Isotope effects on chemical shift are well known, and have been shown to reflect the chemical environment of the observed nucleus. This effect has been exploited to study hydrogen bonding within small molecules in organic solvents. In this project we wish to investigate the utility of the isotope shifts for probing hydrogen bonds within protein in aqueous solution. There has been a great deal of recent interest in the possible role of """"""""strong"""""""" hydrogen bonds in catalysis. Protons believed to be involved in these strong hydrogen bonds have NMR resonances that are strongly downfield shifted. Initial experiments are to examine the isotope effect on chemical shift of some of these strongly shifted protons, and establish the degree of correlations between isotope shift, chemical shift and functional behavior. For small molecules it has often been possible to measure the primary isotope effect on shift by comparing hydrogen and deuterium. The linewidths for deuterium are substantially larger due to its quadrupole coupling, but for short correlation times the lines are sharp enough to accurately determine the peak centers. However for proteins, with their much longer correlation times, deuterium lines are so broad that they cannot be detected, and the comparison must be done between proton and tritium. Since the hydrogens involved are labile (exchanging relatively rapidly with solvent) the measurements must be done in tritiated water (ca. 2% T in H or D). Detection then requires suppression of the bulk tritium signal from water, but this is done in the same way that solvent suppression is normally done for detection of proton signals in protonated water. Initial studies have been done on RNase A and chymotrypsin. For RNase at 10 mM concentration the fairly broad, downfield shifted resonances could easily be detected in the tritium spectrum. Lines at 13.736, 13.162 and 12.470 ppm had isotope shifts of -0.177, -0.116 and -0.099 ppm respectively. Thus there does seem to be a general correlation of chemical shift and isotope shift. In chymotrypsin the sample was only 2 mM, and the proton resonance at about 18.5 ppm in the inhibited enzyme was not visible in the tritium spectrum. Thi s experiment will be repeated with effort to optimize the sensitivity of detection, and with longer acquisition time if needed. An additional experiment is scheduled to do T-15N correlations to detect isotope effects on both T and 15N in a sample of Staphylococcus nuclease.
Showing the most recent 10 out of 28 publications
