The combination of X-ray crystallography and rapid cryo-trapping methods has enabled the visualization of catalytic intermediates in a variety of enzyme systems. However, the resolution of the X-ray experiment is not always sufficient to precisely place the structure on the reaction pathway. In addition, many trapped intermediates are X-ray-sensitive and can decay during diffraction data collection, resulting in a final structure that may not be representative of the initial trapped species. Complementary methods, such as single-crystal spectroscopy, provide a means to precisely identify the cryo-trapped species as well as detect any X-ray-induced changes during diffraction data collection.
- diffraction data collection
- electron-density map
- enzyme system
- intermediate trapping
- single-crystal spectroscopy
- X-ray crystallography
There have been considerable advances in the last 10 years in both the demand for and availability of instrumentation for the spectroscopy of protein crystals and, more recently, nucleic acid crystals. This has been driven by two factors. The first is the observation that many enzymes are capable of catalytic turnover when in the crystalline state, even at the extremes of pH, high viscosity and high ionic strength that are common in crystal stabilizing solutions [1,2]. Owing to the combination of restrictions the crystalline lattice places on protein breathing motions, the different microenvironments experienced by crystallographically independent active sites and the composition of the crystal stabilizing solution, catalysis is often slowed down in the crystal relative to solution. This can enable the build-up of species that are normally extremely transient. Rapid cooling of the crystal to 100 K effectively arrests the reaction, trapping the intermediate species ready for X-ray structure determination . Single-crystal spectroscopy is a vital tool to both follow the course of the reaction in the crystal and determine the optimal moment for cryo-trapping to catch the highest population of the desired intermediate. It can also aid in the assignment of a structure to a particular point on the reaction pathway in cases where the final electron density is ambiguous.
The second factor in the development of single-crystal spectroscopy is the increasing recognition that X-rays can cause rapid changes in redox state and structure during routine X-ray data collection, even at low absorbed X-ray doses [3,4]. These fast X-ray-induced changes are of particular relevance to the structural enzymology community, who seek to probe enzyme function and mechanism through the structures of trapped intermediates. These changes can occur long before the classical signs of X-ray damage, such as loss of diffraction quality, become evident . In addition, owing to their electrophilic nature, enzyme active sites are often the first targets of X-ray-induced changes. This is particularly true for those containing redox-active centres such as metals or cofactors. In addition, heavy atoms, such as metals or phosphate, bound at active sites have a higher probability of absorbing incident X-rays that can then be re-emitted, causing local damage [6,7]. Low-dose X-ray damage has been shown to have a variety of effects, including redox changes in both metals and organic cofactors [5,8] and breakage of radiation-sensitive bonds in both protein and ligands [9,10]. If detected, these X-ray-induced changes can either be minimized by the careful design of diffraction data collection strategies or exploited to drive catalysis and trap intermediate species that would be otherwise unobservable [8,11].
A wide range of spectroscopic instruments have now been made available to the structural biology community. These include instruments for optical spectroscopy, such as UV–visible absorption, fluorescence, (resonance) Raman and IR spectroscopies, as well as EPR and EXAFS spectroscopies (Table 1). This variety, particularly the addition of vibrational spectroscopic methods, means that single-crystal spectroscopy is no longer restricted to systems possessing a visible chromophore. Several of these systems are located at synchrotron sources and are available as a user resource (Table 2).
Here, we review the use of single-crystal spectroscopy to track enzyme catalysis in the crystal and discuss several examples of the recent use of single-crystal spectroscopy to identify trapped intermediates as well as to follow changes in redox state during X-ray exposure. We also discuss new developments in the instrumentation available for single-crystal spectroscopy at synchrotron sources.
Tracking and trapping intermediates in the crystal
As long as there are no large conformational changes in the enzyme related to catalysis, most enzymes are able to undergo catalytic turnover in the crystal without disruption of the lattice and loss of diffracting power. However, this requires the enzyme active site to be accessible to solvent, i.e. not blocked by a crystal contact. Useful information can still be obtained in such cases, but the use of more traditional methods such as co-crystallization with substrates, products or analogues is required. Single-crystal spectroscopy can still be useful in these cases to provide additional information to identify serendipitously trapped species when the resolution of the electron-density map is insufficient to do so; however, we will not discuss this further in the present review. If multiple crystal forms are available, it is often worth investigating all of them, as the distinct packing interactions may stabilize different stages of catalysis [12,13].
