Single molecule study of enzyme kinetics: alpha-Chymotrypsin activity with single turnover resolution

The outstanding catalytic activity and selectivity of enzymes has always interested scientists. Since the discovery of enzymes at the end of the 19th century a large number of different studies has been performed to shed light on the relationship between the enzyme structure and these unique properties. In general, this relationship is well understood. In recent years it has become clear, however, that dynamic changes in the complex three-dimensional structure need to be considered to understand the principles of enzyme activity. It is now well recognized that conformational changes are crucial for the function of many enzymes. The enzyme structure is constantly exposed to fluctuations. These conformational fluctuations can further lead to a distribution of catalytic activities in an enzyme population. Classical ensemble experiments cannot reveal conformational heterogeneities and related activity fluctuations.The use of Single Molecule Fluorescence Spectroscopy allows monitoring dynamic processes in real time with high time resolution and sensitivity. Studying enzymes at the single molecule level thus allows the identification of different conformational states and their interconversion rates. Studies performed with single turnover resolution have suggested that the activity fluctuates over time (dynamic disorder) suggesting that enzyme catalysis might be influenced by conformational dynamics. This dissertation aims to utilize the power of Single Molecule Fluorescence Spectroscopy to gain more detailed insight in the relationship between the catalytic process and the conformational dynamics of the enzyme a-chymotrypsin. This enzyme was chosen as a model system as a-chymotrypsin has been thoroughly studied at the ensemble level and has great applicability in industry.The most general approach for studying enzymes at the single molecule level is based on fluorogenic substrates. Ideally, each enzymatic turnover produces a highly fluorescent product molecule that can be detected. To facilitate the detection of every enzymatic turnover, required for a correct data analysis, the reaction needs to have 1:1 stoichiometry. Many of the commonly used fluorogenic substrates are based on dyes, however, that contain two equal sites for the conjugation of enzyme-cleavable groups. In chapter 2 the synthesis of substrates based on morpholinecarbonyl-Rhodamine 110 is described. These substrates with only one cleavable bond were characterized at the ensemble level using a-chymotrypsin. The kinetic constants were compared with the kinetics of frequently used Rhodamine 110 based substrates with two cleavable peptide bonds. This comparison showed that double-substituted substrates yield only apparent Km and kcat values that are quite different from the real kinetic parameters. Moreover, the investigation of the photophysical properties of morpholinecarbonyl-Rhodamine 110 and peptide-Rhodamine 110 showed that morpholinecarbonyl-Rhodamine 110 has higher fluorescence intensity than peptide-Rhodamine 110 derivatives. This higher intensity does not originate from the earlier proposed equilibrium between a non-fluorescent lactone form and a fluorescent zwitterionic form but depends on the electronic structure of these dye molecules. Overall, due to their relatively high brightness and their 1:1 stoichiometry the newly synthesized substrates are a very useful alternative for the previously used double-substituted Rhodamine 110 based substrates both at the ensemble and at the single molecule level.When using a confocal fluorescence microscope to detect the individual product molecules, the data consists of a sequence of arrival times of individual photons emitted by the fluorescent product. The assignment of fluorescent on- and off-states corresponding to the dwell times of the enzymatic reaction is difficult due to the intrinsic noise in such an experiment. In this dissertation a new method was applied, validated and improved. This new method, that determines intensity change points with a purely statistical approach, is described in chapter 3. Using simulated photon arrival time data, the performance of the new method was compared with the commonly used binning/thresholding approach. This comparison included a systematic investigation of how the accuracy of the analysis is influenced by the bin size and the threshold position as well as the signal-to-noise ratio and the photon count rate. It was shown that the enzymatic dwell time distributions are not always correctly reproduced by the data analysis procedure. In general, the new method is not only more accurate but also reduces the subjective factor of the binning/thresholding approach.The presence of dynamic disorder in the enzymatic activity was previously concluded from a stretched exponential off-time distribution and correlations between consecutive off-times observed in several single enzyme experiments. The analysis of the simulated data in chapter 3 showed that the stretched exponential distributions might originate from inaccuracies of the binning/thresholding method. Chapter 4 describes the results of the analysis of experimental data with both methods. Also this analysis shows that some of the previous observations of dynamic disorder are most likely artifacts introduced by the data analysis procedure rather than an intrinsic property of the enzyme.Single molecule fluorescence spectroscopy with time-correlated single photon counting detection gives access to multiple parameters such as fluorescence intensity, lifetime and polarization. In chapter 5, this multiparameter approach was applied with the goal to try to improve the accuracy of the assignment of fluorescent states in the single enzyme experiments and to investigate the two-step hydrolysis of Rhodamine 110 based substrates with two cleavable peptide bonds. Utilizing differences in the decay times of the peptide-Rhodamine 110 intermediate and Rhodamine 110 it was possible to determine the sequence of events in the two-step cleavage reaction. It was found that only the intermediate is produced at high substrate concentrations proving that in this specific example every enzymatic turnover can be detected.Finally, having optimized the data analysis and knowing that every enzymatic turnover is detectable, experiments were performed to investigate the known pH-dependent conformational changes of a-chymotrypsin and to extract the underlying kinetic scheme. The results, described in chapter 6, show a heterogeneous response of different enzymes to a pH increase. Despite these heterogeneities, a gradual decrease of the activity was observed for most enzymes in agreement with the known transition to an inactive conformation at high pH. No distinct periods of high and low enzyme activity were observed, however, leading to the conclusion that conformational transitions do occur during the measurements but that the rates are faster than or equal to the rates of the catalytic reaction.In summary, this dissertation clearly highlights the potential of single molecule experiments to study enzyme kinetics. A more accurate data analysis procedure has been introduced and strategies have been developed that allow the detection of every enzymatic turnover. These important improvements now open up the way for investigating conformational dynamics and to extract the underlying kinetic scheme from single turnover time series.status: publishe