Search

In a continuing attempt to explore the types of specificity determinants that may affect protein-protein (peptide) interactions, a number of short (2-5 residues) acetylated peptides have been compared as substrates for the enzyme acetylaminoacyl-peptide hydrolase (EC 3.4.19.1). The reference substrate was Ac-AAAA, and most of the other substrates were derived from this basic structure by single amino acid substitutions. The Km and kcat for the different substrates were determined by standard steady-state kinetics, and the corresponding delta delta GT++ value derived from kcat/Km was used for the comparison, setting delta detal GT++ for Ac-AAAA equal to 0. The best substrates were found to be those containing negative charges (Asp > Glu) or aromatic residues in positions 1', 2', or 3' (delta delta GT++ values of 2-5 kJ); the negative charge provided by the C-terminus of the substrate also appears to be important, since the amide and O-Me ester derivatives caused a change in delta delta GT++ values of -7 to -8 kJ from the reference peptide. The stimulating effect of the negative charges is consistent with the inhibitory effect of positive charges in similar peptides (Krishna RG, Wold F, 1992, Protein Sci 1:582-589), and the proposed active site model incorporates subsites for both charge-charge and hydrophobic interactions. In assessing all the data, it is clear that the properties of the individual substrates reflect the total make-up of each peptide and not only the effect of a single residue in a given position.(ABSTRACT TRUNCATED AT 250 WORDS)

The specificity of Golgi-membrane glycoprotein 6-α-[ l -fucosyltransferase [GDP-[ l ]-fucose: 2-acetamido-2-deoxy-β-[ d ]-glucoside (Fuc → Asn-linked GlcNAc) 6-α-[ l ]-fucosyltransferase; EC 2.4.1.68] has been assessed with regard to substrate covalent structures and the effect of a protein matrix on the conformational display of those covalent structures. Specificity was studied by direct comparison of the substrate quality of nine 6-biotinamidohexanoylAsn (= R) derivatives of intermediates and products in the pathway from Man5GlcNAc2-R to a fully sialylated biantennary complex-type glycan. The Man5 derivative and the sialic acid-containing glycans were completely inactive as substrates. The other glycans were all fucosylated; the best substrate was GlcNAcMan3GlcNAc2-R. The protein-matrix effect was studied by comparing the substrate quality of the same 6-biotinamidohexanoylAsn derivatives as well as the corresponding biotinylAsn derivatives free in solution and bound to streptavidin. On the basis of a model derived from the known 3D structure of biotin (biocytin)-saturated streptavidin, it was predicted that the fucosylation site in the substrates would be completely masked in the biotin-binding pocket in the biotinyl derivatives (proximal display), and at least partially masked in the 6-biotinamidohexanoyl derivatives (distal display). The activity measurements were in agreement with these predictions; the glycan structures GlcNAcMan5GlcNAc2-, GlcNAcMan3GlcNAc2-, and GlcNAc2Man3GlcNAc2- were readily fucosylated as derivatives free in solution, but were totally inert in the proximal complex with streptavidin. In the distal complexes the latter two structures were found to be fucosylated very slowly while the former structure was inactive.

The effect of a protein matrix on the processing of glycoprotein glycans by Golgi enzymes from plant seedlings has been determined with an artificial glycoprotein system, comparing the processing rates of glycan-(biotinyl)Asn (or glycan-(biotinamidohexanoyl)Asn) substrates either free or bound to avidin. An analysis of the pooled glycoproteins from the seedlings suggested that the most common glycan structure is a complex one (GlcNAc-Man3GlycNAc2-protein), and consistent with this processing end-product, mannosidases I and II and GlcNAc transferases I and II were all found to be present in the seedling Golgi membrane preparations. The effect of the avidin matrix either in a proximal (biotinyl substrates) or distal (N-(biotinamido)hexonoyl substrates) association with the appropriate glycan substrate for these four enzymes was assessed from the direct comparison of the apparent first-order rate constants for the free and avidin-bound substrate-product conversions. All four plant enzymes were inhibited by the association of the glycan substrates with avidin, but the inhibition was much less pronounced than that observed with the corresponding enzymes from rat liver and hen oviduct. The rate effect shows a progression from 3- to 10-fold rate decreases in the proximal complexes and 2- to 3-fold in the distal complexes in going from the first (mannosidase I) to the fourth (GlcNAc transferase II) enzyme; with the mammalian and avian enzymes the largest effects were for the first ones and much larger absolute rate effects were observed. The results suggest that the nature of the processing enzymes in terms of this response to the avidin glycan substrates may differ in different organisms.

