Many of the initial applications of single molecule spectroscopy in biology have been in the realm of extremely sensitive imaging and analyte detection. While these are sure to continue to be important areas of single molecule work, there have more recently emerged a number of studies that are truly aimed at studying the spectroscopy and dynamics of single molecules and their reactions and interactions with the molecules and complexes that surround them (for recent reviews see ).
There are three major advantages of looking at the dynamics and spectroscopy of single biological molecules. The first is the study of processive or sequential dynamics (Fig. 1). A population of molecules undergoing a series of events, such as a complex assembly process or linear polymerization, become out of phase and the detailed dynamics of individual steps are lost. Of course, on the single molecule level, one watches an individual molecule throughout the course of events, removing the phase problem.
Another advantage of single molecule spectroscopy is that single molecules exist at any given time in particular conformational states with a particular solvent environment. Observing only population averages can hide dynamic or mechanistic features of biological molecules that are important to function. It can also have a profound effect on the optical spectra of molecules, particularly at low temperature where the solvent environment and the conformation are not changing rapidly. Working with single molecules allows one to measure optical spectra in the absence of inhomogeneous broadening, learning both a great deal about the number and identity of overlapping transitions as well as the underlying conformational heterogeneity that gives rise to broad spectra in populations of molecules (Fig. 2).
Finally, when combined with scanning probe microscopy (SPM), single molecule spectroscopy allows the measurement of mechanical or electrical properties of single molecules: binding forces, forces holding together secondary and tertiary structure of molecules, torque, bond strength, conductivity, etc. Thus, a major thrust of this work will be the development of a new, high performance, combined SPM/Single molecule fluorescence microscope for simultaneously manipulating, imaging and electronically probing single molecules while either monitoring the fluorescence or optically exciting the molecules using a laser focussed through a confocal microscope.
In general, it is necessary to obtain information in addition to fluorescence amplitude changes in order to interpret the dynamics of biological molecules at the single molecule level (for example see ). What one wants to do is to look for amplitude changes that are correlated to some physical parameter such as movement, binding or state changes. One of the major goals of this work will be to construct an optics and detector system that will simultaneously collect up to four types of correlated information from each photon: 1) the time at which the photon was collected, 2) the polarization of the photon, 3) the wavelength region (above or below some cutoff) of the photon and 4) the lifetime of the excited state that gave rise to the photon. A diagram of a detector system that should perform this photon by photon analysis is given in Fig. 5. Each microscope will have such a detector system (note that in general either polarization or wavelength will be measured, though it would be possible with the system under design to measure both simultaneously if two additional detectors were borrowed from one of the other microscopes).
Normally, laser scanners are sold as a unit with detectors for standard scanning confocal work. Both the type of detector and the optics preceding it will have to be modified for single molecule fluorescence work. After the scanning mirrors, the fluorescence will be sent to a home built confocal pinhole arrangement followed by either polarization or wavelength selection optics, an avalanche photodiode (APD) detector and a 2-dimensional timing board for recording time of arrival and excited state lifetime, as shown in Fig. 5.
Just as in bulk spectroscopy, polarization of fluorescence provides information about the rotation rate and therefore the size of the molecule and, more importantly, changes in the size or conformation of the molecule as the process in question proceeds. This will be very useful for studying assembly processes at the single molecule level and examples are given in the research section. Recently, time resolved anisotropy changes at the single molecule level have been recorded (e.g. ). Note that by correlating the relative changes in the vertical and horizontal components of the emission, noise that is not correlated to changes in size or conformation is removed.
Correlation of two different wavelength regions of emission is usually done when performing fluorescence resonant energy transfer (FRET) experiments (e.g. ). Here, the emission is again split into two detector channels, but this time based on what wavelength range it is in. Consider the experiment outlined in Fig. 6. As the two fluorophores approach each other, the amount of energy transfer increases. As a result, the fraction of the emission observed from the acceptor fluorophore increases at the expense of the emission from the donor fluorophore. Again, by correlating the changes in amplitude at two wavelength regions, uncorrelated noise is suppressed.
