Strategy for the mass spectrometric verification and correction of the primary structures of proteins deduced from their DNA sequences.

Fast atom bombardment mass spectrometry has been used to confirm and correct regions from the amino acid sequences of three large proteins, glutaminyl- and glycyl-tRNA synthetase from Escherichia coli and methionyl-tRNA synthetase from yeast, whose primary structures had been deduced from the base sequences of their corresponding genes. The strategy is based on a comparison of the molecular weights of the tryptic peptides predicted from all three reading frames of the gene sequences with those determined mass spectrometrically. The experimental molecular weights either match or differ and can be used to assess the correctness of the base sequences, identify errors that lead to frame shifts, premature stop codons, incorrect amino acids, etc., or identify the presence of posttranslational modifications. This method is very fast and requires little material (5-20 nmol).     

Complete amino acid sequence of chicken cartilage link protein deduced from cDNA clones.

cDNA clones coding for chicken cartilage link protein were isolated and sequenced. The DNA sequence for the entire core polypeptide of the mature link protein and the predicted signal peptide consists of 1065 nucleotides. The deduced primary translation product (355 amino acids) has a molecular mass of 40.7 kDa; the calculated molecular mass of the mature link protein core polypeptide (340 amino acids) is 39.06 kDa. The DNA sequence contains two tandemly arranged repeat sequences that may code for repeated functional domains of link protein involved in binding to hyaluronic acid. The mRNAs for chicken link protein are 6.0, 5.8, and 3.0 kilobase pairs, and the difference between the sizes of the RNA species lies in the 3' untranslated region.     

The path of murine serum amyloid A through peritoneal macrophages.

Serum amyloid A (SAA) is a family of proteins encoded by four related genes. Of the four, isoforms 1.1 and 2.1 are acute phase proteins synthesized by the liver. They become major components of the HDL plasma fraction during acute tissue injury and the HDL/SAA complex is readily taken up by macrophages. Herein we investigated the path SAA follows when presented to macrophages as HDL/SAA or in liposomes. Using antibodies specific to SAA and confocal microscopy, or EM autoradiography where only SAA is radio-labeled, we show that HDL/SAA is taken up rapidly by macrophages and within 30 min SAA, or fragments thereof, proceeds through the cytoplasm to the peri-nuclear region and then the nucleus. Within 45-60 min SAA, or fragments thereof, is found back in the cytoplasm and at the plasma membrane where it is subsequently extruded. The observation that SAA, or fragments thereof, traverse the nucleus is a novel finding and may implicate SAA in macrophage gene regulation. It also raises questions by what mechanism SAA enters and leaves the nucleus. We further investigated if both SAA isoforms traffic through the macrophage in a similar manner. Isoform differences were observed. Both isoforms bind well to the plasma membrane of macrophages at 4 degrees C, but at 37 degrees C only SAA2.1 is taken up by the cell in significant quantity, and is observed in the nucleus, suggesting that the two isoforms are handled differently and that they may have discrete physiological roles.     

cDNA sequences of two inducible T-cell genes.

We have previously described a set of human T-lymphocyte-specific cDNA clones isolated by a modified differential screening procedure. Apparent full-length cDNAs containing the sequences of 14 of the 16 initial isolates were sequenced and were found to represent five different species of mRNA; three of the five species were identical to previously reported cDNA sequences of preproenkephalin, T-cell-replacing factor, and a serine esterase, respectively. The other two species, 4-1BB and L2G25B, were inducible sequences found in mRNA from both a cytolytic T-lymphocyte and a helper T-lymphocyte clone and were not previously described in T-cell mRNA; these mRNA sequences encode peptides of 256 and 92 amino acids, respectively. Both peptides contain putative leader sequences. The protein encoded by 4-1BB also has a potential membrane anchor segment and other features also seen in known receptor proteins.     

Complete cDNA sequence of the fourth component of murine complement.