Publisher Summary This chapter describes posttranslational modifications of proteins. A number of short peptides, such as hormones and neuropeptides, are synthesized as multifunctional large polypeptide precursors whose sequences are encoded by mRNA and are assembled by the regular ribosomal synthetic apparatus. There are short polypeptide antibiotics and cell wall constituents that are assembled in step-by-step amino acid activation and condensation catalyzed by specific enzymes in the absence of genetic information and ribosomes. Several of the proteolytic processing steps can be illustrated by a brief review of the biosynthesis of insulin. This disulfide-bonded two-chain structure is encoded as the precursor preproinsulin. It is found that in the early stages of polymerization while the nascent chain is still attached to the membrane-associated polysomes, two cleavages take place. The methylation reactions occur in all prokaryotic and eukaryotic species and involve several different residues such as the side chains of Asp and Glu. It is found that nonmethylated analogs of methylated proteins such as histones, ribosomal proteins, enzymes, binding proteins, and some structural proteins have been produced by mutation or by comparing the same protein produced in different species.

The disulfide bonds and N-glycosylation sites in a glycoprotein from the Rathke's gland secretion of the Kemp's ridley turtle ( Lepidochelys kempi ) have been characterized with respect to peptide sequences and glycan structures. The glycoprotein constitutes about 70% of the total protein in the secretion, and based on partial sequence information, it shows more than 20% identity with both the catalytic (esterases) and the noncatalytic (thyroglobulin) members of the esterase/lipase family of proteins. For the determination of the disulfide locations, the glycoprotein was digested with chymotrypsin, and the three HPLC peptide peaks yielding fluorescent products after treatment with tributylphosphine (Bu 3 P) and 4-(aminosulfonyl)-7-fluoro-2,1,3-benzoxadiazole (ABD-F) were collected. The three fractions were treated with the same reagents in separate experiments, the resulting pairs of ABD-Cys-containing peptides were separated by HPLC, and the sequence of each individual peptide was determined. The peptide identity established that three disulfide bonds existed in the glycoprotein: Cys 65-Cys 91, Cys 254-Cys 265, and Cys 130-Cys 404; the first two of these are conserved in all the members of the esterase family. For the study of the glycosylation sites, the glycoprotein was reduced with Bu 3 P and the SH groups covalently blocked with ABD-F, and the resulting product was digested with chymotrypsin. The glycopeptides were isolated by affinity chromatography, separated by reverse-phase HPLC, and subjected to sequence analysis and fast atom bombardment mass spectrometry before and after separation of the glycans and the peptides through the action of glycoamidase. Three separate glycosylation sites were identified, each containing multiple glycans. The sugar analyses of the hydrolysates of the glycoprotein indicated that only GlcNAc and Man were present as building blocks, and the mass spectrometric data showed that Man 3 GlcNAc 2 -, GlcNAc 2–4 Man 3 GlcNAc 2 -, and possibly GlcNAc 2 Man 2 GlcNAc 2 - were the major glycan structures, distributed differently at the three sites. The three glycosylation sites match three of the nine sites glycosylated in human serum choline esterase, and one of them, Asn 106, is also found as one of two glycosylation sites in the homologous segment of thyroglobulin.

The secretion produced by Rathke's glands of Kemp's ridley sea turtles (Lepidochelys kempi) contains the enzyme gamma-glutamyl transpeptidase. The approximately 200 kDa enzyme contains two different subunits, alpha (54 kDa) and beta (21 kDa), in an unknown stoichiometry. The enzyme transfers gamma-Glu from a number of different donors, such as glutamine, glutathione, S-Me-glutathione, N epsilon(gamma-Glu)-Lys, gamma-Glu-Ala, and other gamma-glutamyl amino acids, either to water or to a variety of acceptor substrates. It appears that a free alpha-amino group is the preferred acceptor. The enzyme is not inhibited by typical sulfhydryl reagents such as N-ethyl-maleimide, p-(chloro)mercuri-benzoate or 5,5'-dithio-bis-(2-nitrobenzoate) or by the active Ser reagent tosyl fluoride. Maleate stimulates the activity of the enzyme, and in the presence of 100 mM maleate 2 mM tosyl fluoride becomes an inactivator of the enzyme. The catalytic and molecular properties of the turtle gamma-glutamyl transpeptidase are similar to those established for mammalian gamma-glutamyl transpeptidase. Neither the physiological role of the enzyme nor the biological function of the secretion in which it occurs is understood at this time.

