Synthesis, characterization, solution stability, and X-ray crystal structure of the thiolatocobalamin gamma-glutamylcysteinylcobalamin, a dipeptide analogue of glutathionylcobalamin: insights into the enhanced Co-S bond stability of the natural product glutathionylcobalamin.

Glutathionylcobalamin (gamma-glutamylcysteinylglycinylcobalamin; gamma-GluCysGly-Cbl) is a natural product which functions as an intermediate in the biosynthesis of the active B(12) coenzymes adenosylcobalamin and methylcobalamin. Of interest to the present studies is glutathionylcobalamin's unique stability in comparison to other thiolatocobalamins, notably the > or =6 x 10(4) fold less stable cysteinylcobalamin, Cys-Cbl. In order to determine which parts of the glutathione tripeptide contribute to the overall stability of glutathionylcobalamin, two cysteine-containing dipeptides, which are truncated versions of glutathione, were used to synthesize their corresponding cobalamins, specifically gamma-glutamylcysteinylCbl (gamma-GluCys-Cbl) and cysteinylglycinylcobalamin (CysGly-Cbl). As with glutathionylCbl, the dipeptide gamma-GluCys-Cbl forms a stable thiolatocobalamin. However and most interestingly, CysGly-Cbl is observed to be unstable much like Cys-Cbl. The results require that the extra stability of glutathionylcobalamin and its congeners, compared to cysteinylcobalamin and its analogues, must be derived from destabilization by the gamma-NH(3)(+) group in cysteinylcobalamin, or stabilization by the gamma-NHC(=O)- amide linkage in glutathionylcobalamin, or both. To probe any ground-state structural basis for the possible stabilization in gamma-GluCys-containing cobalamins, gamma-GluCys-Cbl was crystallized and yielded the first X-ray structural determination of a true thiolatocobalamin, and only the second structure of a cobalamin containing a Co-S bond, the first example being Randaccio and co-workers' 1999 structure of the thioketone complex, thioureacobalamin, (NH(2))(2)CSCbl. Key features of the structure of gamma-glutamylcysteinylcobalamin include (i) a normal Co-S bond length of 2.267(2) A, (ii) a Co-N(axial) bond length of 2.049(6) A, (iii) two alternate conformations of the gamma-glutamylcysteinyl moiety, and (iv) folding of the corrin ring upward by 24.2 degrees, the highest degree of folding yet observed for a cobalamin. These results do not show any strong stabilization (e.g., no shortened Co-S bond), although it is not clear for certain what the effect is (stabilizing or destabilizing) of the elongated Co-N(axial) bond; instead, the crystallographic results suggest that the metastable Cys-Cbl probably has a Co-S cleavage transition state that is stabilized (along with, possibly, any ground-state destabilization of the Co-S bond). Overall, the results strongly suggest that placing a positive charge on the gamma-NH(3)(+) stabilizes the Co-S bond cleavage transition state, thereby setting the stage for the needed full thermolysis product and kinetic studies-as a function of the axial-base on-off equilibrium-that will be required to understand in even greater detail the unique stability of glutathionyl- (gamma-glutamylcysteinylglycinyl-) and gamma-glutamylcysteinylcobalamins.

Thermodynamic and kinetic studies on the reaction between the vitamin B12 derivative beta-(N-methylimidazolyl)cobalamin and N-methylimidazole: ligand displacement at the alpha axial site of cobalamins.

The equilibria and kinetics of substitution of the 5,6-dimethylbenzimidazole at the alpha site of beta-(N-methylimidazolyl)cobalamin by N-methylimidazole have been investigated, and the product, bis(N-methylimidazolyl)cobalamin, has been characterized by visible and 1H NMR spectroscopies. The equilibrium constant for (N-MeIm)Cbl+ + N-MeIm right harpoon over left harpoon (N-MeIm)2Cbl+ was determined by 1H NMR spectroscopy (9.6 +/- 0.1 M(-1), 25.0 degrees C, I = 1.5 M (NaClO4)). The observed rate constant for this reaction exhibits an unusual inverse dependence on N-methylimidazole concentration, and it is proposed that substitution occurs via a base-off solvent-bound intermediate. Activation parameters typical for a dissociative ligand substitution mechanism are reported at two different N-MeImT concentrations, 5.00 x 10(-3) M (DeltaH++ = 99 +/- 2 kJ x mol(-1), DeltaS++ = 39 +/- 5 J x mol(-1) x K(-1), DeltaV++ = 15.0 +/- 0.7 cm3 x mol(-1), and 1.00 M (DeltaH++ = 109.4 +/- 0.8 kJ x mol(-1), DeltaS++ = 70 +/- 3 J x mol(-1) x K(-1), DeltaV++ = 16.8 +/- 1.1 cm3 x mol(-1)). According to the proposed mechanism, these parameters correspond to the equation of (N-MeIm)2Cbl+ and the ring-opening reaction of the alpha-DMBI of (N-MeIm)Cbl+ to give the solvent-bound intermediate in both cases, respectively.

