Michael W. Washabaugh

Purification of aqueous ethylene glycol

Reagent-grade ethylene glycol has been shown to contain substantial amounts of aldehydes, peroxides, iron, and uv-absorbing hydrocarbons. These impurities can be removed by reduction with sodium borohydride, dilution with H2O, passing through a train of four columns, and filtering through a 0.45-micron filter. The product is stable for at least several months and perhaps much longer; storage under nitrogen in acid-washed dark bottles is preferable. Ten liters of 25% (v/v) aqueous ethylene glycol can easily be purified in about 1 week using equipment commonly available in a biochemical laboratory. This purification is also applicable to aqueous glycerol.

Thiamin transport in Escherichia coli: the mechanism of inhibition by the sulfhydryl-specific modifier N-ethylmaleimide

Active transport of thiamin (vitamin B(1)) into Escherichia coli occurs through a member of the superfamily of transporters known as ATP-binding cassette (ABC) transporters. Although it was demonstrated that the sulfhydryl-specific modifier N-ethylmaleimide (NEM) inhibited thiamin transport, the exact mechanism of this inhibition is unknown. Therefore, we have carried out a kinetic analysis of thiamin transport to determine the mechanism of inhibition by NEM. Thiamin transport in vivo exhibits Michaelis-Menten kinetics with K(M)=15 nM and V(max)=46 U mg(-1). Treatment of intact E. coli KG33 with saturating NEM exhibited apparent noncompetitive inhibition, decreasing V(max) by approximately 50% without effecting K(M) or the apparent first-order rate constant (k(obsd)). Apparent noncompetitive inhibition is consistent with an irreversible covalent modification of a cysteine(s) that is critical for the transport process. A primary amino acid analysis of the subunits of the thiamin permease combined with our kinetic analysis suggests that inhibition of thiamin transport by NEM is different from other ABC transporters and occurs at the level of protein-protein interactions between the membrane-bound carrier protein and the ATPase subunit.

Overexpression, purification, and characterization of the periplasmic space thiamin-binding protein of the thiamin traffic ATPase in Escherichia coli

Thiamin (Vitamin B(1)) transport in Escherichia coli occurs by the superfamily of traffic ATPases in which the initial receptor is the periplasmic binding protein. We have cloned the periplasmic thiamin-binding protein (TBP) of the E. coli periplasmic thiamin transport system and purified the overexpressed protein to apparent homogeneity. A subsequent biochemical characterization demonstrates that TBP is a 34.205kDa monomer. TBP also contains one tightly bound thiamin species [thiamin, thiamin monophosphate (TMP), or thiamin diphosphate (TDP)] per monomer (K(D)=0.8 microM) when isolated under conditions that would remove any loosely bound ligands. We also demonstrate that thiamin is readily exchangeable in the presence of exogenous thiamin with a k(off)=0.12s(-1). The biochemical characteristics of the overexpressed, plasmid-derived TBP are indistinguishable from those determined for endogenous TBP purified from E. coli. The overexpression and purification of TBP that we present here allows the rapid isolation of large amounts of pure protein that are required for further mechanistic and structural studies and demonstrates a vast improvement over previously reported purifications.

Mechanism of hydrolysis of a thiazolium ion: General acid-base catalysis of the breakdown of the tetrahedral addition intermediate☆

Abstract Rate constants in the pH range 3–9 for formation of the enethiolate product upon hydrolysis of 3,4-dimethylthiazolium ion, a model for the coenzyme thiamin, have been determined by irreversible iodination of the enethiolate at 25°C and ionic strength 1.0 m in aqueous solution. It is concluded that the rate-limiting step for hydrolysis of 3,4-dimethylthiazolium ion in the pH range 3–11 is breakdown of the neutral tetrahedral addition intermediate (T 0 ) to product: general acid catalysis for enethiolate formation is observed and is inconsistent with rate-limiting formation of T 0 ; buffer catalysis results provide no evidence for a change in rate-limiting step. Bronsted values are α = 0.53 for general acid and β = 0.31 for general base catalysis of the formation of the hydrolysis product by oxygen-containing buffers and primary amines: nucleophilic attack of the buffer bases has been ruled out. Catalysis by buffer acids is formulated as concerted general acid catalysis of the departure of the enethiol from T 0 . The buffer base- and water-catalyzed reactions are formulated as concerted general base catalysis of the expulsion of the enethiolate from T 0 . It is suggested that these mechanisms are general for hydrolysis reactions of 3-substituted-4-methylthiazolium ions where the substituent on the nitrogen atom of the thiazolium ring is not an intramolecular nucleophilic catalyst.

