Time-resolved structural methods capture macromolecular structures at defined time points during a biological process, providing atomic-resolution movies of conformational changes, catalytic cycles, and ligand binding. Standard crystallography and cryo-EM produce static, time-averaged structures — they reveal where atoms are but not how they move. Time-resolved approaches overcome this limitation through several strategies: serial femtosecond crystallography (SFX) at X-ray free-electron lasers (XFELs) collects diffraction from microcrystals before radiation damage occurs (the "diffraction-before-destruction" principle), enabling room-temperature structures and pump-probe experiments where a light pulse or substrate triggers the reaction and the XFEL pulse captures the structure at a defined delay time. Time-resolved cryo-EM captures intermediates by rapid mixing or photolysis followed by plunge-freezing at controlled time points. These methods have revealed catalytic intermediates in enzymes, light-driven conformational changes in photoreceptors, and the structural dynamics of molecular machines in real time.
For most of its history, structural biology has produced static pictures of molecules. A crystal structure shows where atoms are on average; a cryo-EM map shows a frozen snapshot. But biological function is inherently dynamic: enzymes catalyze reactions through sequences of conformational changes, molecular machines like the ribosome and ATP synthase cycle through multiple structural states, and signaling proteins switch between active and inactive conformations. Time-resolved structural methods aim to add the dimension of time — capturing not just where atoms are, but how they move during biological processes.
X-ray free-electron lasers (XFELs) represent the most dramatic advance in time-resolved structural biology. An XFEL generates X-ray pulses of extraordinary brightness (10^12 photons per pulse) and ultrashort duration (10-50 femtoseconds). These pulses are so intense that they vaporize any crystal they hit — but the diffraction pattern is recorded before the crystal is destroyed, because the pulse duration is shorter than the timescale of radiation-induced atomic motion (the diffraction-before-destruction principle). This has two transformative consequences. First, data can be collected at room temperature rather than the cryogenic temperatures (100 K) required at synchrotrons, capturing proteins in their native conformational ensemble rather than cryo-trapped states. Second, pump-probe experiments become possible: a laser pulse (the pump) triggers a reaction in the crystal (e.g., photoisomerization of a chromophore), and the XFEL pulse (the probe) captures the structure at a precisely controlled delay time (from femtoseconds to seconds). By varying the delay, a molecular movie is assembled frame by frame.
Serial femtosecond crystallography (SFX) is the data collection strategy that makes XFEL crystallography practical. Since each crystal is destroyed by one pulse, a continuous stream of microcrystals (1-30 micrometers) is injected across the XFEL beam. Each crystal diffracts in a random orientation, producing a single still image (no oscillation). Tens of thousands of such images are merged using algorithms (CrystFEL, cctbx.xfel) that index each pattern and scale the reflections, reconstructing a complete dataset from the partial observations. The requirement for large numbers of microcrystals is both a challenge (growing sufficient microcrystals is nontrivial) and an advantage (many proteins form microcrystals more readily than the large single crystals needed for synchrotron work). For time-resolved experiments, the pump laser illuminates the crystal stream microseconds to seconds before the XFEL pulse, and different delay times are interleaved during the experiment.
Time-resolved cryo-EM takes a complementary approach suited to larger conformational changes and non-crystalline samples. The strategy involves rapidly mixing the macromolecule with its substrate or trigger using microfluidic devices (achieving mixing times of ~1 millisecond), allowing the reaction to proceed for a controlled interval (milliseconds to seconds), and then plunge-freezing to trap the intermediate state. Because cryo-EM works on individual particles in solution, there is no crystal lattice to constrain conformational changes, and the computational classification methods developed for single-particle analysis can separate a mixed population of intermediates into distinct structural classes — effectively performing the temporal sorting after data collection rather than requiring temporal synchrony in the sample. This approach has captured ribosome dynamics during translocation, ATP-driven conformational changes in chaperonins, and catalytic intermediates in spliceosomes. Together, XFEL serial crystallography and time-resolved cryo-EM are fulfilling structural biology's ambition of watching molecular machines in action, providing atomic-resolution understanding of how structure changes drive biological function.
No topics depend on this one yet.