We have used the method of Okayama and Berg to construct cDNA clones that span the entire length of the fourth component of murine complement (C4) from mouse strain B10.WR. The cDNA sequence spans 5372 nucleotides and encodes a pre-pro-C4 molecule 1738 amino acids in length; it includes 56 and 99 untranslated bases at the 5' and 3' ends, respectively, of the C4 mRNA. The deduced pre-pro-C4 molecule includes a 19 amino acid signal peptide and two highly basic interchain regions that presumably are excised during maturation of the protein. The nucleotide sequence shows 79% identity with the human C4 cDNA sequence over the length of the protein-coding region. A comparison of the B10.WR C4 sequence with those of C4 and sex-limited protein (Slp) from mouse strain FM (i) suggests that the difference in hemolytic activity between the C4 proteins from B10.WR and FM is due to structural changes distant from both the C1 cleavage site and the internal thioester site and (ii) raises the possibility that the C4 gene from B10.WR has undergone genetic exchange with its adjoining Slp genes.     

Complete nucleotide sequence of cDNA and deduced amino acid sequence of rat liver catalase.

We have isolated five cDNA clones for rat liver catalase (hydrogen peroxide:hydrogen peroxide oxidoreductase, EC 1.11.1.6). These clones overlapped with each other and covered the entire length of the mRNA, which had been estimated to be 2.4 kilobases long by blot hybridization analysis of electrophoretically fractionated RNA. Nucleotide sequencing was carried out on these five clones and the composite nucleotide sequence of catalase cDNA was determined. The 5' noncoding region contained 83 bases and was followed by 1581 bases of an open reading frame that encoded 527 amino acids. The 3' noncoding region was 831 bases long and contained long repeats of the unit AC. The amino acid sequence deduced from the nucleotide sequence of the cDNAs showed about 90% homology with the reported primary structure of bovine liver catalase. The molecular weight of rat liver catalase was calculated to be 59,758 from the predicted amino acid sequence. The amino acid residues in contact with the heme group are completely identical for bovine liver and rat liver catalases. The amino acid sequence at the COOH terminus was confirmed by the results of carboxypeptidase P treatment of the protein purified from rat liver in the presence of leupeptin. Rat liver catalase has no cleavable signal peptide for translocation of the enzyme into peroxisomes.     

Carboxyl-terminal sequence of entactin deduced from a cDNA clone.

Entactin is a widely distributed basement membrane sulfated glycoprotein of approximately equal to 150 kDa. The entactin gene is expressed early in mouse embryogenesis. Two cDNA clones complementary to rat entactin mRNA were isolated by antibody screening of an oligo(dT)-primed cDNA library constructed in the lambda gt11 expression vector. One of the clones, lambda 1E, was subcloned into plasmid pBR322 and further characterized. The clone contained sequences complementary to an mRNA species 6 kilobases in length. This mRNA was translated in rabbit reticulocyte lysates to yield a polypeptide of 143 kDa that was precipitated with anti-entactin antiserum. The cDNA insert, 1328 base pairs long, was sequenced and found to contain an open reading frame of 729 base pairs that coded for 243 amino acids at the carboxyl terminus of entactin. Analysis of the peptide revealed no extended alpha-helical or beta-sheet secondary structures. Radiolabeled probes prepared by nicktranslation of p lambda 1E were used to monitor the steady-state levels of entactin mRNA in F9 embryonal carcinoma cells that were induced to differentiate by exposure to retinoic acid and dibutyryl cyclic AMP. The increase in steady-state levels of entactin mRNA lagged behind the increase in mRNA for the B2 chain of laminin, suggesting that laminin and entactin are independently rather than coordinately regulated.     

Complete primary structures of the E beta chain and gene of the mouse major histocompatibility complex.

Using the cross-hybridization with plasmid pDC beta-1, containing the cDNA coding for the DC beta chain of the human major histocompatibility complex class II molecules, we have cloned and subjected to sequence analysis both the cDNA and genomic gene for the E beta chain of the BALB/c (d haplotype) mouse. The nucleotide sequences of the cDNA and genomic DNA clones permitted us to deduce the entire primary structure of the E beta chain and the complete exon-intron structure of the E beta gene. Unlike alpha chain genes that contain five exons, the E beta gene consists of six exons corresponding to the six functional domains--the leader, beta 1 and beta 2 domains, transmembrane peptide, intracytoplasmic peptide, and 3' untranslated region. In addition, two short blocks of sequences common to alpha and beta chain genes were identified in the 5' flanking regions. We propose that these sequences are involved in the coordinate expression of alpha and beta chains.     