Normal acid behavior for C(α)-proton transfer from a thiazolium lon☆

Abstract Rate constants for C(α)-proton transfer from racemic 2-(1-hydroxyethyl)-3,4-dimethylthi-oazolium ion catalyzed by lyoxide ion and various oxygen-containing and amine buffers were determined by iodination at 25°C and ionic strength 1.0 m in H2O. Thermodynamically unfavorable C(α)-proton transfer to oxygen-containing and amine bases shows general base catalysis with a Bronsted β value of ≥0.92 for bases of pKa′ ≤ 15; this indicates that the thermodynamically favorable protonation reaction in the reverse direction has a Bronsted α value ≤0.08, which is consistent with diffusion-controlled reprotonation of the C(α)-enamine by most acids. General base catalysis is detectable because there is an 85-fold negative deviation from the Bronsted correlation by hydroxide ion. Primary kinetic isotope effects of ( k H k D ) obsd = 1.0 for thermodynamically unfavorable proton transfer to buffer bases and hydroxide ion (ΔpKa ≤ −6) and a secondary solvent isotope effect of k DO − k HO − = 2.3 for C(α)-proton transfer are consistent with a very late, enamine-like transition state and rate-limiting diffusional separation of buffer acids from the C(α)-enamine in the rate-limiting step, as expected for a “normal” acid. The second-order rate constants for catalysis by buffer bases were used to calculate a pKa′ of 21.8 for the C(α)-proton assuming a rate constant of 3 × 109 m −1 s−1 for the diffusion-controlled reprotonation of the C(α)-enamine by buffer acids in the reverse direction. It is concluded (i) that C(α)-proton removal occurs at the maximum possible rate for a given equilibrium constant, and (ii) that C(α)-enamines can have a significant lifetime in aqueous solution and on thiamin diphosphate-dependent enzymes.

General acid catalysis of the cleavage of 2-(1-hydroxybenzyl)thiamin by a preassociation mechanism☆

Abstract Cleavage of racemic 2-(1-hydroxybenzyl)thiamin (HBT) to benzaldehyde and thiamin in aqueous solution, a retrograde aldol-type reaction, is catalyzed by substituted acetate ions and other oxygen-containing buffer bases at 40°C and ionic strength 1.0 m (KCl). The Bronsted β value is 0.61 for N(1′)-protonated HBT, but there is no significant solvent deuterium isotope effect for catalysis by acetate ion. The water and buffer base-catalyzed reactions are formulated as general acid catalysis of the departure of thiamin from the alcoholate anion (pKa′ROH = 10.7) of HBT (general base catalysis of thiamin attack in the reverse direction). It is concluded that this reaction proceeds by a concerted mechanism in aqueous solution that is determined by the short lifetime of the thiazolium C(2)-ylide even though the carbanion has a significant lifetime in aqueous solution and a stepwise pathway for the aldol-type addition reaction of the C(2)-ylide must exist. It is suggested that thiamin diphosphate-dependent enzymes could also use the lower energy, preassociation pathway.

Synthesis and purification of racemic 2-(1-hydroxyethyl)thiamine revisited

Abstract A procedure is reported for the rapid (≤1 day) and reproducible purification of 2-(1-hydroxyethyl)thiamine (HET) from contaminating thiamine (≤50%) using aqueous cation exchange chromatography. The unhydrated product contains ≤0.2% thiamine on the basis of its 1H NMR spectrum, visualization on silica gel TLC by fluorescence quenching after development in 60% aqueous ethanol, and the failure to observe the hydrolysis of thiamine using an iodine trapping assay under conditions where hydrolysis of ≤0.2% contaminating thiamine would have been detected. This purification procedure is also applicable to the resolved R and S stereoisomers of HET. Several literature procedures for the synthesis and purification of racemic HET are discussed and critically evaluated.

Retrograde aldol-type reactions involving thiamin in aqueous solution : evidence for changes in transition-state structure

Abstract Cleavage of racemic 2-(1-hydroxyethyl)- and 2-(1-hydroxyaryl)-3-R-4-methylthiazolium ions to the corresponding aldehyde and thiazolium ion in aqueous solution is catalyzed by oxygen-containing and amine buffer bases at 40°C and ionic strength 1.0 m (KCl). The buffer base-catalyzed reactions are formulated as general acid catalysis of the departure of thiazolium ion from the alcoholate anion of the substrate (general base catalysis of thiazolium ion attack on the aldehyde in the reverse direction). Bronsted α values decrease from 0.38 to 0.21 for general acid catalysis of the cleavage of (2-1-hydroxyethyl)-R-4-methylthiazolium ions (R = C6F5CH2, 4-aminopyrimidinyl, Bzl). The decrease in α with decreasing pKa of C(2)-H in the leaving thiazolium ion is described by a positive interaction coefficient p xy′ = ∂α ∂ p K lg = 0.3 ± 0.1 . Bronsted α values increase from 0.39 (R′ = Ph) to 0.48 (R′ = 4-CF3-Ph) for the corresponding reactions with 2-(HOCH(R′))thiamin. The increase in α as the carbon electrophile becomes less stable is described by a positive interaction coefficient p xy = sol∂α ∂σ para = 0.2 ± 0.1 . These positive interaction coefficients support a nonenforced concerted reaction mechanism with an important component of proton transfer in the transition state. Mechanistic implications for thiamin diphosphate-dependent enzymes are discussed.