Amine boranes as alternative reducing agents for reductive alkylation of proteins

The use of several commercially available amine-borane complexes was investigated for the reductive methylation of amino groups of several proteins. An earlier study in our laboratory, using turkey ovomucoid as the model protein, showed that dimethylamine borane is a slightly weaker reducing agent, but a good, less toxic substitute for sodium cyanoborohydride (K. F. Geoghegan, J. C. Cabacungan, H. B. F. Dixon, and R. E. Feeney, 1981, Int. J. Peptide Protein Res.17, 345–352). N-α-Acetyl-l-lysine, poly-l-lysine, turkey ovomucoid, bovine serum albumin, chicken ovalbumin, β-lactoglobulin, casein, and soybean protein were reductively methylated with dimethylamine borane and trimethylamine borane. The latter produced a consistently lower degree of modification even in the presence of sodium dodecyl sulfate. In a comparison that included the boranes triethylamine, t-butylamine, morpholine, and pyridine, pyridine borane was found to be slightly stronger than sodium cyanoborohydride. In a pH 7 solution containing 2 mmN-α-acetyl-l-lysine and 20 mm formaldehyde, complete dimethylation was achieved with about 10 mm pyridine borane after 2 h incubation at 22°C, while more than 15 mm was necessary with sodium cyanoborohydride. Like dimethylamine borane, both pyridine borane and triethylamine borane showed a reducing capacity at pH 7 which was as high as that at pH 9. Reductive alkylation under neutral to mild acid conditions allows modification of alkaline labile proteins and also limits the side reactions between proteins and formaldehyde.     

In vitro labeling of proteins by reductive methylation: Application to proteins involved in supramolecular structures

Actin and tropomyosin, purified from both muscle and brain, and α-actinin, purified from muscle, have been labeled in vitro by reductive methylation to specific activities of greater than 10⁵ dpm/μg protein. Actin so modified bound DNase I and polymerized identically to unmodified actin. Furthermore, the spectral properties of actin did not change after labeling. The interactions of labeled tropomyosin and α-actinin with F-actin were nearly identical to those of the unmodified proteins. These modified proteins comigrated with their unmodified counterparts in both SDS-containing polyacrylamide gels and isoelectric focusing gels. The labeled actin was quantitatively extracted from SDS-containing polyacrylamide gels (yield > 98% of radioactivity applied demonstrating that all of the radioactivity was protein bound. The reductive methylation procedure worked well at pH 8.0–8.5 in either pyrophosphate buffer or Bicine buffer using formaldehyde with [³H]-sodium borohydride as the reducing agent. The procedure could also be performed at pH 7.0 in phosphate buffer using [¹⁴C]-formaldehyde with sodium cyanoborohydride as the reducing agent. Proteins so labeled are ideal for use in quantitative experiments involving protein-protein interactions.     

Immobilization of proteins by reductive alkylation with hydrophobic aldehydes

A new method is described for the immobilization of enzymes and other proteins onto hydrophobic gels. Trypsin, for example, can be quantitatively immobilized by reaction with sodium cyanoborohydride and dodecyladehyde-coated Octyl-Sepharose. Noncovalent interactions between the hydrophobic gel and approximately 10 resulting dodecylamino groups in the modified enzyme bind it very strongly but do not affect its ability to hydrolyze benzolarginine ethyl ester. The same procedure can also be used to immobilize E. Coli asparaginase and yeast alcohol dehydrogenase in high yield.     

