Time-resolved fluorescence using time-correlated single-photon counting is a versatile tool for biophysical studies. By resolving the nanosecond timescale decay of fluorescence we can obtain information about the local environment of fluorophores (via time-resolved anisotropy, for example) and, with suitable donor and acceptor fluorophores, distance distributions in biological macromolecules. By resolving the lifetime one can also discern sub-populations in heterogeneous samples. For example, a marginally stable protein or peptide may populate both a high-FRET and a low-FRET state. This will be very apparent in thetime-resolved fluorescence decay of the donor excited state. In other words, rather than measuring an averaging distance (which may not reflect any meaningful distance in a heterogeneous sample) one measures the distribution of distances. One way to visualize the information content of a time-resolve decay is as follows:
Each colored block represents one photon from a single molecule. The color corresponds with the energy transfer efficiency – the early arriving photons are statistically more likely to come from donors that have a closer acceptor. In this way one can do a Laplace transform and obtain a distance distribution of the donor-acceptor distance. The least biased way to do this is by maximum entropy and details of this approach can be found in the Software section of my web site.
To do a time-resolved fluorescence experiment one needs a pulsed laser, fast detectors and fast electronics capable of storing the arrival time or each photon with ps accuracy. Our time-resolved fluorescence instrument is setup for performing both UV and VIS fluorescence experiments. Our excitation source consists of a Ti:Sapphire laser (Coherent Verdi V10 pumped Mira 900D) with a pulse picker and harmonic generation optics. The excitation source is shown here:
Our sample compartment can be a microplate (used for high-throughput screening), a synthetic fused silica flow cell (for autosampler based measurements), a microfluidic chip or a static cuvet. We use Becker and Hickl electronics for all of our TCSPC applications – they are very versatile and integrate well with our custom built instruments.
The TCSPC experiment is typically run using 3 programs: The TCSPC acquisition program from Becker-Hickl (shown on the right, below), the sample controller (shown in the middle, below) and the titration or autosampler program (shown on the left, below). One could run everything manually, too, of course. The reason why I like to use the autosampler, even if running one sample at a time, is because the sample can be flowed in a controlled manner. This significantly minimizes photobleaching and gives very clean and reproducible data.
The data analysis is typically done using Savuka and its associated utilities.