Computer-aided design of a catalyst for Edman degradation utilizing substrate-assisted catalysis

Molecular biology has been revolutionized by the miniaturization and parallelization of DNA sequencing assays previously performed on bulk samples. Many of these technologies rely on biomolecular reagents to facilitate detection, synthesis, or labeling of samples. To aid in the construction of analogous experimental approaches for proteins and peptides, we have used computer-aided design to engineer an enzyme capable of catalyzing the cleavage step of the Edman degradation. We exploit the similarity between the sulfur nucleophile on the Edman reagent and the catalytic cysteine in a naturally occurring protease to adopt a substrate-assisted mechanism for achieving controlled, step-wise removal of N-terminal amino acids. The ability to expose amino acids iteratively at the N-terminus of peptides is a central requirement for protein sequencing techniques that utilize processive degradation of the peptide chain. While this can be easily accomplished using the chemical Edman degradation, achieving this activity enzymatically in aqueous solution removes the requirement for harsh acid catalysis, improving compatibility with low adsorption detection surfaces, such as those used in single molecule assays.     

Ring cleavage of 2-iminothiazolidine-4-carboxylates by catalytic reduction. A potential method for unblocking peptides formed by specific chemical cleavage at half-cystine residues

Specific chemical cleavage of proteins at S-cyanocysteine residues leads to formation of iminothiazolidinecarboxylyl peptides which, because their amino termini are blocked, are not susceptible to Edman degradation. A catalyst prepared from NiCl₂ and NaBH₄ converts 2-iminothiazolidine-4-carboxylate to alanine and iminothiazolidine carboxylylglycine to Ala-Gly in good yield. Unmodified proteins treated with this catalyst in 8M guanidinium chloride are recovered in good yield, with quantitative conversion of methionine to amino-butyrate and half-cystine to alanine by desulfuration. The catalyst also induces cleavage of a small subclass of peptide bonds, probably Phe-Thr and Phe-Ser sequences, producing discrete fragments.     

Partial structural analysis of a highly basic low molecular weight protein from rat testis

Automated Edman degradation of a testis-specific basic protein isolated from the rat gave the following NH₂-terminal sequence of amino acids:

Cleavage of the native protein with cyanogen bromide produced two fragments which were purified by gel filtration. Amino acid analysis of the smaller fragment revealed it to be the NH₂-terminal undecapeptide resulting from cleavage at Met₁₁. The partial sequence analysis of the intact protein coupled with compositional analyses of these cyanogen bromide peptides indicate that the basic testis protein contains 24 basic amino acids and a single methionine in a sequence of 54 amino acids.     

Crotalus adamanteus phospholipase A2-α: Subunit structure,NH2-terminal sequence,and homology with other phospholipases

Automated Edman degradation of reduced and carboxymethylated phospholipase A₂-α from Crotalus adamanteus venom revealed a single amino acid sequence extending 30 residues into the protein from the amino terminus. The singularity of the sequence and the yields of the phenylthiohydantoin amino acids thus obtained indicate that the subunits comprising the phospholipase dimer are identical. Further chemical evidence in support of subunit identity was obtained by cleavage of phospholipase A₂-α with cyanogen bromide. Compositional analysis of the protein revealed one residue of methionine per monomer and the sequence determination placed this amino acid at position 10 in the sequence of 133 amino acids. Cyanogen bromide cleavage of the protein, followed by reduction and carboxymethylation afforded the expected 2 fragments: an NH₂-terminal decapeptide (CNBr-1) and a larger COOH-terminal fragment of 123 residues (CNBr-II). Automated Edman degradation of the latter has extended the sequence analysis to 54 residues in the NH₂-terminal segment of the monomer chain. Comparison of this sequence with those derived for phospholipases from other snake venoms, from bee venom, and from porcine pancreas has revealed striking homologies in this region of the molecules. As expected on the basis of their phylogenetic classification, the phospholipases from the pit vipers C. adamanteus and Agkistrodon halys blomhoffii are more similar to one another in sequence than to the enzyme from the more distantly related viper, Bitis gabonica. Furthermore, the very close similarities in sequence observed among all of these phospholipases in regions corresponding to residues 24 through 53 in the C. adamanteus enzyme suggest that this segment of the polypeptide plays an important role in phospholipase function and probably constitutes part of the active site.     