Two acyl-coenzyme A oxidases in peroxisomes of the yeast Candida tropicalis: primary structures deduced from genomic DNA sequence.

We report the complete nucleotide sequence of two genes encoding major peroxisomal polypeptides (PXPs) of Candida tropicalis. One, POX4, encodes PXP-4, which is the most abundant polypeptide in cells grown on oleic acid, and the other, POX5, is the gene for PXP-5. Each of the two polypeptides was found to be the subunit of a distinct long-chain acyl-coenzyme A oxidase: acyl-CoA oxidase II (PXP-4) or acyl-CoA oxidase I (PXP-5). Both the genes had no intron and gave a single open reading frame. The NH2-terminal sequences, except the initiator methionine, and the calculated molecular weights of the deduced polypeptides were consistent with those of the respective PXPs. Well-conserved sequences of 12 and 16 hydrophobic amino acids were present in the middle of the polypeptide, instead of at the NH2 terminus, and may be internal signal sequences for the peroxisomal location of PXPs. Although the two polypeptides were significantly homologous throughout their sequences, the local homologies in two regions out of five were markedly diverged from the average (63%); the homology in the second region was 93%, whereas that in the fourth one was only 24%. The implications of this finding are discussed in respect to the multiplicity of peroxisomal enzymes and the presence of multifunctional proteins in peroxisomes.     

Amino acid sequence of mammalian elongation factor 2 deduced from the cDNA sequence: homology with GTP-binding proteins.

Complementary DNA clones, pHEW1 and pRE2, coding for hamster and rat polypeptide chain elongation factor 2 (EF-2), respectively, were isolated and sequenced. It was shown that the cDNA insert in pHEW1 contains a 2574-base-pair open reading frame coding for an 857-amino acid polypeptide with Mr 95,192, excluding the initiation methionine. Comparative studies of sequence homology among EF-2 and several GTP-binding proteins show that five regions in the amino-terminal position of EF-2, corresponding to about 160 amino acids, show homology with GTP-binding proteins, including protein synthesis elongation and initiation factors, mammalian ras proteins, and transducin. The carboxyl-terminal half of EF-2 contains several regions that have 34-75% homology with bacterial elongation factor G. These results suggest that the amino-terminal region of EF-2 participates in the GTP-binding and GTPase activity whereas the carboxyl-terminal region interacts with ribosomes. Finally, the sequence provides direct evidence that diphthamide (2-[3-carboxy-amido-3-(trimethylammonio)propyl]histidine), the site of ADP-ribosylation by diphtheria toxin, is produced by post-translational modification of a histidine residue in the primary translational product.     

Epitopes of human testis-specific lactate dehydrogenase deduced from a cDNA sequence.

The sequence and structure of human testis-specific L-lactate dehydrogenase [LDHC4, LDHX; (L)-lactate: NAD+ oxidoreductase, EC 1.1.1.27] has been derived from analysis of a complementary DNA (cDNA) clone comprising the complete protein coding region of the enzyme. From the deduced amino acid sequence, human LDHC4 is as different from rodent LDHC4 (73% homology) as it is from human LDHA4 (76% homology) and porcine LDHB4 (68% homology). Subunit homologies are consistent with the conclusion that the LDHC gene arose by at least two independent duplication events. Furthermore, the lower degree of homology between mouse and human LDHC4 and the appearance of this isozyme late in evolution suggests a higher rate of mutation in the mammalian LDHC genes than in the LDHA and -B genes. Comparison of exposed amino acid residues of discrete antigenic determinants of mouse and human LDHC4 reveals significant differences. Knowledge of the human LDHC4 sequence will help design human-specific peptides useful in the development of a contraceptive vaccine.     

Bovine cone photoreceptor cGMP phosphodiesterase structure deduced from a cDNA clone.