The use of the reagent tributyl phosphine (Bu 3 P) to reduce disulfides (Ruegg, U. T., and Rudinger, J., Methods Enzymol. 47, 111-116, 1977) and of 4-(aminosulfonyl)-7-fluoro-2, 1 ,3-benzoxadiazole (ABD-F) to block free sulhydryl groups (Toyo′oka, T., and Imai, K. Anal. Chem. 56, 2461-2464, 1984) is well established in the literature. Since the two reagents apparently do not react with each other, their combination offers a convenient and quite general method for the complete characterization of free Cys (SH) and crosslinked Cys (S-S) in proteins (Kirley, T. L., J. Biol. Chem. 264, 7185-7192, 1989). We review some of the characteristics of the reaction of these reagents with Cys in peptides and proteins and some of the properties of the ABD-Cys derivatives. The review includes reactions with model compounds (e.g., Cys and glutathione), proteins such as enolase from yeast and rabbit muscle, containing only free Cys, a protein, fetuin, containing only crosslinked Cys, and a protein, superoxide dismutase, containing both free and crosslinked Cys. In all cases the direct comparison of the tryptic or chymotryptic peptides derived from the products of parallel reactions of the protein with ABD-F alone and with ADB-F together with Bu 3 P permitted the determination of both free and total Cys in the protein. Sequencing the fluorescent ABD-peptides established the position of the Cys residues in the primary sequence. To determine the pairing of Cys residues in disulfide bonds, the individual peptides from a nonreduced protein sample, separated and collected by IIPLC, were treated simultaneously with ADB-F and Bu 3 P, and the resulting single or paired fluorescent peptides, separated and collected after rechromatography, were identified by sequencing. The ability to reduce disulfides and block the resulting thiols simultaneously in a single reaction, the stability of ABD-Cys to acid hydrolysis, and the recovery, albeit in low yield, of PTH-(ABD-)Cys on sequencing, along with the characteristic fluorescence of ABD-thiol derivatives, are all very favorable features of this methodology. The apparently strong effect of environmental factors on the relative fluorescence intensity of ABD-Cys peptides represents a minor disadvantage; Cys-containing peptides with very low fluorescence may be encountered and must be anticipated in the analysis of unknown proteins.

We have isolated individual oligosaccharyl-asparagine derivatives from the total soluble glycoproteins from kidney beens ( Phaseolus vulgaris ) and from lima beans ( Phaseolus limensis ). The protein/glycoprotein mixture was digested exhaustively by pronase, and the glycan-containing fractions were separated from free amino acids and peptides by gel filtration. The oligosaccharyl-asparagine derivatives were finally fractionated on Dowex 50 (C. C. Huang, H. E. Meyer, and R. Montgomery, Carbohydr. Res. 13, 127-137, 1970), and the individual fractions were characterized by mass spectrometry, NMR, and ion exchange chromatography. With the procedures described, only oligomannose derivatives were obtained from the beans. In the case of kidney beans, six different derivatives were observed and characterized, Man 9 GlcNAc 2 Asn, two positional isomers of Man 8 GlcNAc 2 Asn, two positional isomers of Man 7 GlcNAc 2 Asn, and Man 6 GlcNAc 2 Asn. Under identical conditions the lima beans yielded primarily the Man 9 GlcNAc 2 Asn derivative along with a small amount of the two Man 8 GlcNAc 2 Asn derivatives. The oligomannose structures can be isolated in reasonable quantities (2-20 mg) from about 200 g of dry beans.

The purpose of this chapter is to attempt to compile a list of the amino acid derivatives known to exist in proteins in a manner that may be of use to protein chemists concerned with protein sequencing and with the prediction and elucidation of the complete covalent structure of proteins. With the many reviews of the area of post-translational modifications of proteins written in the course of the last 15 years (Alix and Hays, 1983; Krishna and Wold, 1992; 1993; Uy and Wold, 1977; Whitaker, 1977; Wold, 1981; 1983), there is little justification for just another review; however, with the rapidly increasing use of mass spectrometry in the elucidation of protein structure (Biemann, 1992), and with mass spectrometry as the most obvious tool toward recognizing and identifying unusual amino acids in proteins, it has been suggested that a list of most of the known derivatives along with their molecular masses might be of some use, especially to the individuals at this meeting.