Studies on the mechanism of the reaction between coenzyme B12 and cyanide: direct 1H NMR spectroscopic evidence for a (beta-5'-deoxyadenosyl)(alpha-cyano)cobalamin intermediate.

The reaction between coenzyme B12 (5'-deoxyadenosylcobalamin, AdoCbl) and tetrabutylammonium cyanide to give dicyanocobalamin, adenine, and 1-cyano-D-erythro-2,3-dihydroxy-4-pentenol has been examined in 92% N,N-dimethylformamide (DMF)/8% D2O. Under these conditions rate-determining Co-C heterolytic cleavage is preceded by rapid addition of cyanide to AdoCbl to form an intermediate, (beta-5'-deoxyadenosyl)(alpha-cyano)cobalamin ((beta-Ado)(alpha-CN)Cbl-), identified by 1H NMR spectroscopy. Rate constants have been determined by both 1H NMR and visible spectroscopies, with the latter showing saturation kinetics. The observed rate constant is pH-independent in the pH region studied, and replacing D2O by H2O increases it by ca. 10%. Increasing the percentage of D2O in the DMF/D2O solvent mixture also increases the reaction rate, and for D2O > or = 50% there is a change in the rate-determining step, with formation of the (beta-Ado)(alpha-CN)Cbl- intermediate becoming rate-determining. A mechanism in 92% DMF/8% D2O is proposed which involves rapid reversible formation of (beta-Ado)(alpha-CN)Cbl- from base-off AdoCbl plus cyanide, followed by rate-determining solvent-assisted cleavage of the Co-C bond of the intermediate and subsequent rapid addition of a second cyanide to give the products.

Synthesis and characterization of isolable thiolatocobalamin complexes relevant to coenzyme B12-dependent ribonucleoside triphosphate reductase.

The syntheses, isolation and characterization of cyclohexylthiolatocobalamin (C6H11SCbl), glutathionylcobalamin (GluSCbl), and cysteinylcobalamin (CysSCbl) are reported in 75, 55, and 65% yield, respectively. Characterization was achieved using elemental analyses, L-SIMS (liquid secondary ion mass spectrometry), UV-visible spectroscopy and, for the more stable C6H11SCbl and GluSCbl, our recently established 1H NMR method (which emphasizes the readily interpreted aromatic region of the cobalamin's 1H NMR spectrum). Preliminary evidence is presented for clean homolysis of the RS-Co bond in C6H11SCbl, GluSCbl, and CysSCbl to give RS. and .Co(II)Cbl radical pairs analogous to those that are intermediates in ribonucleoside triphosphate reductase (RTPR). A summary is provided which emphasizes the seven variables identified to date, underlying the successful syntheses and isolation of thiolatocobalamins, variables which make the one-step syntheses of RSCbls considerably more complex than they initially appear. Also briefly discussed are the analogous protein-S-Cbl complexes that are seen as side-products in RTPR, and the probability that such side-products are formed when HOCbl.HX is used as a possible 'active-site inhibitor' complex with B12-dependent enzymes.

A simple, convenient and direct method for assessing the purity of cobalamins.

A straightforward and simple, but powerful and direct, method is presented for both the detection and quantitation of cobalamin impurities in either commercial cobalamins or in metastable cobalamins (Cbls), such as RSCbls. The method is, quite simply, the use of the aromatic region of the 1H NMR of cobalamins; it is a method developed as an outgrowth of our work preparing metastable thiolatocobalamins (RSCbls) and is a method that proved necessary for characterizing those (and by inference other) cobalamins unstable to HPLC separation conditions (i.e., and, therefore, where the normally powerful HPLC method so commonly used in cobalamin chemistry fails). Despite considerable, prior, modern multidimensional NMR literature on cobalamins, the present method has not yet been indicated explicitly, nor has anyone reported previously the NMR data required to prove that the method works (i.e., the data for a series of cobalamins and their common impurities proving that they have different chemical shifts in the aromatic region of their 1H NMR when examined under identical NMR solvent, pH and other conditions). The direct NMR method is easy to perform, readily quantitated and applicable to species unstable to the HPLC conditions required to separate cobalamin impurities. The results have allowed quantitation of the 5-11% impurities in, for example, commercial HOCb1.HX, results which document that some commercially available cobalamins are not as pure as the manufacturers' claims.