No abstract

1H NMR chemical shift assignments were established for Nδ1H (16.9 ppm) and Nε2H (16.1 ppm) of the active-center His57 for the complex of MeOSuc-Ala-Ala-Pro-boroPhe (BoroPhe) with chymotrypsin and for the Cε1H proton (9.2 ppm at low pH and 8.5 ppm at high pH) of His57 in uncomplexed chymotrypsin. The assignment for Cε1H corrects previous assignments and reveals an unusual environment of this carbon-bound proton. The relative NH assignments are reversed from the order of NH assignments previously found for α-lytic protease complexes with boronate inhibitors. Isotopic fractionation factors (H/D) were determined using 1H NMR for hydrogen bonds to the active site histidine in BoroPhe complexes with chymotrypsin and subtilisin E, and for uncomplexed chymotrypsin. Measured fractionation factors accurate to about ±0.1 were 0.82 (pH 10) and 0.64 (pH 3) for the Nδ1H proton of uncomplexed chymotrypsin. In the presence of BoroPhe at pH 6.5, the Nδ1H fractionation factors were 0.65 for the chymotrypsin−inhibitor complex, and 0.53 for the subtilisin−inhibitor complex. Measurements for the Nε2H fractionation factor were 1.05 (uncomplexed chymotrypsin at pH 10), 0.93 (BoroPhe−chymotrypsin at pH 6.5), and 0.76 (BoroPhe−subtilisin at pH 6.5). Both model calculations of isotopic fractionation factors and experimentally determined inhibition constants were used in the analysis of the fractionation-factor results.

Vibrational analysis calculations have been conducted in an effort to understand the origins of apparently anomalous relationships reported between kH/kT and kdkT kinetic isotope effects. The actual isotopic substitutions used in previous work are more complex than a simple comparison between H/T and D/T isotope effects. The relationships between the actual isotope effects determined in these studies amount to tests of two standards of conventional isotope effect theory, the rule of the geometric mean (or the lack of isotope effects on isotope effects) and the Swain-Schaad rule. Model calculations illustrate the importance of violations of the rule of the geometric mean (over Swain-Schaad deviations) arising in models that incorporate both explicit reactioncoordinate coupling of two isotopic sites and reaction-coordinate tunneling. Implications for experimental studies of tunneling using apparent Swain-Schaad exponents are discussed.

S-Adenosylmethionine, generated enzymically in chirally pure form (S configuration at sulfur), undergoes simultaneous irreversible conversion to 5'-deoxy-5'-(methylthio)adenosine and homoserine with a rate constant of 6 X 10(-6) s-1 at pH 7.5 and 37 degrees C and reversible conversion to an enzymically inactive stereoisomer (R configuration at sulfur) with a forward rate constant of 8 X 10(-6) s-1 at pH 7.5 and 37 degrees C. These forms of instability require small turnover times and/or stabilization through macromolecular binding for S-adenosylmethionine, if organisms that utilize it are to avoid losses of metabolic energy.