In an attempt to explore how specific features of the substrate's primary structure may affect the activity of rabbit muscle acylaminoacyl-peptide hydrolase (EC 3.4.19.1), a number of acetylated peptides containing specific amino acid replacements in specific positions were prepared and compared as substrates for the hydrolase. The principal variants were D-Ala, Pro, and positive charges (His, Arg, Lys); in addition, the effect of the length of the peptide was also investigated in a less systematic manner. The substrates were either prepared by direct acetylation of peptides, by extension of the N-terminus with acetylamino acids or acetylpeptides, activated as N-hydroxysuccinimide esters, or by isolation of the N-terminal peptides from naturally occurring acetylated proteins. It was found that D-Ala on either side of the bond to be cleaved (positions 1 and 2) completely inhibited the enzymatic activity, whereas acetylated peptides with D-Ala in positions 3 or 4 were as good substrates as those containing L-Ala. Peptides with Pro in positions 2 were also inactive, and most of the peptides with Pro in the third position were very poor substrates; only the peptide Ac-AAP gave reasonably high activity (30% of Ac-AAA), which was reduced to 1-2% if additional residues were present at the C-terminus (Ac-AAPA, Ac-AAPAA). The presence of a positive charge in positions 2, 3, 4, 5, and 6 gave strong reduction in hydrolase activity varying with the charge's distance from the N-terminus from 0 to 15-20% of the rates obtained with the reference peptides without positive charges.(ABSTRACT TRUNCATED AT 250 WORDS)

The enzyme acylaminoacyl-peptide hydrolase represents an attractive reagent for the removal of acetylamino acids from the N-terminus of proteins prior to sequencing. However, the enzyme will not accept intact proteins as substrates, and a blocked protein must consequently be fragmented to generate a relative short blocked peptide, and all the newly generated amino termini must be blocked with an hydrolase-resistant reagent before the enzyme can be used to specifically unblock the N-terminus. When a number of N-acetylated proteins (enolase, α-crystallin, ovalbumin, cytochrome c, parvalbumin, superoxide dismutase, and myelin basic protein) were subjected to fragmentation with proteases or cyanogen bromide, treatment with succinic anhydride and exhaustive extraction with ether, and the resulting salt-free, succinylated peptides were incubated with the hydrolase, the N-terminal sequence was specifically unblocked. An aliquot of the entire peptide mixture was applied to the protein sequencer, and a single sequence, corresponding to the known N-terminal sequence starting at residue 2, was obtained. When another aliquot of the same hydrolase-treated peptide mixture was treated with the enzyme acylase I, the liberated acetylamino acid was cleaved, and the N-terminal amino acid (residue 1) could be identified by amino acid analysis. The amount of sequence information obtained from different proteins with different fragmentation methods varied considerably; in the case of parvalbumin a sequence of 12 residues was obtained, while for myelin basic protein, only 3 residues could be identified; the other proteins yielded from 5- to 9-residue sequences. Rabbit muscle acetylaminoacyl-peptide hydrolase, the enzyme used in the present studies, is itself a blocked protein and was subjected to chymotrypsin digestion, succinylation, and hydrolase digestion. Subsequent sequencing established the sequence ERQVL-, and acylase treatment gave M as the only free amino acid, demonstrating that the N-terminal sequence of this enzyme is Ac-Met-Glu-Arg-Gln-Val-Leu-, identical to the sequences deduced for other hydrolases.

Glycopeptides representing each individual N-glycosylation site in six animal and plant glycoproteins (ovoinhibitor and ovotransferrin, orosomucoid, antitrypsin, phaseolin, and phytohemagglutinin) have been isolated and compared by mass spectrometric analysis. Since the isolation step separates each individual peptide regardless of the nature of the glycan attached to it, it is possible to observe the entire spectrum of glycans associated with each site from the mass spectrum of the corresponding glycopeptide. The three glycosylation sites in ovoinhibitor have very similar but not identical glycans; they are significantly different from those observed in the single site of ovotransferrin. The three sites in serum antitrypsin also have quite similar glycans, whereas the five sites in orosomucoid show considerable variation in both the nature and the relative amount of glycans. The two plant glycoproteins each have two sites with very different glycan structures. Except for the first and third glycosylation sites of antitrypsin which were found to have remarkably homogeneous glycans (97 and 90% of a biantennary complex structure), all the individual glycosylation sites contained heterogeneous mixtures of glycan structures. The results support the proposition that each N-linked glycan in a glycoprotein is affected by its unique protein environment to such an extent that each one may be displayed to the processing enzymes as a unique structural entity. On the basis of a limited number of observations of the glycan interfering with chymotryptic but not tryptic cleavage in the proximity of the glycan attachment site, it is proposed that hydrophobic interactions between the protein and the glycan may be involved in the conformational modulation of the glycans.

