Principal Investigator Peter So
With the development of dual-color FCSand, more recently, tri-color FCS, it is now possible to analyze the dynamics of multi-component processes such as nucleic acid single strand annealing, antibody-antigen interactions, or ligand-receptor binding. In these studies, the cross correlation signal relies on good spectral segregation of the individual dyes. Presently, no two-color FCS system completely eliminates spectral bleed through or crosstalk in individual detection channels. Thus there exists a lower limit for detectable associated two-color product, which limits the visualization of important bimolecular processes.
With the group's recent development of a modified multi-anode PMT, we can spectrally monitor the emission of individual fluorophores simultaneously. Using multi-channel cross correlation analysis, we are attempting to use this technique to experimentally resolve three and more molecular species labeled with spectrally distinct chormophores.
Magde, Elson, and Webb developed Fluorescence Correlation Spectroscopy (FCS) in 1972 at Cornell University. Previous to FCS development, chemical kinetic constants and diffusion coefficients were measured by introducing perturbations to a chemical equilibrium and observing the system’s response. With FCS, fluctuations in a system can be measured in without perturbing its equilibrium.
Fluorescent particles emit light as they diffuse into and out of a defined open volume of light in a given time interval. Since the distribution of particles in the volume at any time is Poissonian, the fluctuation of the particles is given by N^(-1/2) where N is the number of molecules. Thus, FCS signal depends on the number of molecules present at a given time interval. As the number of molecules increase in a a defined open volume of light, fluorescent fluctuations from the average fluorescence for a given time interval decrease. On the other hand, as the number of molecules decreases, fluctuations become larger. When these fluctuations are correlated for a defined time interval, t, they give rise to a correlation signal. This correlation signal is directly proportional to molecular aggregation, molecular flow, receptor ligand binding, chemical kinetics, and diffusion.
Combining the advantages of improved excitation sources, optics, and photodetectors with readily available high quantum yield dyes with good spectral separation, multi-color FCS has emerged as a technique that already plays an integral role in biotechnology.
Although this independent signal for multi-labeled species improves their detection, a lower detection limit for multi-labeled species exists because of the of crosstalk between detection channels. If dyes have spectra that are not well resolved (as seen with Alexa 488 and 546, Molecular Probes, Eugene, OR), dichroics and filter choices become limited, photon counts are lost, and FCS signal and results are compromised.
In an attempt to provide a solution to this problem, we have combined a 16-channel multi-anode PMT with a 16-channel single photon counting card and a spectrograph to spread the fluorescent emission spectrum into the 16-channel array detector.
Since we are single photon counting, we can perform FCS or SRCS. The advantages to this configuration are (1) all of the emitted photons are collected along with (2) the spectral information. We capitalize on these advantages by using a different global fitting algorithm for data analysis (shown below).
For a 4 nM yellow-green (505/515) 20 nm bead solution (Molecular Probes), we took data at 10-microsecond time intervals for approximately 30 seconds.
Using mixed bead solutions, we attempt to recover their individual ratios using a global fitting algorithm. We attempt to recover the respective bead ratios for each solution using a global fitting algorithm shown below. This approach is compared with the traditional FCS fitting cross correlation data fitting algorithm.
In conclusion, the FCS measurements suggest that the combination of our unique detection system and our global fitting analysis routine improve upon standard FCS techniques. Although we have been capable of improving these FCS measurements, our detection sensitivity remains an inherent difficulty with our technique. Our attempt to solve this problem evolves around the redesign of the spectral detection scheme. We plan to replace the de-scanning optics and spectrograph with a silver mirror and AR coated Brewster cut prism to improve our signal to noise ratios.