Amino acid sequence of p15 from avian myeloblastosis virus complex

The complete amino acid sequence of the p15 gag protein from avian myeloblastosis virus (AMV) complex has been determined by sequential Edman degradation of the intact molecule and of peptide fragments generated by limited tryptic cleavage, cleavage with staphylococcal protease, and cyanogen bromide cleavage. AMV p15 is a single-chain protein containing 124 amino acids. The charged amino acids tend to be clustered in the primary structure. p15 contains a single cysteine at position 113 which may be essential for the p15 associated proteolytic activity. However, p15 shows no appreciable sequence homology with papain or other classical thiol proteases.     

Fluorescence detection of amino acids in the postcleavage conversions for manual sequencing of a peptide

A modified Edman degradation method where fluorescent derivatives of amino acids were generated from the postcleavage products of a peptide is described. In the method, the target peptide was applied onto double glass fiber membranes in a small filter disk (4 mm i.d.) and then treated with small amounts of reagents for the manual sequencing of the peptide. The anilinothiazolinone (ATZ) of N-terminus amino acid residue after the isolation from the solid-phase membranes was reacted with a primary amine, 4-(1′-cyanoisoindolyl)aniline (CIA), to form a more stable and sensitive fluorescent derivative, phenylthiocarbamoyl-CIA. An average yield of 85% was obtained in neutral pH conditions for the CIA reaction. The ATZ-CIA-amino acids were separated by reversed-phase liquid chromatography and detected by fluorometry. The lower limits of the detection for amino acids after the Edman degradation were 0.16 to 0.52 pmol (signal/noise ratio = 3) on the column. The sensitivity was approximately 10 times higher than ultraviolet absorbance detection of phenylthiohydantoin products in the conventional Edman degradation. The suitability of the method was demonstrated by the sensitive manual sequencing of insulin chain B composed of 30 amino acids.     

Studies on the structure of rabbit muscle aldolase. 3. Primary structure of the BrCN peptide containing the active site

The four peptide segments obtained from rabbit muscle aldolase by cleavage with BrCN and separation with gel-filtration chromatography (1) have been redesignated according to their positions in the molecule, N-A-B-C. The primary structure of segment A, containing 66 amino acid residues, including the Schiff base-forming lysine at the active site, has been elucidated by isolation and sequence analyses of the proteolytic subfragments. Preliminary separation of tryptic peptides containing 7–25 residues was achieved by chromatography on Sephadex G-25 which facilitated subsequent purification. For the study of the tryptic peptide of 25 residues further fragmentation with pepsin then subtilisin (Nagarse) was employed. Edman degradation directly after subtilisin cleavage of a peptide was found useful in avoiding deamidation of a glutamine NH₂-terminus newly formed in the proteolysis. The sequence of 90 amino acids in the center region of the polypeptide chain of rabbit muscle aldolase has now been established.     

Amino acid sequence of ferredoxin II from the phototroph Rhodospirillum rubrum: Characteristics of a 7Fe ferredoxin

The complete sequence of amino acids of ferredoxin II (FdII) from Rhodospirillum rubrum was determined by repetitive Edman degradation using pyridylethylated-ferredoxin and oxidized, denatured ferredoxin. Peptides derived from trypsin, pepsin, Glu-C endoproteinase, Arg-C endoproteinase, tryptophan specific cleavage and partial acid hydrolysis and C-terminal sequence from carboxypeptidase digestion were used to construct the total sequence. RrFdII is a polypeptide of 104 amino acids having a calculated molecular weight of 11556 excluding the iron and sulfur atoms. The complete amino acid sequence was: PYVVTENCIKCKYQDCVEVCPVDCFYEGENFLVINPDECIDCGVCNPECPAEAIAGKWLEINRKFADLWPNITRKGPAL ADADDWKDKPDKTGLLSENPGKGTV. Sequence comparisons, EPR characteristics and iron analyses indicate that RrFdII has structural features in common with ferredoxins containing [3Fe-4S], [4Fe-4S] centers. Of 104 amino acids, 60 (58%) including all 9 cysteines, are found in identical locations in the 7Fe ferredoxin prototype, Azotobacter vinelandii FdI.The protein sequence data reported in this paper will appear in the SWISS-PROT database and EMBL Data Library under the accession number P80448.     

