Approach

Despite a commonly held belief that the spatial resolution in wide-field fluorescence microscopy is constrained by the diffraction limit, CBST researchers, led by Drs. Gustafsson, Sedat, and Agard at UCSF and including participants from UCD and LLNL, have now shown that the resolution of fluorescently labeled objects can in fact be improved using a structured illumination (SI) technique.^{2, 3}  Currently in our experimental setup a 1-D linear illumination pattern (see sigure) combines with high-spatial frequencies in a sample to produce low-frequency moiré patterns that are visible to the microscope.  Mathematical techniques are then used to form a real image at enhanced resolution from multiple individual moiré images acquired using different illumination pattern orientations.  The resolution enhancement of SI microscopy depends on the spatial frequencies contained in the illuminating pattern but the maximum frequency is limited by diffraction and typically results in a resolution enhancement of no more than a factor of ² (linear structured illumination).² If the fluorescent probes in the sample produce a nonlinear response, higher-frequency harmonics are introduced into the effective illumination pattern, allowing even greater resolution increases (nonlinear structured illumination). ^{3, 4}

Systems/Experiments

One technique for obtaining nonlinear response from fluorescent probes is to increase the illumination intensity to the point where the fluorescence emission rate starts to become saturated. Using this approach we have obtained images of fluorescent beads with a point resolution of <50 nm, 5.5 times better than the diffraction limit.³ (See figure)

Another class of nonlinear phenomena involves photo-switchable molecules, molecules whose fluorescence can be turned on and off reversibly by exposure to different wavelengths of light.  Intriguingly, some of the most promising are proteins related to the green fluorescent protein (GFP).^5-7  GFP has revolutionized cell biology by its use as a genetically encodable fluorescent marker. Switchable proteins hold the promise of bringing the power of genetically encodable markers into the realm of nano-scale spatial resolution and could have a substantial impact on cell biology. ^8-10

We have measured the switching properties of two mutants of the switchable protein asulCP and found that they retain their switching ability through several hundred cycles in a living cell. This approach carries the promise of resolution levels well beyond the 50 nm that was achieved with the saturation method, even before any genetic optimization. Ultimately the resolution is limited by photo-bleaching and the degree to which the fluorescent molecules can be switched off.  Simulations suggest that 25 nm resolution (10X diffraction limit) is possible under optimum conditions.¹¹

During the past year we developed computer simulation techniques for SI microscopy and performed numerical simulation and analytic studies of imaging with 2-D (2 superposed orthogonal 1-D patterns) SI patterns. We demonstrated that these 2-D patterns can be used to obtain high resolution images while avoiding the time-consuming step of rotating the diffraction grating, required with 1-D illumination patterns, while only moderately increasing the effect of the noise in the reconstructed images.

Accomplishments

Future Directions

To demonstrate 2-D SI resolution extension based on asulCP7 and another switchable protein dronpa5, a new tabletop microscope system has been designed and is being assembled.
Through a collaboration between two major Center projects (Advanced Microscopy and Phytochrome Engineering), we plan to improve the brightness, photostability, switching, and spectroscopic properties of the molecules by optimization, as has been done with other fluorescent proteins.10 We are also examining the feasibility of engineering switchable phytochrome proteins (Section 5.2.1) developed by directed evolution for SI microscopy.
A prototype microscope system, called OMX, will be used to translate these new imaging techniques to the biology community. The first such system has been fabricated at UCSF with CBST assistance and is now being tested and debugged. This system is capable of 4-color real-time imaging, z-sectioning at 30 sections/s, and both linear and saturated structured illumination. We expect to fabricate an OMX microscope at the CBST laboratories in Sacramento, have already assembled an enthusiastic user community at UCD and LLNL, and expect many other Center participants become users. CBST affiliated researchers are particularly interested in applying OMX to problems related to chromatin structure, neuron imaging, structural differences in stem cells, and analysis of targeted drug delivery using nanotechnology delivery devices. Approximately 10 systems are being sought by bioscience centers throughout the World. CBST will help facilitate the manufacture of these systems to provide a means for introducing SI microscopy into the research community, thus moving forward with the Center’s long-term goal of revolutionizing cellular light microscopy by bringing these techniques into widespread usage.

People

Mats Gustafsson
Goran Johansson
Lin Shao
Lukman Winoto
Peter Kner
David Agard
John Sedat
Eugene Ingerman
Richard London

References

1.    Lippincott-Schwartz, J. & Patterson, G.H. Development and use of fluorescent protein markers in living cells. Science 300, 87-91 (2003).
2.    Gustafsson, M.G. Surpassing the lateral resolution limit by a factor of two using structured illumination microscopy. J Microsc 198 ( Pt 2), 82-87 (2000).
3.    Gustafsson, M.G. Nonlinear structured-illumination microscopy: wide-field fluorescence imaging with theoretically unlimited resolution. Proc Natl Acad Sci U S A 102, 13081-13086 (2005).
4.    Heintzmann, R., Jovin, T.M. & Cremer, C. Saturated patterned excitation microscopy--a concept for optical resolution improvement. J Opt Soc Am A Opt Image Sci Vis 19, 1599-1609 (2002).
5.    Ando, R., Mizuno, H. & Miyawaki, A. Regulated fast nucleocytoplasmic shuttling observed by reversible protein highlighting. Science 306, 1370-1373 (2004).
6.    Irie, M., Fukaminato, T., Sasaki, T., Tamai, N. & Kawai, T. Organic chemistry: a digital fluorescent molecular photoswitch. Nature 420, 759-760 (2002).
7.    Lukyanov, K.A. et al. Natural animal coloration can be determined by a nonfluorescent green fluorescent protein homolog. J Biol Chem 275, 25879-25882 (2000).
8.    Bulina, M.E., Verkhusha, V.V., Staroverov, D.B., Chudakov, D.M. & Lukyanov, K.A. Hetero-oligomeric tagging diminishes non-specific aggregation of target proteins fused with Anthozoa fluorescent proteins. Biochem J 371, 109-114 (2003).
9.    Campbell, R.E. et al. A monomeric red fluorescent protein. Proc Natl Acad Sci U S A 99, 7877-7882 (2002).
10.    Wang, L., Jackson, W.C., Steinbach, P.A. & Tsien, R.Y. Evolution of new nonantibody proteins via iterative somatic hypermutation. Proc Natl Acad Sci U S A 101, 16745-16749 (2004).
11.    Personal communication from Eugene Ingerman, Ph.D. - Publication in preparation.