A full-length cDNA clone encoding the alpha' subunit of cGMP phosphodiesterase (PDE) from bovine cone photoreceptors was selected by probing a retinal library with a DNA fragment encoding the catalytic core of the rod cGMP PDE alpha subunit. Identity of the clone was confirmed by comparing its deduced sequence with cone PDE peptide sequences determined by Charbonneau et al. [Charbonneau, H., Prusti, R. K., LeTrong, H., Sonnenburg, W. K., Mullaney, P. J., Walsh, K. A. & Beavo, J. A. (1990) Proc. Natl. Acad. Sci. USA, pp. 288-292]. The cone PDE alpha' and the rod PDE alpha and beta subunits are encoded by distinct genes. cGMP PDE subunits share a common ancestry with cAMP PDEs and cyclic nucleotide-binding proteins. Sequence comparisons predict the presence of a catalytic core and possible secondary sites for noncatalytic cGMP binding. The presence of a C-terminal CAAX (Cys-aliphatic-aliphatic-Xaa) motif suggests the cone enzyme may be posttranslationally modified by proteolysis, methylation, and isoprenylation.     

Complete amino acid sequence of rat brain hexokinase, deduced from the cloned cDNA, and proposed structure of a mammalian hexokinase.

The complete amino acid sequence for the type I isozyme of hexokinase from rat brain has been deduced from the nucleotide sequence of cloned cDNA. The nucleotide sequence of 91 bases in the 5' untranslated region as well as that of the entire 3' untranslated region preceding the poly(A) sequence have also been determined. The N- and C-terminal halves of brain hexokinase show extensive sequence similarity to each other and to yeast hexokinase. These results provide direct support for the proposal that the mammalian hexokinases of approximately 100 kDa have evolved by a process of duplication and fusion of a gene encoding an ancestral hexokinase similar to the yeast enzyme of approximately 50 kDa. Taking this similarity in sequence to indicate basic similarity in structure between the N- and C-terminal regions of brain hexokinase and the yeast enzyme, a proposed structure for the mammalian hexokinase has been developed by fusing two molecules of yeast hexokinase, whose structure has previously been determined by x-ray crystallographic studies. Various features of the model are shown to be consistent with experimental observations bearing on the structure of the brain enzyme.     

Amyloid formation in the rat: adenoviral expression of mouse serum amyloid A proteins.

Serum amyloid A (SAA) proteins are acute-phase apolipoproteins that are associated with high-density lipoprotein (HDL) particles: SAA proteins are precursors to secondary amyloid fibril proteins and under certain conditions of chronic or recurrent inflammation these proteins are deposited as amyloid fibrils. Of two isotypes found in mouse, SAA1.1 and SAA2.1, only SAA1.1 is deposited into amyloid. The CE/J mouse is unique, in that the only isoform identified is a hybrid between SAA1.1 and SAA2.1 and the mouse does not show amyloid deposition. In the rat, a deletion in the SAA1/SAA2 gene is associated with the absence of protein in the plasma and subsequently no amyloid deposition is detected. We have generated adenoviral vectors to study the expression of SAA proteins on HDL metabolism and amyloid formation. Injection of SAA viruses into rats resulted in expression of the mouse SAA proteins in the plasma with specific association of the SAA with HDL particles. The induction of SAA proteins was comparable to that seen in mice presented with the inflammatory agent, bacterial lipopolysaccharide (LPS). Adenoviral induced SAA levels were maintained for up to several weeks without a significant decrease in SAA expression. Injection of rats with the mouse SAA1.1 adenoviral vector, followed by amyloid enhancing factor (AEF) and silver nitrate resulted in the deposition of amyloid fibrils in the spleen. After 2 weeks, amyloid could be detected in other tissues, including the heart, liver, kidneys and lungs. When animals were injected with null or the SAA2.2 virus no amyloid was detected. These studies demonstrate that the inability of the rat to develop AA amyloid is due to the lack of synthesizing an amyloidogenic SAA protein. Furthermore, the expression of the adenoviral SAA protein from the liver and incorporation onto HDL particles further supports the hypothesis that AA amyloid is derived from circulating SAA protein. The ease of use of the adenoviral vectors and the rat provide an excellent model to study the function of SAA proteins.     

Serum amyloid A1 alleles and plasma concentrations of serum amyloid A.

