Molecules in solution change their conformations so quickly that no method has been able to record the process. This looks set to change, as infrared spectroscopy rises to the challenge.
The science of high-speed photography was created by Eadweard Muybridge in 1887, when he published his famous photographic sequence of a galloping racehorse. Scientists have since sought analogous microscopic and spectroscopic techniques to enable them to watch the motion of biological molecules in real time. As reported on page 469 of this issue1, Hamm and co-workers have transferred the principles of high-speed photography to two-dimensional infrared spectroscopy. In this way, events that occur over a few hundred picoseconds can be studied by taking a series of snapshots — that is, two-dimensional infrared spectra — of the molecules concerned.
Hamm and colleagues' technique is not just an ultra-high-speed version of Muybridge's method that uses infrared radiation; the smaller scale requires notable differences. The subject of the authors' study1 is a loop of four amino acids, known as a cyclic tetrapeptide, that is a billion times smaller than a horse. More pertinently, this tetrapeptide is 100 to 1,000 times smaller than the wavelengths of infrared or visible radiation, so it is impossible simply to take a photograph of it. Instead, the authors recorded a spectrum in which the absorption bands are associated with polarization changes induced by interactions of infrared radiation with the molecule. Many theoretical tools have been developed for transforming two-dimensional optical spectra into molecular structures. These tools have been successfully applied to provide information about a variety of processes, such as the making and breaking of hydrogen bonds between solute and solvent molecules2,3,4,5, energy transfers in a photosynthetic light-harvesting protein6, and changes in distance between oscillating groups found in biomolecules1,7.
Throughout these theoretical and experimental efforts, a recurrent theme has emerged: the effects of interactions between molecular groups are tiny. One-dimensional optical spectroscopy is used to measure the so-called primary properties of vibrating groups in molecules, such as their oscillation frequencies and dipole strengths. But interactions between different oscillators — such as couplings between vibrational or electronic oscillations — are known as secondary properties, and are up to 1,000 times smaller than the primary properties. Secondary properties cannot be selectively measured by most conventional spectroscopic techniques, because they are usually hidden under the dominant primary features. But it is the secondary properties that are of most interest when studying the way that a biomolecule folds and unfolds. The precise three-dimensional structure of a molecule can have a profound effect on the secondary properties of vibrating groups within that molecule, because these are often proportional to 1/Rn (where R is the distance between two oscillators and n is greater than or equal to 3).
Fortunately, two-dimensional optical spectroscopy is a superb tool for measuring the secondary properties of molecular oscillators, using a process known as differencing. In this technique, one spectrum is subtracted from another that has been obtained under alternative conditions, so that the observed differences can be correlated with secondary properties. The transient two-dimensional infrared spectroscopy used by Hamm and colleagues1 is actually a double differencing technique. To remove the obscuring layer of primary contributions, they recorded a 'pump–probe' spectrum; this is the difference between a spectrum obtained using a pump pulse of infrared (which excites the molecule) and a spectrum obtained using a probe pulse (which records the temporal evolution of the excited state). To obtain an even sharper glimpse of the temporal changes in secondary properties, differences between spectra recorded at different times are also calculated.
The authors used their technique to watch a biomolecule unfolding in solution, by observing the breakage of a hydrogen bond. They prepared a cyclic tetrapeptide containing a weak chemical bond known as a disulphide bridge, that holds the molecule in a folded conformation. They then used a picosecond pulse of ultraviolet light to break the disulphide bridge. This triggering reaction allowed the unrestricted molecule to unfold, and the authors recorded its movement as a series of two-dimensional infrared spectra using femtosecond pulses of infrared. One could compare Hamm's team to film directors, cueing the molecular action with a burst of ultraviolet and using infrared spectra as their movie frames; the shutter speed of their 'camera' is the duration of an infrared pulse, and the exposure time for each frame is the period between pairs of pulses.
By subtracting a reference spectrum from the transient ones, the authors revealed the secondary properties associated with the molecule's motion as cross-peaks in the spectra (Fig. 1). They identified four oscillators in the tetrapeptide, and with the help of computer simulations they were able to describe the temporal evolution of the cross-peaks' amplitude in terms of the time-dependent changes of length for a particular hydrogen bond. In other words, they were able to 'watch' the dynamics of the hydrogen bond as it stretched and broke.
Two-dimensional infrared spectroscopy is currently limited to studying certain types of oscillator in proteins, as only a narrow range of oscillation frequencies is visible using this technique. Furthermore, it faces increasing competition from an analogous technique in which optical spectroscopy is used to study coupling between electrons6. Although two-dimensional infrared spectroscopy has not yet burgeoned into a systematic discipline, its applications — such as studying the ultrafast dynamics of biomolecular and chemical reactions, or of protein interactions with other molecules — may be waiting just around the corner. With this approach, Hamm and colleagues have shown that the movements of participants in molecular dramas can be recorded in vivid detail.