Immobilization of proteins on aldehyde-activated polyacrylamide supports

A method has been developed for the immobilization of proteins on derivatized polyacrylamide gels. Aminoethyl Bio-Gel P-150 was converted to its stable N-2,3-dihydroxypropyl derivative by borohydride reduction of the Schiff base formed with glyceraldehyde. Periodate oxidation of the modified gel provided a reactive aldehyde, which was subsequently coupled to protein by reductive amination with sodium cyanoborohydride. Coupling efficiencies were found to be >90% for concanavalin A and bovine serum albumin, and the gels contained as much as 5 and 20 mg of protein/ml of gel, respectively. Immobilized concanavalin A retained 89% of its binding capacity and was demonstrated to be chemically stable with variations in pH, and changes in concentrations of Triton X-100 and sodium dodecyl sulfate (at concentrations <0.1%). Bovine β-hexosaminidase and β-glucuronidase, higher molecular weight proteins, were also bound with retention of activity, but with less efficiency. This procedure provides an efficient method for the covalent immobilization of proteins.     

Stereochemistry of octopine and of its isomers and their enzymatic properties

The four isomers of octopine were prepared from pyruvic acid and l- or d-arginine and from α-keto δ-guanidinovaleric acid and l- or d-alanine by reduction with sodium cyanoborohydride. The absolute configuration of d-octopine, the natural occurring isomer being S(l) at the arginine center, and R(d) at the alanine center, was confirmed enzymatically. d-Octopine is the only isomer oxidized by NAD⁺ in the presence of octopine dehydrogenase from Pecten maximus L. The isomer with configuration S(l) at the alanine center is found to be a competitive inhibitor. Isomers with R(d) configuration at the arginine center show no detectable effect on the enzymatic reaction.     

In vitro studies of histone glycation

Reactive amination of histone H1 by [U-¹⁴C]glucose was performed in the presence of sodium cyanoborohydride and was approximately proportional to the glucose concentration. Lysine was the principal amino acid substituted. Glycation also occurred in the absence of cyanoborohydride. Browning reactions of histones were monitored by ΔA₃₂₅ whereby it was shown that glucose 6-phosphate was more reactive than glucose and that each of the histone fractions reacted with glucose 6-phosphate giving the browning reaction.     

Coupling of polysaccharides activated by means of chloroacetaldehyde dimethyl acetal to amines or proteins by reductive amination

A novel method has been developed for the coupling of modified polysaccharides to proteins or other amines. Chloroacetaldehyde dimethyl acetal has been used for the introduction of O-(2,2-dimethoxyethyl) groups into amylose, dextran, and a linear (1→3)-β-d-glucan. In amylose and the (1→3)-β-d-glucan, these groups were attached preponderantly at O-6 and in dextran at O-2. Mild treatment with acid then gave polysaccharide derivatives substituted with aldehyde groups which were coupled in good yields to proteins and other amines by reductive amination with sodium cyanoborohydride in aqueous solution at pH 7. An aminated (1→3)-β-d-glucan derivative that induced antitumor activity in mouse macrophages in vitro is reported.     

Modification of available arginine residues in proteins by p-hydroxyphenylglyoxal

p-Hydroxyphenylglyoxal reacts with arginine residues in proteins to give a single product which can be quantitated spectrophotometrically. The reaction takes place under mild conditions, pH 7–9 and 25°C. Under these conditions up to complete modification of N^α-citraconyl-l-arginine was obtained within 60 min with less than 5% modification of other common amino acid side chains. The extent of modification in a protein can be determined at 340 nm using the molar absorption coefficient of 1.83 × 10⁴m^?1 cm^?1 for the product at pH 9.0 and 25°C following removal of excess reagent by gel filtration. Several proteins, previously shown to have essential arginines, were modified by p-hydroxyphenylglyoxal and the losses in arginines were determined spectrophotometrically. These results were in close agreement with those of previous investigators. Rhea ovomucoid, a glycoprotein without arginines but containing an essential lysine, was relatively unaffected.     

