UBC Theses and Dissertations
Magnetic resonance studies of trypsin Kang, Shyue-yue
A great advance in the understanding of the mechanism for enzymatic reactions on a molecular basis has resulted from knowledge of the three dimensional structure of several enzymes from x-ray diffraction methods. It is not possible, however, to determine the enzyme mechanism only by knowing its three dimensional structure. The dynamic aspect of the enzymatic reaction is required to understand its mechanism. Nuclear magnetic resonance (NMR) and electron spin resonance (ESR) are physical methods which contain both structural and dynamic information. This thesis presents studies of the interactions between the macromolecule, trypsin, and the small molecules, substrate like inhibitors, or ions, Ca⁺⁺, Mn⁺⁺ by magnetic resonance. The theories of nuclear magnetic relaxation in the presence of chemical exchange, relaxation mechanisms, and the methods of measurement of relaxation are presented in Chapter 1. Here are discussed the equations relating measured relaxation times to chemical exchange rates, chemical shifts and relaxation times of nuclei on small molecules in dynamic chemical exchange to a macromolecular site. Different exchange limits are discussed and means for distinguishing various limits are provided. In order to understand the effect of divalent ions (Ca⁺⁺, Mn⁺⁺, etc.) on the properties of trypsin, a study of Mn⁺⁺ binding to trypsin is described in Chapter II. Mn⁺⁺ was chosen as a model for Ca⁺⁺ binding, since Mn⁺⁺ is paramagnetic. Although all previous attempts to use NMR to interpret Mn⁺⁺ binding were based on use of "enhancement factors", I found that a more straightforward and clearer approach was to use the NMR relaxation times directly. The existing theory for effect of chemical exchange on NMR T-₁'s was extended to the case of three distinct chemical sites with all possible mutual inter-conversions, and applied the result to the binding of water to free Mn⁺⁺ and Mn⁺⁺ :enzyme complex. An exact treatment of the correction for the internal rotation of water at the Mn⁺⁺ binding site is also presented. The main conclusions were that Mn⁺⁺ binds strongly only on active trypsin and at just one site, and that the water bound to Mn⁺⁺ at that site can rotate rather freely, suggesting that the site must lie in an open region of the tertiary structure. These facts are most consistent with the binding of Mn⁺⁺at Asp 71 and Asp 153 (or Glu 77), where it has been suspected (but not shown) that Ca⁺⁺ may act to hold two loops of the enzyme together. The next two chapters are devoted to the study of the active site of trypsin by UV and NMR. This basic approach was to choose a homologous series of substrate-like inhibitors of trypsin, and study their binding to trypsin both by steady- state (uv) kinetics and also by NMR relaxation time measurements. This work would provide for the first time a direct comparison between the strength of binding (as measured by the binding constant from uv data) and the rigidity with which the inhibitor was bound to the enzyme (from NMR data). Any correlation, or lack of it, between these two parameters should provide more insight into theories of enzyme action. For rigorous NMR analysis, it was desirable to have inhibitors with - OCH₃ groups, to obtain a single, sharp NMR signal well-separated from other parts of the NMR spectrum. This requirement entailed the synthesis of a number of inhibitors, and in most cases, a given synthesis was not in the literature and had to be devised individually. This is described in the experimental section of Chapter III. In order to locate the binding sites of each inhibitor, and to obtain an accurate dissociation constant for each inhibitor, uv steady-state trypsin assays using D,L-BAPA as substrate were carried out. As a result, five of the interesting inhibitors gave Dixon plots with intersections below the x-axis, a result which cannot be explained by previous trypsin inhibitor work. The difficulty was eventually resolved by taking into account the interaction between the D-BAPA and my inhibitors. Although apparently a complication, the algebraic consequences showed that my seven trypsin inhibitors could be classified according to whether their binding was competitive, repulsive, non-competitive, or cooperative with the binding of D-BAPA. This then gave a rather complete picture of the inhibitor binding. The data definitely showed the presence of at least one secondary binding site, which is consistent with a number of unpublished X-ray results, and the secondary binding site exhibits some cooperative effect toward binding of a substrate analog. This had been observed on TAME substrate activation at high concentration. Chapter IV presents the measurements of the bound relaxation time of each of the inhibitors on trypsin by selective pulsed high resolution NMR (The measurement was made on the single sharp line of the methyl protons of the inhibitors). A special-purpose pulse unit is described briefly and the advantages and limitation of selective determination of relaxation time on high resolution NMR was also discussed. The relaxation time for the rigidly bound inhibitor was calculated and a correction for the effect of internal rotation of methyl group was made. The results strongly suggest that for the inhibitors of the same category (from uv), correlation between strength of binding and rigidity of binding can be demonstrated. The resultant implications for theories of enzyme catalysis was also discussed. In the last chapter, an ESR "spin label" to the active site serine of trypsin, with the intent of using the effect of this spin label on the NMR lineshape of my inhibitors has been used as a "ruler" to determine the distance from the active site to the secondary binding sites. Hopefully, the interaction between the "spin label" and Mn⁺⁺ on the enzyme is strong enough so that the distance, between them can also be estimated. The conformational change of the active site region with different purtubations (pH, Ca⁺⁺, inhibitors, etc) was also expected to be monitored through the changes of the ESR signal of the attached spin label. In order to ensure the protection of the spin labelled trypsin from autoproteolysis during the NMR experiments, the active trypsin-free, spin labelled trypsin was prepared successfully by a new process using soybean trypsin inhibitor. From the NMR measurements, it can be estimated that the distances between the spin label and the secondary binding sites are all around 9 to 10 Å. With the help of X-ray data, the location of the secondary sites may be speculated upon and the consequence evaluated. Due to the long distance between the spin label and the Mn⁺⁺ ion binding site, the interaction between these two paramagnetic species can not be observed. In addition, the ESR signal of the ser-195 spin label was not sensitive to the minor conformational changes induced by the various perturbations added.
Item Citations and Data