Publisher Summary Biotinylated glycans bound to avidin or streptavidin represent useful glycoprotein models for the study of both glycan processing by Golgi enzymes and glycoprotein interactions with lectins/receptors. The radiolabeled derivative with the 6-carbon extension arm can be prepared by reaction of 6-aminohexanoate with the activated radiolabeled biotin derivative. The product can be further activated with N -hydroxysuccinimide. An alternative for the latter preparation is to modify the α-amino group of the glycopeptide with N -blocked aminohexanoic acid and then, after unblocking, to react with activated biotin. The avidin-bound derivative with the extension arm (biotinamidohexanoyl) shows a slow cleavage under the same conditions. Similarly, exposure of the free and bound biotinylated glycans to α-mannosidase shows significant effects of the protein matrix on both the rate and the products of the mannosidase action. A major area of use for these derivatives has been in the investigation of the effect of the protein matrix on the individual processing enzymes from Golgi membranes.

Four proteases, crude acid protease from Aspergillus , pronase, amino-peptidase M, and prolidase, have been covalently attached to activated agarose and to amino propyl glass beads. The matrix-bound enzymes have been tested as catalysts for the complete hydrolysis of protein substrates, with the primary goal to isolate unstable amino acid derivatives present in the substrate protein. Under conditions used in the present work, the total amino acid release from the protease-catalyzed hydrolysis of four substrate proteins (pancreatic ribonuclease, egg white lysozyme, yeast enolase, and bovine insulin) was 95–103% of that observed in standard acid hydrolysis. Recovery of individual amino acids showed greater deviation from the theoretical values, but cystine was the only amino acid recovered in low yields (42–77%) from all four proteins. Derivatized amino acids, such as methionine sulfoxide, O -(butylcarbamoyl)-serine, and N -glycosyl asparagine have been obtained from chemically modified proteins or from unmodified glycoprotein in good yield, and normal amino acid constituents of proteins which cannot be quantified after acid hydrolysis (tryptophan, asparagine, and glutamine) have also been determined either directly after proteolysis or after proteolysis in conjunction with acid hydrolysis.

New types of neoglycoproteins, β-caseins coupled with ovalbumin-derived asparagine oligosaccharides (AO), aspartate aminotransferase-phosphopyridoxylated AO complex (AAT-PG), and streptavidin-biotinylated AO complex (SA-BAO), were tested for their inhibitory effect on binding of bovine serum albumin derivatized with thiomannoside, Man-AI-BSA [Lee YC, Stowell CP, Krantz MJ (1976) Biochemistry 15:3956–63] by rabbit alveolar macrophages. The β-casein derivatives and the AAT-PG complex increased binding affinity as the number of oligosaccharide chains attached was increased. Their inhibitory potencies were closely related to those of the Man-Al-BSA derivatives [Hoppe CA, Lee YC (1983) J Biol Chem 258:14193–99] on the basis of terminal mannose density. The SA-BAO complex containing three AO chains gave stronger inhibitory potency than the β-casein derivative with three AO residues, suggesting that proper orientation of the oligosaccharides on the protein can affect the receptor-ligand interaction.

A lectin has been isolated from the roots of 5-day soybean (Glycine max (L) cultivar Chippewa) seedlings, and its properties have been compared to those of the soybean seed lectin. The sugar-binding activities of the two lectins, both in terms of specific hemagglutinating activity and sugar specificity, are indistinguishable. Molecular properties of the two lectins, measured as relative molecular weights, isoelectric and electrophoretic patterns, amino acid compositions, immunochemical cross-reactivity, and chromatographic behavior on Sepharose-concanavalin A adsorbents suggest that the seed and the root lectin are very similar but not identical. On the basis of these comparisons, we conclude that models regarding biological functions of soybean lectin derived from studies using the seed lectin can be extended to include the root lectin in this cultivar. Studies on the distribution of the lectin in the root tissue suggest that it is associated with the outer surface of the root and is concentrated in the segments of the root at which hair and early secondary roots are observed. Since this is the region at which Rhizobium binding occurs and at which nodulation probably is initiated, all the reported observations on the root lectin are consistent with its proposed role in the specific interaction of the developing soybean with its symbiont.