If a high repetition rate, mode-locked laser is used for excitation, time correlated photon counting can be used to give direct information about excited state lifetimes . Excited state lifetimes are amplitude independent, and thus immune to amplitude noise. Lifetime information is useful both in conjunction with the anisotropy change measurements (it allows a more direct measurement of the rotational time) and fluorescence resonance energy transfer experiments (the lifetime of the excited state will vary with quenching due to energy transfer, see Fig. 6). Complete boards are now available for photon counting that allow one to collect individual photons and associate with them both an excited state lifetime and a time of arrival (PicoQuant). Note that as long as a mode-locked laser source is used for excitation, each photon detected by each polarization or wavelength channel can be time correlated with the excitation pulse giving the excited state lifetime. This can be done routinely for every photon detected and it provides an additional very useful dimension of information.
Part of a Study performed with the DNA intercalating dye, thiazole orange (TO) is shown above. Here we compared the binding of TO to DNA by itself or covalently attached to a zinc finger DNA binding protein. In both cases, the TO only fluoresces when bound to DNA. Thus one is watching the TO bind and leave the DNA with time. One can see that the Zn finger-TO complex binds more often and longer than does the free TO. However, on average, the fluorescence intensity from the complex is less, probably due to constrains placed on the TO by the covalent attachment.
1. JK Gimzewski & C Joachim (1999) Nanoscale science of single molecules using local probes. Science 283: 1683-1688
2. AD Mehta, M Rief, JA Spudich, DA Smith & RM Simmons (1999) Single-molecule biomechanics with optical methods. Science 283: 1689-1695
3. WE Moerner & M Orrit (1999) Illuminating single molecules in condensed matter. Science 283: 1670-1676
4. S Weiss (1999) Fluorescence spectroscopy of single biomolecules. Science 283: 1676-1683
5. SM Nie, DT Chiu & RN Zare (1994) Probing individual molecules with confocal fluorescence microscopy. Science 266: 1018-1021
6. DC Daniel, M Thompson & NW Woodbury (1999) Fluorescence intensity fluctuations of individual labeled DNA fragments and a DNA binding protein in solution at the single molecule level: A comparison of photobleaching, diffusion, and binding dynamics. J. Phys. Chem. B (submitted):
7. L Lee, C-H Chen & L Chiu (1986) Thiazole orange - A new dye for reticulocyte analysis. Cytometry 7: 508-517
8. M Thompson & NW Woodbury (1999) Fluorescent and photochemical properties of a single zinc finger crosslinked with a fluorescent DNA-binding probe. Biochemistry (submitted):
9. AA Deniz, M Dahan, JR Grunwell, T Ha, AE Faulhaber, DS Chemla, S Weiss & PG Schultz (1999) Single-pair fluorescence resonance energy transfer on freely diffusing molecules: Observation of Förster distance dependence and subpopulations. Proc. Natl. Acad. Sci. USA 96: 3670-3675
10. T Ha, TA Laurence, DS Chemla & S Weiss (1999) Polarization spectroscopy of single fluorescent molecules. J. Phys. Chem. B 103: 6839-6850
11. T Ha, AY Ting, J Liang, WB Caldwell, AA Deniz, DS Chemla, PG Schultz & S Weiss (1999) Single-molecule fluorescence spectroscopy of enzyme conformational dynamics and cleavage mechanism. Proc. Natl. Acad. Sci. USA 96: 893-898
12. J Schaffer, A Volkmer, C Eggeling, V Subramaniam, G Striker & CAM Seidel (1999) Identification of single molecules in aqueous solution by time-resolved fluorescence anisotropy. J. Phys. Chem. A 103: 331-336
13. S Wennmalm & R Rigler (1999) On death numbers and survival times of single dye molecules. J. Phys. Chem. B 103: 2516-2519
14. RM Dickson, DJ Norris, Y-L Tzeng & WE Moerner (1996) Three-dimensional imaging of single molecules solvated in pores of poly(acrylamide) gels. Science 274: 966-968
15. AM Van Oijen, M Ketelaars, J Köhler, TJ Aartsma & J Schmidt (1999) Unraveling the electronic structure of individual photosynthetic pigment-protein complexes. Science 285: 400-402
16. AM Van Oijen, M Ketelaars, J Köhler, TJ Aartsma & J Schmidt (1998) Spectroscopy of single light-harvesting complexes from purple photosynthetic bacteria at 1.2 K. J. Phys. Chem. B 102: 9363-9366