O-carboxamidomethyl tyrosine as a reaction product of the alkylation of proteins with iodoacetamide

Iodoacetamide-1-¹⁴C was found to be rapidly incorporated into RBP, a protein devoid of free SH, under conditions similar to those normally employed for the derivatization of SH. The reaction extent is pH dependent and at pH 10.0 one mole of acetamide was incorporated per mole of tyrosine residue. The amino acid composition of the alkylated RBP was found to be identical to RBP except for a loss of tyrosine and the appearance of a new dicarboxylic acid peak which was converted to tyrosine by prolonged acid hydrolysisUnder similar reaction conditions, phenol and p-cresol reacted rapidly to form phenoxyacetamide and p-cresoxyacetamide. These phenolic ethers as well as anisole and phenetole were found to be readily hydrolized under the conditions normally used to hydrolyze protein.The incorporation of acetamide into RBP did not effect its riboflavinbinding capacity or its immunological reactivity to RBP antibody. The ¹⁴C alkylated RBP has been found to be a convenient tool for biological half-life studies.The tyrosine residues of glucagon react with iodoacetamide in a similar fashion and the use of ¹⁴C-iodoacetamide may prove to be a convenient means of introducing ¹⁴C into proteins. Iodoacetamide in the pH 7–10.0 range will derivatize the cysteine and tyrosine groups of proteins at comparable rates.     

Fluorescent detection of α-aminoadipic and γ-glutamic semialdehydes in oxidized proteins

The oxidative modification of proteins is believed to play a critical role in the etiology and/or progression of several diseases. α-Aminoadipic semialdehyde (AAS) and γ-glutamic semialdehyde (GGS) residues represent major oxidized amino acids generated in oxidized proteins. This paper describes a novel procedure for the specific and sensitive determination of AAS and GGS after their reductive amination with sodium cyanoborohydride and p-aminobenzoic acid, a fluorescence reagent, to their corresponding derivatives, followed by a high-performance liquid chromatography (HPLC) analysis. This fluorescent labeling of protein-associated aldehyde moieties is a simple and accurate technique that may be widely used to reveal increased levels of oxidatively modified proteins with reactive oxygen species during aging and disease.     

ADP-ribosylation reactions in Sulfolobus solfataricus,a thermoacidophilic archaeon

An ADP-ribosylating system was detected in a crude homogenate from Sulfolobus solfataricus, a thermophilic archaeon, optimally growing at 87°C. The archaeal ADP-ribosylation reaction was time-, temperature- and NAD-dependent. It proved to be highly thermostable, with about 30% decrease of ¹⁴C incorporation from [¹⁴C]NAD on incubation at 80°C for up to 24 h. The main reaction product was found to be mono-ADP-ribose. Testing both [adenine- ¹⁴C(U)]NAD and [adenine- ¹⁴C(U)]ADPR as substrates, it was found that acceptor proteins were modified by ADP-ribose both enzymatically, via ADP-ribosylating enzymes, and via chemical attachment of free ADP-ribose, likely produced by NAD glycohydrolase activity. The synthesis of ADP-ribose-protein complexes was shown to involve mainly acceptors with molecular masses in the 40–100 kDa range, as determined by electrophoresis on polyacrylamide gel in the presence of sodium dodecyl sulphate.     

On the interaction of 2-keto-3-deoxygalactonate-6P with 2-keto-3-deoxygluconate-6P aldolase ofPseudomonas putida. Evidence for enzyme-mediated,noncatalytic C-C bond turnover