Cytochrome c from Aspergillus niger consists of two forms, a major one (80%) with 111 amino acid residues and a minor one (20%) with 108 residues, missing the three N-terminal residues of the major one. The primary sequence of A. niger cytochrome c was determined by standard spinning-cup Edman degradation of purified peptides and of pairs of peptides, from which the desired sequence was readily deduced by subtraction of common sequencies. Except for the extension and some variability at the N-terminal sequence, the A. niger protein conforms well with other cytochrome c structures.

A computer program, Pattern Learn, can distinguish between acetylated and non-acetylated proteins by comparing the first 40 residues for their statistical assignment as secondary-structure formers (+), breakers (−) or neutrals (0). The major distinguishing features between acetylated and non-acetylated proteins appear in the sequences ranges 1–10, 16–24 and 30–40, but the precise nature of the distinguishing features is not known.

Transglutaminase has been isolated from human erythrocytes, and some of its molecular and catalytic properties have been determined. An enzyme preparation of about 15% purity is readily obtained in about 25% yield after DEAE-cellulose fractionation and gel filtration. In order to achieve this yield of enzyme it is essential to add to the buffers a dialyzable stabilizing factor which is present in the early enzyme fractions. This natural factor can be partly replaced by chelating compounds and totally replaced by ATP, and in practice, the purification of the enzyme is best carried out with ATP present in the buffers. The role of ATP in stabilizing the enzyme is unknown. Complete purification of the erythrocyte transglutaminase can be accomplished by preparative acrylamide gel electrophoresis. The pure enzyme has a molecular weight of 82 000 ± 5000, as established by gel filtration and SDS gel electrophoresis, and its catalytic properties are essentially identical to those of guinea pig liver transglutaminase. The guinea pig liver and the erythrocyte enzymes have also been compared as catalysts for protein modification reactions, and have been found to have quite similar specificity requirements for protein substrates. Both enzymes catalyzed significant incorporation of amines into 4 of 20 soluble proteins tested and into proteins 1, 2 and 3 of the red cell membrane. The partially purified erythrocyte enzyme has been found to be completely satisfactory for protein modification experiments, and the ready availability of outdated human blood and the simple purification procedure should make this enzyme a convenient protein-modifying or crosslinking reagent.

Hemagglutinating proteins were isolated by affinity chromatography from seeds of each of five cultivars of soybeans (Clycine max (L.) Merr.) previously reported to lack detectable lectin (S.P. Pull et al., 1978; Science 200, 1277). Quantities were between 1,000 and 10,000 times less than that found in the seeds of the reference cultivar, Chippewa. The sensitivity of the hemagglutinating assay was 0.05 μg ml(-1). Hemagglutinating activity was demonstrated in affinity-purified fractions from bulk seeds and seeds from individual plants in two cultivars, 30-70% ammonium-sulfate-precipitable fractions of seeds from individual plants of all five cultivars, and in whole crude extracts of individual seeds from each cultivar. In all instances, hemagglutinating activity was inhibited by galactose, anti-soybean agglutinin (SBA), and lectin-binding polysaccharide produced by Rhizobium japonicum. Affinity-purified lectin from seeds of a single Columbia plant was labeled with fluorescein isothiocyanate (FITC) and observed by fluorescence microscopy to bind to R. japonicum cells with specificity, intensity and localization indistinguishable from FITC-SBA. Lectins from distinguishable from FITC-SBA. Lectins from three cultivars in sufficiently high concentration for study had molecular properties very similar to Chippewa SBA.

A symposium at the FASEB meeting in Las Vegas in May 1988 will be devoted to the role of cofactors (vitamins, coenzymes, prosthetic groups) in and as posttranslational protein modifications; the symposium is part of a thematic focus on metabolic regulation. In planning the symposium, we decided to consider metabolic regulation in its broadest context, which should include both the short-term activity modulations in the life of contemporary organisms and the adaptations of special molecular strategies over evolutionary time. We further decided to focus the symposium context on the involvement of cofactors both as catalytic participants in and as substrates or end products of posttranslational modifications. As a preview of the actual symposium, the present discussion is an attempt to enumerate cases of cofactor involvement in these different categories: 1) essential nutrients as participants in posttranslational modifications; 2) cofactors as donor substrates in reversible, regulatory modifications; and 3) cofactor incorporation or generation as covalent constituents of proteins. The actual symposium topics are taken from category 1: vitamin C and protein hydroxylation (K. I. Karivikkio) and vitamin K and protein carboxylation (J. W. Suttie) and category 3: biotinylation (H. G. Wood), phycobiliproteins (A. Glazer), and pyruvoyl enzymes (W. Dowhan).

