein HPI1, a putative precursor of human sperm protamines HP2 and HP3

The largest intermediate basic protein HPI1 (101 residues) from human sperm chromatin was isolated and characterized. The amino acid composition and sequence analysis of the protein and of tryptic peptides together with peptide mapping of endoproteinases Lys-C and Glu-C hydrolysates showed that the C-terminal region (residues 45-101) of HPI1 is identical to protamine HP2. These structural data strongly suggest that protein HPI1 is a precursor of human sperm protamines HP2 and HP3 (57 and 54 residues, respectively) as well as of two other intermediate basic proteins HPS1 and HPS2 (69 and 66 residues, respectively) sequenced previously.     

Characterization of the C-terminal flanking peptide of human beta-preprotachykinin

The nucleotide sequence of cDNA encoding the common biosynthetic precursor of substance P, neurokinin A and neuropeptide K (beta-preprotachykinin) predicts that, in the human, the precursor contains a C-terminal flanking peptide of 19 amino acid residues [beta-preprotachykinin(111-129)-peptide]. Using an antiserum raised against synthetic human beta-preprotachykinin(117-126)-peptide in radioimmunoassay, we have demonstrated that an extract of a human neuroendocrine tumor of the adrenal medulla contained approximately equimolar concentrations of C-terminal preprotachykinin immunoreactivity (C-PPT-IR), substance P and neurokinin A. The C-terminal preprotachykinin flanking peptide was purified to homogeneity and its primary structure was determined. The amino acid sequence of the peptide, Ala-Leu-Asn-Ser-Val-Ala-Tyr-Glu-Arg-Ser-Ala-Met-Gln-Asn-Tyr-Glu, indicates identity with beta-preprotachykinin(111-126)-peptide. The data suggest that the C-terminal flanking peptide, like the tachykinins, is packed into secretory storage vesicles but the Arg127-Arg128-Arg129 residues in human beta-preprotachykinin are removed from the peptide by the action of endogenous processing enzyme(s).