A key step when carrying out reactions in crystals is the full characterization of the system in both the solution and crystalline phases. Owing to the restrictions imposed by the crystal lattice and the properties of crystal stabilizing solutions, rates of catalysis can be considerably altered, particularly in those steps that require some structural rearrangement. Relative rates can also change and this can result in the trapping of off-pathway, non-physiologically relevant species, although these can potentially yield useful information on the mechanism. This phase characterization can be done either in capillary flow cells or in humidified droplets and detailed procedures for this can be found elsewhere [14,15]. Once the initial characterization has been carried out, two trapping strategies exist to obtain maximum build-up of the desired intermediate in the crystal, and these are detailed below.
By far the easiest method of trapping reaction intermediates in the crystalline state is to carry out a single turnover reaction while adjusting the reaction conditions to arrest catalysis at a specific point. This results in a build-up of a single species in most of the crystal that will subsequently dominate the final electron-density map. Changing the reaction conditions can be done in a variety of ways. For example, if catalysis proceeds via a deprotonation step, lowering the pH can gate the reaction, resulting in build-up of the preceding intermediate [16,17]. Alternative methods include controlling the availability of electron donors and acceptors [16,17], the use of non-native substrates that cannot be completely turned over  and controlling the availability of ions that promote or retard catalysis .
Unlike mechanistic trapping, kinetic or on-the-fly trapping involves the rapid cryo-cooling of a crystal at various time points after the addition of substrate to the crystal-stabilizing solution . As rates are considerably lower in the crystal than in the solution, it can often take seconds or even minutes for all the molecules to turn over, allowing the reaction in the crystal to be halted by rapid cooling to 100 K in liquid nitrogen. The simplest form of kinetic trapping is to drive the system into a steady state in which most of the molecules should have accumulated at the intermediate preceding the slowest step in catalysis . However, it must be noted that the rate-determining step in the crystal should not be assumed to be the same as that observed in solution. In order to trap non-steady-state species, single-crystal spectroscopy is key to defining the time points at which maximal build-up of each intermediate occurs [8,11,19]. The spectroscopic data also provide information on the relative abundance of different species that can be extremely useful when deconvoluting an electron-density map that contains several different species .
Detecting and tracking X-ray-induced changes
Single-crystal spectroscopy has also been crucial in the design of data collection strategies to deal with the problem of rapid X-ray-induced changes in oxidized and/or trapped species. A large number of photoelectrons are generated in the crystal as soon as X-rays are incident on the sample , and even at cryotemperatures these can propagate through the crystal. An increasing number of studies have shown that electrophilic sites are specifically and rapidly damaged during X-ray exposure, resulting in redox changes and bond breakages within a fraction of the time required to collect a complete dataset [9,10,22]. This means that the electron-density map calculated at the end of data collection may be dominated by a radiation-damaged structure. One way to reduce this problem is to collect wedges of data from multiple crystals to generate a composite dataset. As a rule of thumb, the absorbed X-ray dose for each wedge should be such that not more than 20% change in the site of interest has occurred (if a subpopulation is present at less than 20% occupancy, it is unlikely to be visible in the electron-density map unless very high-resolution diffraction data are obtained). Single-crystal spectroscopy has been extremely useful in determining the maximum X-ray dose that can be tolerated per wedge before this limit is reached [5,8,23].
Several studies have taken advantage of the specific X-ray-induced changes at enzyme active sites to generate and structurally characterize extremely transient species. This is best exemplified by work on haem enzymes such as cytochrome P450cam and horseradish peroxidase [8,23]. In both cases, single-crystal spectroscopy showed that oxygen activation could be driven by X-ray exposure, with several intermediates forming sequentially as the X-ray dose increased. Compilation of multicrystal composite datasets from wedges corresponding to the maximal occupancy of each intermediate, as determined by spectroscopy, allowed structure determination of these intermediates, revealing the oxygen activation mechanism in exquisite detail.
Single-crystal spectroscopy is a vital tool for the structural enzymologist and is increasingly available at synchrotron sources as a user resource. In recent years, there has been a considerable expansion in the spectral ranges accessible beyond simple UV–visible spectroscopy. This has opened the door to the application of the spectroscopically informed trapping methodologies described above to a much wider range of enzyme systems.
Enzyme Mechanisms: Fast Reaction and Computational Approaches: Biochemical Society Focused Meeting held at Manchester Interdisciplinary Biocentre, U.K., 9–10 October 2008. Organized and Edited by Nigel Scrutton and Andrew Munro (Manchester, U.K.).
- © The Authors Journal compilation © 2009 Biochemical Society