Amino acid sequence of the Pseudomonas putida cytochrome P-450. I. Sequences of tryptic and clostripain peptides

In order to elucidate the complete amino acid sequence of Pseudomonas putida cytochrome P-450, tryptic digestion was performed on the S-carboxymethylated enzyme. Although cleavage did not occur at every lysyl and arginyl bond, 31 tryptic peptides ranging in size from 1 to 55 residues were isolated. These were sequenced by manual Edman degradation and carboxypeptidase digestion. Overlaps of some od these tryptic peptides were obtained by data obtained from partial Edman degradation and amino acid composition of the clostripain cleavage products. These results, together with data from the cyanogen bromide and acid cleavage peptides reported in the accompanying paper, established the complete amino acid sequence of P. putida cytochrome P-450.     

Lysosomal degradation of Asn-linked glycoproteins

Catabolism of Asn-linked glycoproteins to monosaccharides and amino acids occurs in lysosomes. Break-down must be complete to avoid lysosomal storage diseases that occur when fragments as small as dimers are left undigested. Recent results have clarified several aspects of Asn-linked glycoprotein catabolism in mammals. First, degradation of the oligosaccharide portion is accomplished by exo-glycosidases, which act only from the nonreducing end of chains to release sugar monomers as products. In contrast, proteolysis can proceed from both end and internal points along the polypeptide to eventually yield free amino acids. A second important feature of the glycoprotein disassembly pathway is that the hydrolytic steps can be grouped into two sets of ordered reactions: I) stepwise hydrolysis of the major portion of the oligosaccharide chains by a set of exoglycosidases, and II) ordered disassembly of the protein and the oligosaccharide-to-protein linkage region. Process II can vary at a single reaction step depending on the species in which degradation takes place. Thus, the last step of reaction sequence II can be either: 1) hydrolysis of the actual peptide-to-carbohydrate linkage, or 2) removal of the reducing-end GlcNAc from a previously freed oligosaccharide. The latter cleavage is catalyzed by the lysosomal glycosidase chitobiase. Chitobiase has been found only in humans and rats and not in other mammals (dogs, cats, goats, sheep, cats, or cattle). The hydrolytic mechanism of this enzyme is unique as it appears to be a reducing-end glycosidase and can be viewed as an accessory step in the human and rat digestive pathways. The species that lack this enzyme likely rely on exo-beta-D-glucosaminidase to cleave GlcNAc from both outer chain residues and the chitobiose moiety at the protein-to-carbohydrate linkage.     

High-performance liquid chromatography of side-chain-protected amino acid phenylthiohydantoins

Eighteen side-chain-protected amino acids, routinely employed in solid-phase peptide synthesis, were derivatized to their phenylthiohydantoins (PTH) by one cycle of the Edman degradation. All of these side-chain-protected PTH amino acids elute, with almost-baseline resolution, in less than 18 min by high-performance liquid chromatography, utilizing a biphasic gradient of acetonitrile in 0.01 n sodium acetate, pH 4.5, or a linear gradient of 0 to 100% acetonitrile with the exception of the coelution of a O-benzyl-threonine and carbobenzoxy-lysine phenylthiohydantoin amino acids. The derivatized amino acids were subjected to reverse-phase chromatography on a Zorbax ODS column and monitored at 254 nm. None of the PTH amino acids coelute with side-chain-protected PTH amino acid counterparts, although PTH-tosyl-histidine undergoes deprotection to PTH-histidine in the Edman degradation. A protected decapeptide attached to a chloromethylated polystyrene resin was degraded on a solid-phase sequencer in 16 h. The PTH amino acids resulting from the automated Edman degradation on the decapeptide were fully resolved and quantified in less than 3 h demonstrating that automated high-performance liquid chromatography can keep pace with both the automated sequencer and synthesizer which requires minimally 2–3 h for attachment of each residue to the growing peptide chain.     

