Search

We have investigated the effect of pressure and temperature on the structural and thermodynamic stability of a protein dihydrofolate reductase from a deep-sea bacterium Moritella profunda in its folate-bound form in the pressure range between 3 and 375 MPa and the temperature range between -5 and 30 degrees C. The on-line cell variable pressure 1H NMR spectroscopy has been used to analyze the chemical shift and signal intensity in one-dimensional 1H NMR spectra. Thermodynamic analysis based on signal intensities from protons in the core part indicates that the thermodynamic stability of Moritella profunda DHFR is relatively low over the temperature range between -5 and 30 degrees C (deltaG0=15.8 +/- 4.1 kJ/mol at 15 degrees C), but is well adapted to the living environment of the bacterium (2 degrees C and 28 MPa), with the maximum stability around 5 degrees C (at 0.1 MPa) and a relatively small volume change upon unfolding (deltaV= 66 +/- 19 ml/mol). Despite the relatively low overall stability, the conformation in the core part of the folded protein remains intact up to approximately 200 MPa, showing marked stability of the core of this protein.

High-pressure (15)N/(1)H NMR techniques were used to characterize the conformational fluctuations of hen lysozyme, in its native state and when denatured in 8 M urea, over the pressure range 30--2000 bar. Most (1)H and (15)N signals of native lysozyme show reversible shifts to low field with increasing pressure, the average pressure shifts being 0.069 +/- 0.101 p.p.m. ((1)H) and 0.51 +/- 0.36 p.p.m. ((15)N). The shifts indicate that the hydrogen bonds formed to carbonyl groups or water molecules by the backbone amides are, on average, shortened by approximately 0.02 A as a result of pressure. In native lysozyme, six residues in the beta domain or at the alpha/beta domain interface have anomalously large nonlinear (15)N and (1)H chemical-shift changes. All these residues lie close to water-containing cavities, suggesting that there are conformational changes involving these cavities, or the water molecules within them, at high pressure. The pressure-induced (1)H and (15)N shifts for lysozyme denatured in 8 M urea are much more uniform than those for native lysozyme, with average backbone amide shifts of 0.081 +/- 0.029 p.p.m. ((1)H) and 0.57 +/- 0.14 p.p.m. ((15)N). The results show that overall there are no significant variations in the local conformational properties of denatured lysozyme with pressure, although larger shifts in the vicinity of a persistent hydrophobic cluster indicate that interactions in this part of the sequence may rearrange. NMR diffusion measurements demonstrate that the effective hydrodynamic radius of denatured lysozyme, and hence the global properties of the denatured ensemble, do not change detectably at high pressure.

In our continuing search to find systemically bioavailable ACAT (acyl-CoA:cholesterol O-acyltransferase) inhibitors with more potent antiatherosclerotic effect than N-[2-(dimethylamino)-6-[3-(5-methyl-4-phenyl-1H-imidazol-1-yl)propoxy] phenyl]-N'-pentylurea (3), a series of phenylureas linked to 4-phenylimidazole were synthesized and evaluated for in vitro inhibitory activity toward both aortic and intestinal ACATs, and for in vivo hypocholesterolemic activity. The structure-activity relationships (SARs) were studied by strategic modification of five regions in the molecule of 3, i.e., by introducing functional groups or exchanging carbon atoms for heteroatoms. The SAR studies allowed us to select optimum substituents in the five regions, as follows. (1) Dimethylamino was convertible into nitro, methyl, ethyl, propyl, isopropyl, and chloro. On the basis of preliminary pharmacokinetic studies, the methyl group in the ortho-position of the phenylurea was selected. (2) Butyl, pentyl, isopentyl, and neopentyl were better substituents in the urea moiety. (3) Propoxy was the optimal moiety in the bridging portion. (4) Proton, methyl, ethyl, isopropyl, hydroxymethyl, and chloro were better substituents at the 5-position of the imidazole moiety. (5) An unsubstituted phenyl ring was selected as the phenyl group of phenylimidazole. The subsequent comparison studies of compounds containing various combinations of the optimum substituents in each region resulted in the selection of two compounds (67, 68) for further pharmacological and toxicological testing. These compounds were orally bioavailable, and possessed potent in vitro aortic ACAT inhibitory activity (IC50 = 0.16 and 0.012 microM, respectively) and in vivo cholesterol lowering effect (46% and 52% at 1 mg/kg po, respectively). In particular, 68 was 10-fold more potent in the in vitro aortic ACAT assay and 5-fold more potent with respect to hypocholesterolemic activity in vivo than 3.