B-11 NMR-SPECTROSCOPY OF PEPTIDE BORONIC ACID INHIBITOR COMPLEXES OF ALPHA-LYTIC PROTEASE - DIRECT EVIDENCE FOR TETRAHEDRAL BORON IN BOTH BORON HISTIDINE AND BORON SERINE ADDUCT COMPLEXES

We have previously shown, using N-15 and H-1 NMR spectroscopy, that MeOSuc-Ala-Ala-Pro-boroPhe and certain other boronic acid inhibitors form boron-histidine adducts with alpha-lytic protease instead of transition-state-like tetrahedral boron-serine adducts as is generally supposed [Bachovchin, W. W., Wong, W. Y. L., Farr-Jones, S., Shenvi, A. B., & Kettner, C. (1988) Biochemistry 27, 7689-7697]. An X-ray crystallographic study of the MeOSuc-Ala-Ala-Pro-boroPhe complex with alpha-lytic protease [Bone, R., Frank, D., Kettner, C. A., & Agard, D. A. (1989) Biochemistry 28, 7600-7609] has confirmed the existence of the boron-histidine bond but has concluded that the boron atom is trigonal rather than tetrahedral. Here we report a B-11 NMR study at 160.46 MHz of this histidine adduct complex and of two other complexes known to be serine adducts: alpha-lytic protease with MeOSuc-Ala-Ala-Pro-boroVal and chymotrypsin with MeOSucAla-Ala-Pro-boroPhe. The B-11 NMR chemical shifts demonstrate that the boron atom is tetrahedral in both the histidine and serine adduct complexes. A comparison of our results with those of a previous B-11 NMR study of chymotrypsin complexed to MeOSuc-Ala-Ala-Pro-boroPhe at a lower magnetic field strength (64.21 MHz) [Zhong, S., Jordan, F., Kettner, C. A., & Polgar, L. (1991) J. Am. Chem. Soc. 113, 9429-9435] indicates that the motional properties of the enzyme-bound B-11 nucleus place it outside of the extreme narrowing region and that the difference in resonance frequency between the free and the enzyme-bound tetrahedral boron can largely be attributed to a second-order dynamic frequency shift.