The nucleotide sequence of the aroF gene of Escherichia coli and the amino acid sequence of the encoded protein, the tyrosine-sensitive 3-deoxy-D-arabino-heptulosonate 7-phosphate synthase

The translated sequence of aroF, the first structural gene of the tyrosine operon of Escherichia coli, has been determined. The 1068 nucleotides encode the 356 amino acids that form the subunit of the dimeric tyrosine-sensitive 3-deoxy-D-arabino-heptulosonate 7-phosphate synthase. The primary structure of this enzyme has been confirmed by automated Edman degradation of peptide fragments produced by cleavage with cyanogen bromide, limited trypsin digestion, Staphylococcus aureus strain V8 protease, or mild acid hydrolysis. The amino acid sequence of this enzyme is compared with the sequence of the phenylalanine-sensitive 3-deoxy-D-arabino-heptulosonate 7-phosphate synthase, deduced from the aroG DNA sequence (Davies, W. D., and Davidson, B. E. (1982) Nucleic Acids Res. 10, 4045-4058).     

Partial amino acid sequence of the preprotein form of the alpha subunit of human choriogonadotropin and identification of the site of subsequent proteolytic cleavage

Messenger RNA isolated from first trimester placentae was translated using radiolabeled amino acids in both the wheat germ and the ascites cell-free systems. The choriogonadotropin α subunit product was purified by immunoprecipitation with a subunit specific antiserum. Its amino acid sequence was partially determined by automated Edman degradation analysis. An NH₂-terminal extension of 24 amino acids was found and its partial sequence is:

The preprotein form of the subunit was cleaved by the addition of microsomal membranes resulting in a homogeneous NH₂-terminal product. Hence, it is unlikely that this processing step accounts for the heterogeneity that has been observed previously in the structure of this region of the subunit.     

A study of human furin specificity using synthetic peptides derived from natural substrates, and effects of potassium ions

We explored furin substrate requirements in addition to the motif R-X-K/R-R using synthetic fluorescent resonance energy transfer (FRET) decapeptides. These decapeptides were derived from furin cleavage sites in viral coat glycoproteins and human and bacterial protein precursors. The hydrolysis by furin of most substrate was activated by K⁺ ion, whereas kosmotropic anions of the Hofmeister series were inhibitors. The analysis of furin hydrolytic activity showed that its efficiency is highly dependent on the particular combinations of amino acids at different substrate positions. There is a clear interdependence of furin subsites that must be taken in account in determining its specificity and also for the design of inhibitors. However, clear preferences were detected for substrates with S at P₁′, and V at P₂′, at P₃′ the amino acids D, S, L and A are almost equally frequent. In the non-prime subsites the best substrates presented S and H at P₆; basic amino acids at P₅; and no clear tendency at P₃. Interestingly, two amino acid substitutions on the prime side of the peptide derived from H5N1 influenza hemagglutinin furin processing site highly improved its hydrolysis. These modifications are possible by single point mutations, suggesting a potential yield of a more infectious virus.     

Structural studies on equine glycoprotein hormones. Amino acid sequence of equine lutropin beta-subunit

The amino acid sequence was determined for equine lutropin beta (eLH beta). Large fragments were derived from reduced, carboxymethylated eLH beta by digestion with Staphylococcus aureus V8 protease, by cyanogen bromide cleavage, and by cleavage of acid-labile Asp-Pro bonds. The fragments were purified by gel filtration and high performance liquid chromatography (HPLC). The fragments were sequenced by automated Edman degradation to establish the primary structure of eLH beta. Some peptides were further digested with chymotrypsin and the resulting peptides purified by HPLC. In addition to sequencing by automated Edman degradation, these were also sequenced by the complementary 5-dimethylaminonaphthalene-1-sulfonyl-Edman procedure which enabled us to directly identify glycosylated amino acids. The eLH beta subunit is a glycoprotein of 149 amino acids containing both N- and O-linked oligosaccharides. It possesses a COOH-terminal extension similar to that seen in human chorionic gonadotropin. Carboxypeptidase Y digestions suggest that the COOH terminus is blocked by glycosylation. Interestingly, the amino acid sequence of eLH beta is identical to that of equine chorionic gonadotropin beta (Sugino, H., Bousfield, G. R., Moore, W. T., and Ward, D. N. (1987)J. Biol. Chem. 262, 8603-8609).     

