Metal Fluorides as Probes for Enzyme Catalysed Phosphoryl Transfer

The work described in this thesis is based on the use of multinuclear NMR and protein crystallography as two powerful and symbiotic biological structural tools to study the catalytic mechanisms of five phosphoryl transfer enzymes: β-phosphoglucomutase, phosphoserine phosphatase, cAMP-dependent protein kinase, small G protein RhoA-
RhoGAP, and UMP/CMP kinase. The central feature is the use of magnesium and aluminium fluorides as transition state analogues for phosphoryl group transfer.
β-Phosphoglucomutase is first examined using stable βG1P phosphonate analogues of which (S)-βCHFG1P and βCH2G1P form MgF3– and AlF4– TSA complexes with βPGMwt. Next, mutant studies show D10N and D8E form AlF4– complexes while D8N does not. Only D10N forms a fluoroberyllate complex on the nucleophilic Asp 8, which can also be phosphorylated by acetyl phosphate.
The general acid-base role of D13 in PSP is further established by showing the inactivity of PSPD13N, a mutant that forms MgF3– and AlF4– TSA complexes with water
but not serine. Fundamental functional differences between PSP and βPGM are rationalised.
cAPK forms MgF3– and AlF4– TSA complexes, disproving the ‘AlF30’ assignment in PDB: 1L3R. The binding constant of cAPK for its MgF3– TSA is only ~10 fold weaker than for the AlF4– TSA complex. This suggests that fluoride inhibition of a wide range of signalling proteins may be dominated by MgF3– rather than by the more widely
studied AlF4– with significant physiological implications.
19F NMR data confirm the RhoA-GDP-MgF3–-OH2-RhoGAP TSA solid state structure, and identify an AlF4– TSA complex that obeys the Charge Balance Hypothesis. A GAP-free octahedral RhoA-GDP-AlF3(H2O)0 complex is formed that also conforms to the CBH. NMR and crystallography show RhoA-GDP-MgF3–-RhoGAPR85A and RhoA-GDP-AlF4–-RhoGAPR85A TSA complexes maintain negative charge on the metal fluoride moieties via Tyr 34 coodination. Finally, substrate analogue GTPγF is remarkably stable to hydrolytic acitivity by RhoA alone, or with RhoGAP with or minus its arginine finger.
19F NMR data disprove the earlier assignment of tbp ‘AlF30’ in UMP/CMP kinase structures crystallised at pH 8.5 and show MgF3– is the actual TSA at high pH.
NMR/pH titration reveals a new intermediate aluminium TSA species, AlF3Z, not hitherto identified. NMR analysis endorses earlier solid state observations on a UMPK-ADP-BeF2
0-UDP TSA complex and demonstrates the catalytic promiscuity of UMPKdicty. AMPCP is a stable substitute for ADP in TSA complex formation while AMPCF2P or ADPβF fail unexpectedly. Similarly, UMP cannot be replaced by
5FdUMP.
This thesis is a prime example of the symbiotic use of
19F NMR and protein crystallography to study enzyme mechanisms. 19F NMR can identify TSA complex formation in solution and hence can direct crystal trials. Crystal structures are more accurately assigned using NMR data while, in turn, they give further support to detailed
interpretation of NMR signals.