2-Keto-3-deoxygluconate-6P (KDPG) aldolase ofPseudomonas putida mediates the cleavage of, as well as the condensation of, pyruvate andd-glyceraldehyde-3P (GaP) yielding, 2-keto-3-deoxygalactonate-6P (KDPGal) as side reactions of normal catalysis. These are visualized at high levels of aldolase. KDPGal cleavage occurs with aV _max that is 1/5000 that for KDPG cleavage. TheKm for KDPGal is 0.2 mM, with aK _i of 0.85 mM. The E-KDPGal complex is reductively inactivated having aKd of 0.55 mM. TheV/K value for KDPG cleavage is 2.0×10⁸ sec^?1, while the value for KDPGal cleavage is 1220 sec^?1. The difference in first-order rate constants of 164,000-fold argues that a step in the cleavage of KDPGal mediated by the enzyme is uncatalyzed. The enzyme is reductively inactivated by trapping the E-pyruvate, E-KDPG, or E-KDPGal complex. The enzyme can also be inactivated by reductive trapping of a catalytically nonproductive E-glyceraldehyde-3P complex. This latter occurs with aKd for GaP of 20 mM and a rate constant equivalent to a limiting half-time of 1110 sec at 1 mM cyanoborohydride. Reductive inactivation half-times in the presence of high GaP/KDPG ratios were the sum of both E-GaP and E-KDPG trapping by cyanoborohydride so that the inactivation rate due to KDPG could be determined. It was found at 1 mM cyanoborohydride that the limiting half-time for the E-KDPG complex was 2382 sec. The corresponding value for the E-KDPGal complex was 215 sec. Consequently, the E-KDPGal complex is 11 times more sensitive to reductive derivativation than is the E-KDPG complex. This is interpreted to show that the enzyme binds the KDPGal in a “normal” step forming a ketimine. However, turnover to the eneamine with resultant C-C bond cleavage is uncatalyzed. For the case of KDPGal synthesized by KDPG aldolase, it is argued that the pyruvate eneamine is bound to the active site, which can be attacked by GaP with its aldehyde carbon in the catalytically nonproductive conformation as a side reaction, presumably forming a tertiary complex. Spontaneous protonation of the resultant alcoholate anion would generate KDPGal. The data are interpreted to support an argument that catalytic proton turnover at the OH of C-4 of KDPG is required for normal catalysis, and that this step, which catalytically interconverts ketimine/eneamine, imposes steric constraints controlling the overall stereochemistry of the reaction.     

Chelating derivatives of chitosan obtained by reaction with ascorbic acid

Ascorbic acid immediately dissolves Euphausia superba chitosan upon mixing and forms chitosan ascorbate; during the 6-h period after dissolution in water at pH 5–7, ascorbate is oxidized to dehydroascorbate which undergoes Schiff reaction with the amino groups of chitosan, thus yielding a viscous solution of a polymeric ketimine. The latter is characterized by infrared spectrometry, circular dichroism spectropolarimetry, viscometry and alkalimetry. When brought into contact with transition metal ions, the chitosan ascorbate ketimine yields insoluble metal chelates. Upon reduction with sodium cyanoborohydride, the water-insoluble N-[2-(1,2-dihydroxyethyl)tetrahydrofuryl] chitosan (NDTC) is obtained, which shows enhanced capacity for uranium, up to 800 mg U/g from solutions at pH 4·5.     

Proteins containing reductively aminated disaccharides: Synthesis and chemical characterization

Synthetic glycoproteins can be prepared by reductive amination of protein and reducing disaccharide in the presence of sodium cyanoborohydride. The reaction proceeds readily in aqueous solutions over a broad pH range to give high degrees of substitution. The degree of substitution can be determined by amino acid analysis, as the secondary amine linkage formed by reductive amination in stable to acid-catalyzed protein hydrolysis conditions. In order to demonstrate that coupling occurs to lysine residues, synthetic α-N-1-(1-deoxyglucitol)-lysine and ?-N-1-(1-deoxyglucitol)-lysine were prepared and compared with bovine serum albumin conjugates of maltose, cellobiose, lactose, and melibiose by amino acid analysis after acid hydrolysis. These studies demonstrate that the expected secondary amine linkages are formed with the ?-amino groups of lysine.     

Recombination of S-peptide with S-protein during folding of ribonuclease S. II. Kinetic characterization of a stable folding intermediate shown by S-protein at pH 1.7.