Serum amyloid A1 (SAA1), the predominant isotype of acute phase SAA in plasma and the predominant precursor of fibrillar deposits in reactive amyloidosis, is encoded by a gene, for which six allelic variants have been described. Recent studies proposed that the allele SAA1.3 was positively correlated with the development of reactive amyloidosis in Japanese. This study examined whether the plasma concentration of total SAA is influenced by specific SAA1 alleles. Two hundred and eighty healthy Japanese subjects were examined to determine the allelic distribution of SAA1 and SAA2 genes by the PCR-RFLP method, and to measure the total plasma SAA concentrations. SAA concentrations were significantly higher (p < 0.001) in subjects with the allele SAA1.5 than those without it, suggesting that SAA1.5 may have a distinctive feature in the process of synthesis or catabolism. Subjects with the allele SAA1.3 had lower SAA concentrations, though not statistically significant, than those with SAA1.1. There was not significant correlation of SAA2 alleles with SAA concentrations. These results are discussed in terms of amyloidogenicity.     

The path of murine serum amyloid a through peritoneal macrophages

Serum amyloid A (SAA) is a family of proteins encoded by four related genes. Of the four, isoforms 1.1 and 2.1 are acute phase proteins synthesized by the liver. They become major components of the HDL plasma fraction during acute tissue injury and the HDL/SAA complex is readily taken up by macrophages. Herein we investigated the path SAA follows when presented to macrophages as HDL/SAA or in liposomes. Using antibodies specific to SAA and confocal microscopy, or EM autoradiography where only SAA is radio-labeled, we show that HDL/SAA is taken up rapidly by macrophages and within 30 min SAA, or fragments thereof, proceeds through the cytoplasm to the peri-nuclear region and then the nucleus. Within 45–60 min SAA, or fragments thereof, is found back in the cytoplasm and at the plasma membrane where it is subsequently extruded. The observation that SAA, or fragments thereof, traverse the nucleus is a novel finding and may implicate SAA in macrophage gene regulation. It also raises questions by what mechanism SAA enters and leaves the nucleus. We further investigated if both SAA isoforms traffic through the macrophage in a similar manner. Isoform differences were observed. Both isoforms bind well to the plasma membrane of macrophages at 4°C, but at 37°C only SAA2.1 is taken up by the cell in significant quantity, and is observed in the nucleus, suggesting that the two isoforms are handled differently and that they may have discrete physiological roles.     

Pancreatic preproglucagon cDNA contains two glucagon-related coding sequences arranged in tandem.

We have constructed and cloned in bacteria recombinant plasmids containing DNA complementary to the mRNA encoding a pancreatic preproglucagon (Mr 14,500), a product of cell-free translation of angler fish islet mRNAs shown previously by immunoprecipitation analyses to be a precursor of glucagon. cDNAs of 630, 180, and 120 base pairs were isolated and correspond to most of the mRNA for the preproglucagon (650 bases). The cDNAs contain a protein coding sequence of 372 nucleotides and 5'- and 3'-untranslated regions of 58 and 206 nucleotides, respectively. From the coding sequence of the cDNAs, we find that the sequence of glucagon, identical to mammalian glucagon in 20 of 29 positions, resides in the preproglucagon of 124 amino acids flanked by NH2- and COOH-peptide extensions of 52 and 43 amino acids, respectively. The peptide extensions are linked to the glucagon by Lys-Arg sequences characteristic of the sites that are cleaved during the posttranslational processing of prohormones. Notable is the finding that, following the initial Lys-Arg sequence in the COOH-peptide extension is a pentapeptide. Ser-Gly-Val-Ala-Glu, followed by another Lys-Arg and a sequence of 34 residues that shows striking homology with glucagon and the other peptides of the glucagon family--gastric inhibitory peptide, vasoactive intestinal peptide, and secretin. Thus, the preproglucagon mRNA contains two glucagon-related coding sequences arranged in tandem. The finding of Lys-Arg sequences flanking the glucagon and glucagon-related sequences suggests that these two peptides and a pentapeptide are formed in vivo by posttranslational cleavages of a common precursor.     

Common attributes of native-state structures of proteins, disordered proteins, and amyloid

We show that a framework derived from the common character of globular proteins can be used to understand the design of protein sequences, the behavior of intrinsically unstructured proteins, and the formation of amyloid fibrils in a unified manner. Our studies provide compelling support for the idea that protein native-state structures, the structures adopted by intrinsically unstructured proteins on binding as well as those of amyloid aggregates, all reside in a physical state of matter in which the free energy landscape is sculpted not by the specific sequence of amino acids, but rather by considerations of geometry and symmetry. We elucidate the key role played by sequence design in selecting the structure of choice from the predetermined menu of putative native-state structures.