NUCLEAR MAGNETIC RESONANCE STUDIES OF ALPHA-LYTIC PROTEINASE, A BACTERIAL SERINE PROTEINASE

WILLIAM MILO WESTLER, Purdue University

Abstract

I have resolved and assigned a proton nuclear magnetic resonance (('1)H NMR) peak in 360 MHz spectra of (alpha)-lytic proteinase ((alpha)LPase) from the C('(epsilon))-H of the active site histidine residue (histidine-57 in the chymotrypsinogen numbering system). The enzyme, (alpha)LPase, is a serine proteinase isolated from the soil bacterium Lysobacter enzymogenes which contains only a single histidine residue. I was able to resolve the histidine C('(epsilon))-H peak by using spin echo techniques to eliminate the interfering broader resonances arising from nonexchanging N-H groups. I assigned the peak to the histidine C('(epsilon))-H on the basis of its chemical shift and pH* dependence. This assignment was confirmed by ('1)H NMR spectroscopy of (alpha)LPase in which the histidine C('(epsilon)) was selectively labeled with 90% ('13)C. At a given pH*, I found that the histidine C('(epsilon))-H peak could be found at two different chemical shifts depending on the history of the sample. I label the two peaks Ha and Hb and define (alpha)LPase-a and (alpha)LPase-b, respectively, as two conformational species of the enzyme that interconvert slowly. The pK('')(,a) of histidine-57 in the two conformational forms of (alpha)LPase, obtained from the pH* dependence of the chemical shifts of peaks Ha and Hb, differ by 0.6 pH units. Histidine-57 in (alpha)LPase-a has a pK('')(,a) of 6.52 (+OR-) 0.07 whereas in (alpha)LPase-b it has a pK'a of 5.88 (+OR-) 0.04. The magnitude of the titration shift and the chemical shift at a given pH* of the C('(epsilon))-H peak in each form of the enzyme is relatively normal. The data indicate tha the pK('')(,a) of histidine-57 is higher than that of aspartate-102 rather than the reverse predicted for the "charge relay" mechanism of hydrolysis by serine proteinases. (alpha)LPase-a is prepared upon exhaustive dialysis of the enzyme against deionized water and subsequent lyophilization. (alpha)LPase-b is prepared from (alpha)LPase-a by incubation in ('2)H(,2)O or ('1)H(,2)O at pH* 8.0 for 12 to 14 h. The half time of conversion is about 2.5 h. (alpha)LPase-b can be reconverted to (alpha)LPase-a by dialysis and lyophilization. A gel filtration study of (alpha)LPase-a and (alpha)LPase-b on a high pressure column showed no difference in the retention times for the two species, ruling out a monomer-dimer equilibrium as the cause of the two forms. The active site serine-195 of (alpha)LPase was chemically modified by reaction with diisopropylphosphofluoridate (iPr(,2)P-F), a transition state analog inhibitor. The pK('')(,a) of histidine-57 in this derivative may reflect the pK('')(,a) value of this residue in the transition state during the catalytic hydrolysis of substrates. The pK('')(,a) of histidine-57 in iPr(,2)P-(alpha)LPase determined by the pH* dependence of the proton chemical shift of the C('(epsilon))-H is 8.16 (+OR-) 0.03. The phosphorous-31 NMR peak of the iPr(,2)P group exhibits a pH* dependent spectroscopic perturbation caused by the titration of histidine-57; the pK('')(,a) derived from the phosphorous-31 NMR data is 7.9 (+OR-) 0.2. The kinetic parameters for the hydrolysis of succinyl-tri-L-alanyl-p-nitroanilide (STAPNA) by (alpha)LPase-a and (alpha)LPase-b are different. The k(,cat)/K(,M) value for (alpha)LPase-b is a factor of two greater than that for (alpha)LPase-a. Several NMR spectral features of (alpha)LPase have time dependent intensity changes during the conversion of (alpha)LPase-a to (alpha)LPase-b. The rate constant for these changes, assuming pseudo first order kinetics, is (TURN)7 x 10('-5) s for all the observed peaks. The C('(epsilon))-H peak has about the same rate constant for the conversion of one chemical shift form to the other, this suggests that the conversion of (alpha)LPase-a to (alpha)LPase-b is a one step transition.

Degree

Ph.D.

Subject Area

Biochemistry

COinS