Determination of the chirality of amino acid residues in the course of subtractive Edman degradation of peptides.

A chiral reagent, 1-fluoro-2,4-dinitro-5-L-alanine, was synthesized for the analysis of enantiomeric mixtures of amino acids after precolumn derivatization. The resulting diastereomers can be separated and quantitated by microbore RP-HPLC. These derivatives are relatively stable under the conditions used for acid hydrolysis of peptide bonds. Thus, this reagent was included in the protocol of a subtractive Edman degradation procedure of peptides to determine the sequence position of amino acid residues with concomitant identification of their chirality at a nanomolar level.     

Purification and Partial Characterization of a Murein Hydrolase, Millericin B, Produced by Streptococcus milleri NMSCC 061

Streptococcus milleri NMSCC 061 was screened for antimicrobial substances and shown to produce a bacteriolytic cell wall hydrolase, termed millericin B. The enzyme was purified to homogeneity by a four-step purification procedure that consisted of ammonium sulfate precipitation followed by gel filtration, ultrafiltration, and ion-exchange chromatography. The yield following ion-exchange chromatography was 6.4%, with a greater-than-2,000-fold increase in specific activity. The molecular weight of the enzyme was 28,924 as determined by electrospray mass spectrometry. The amino acid sequences of both the N terminus of the enzyme (NH₂ SENDFSLAMVSN) and an internal fragment which was generated by cyanogen bromide cleavage (NH₂ SIQTNAPWGL) were determined by automated Edman degradation. Millericin B displayed a broad spectrum of activity against gram-positive bacteria but was not active against Bacillus subtilis W23 or Escherichia coli ATCC 486 or against the producer strain itself. N-Dinitrophenyl derivatization and hydrazine hydrolysis of free amino and free carboxyl groups liberated from peptidoglycan digested with millericin B followed by thin-layer chromatography showed millericin B to be an endopeptidase with multiple activities. It cleaves the stem peptide at the N terminus of glutamic acid as well as the N terminus of the last residue in the interpeptide cross-link of susceptible strains.     

Characterization of a single glycosylated asparagine site on a glycopeptide using solid-phase Edman degradation

The characterization of site-specific glycosylation is traditionally dependent on the availability of suitable proteolytic cleavage sites between each glycosylated residue, so that peptides containing individual glycosylation sites are recovered. In the case of heavily glycosylated domains such as theO-glycosylated mucins, which have no available protease sites, this approach is not possible. Here we introduce a new method to gain site-specific compositional data on the oligosaccharides attached to a single amino acid. Using a model glycopeptide from a mutant human albumin Casebrook, glycosylated PTH-Asn was recovered after sequential solid-phase Edman degradation, subjected to acid hydrolysis and the sugars were identified by high performance anion exchange chromatography with pulsed amperometric detection. The PTH-Asn(Sac) derivative was further characterized by ionspray mass spectrometry. Comparison between an endoproteinase Glu-C glycopeptide and a tryptic glycopeptide showed that the oligosaccharide attached to Asn494 was stable after at least 10 cycles of Edman degradation.     

A simple,rapid, manual peptide microsequencing procedure

A method is described for manual Edman degradation at the nanomole level. The method is simple, requiring only two extraction steps which minimize extraction loss, time, and there is no need for a conversion step of the thiazolinones to the thiohydantoins. Two different back hydrolysis procedures are compared and their relative merits discussed. The procedure requires no specialized reagents or equipment other than an amino acid analyzer.