At pH 1.7 S-peptide dissociates from S-protein but S-protein remains partly folded below 30 °C. A folded form of S-protein, labeled I₃, is detected and measured by its ability to combine rapidly with S-peptide at pH 6.8 and then to form native ribonuclease S. The second-order combination reaction (k = 0.7 × 10⁶m^?1s^?1 at 20 °C) can be monitored either by tyrosine absorbance or fluorescence emission; the subsequent first-order folding reaction (half-time, 68 ms; 20 °C) is monitored by 2′CMP ² binding. Combination with S-peptide and folding to form native RNAase S is considerably slower for both classes of unfolded S-protein (see preceding paper).I₃ shows a thermal folding transition at pH 1.7: it is completely unfolded above 32 °C and reaches a limiting low-temperature value of 65% below 10 °C. The 35% S-protein remaining at 10 °C is unfolded as judged by its refolding behavior in forming native RNAase S at pH 6.8. The folding transition of S-protein at pH 1.7 is a broad, multi-state transition. This is shown both by the large fraction of unfolded S-protein remaining at low temperatures and by the large differences between the folding transition curves monitored by I₃ and by tyrosine absorbance.The fact that S-protein remains partly folded after dissociation of S-peptide at pH 1.7 but not at pH 6.8 may be explained by two earlier observations. (1) Native RNAase A is stable in the temperature range of the S-protein folding transition at pH 1.7, and (2) the binding constant of S-protein for S-peptide falls steadily as the pH is lowered, by more than four orders of magnitude between pH 8.3 and pH 2.7, at 0 °C. The following explanation is suggested for why folding intermediates are observed easily in the transition of S-protein but not of RNAase A. The S-protein transition is shifted to lower temperatures, where folding intermediates should be more stable: consequently, intermediates in the folding of RNAase A which do not involve the S-peptide moiety and which are populated to almost detectable levels can be observed at the lower temperatures of the S-protein transition.     

Concentration-dependent hydrogen exchange kinetics of 3H-labeled S-peptide in ribonuclease S.

The hydrogen exchange kinetics of the S-peptide in ribonuclease S can be measured by first tritiating the S-peptide in the absence of S-protein and then allowing it to recombine rapidly with S-protein. Afterwards the exchange reactions of this specific segment of ribonuclease S can be studied. The exchange kinetics of bound S-peptide are complex, indicating that different protons exchange at markedly different rates. The terminal exchange reaction, involving at least five highly protected protons, has been studied as a function of pH.At low concentrations of ribonuclease S the exchange kinetics become concentration-dependent, owing to the dissociation of the S-peptide. Although the fraction of free S-peptide is always very small, its rate of exchange is several orders of magnitude faster than that of bound S-peptide, and the concentration dependence of the exchange kinetics is readily measurable. It provides a highly sensitive method for determining small dissociation constants (K_D). Values of K_D ranging from 10^?6m at pH 2.7, 0 °C, to 2 × 10^?10m at pH 7.0, 0 °C, are reported here. Our value for K_D at pH 7.0, 0 °C, confirms the data and extrapolation to 0 °C of Hearn et al. (1971).At high concentrations of ribonuclease S the terminal exchange reaction is independent of concentration. It probably results from a local unfolding reaction of the bound S-peptide. Above pH 4 the strong pH dependence of K_D closely resembles that of the apparent equilibrium constant for this local unfolding reaction. The latter may be one step in the dissociation process and we present such a model for ribonuclease S dissociation.Measurement of concentration-dependent exchange kinetics should provide a useful method of determining small dissociation constants in other systems: for example, in studies of protein-nucleic acid interactions.     