The circulating half-lives of the four isozymes of bovine pancreatic ribonuclease (RNases A, B, C, and D) have been determined in normal and in nephrectomized rats. The isozymes differ only in their glycosyl content. While A contains no sugars, B has a simple oligosaccharide (GlcNAc2 Man4-5),and C and D each have a complex oligosaccharide (GlcNAc4 Man 2-3 Gal2 Fuc NeuAc2, and GlcNAc4 Man3 Gal2 Fuc NeuAc4, respectively) attached to Asn-34 of the polypeptide chain. All four isozymes were cleared rapidly in normal rats (t 1/2 = 2 to 3 min), as expected on the basis of the established role of the kidneys in removing low molecular weight proteins from circulation. In nephrectomized rats, however, a much slower clearance was observed, thus permitting the evaluation of the role of the carbohydrate chains in the catabolism of the isozymes. The clearance curves can be analyzed in terms of two processes, a rapid initial one, shown to represent the equilibration of the injected enzyme into extravascular space, and a second one which is interpreted as the catabolic clearance of the enzyme. The haf-life of the RNase isozymes was calculated from this second process and found to be in the range 528 to 577 min for RNase A, 15 min for RNase B, 681 to 862 min for RNase C, and 839 to 941 min for RNase D. The rapidly cleared RNase B was treated with alpha-mannosidase to remove 3 of the 4 mannosyl residues, leaving only a trisaccharide (GlcNAc2-betaMan) attached to the protein. The half-life of this RNase B derivatives was found to be in the range 616 to 733 min. From these results it is concluded (a) that the addition of complex oligosaccharides to a protein does not have any significant direct effect on its circulating half-life (RNases C and D compared to RNase A), and (b) that in the rat there exists a mechanism for clearing glycoproteins based on specific recognition of exposed alpha-mannosyl residues (RNase B compared to the other isozymes and to alpha-mannosidase-treated RNase B).

A liver cytochrome P-450 isozyme has been purified to homogeneity from protein-energy malnourished rats induced with beta-naphthoflavone (beta-NF). The purification steps included chromatography on DEAE-Sephadex-A-25, DEAE-cellulose (DE-53), hydroxylapatite (HA) and carboxymethyl-sephadex (CM) columns. The reduced carbon monoxide difference and absolute spectra showed a Soret peak at 446.5 nm. The wavelength maxima for the oxidized and reduced spectra were at 416 and 408 nm, respectively. Cytochrome P-446 appears to have a predominantly low spin ferric iron, migrates as a single band of molecular weight 56,000 in sodium dodecyl sulfate polyacrylamide gels and has a specific content of 14 nmol/mg of protein. P-446 oxidized various substrates at different rates in a reconstituted system with NADPH-cytochrome P-450 reductase and dilauroyl-phosphatidylcholine. In this system turnover rates for benzo[alpha]pyrene, testosterone and benzphetamine oxidation were: 81.10; 1.85 and 1.42 nmoles product/min/nmol P-446 respectively. While NH2 terminal amino acid sequence analysis of 18 of the first 20 residues suggests that the cytochrome P-446 isolated from malnourished rats is identical with form c, the catalytic activities suggest that this isozyme may be a more effective or efficient catalyst for some substrates.

PERSPECTIVES AND SUMMARY . INTRODUCTION . INDIVIDUAL REACTIONS . Modifications Involving the Peptide Bonds . Modifications Involving the Carboxyland Amino-Terminals . Derivatipes of the a-corboxy/ group . DerivatiPe!J of the a-omino group .

Methods were investigated for the stabilization of human splenic α-galactosidase A (α- d -galactoside galactohydrolase, EC 3.2.1.22). Anti-α-galactosidase A antiserum was produced in a goat by repeated immunization with highly purified α-galactosidase A. When this antiserum was incubated at 37°C with α-galactosidase A in varying enzyme: antiserum ratios, a significant increase in the thermal (50 °C) stability and in the resistance to protease digestion of these mixtures was observed compared to appropriate controls. The enzyme was also treated with the bifunctional cross-linking reagent, hexamethylene diisocyanate, and the thermal stability and protease resistance of the cross-linked derivative were increased compared to the native enzyme treated with the monofunctional reagent, butyl isocyanate.

A search for derivatized amino acids in proteins has shown that the extent of posttranslational modification of proteins is quite substantial. While only 20 primary amino acids are specified in the genetic code and are involved as monomer building blocks in the assembly of the polypeptide chain, about 140 amino acids and amino acid derivatives have been identified as constituents of different proteins in different organisms. A brief consideration of the questions about where and when the derivatization reactions occur, how the specificity of the reactions is established, and how the posttranslational modifications can facilitate biological processes, reveal a need for more information on all these points. Answers to these questions should represent significant contributions to our understanding of biochemistry and cell biology.