Studies on enzymes related to diacylglycerol production in activated platelets: I. phosphatidylinositol-specific phospholipasec: Further characterization using a simple method for determination of activity

A simple method of determination of phosphatidylinositol-specific phospholipase C activity in soluble platelet extracts has been devised. It is based on the use of a total lipid extract from rat liver microsomes incubated with [³H]inositol in the presence of MnCl₂. Phosphatidylinositol hydrolysis can thus be detected by determining hydrosoluble radioactivity formed upon incubation with enzyme fractions. Owing to the presence of other phospholipids in the assay system, phospholipase C was inhibited. However, activity was restored by sodium deoxycholate (0.1%, w/v). Optimal conditions also included calcium (1–10 mM) and a pH between 5 and 7, allowing the detection of phospholipase C without the need for purifying the substrate. Using this simplified procedure, platelet phospholipase C was submitted to preparative electrofocusing and to gel filtration chromatography on Sephacryl S-200. Phospholipase C focused in one single peak at pH 6.1. An M_r of 86000 was found upon gel chromatography of a crude extract, against 68 000 when phospholipase C had been previously purified by electrofocusing. These data indicate that phospholipase C might be associated with lipids or with an M_r 20 000 protein, the significance of which is discussed.     

Chemical probes of extended biological structures: Synthesis and properties of the cleavable protein cross-linking reagent [35S]dithiobis(succinimidyl propionate)

The synthesis and properties of a new cleavable protein cross-linking reagent, [³⁵S]dithiobis(succinimidyl propionate), are detailed. Free primary and secondary aliphatic amino groups are quantitatively acylated by the reagent in either organic or aqueous media within two minutes at 23 °C. By contrast, the half-time for hydrolysis of the active ester termini in buffer at pH 7 is four to five hours, so that protein cross-linkage can be optimized by application of low concentrations of reagent. Accessible amino groups of hemoglobin are acylated with extreme rapidity of 0 °C in pH 7 buffer when [³⁵S]dithiobis(succinimidyl propionate) is applied in 0.4 to 9-fold molar excess. Submicrogram quantities of the cross-linked hemoglobin subunits which result are detectable by monitoring the ³⁵S distribution in sodium dodecyl sulfate-polyacrylamide gels. In addition to amine acylation, two of the six thiol groups in hemoglobin, tentatively located at cysteine 93 of the β chains, are reversibly modified at 0 °C by mercaptan-disul-fide interchange with the reagent or its bis amide analogs. This equilibrium-controlled, pH-dependent reaction occurs at a slower rate than acylation, and is blocked by short preincubation of the protein with N-ethylmaleimide or by addition of 3,3′- dithiodipropionamide (or other disulfides) to the reaction mixture. Disulfides introduced into hemoglobin by acylation and interchange are quantitatively cleaved by reduction for 30 minutes at 37 °C with 10 mm-dithioerythritol buffered at pH 8.5.The properties of high reactivity under mild conditions, long solution half-life, and the radioactive label make [³⁵S]dithiobis(succinimidyl propionate) a particularly useful and versatile probe of extended structures in a variety of biological systems.     

Determination of the molecular weight of proteins in heterogeneous mixtures: use of an air-driven ultracentrifuge for the analysis of protein--protein interactions.

A high-speed air-driven ultracentrifuge (Airfuge) has been used to study the molecular weights of proteins in heterogeneous mixtures. The method is based on previous studies (M. A. Bothwell, G. J. Howlett, and H. K. Schachman, 1978, J. Biol. Chem., 253, 2073–2077) which showed that at sedimentation equilibrium in the Airfuge the fraction of a protein remaining in an upper fraction of the Airfuge tube is almost linearly related to the exponential of the reduced molecular weight of the protein. In this study the total fraction of each particular protein remaining in an upper fraction of the Airfuge tube is determined by quantitative sodium dodecyl sulfate-gel electrophoresis. This procedure allows a wide range of proteins to be analyzed in a single Airfuge experiment. The method yields the “native” molecular weights of the protein components and is independent of the shape of the macromolecules being studied. Interactions occurring between the components in solution can be detected from the Airfuge data, and procedures are described which allow the experimental data for such interactions to be analyzed in terms of an equilibrium constant for the interaction. Results obtained for the electrostatic interaction at neutral pH between lysozyme and ovalbumin (K = 1.1 × 10⁵, m^?1) and lysozyme and bovine serum albumin (K = 1.0 × 10⁵, m^?1) agree well with literature values.