In the biosynthesis of glycoproteins containing asparagine-linked glycans, a number of regulatory factors must be involved in converting the single glycan precursor into the variety of different final structures observed in different eukaryotic species. Among these factors are the kind of glycan-processing enzymes available in the Golgi apparatus of different cells, the specificity and regulatory properties of these enzymes, and the unique properties of the protein matrix in which a given glycan resides during the biosynthetic processing. In examining the role of this latter regulatory factor, we have considered a simplified model in which a few key steps are common to all cells, regardless of the nature of the processing enzymes available. The protein-bound oligomannose precursor Man8GlcNAc2-, arriving in the Golgi after the initial trimming in the endoplasmic reticulum (ER), first undergoes a series of preprocessing steps to yield Man5GlcNAc2- in animals and plants or Man13-15GlcNAc2- in yeast. At this stage the key commitment step--to process or not to process--determines whether the above intermediates will remain as unprocessed oligomannose structures or be initiated into a new series of reactions to yield processed structures characteristic of the organisms involved (complex or hybrid for vertebrates, polymannose for yeast, xylosylated glycans for plants and some invertebrates, or Man3GlcNAc2- structures for other invertebrates). It is proposed that this commitment step, along with the obligatory preprocessing steps, is regulated primarily by each glycan's unique exposure on its protein matrix. Subsequent processing steps leading to complex or hybrid structures, fucosylation, extent of branching, and specific structures at the nonreducing terminals are most likely determined primarily by the enzyme makeup of the individual processing machineries, but with the protein matrix still playing a significant role.

The structural model of yeast enolase has undergone substantial modification in recent years, and the quantitative end group analysis of the enzyme was undertaken as a means of establishing the most recent two-subunit model by a chemical method. Carboxypeptidase digestion and hydrazinolysis gave 1.95 and 1.85 moles, respectively, of carboxyl-terminal leucine per mole of enzyme, and amino-terminal analysis by the cyanate method gave 1.8 moles of amino-terminal alanine per mole of enzyme, thus confirming the model of yeast enolase as an 88,000-dalton protein consisting of two 44,000-dalton subunits. Contrary to reports in the literature, pure enolase was found to be very resistant to endopeptidase-free exopeptidase digestion. The earlier findings that large segments of both the amino-terminal and carboxyl-terminal sequences can be removed by exopeptidases without loss of enolase activity can thus not be reproduced in this laboratory. Some possible reasons for this discrepancy are discussed.

Enolase has been isolated from lobster muscle by acetone fractionation, heat treatment, ammonium sulfate fractionation, gel filtration, and ion-exchange chromatography. Preliminary characterization of the pure enzyme shows that the catalytic properties are very similar to those of the enolases from rabbit and fish.

Enolase has been purified from aqueous extracts of Escherichia coli acetone powder by (a) heat treatment, (b) fractionation with acetone, (c) TEAE-cellulose chromatography, (d) Sephadex G-100 chromatography, and (e) crystallization. The purified, crystalline enzyme migrates as a single band in disc gel electrophoresis and is homogeneous by ultracentrifugal analysis. The molecular weight of the enzyme is approximately 90,000, as determined by sedimentation velocity and sedimentation equilibrium experiments. The subunit molecular weight estimated by electrophoresis in polyacrylamide gels containing sodium dodecyl sulfate is 46,000, suggesting that the enzyme is composed of two subunits of equal size. Functionally there are many similarities between E. coli enolase and other enolases studied. Thus, the dependence on Mg2+ for activity and the inhibition by fluoride in the presence of phosphate are quantitatively very similar for all enolases. Other catalytic parameters (Km, Vmax, and pH optimum) are also similar, but minor quantitative distinction indicates that E. coli enolase is more closely related to yeast enolase than to enolases from vertebrate muscle.

Dimethyl adipimidate reacts with lysine residues of proteins to form covalent cross-links. When human erythrocyte membranes are treated with dimethyl adipimidate the percentage of protein which can subsequently be solubilized by treatment with aqueous pyridine is reduced. Evidence is presented that this is due to the formation of cross-links between soluble and insoluble protein molecules. The solubility distribution of protein-bound sialic acid, hexose, and hexosamine is altered in a parallel manner. This effect is not produced by treatment of the membranes with methyl butyroimidate, a monofunctional analogue.