Fluorescence polarization of gfp crystals.
Green fluorescence protein (GFP), isolated from the jellyfish, Aequorea, purified by column chromatography, and dialyzed against distilled water to remove salts, forms elongated needle-shaped crystals, about 3-[micro]m to less than 1-[micro]m wide and some 20to 100-[micro]m long (1). Examined with a polarizing microscope in visible light of wavelength greater than 450 nm, these crystals show a very weak negative birefringence; i.e., their slow axis (larger refractive index) lies perpendicular to the long crystal axis.
They also show a weak, but distinct, blue-green dichroism (Fig. 1A).
Illuminated with blue light of less than 450-nm wavelength, the same crystals show a very bright, green fluorescence when viewed through a 527 [+ or -] 15-nm band-pass filter. Surprisingly, the brightness of the fluorescence varied by a ratio of as much as 6:1 when the crystals were illuminated with polarized blue light and observed in the absence of an analyzer (Fig. 1B. 1C). The fluorescence was greatest when the long axis of the crystal lay parallel to the transmission direction of the polarizer E-vector. In other words, the absorption for the exciting light is six times greater with its E-vector polarized parallel to the length of the crystal axis than across. A similarly high ratio and orientation dependence was observed when the crystals were observed with non-polarized illumination but through an analyzer. In other words, the fluorescence emitted by crystals illuminated with non-polarized light is again some six times greater for polarization parallel to the long crystal axis. Between parallel polarize r and analyzer, the orientation-dependent fluorescence ratio becomes as high as 20:1 to 30:1, apparently the product of the excitation and emission anisotropies. These extremely high polarization ratios show that the resonance vectors of the dichroic fluorophores are oriented parallel to the long crystal axis, and that there is little loss of energy or alignment during fluorescence excitation and emission. [Even the nucleotide bases in oriented strands of B-form DNA show dichroic ratios of only 4:1 over the wavelength range 240 to 380 nm (2, 3).]
According to detailed X-ray analyses (4, 5), the 11 beta sheets that make up the barrel-shaped exterior of the GFP molecule are arranged helically around the barrel axis, with the barrel length somewhat greater than the diameter. The beta sheets lie at an angle slightly less than 45 degrees to the barrel axis. Thus, in the negatively birefringent GFP crystals, it is likely that the long axes of the barrel-shaped GFP molecules lie more or less across the length of the long crystal axis. In addition, X-ray data show that the chromophore responsible for the fluorescence lies within the beta barrel and is tilted approximately 60 degrees to the long axis of the barrel. The fluorescence polarization that we observe strongly indicates that the fluorophore is arranged with its major absorbing and emission resonance planes (dipoles) oriented parallel to the long axis of the crystal. Combining the data on fluorescence polarization and X-ray analysis, we propose that the beta barrels are regularly packed with the barrel axes tilted some 60 degrees to the length of the crystal, and possibly wound as concentric cylinders around the core of the needle-shaped crystal.
The very high degree of anisotropy for excitation and fluorescence of GFP, as well as the dramatic elevation of the orientation-dependent fluorescence polarization ratio by observation between parallel polars, suggest their potential use as indicators of the orientation of molecules with which the GFP or related chromophores are tightly bound. Our observations may also prove important in using GFP and related compounds in the application of FRET (fluorescence resonance energy transfer) and other measurements of molecular distances and orientations, because the interpretation of these measurements relies on the knowledge, or assumptions, of the orientation-dependent polarizability of the fluorophores. The observations may also be relevant in explaining the efficient energy transfer between aequorin and GFP in the light-emitting organ of the jellyfish itself.
We thank Dr. Osamu Shimomura (Marine Biological Laboratory, Woods Hole) for discussions and providing the pure native GFP crystals, Dr. Kensal Van Holde (Oregon State University) for discussions on an early version of this manuscript, and Dr. Yoshinori Fujiyoshi (Kyoto University) for generous support of this project.
(1.) AIST, Tokyo, Japan.
(1.) Morise, H., O. Shimomura, F. H. Johnson, and J. Winant. 1974. Biochemistry 13: 2656-2662,
(2.) Seeds W. E., and M. H. F. Wilkins. 1950. Faraday Discuss. Chem. Soc. 9: 417-423.
(3.) Inoue, S., and H. Sato. 1964. Pp. 209 -248 in Molecular Architecture in Cell Physiology, T. Hayashi, and A. G. Szent-Gyorgyi eds. Prentice-Hall, Englewood Cliffs, NJ.
(4.) Ormo, M., A. B. Cubitt, K. Kallio, L. A. Gross, R. Y. Tsien, and S. J. Remington. 1996. Science 273: 1392-1395.
(5.) Yang, F., L. G. Moss, and G. N. Phillips, Jr. 1996. Nature Biotechnol. 14: 1246-1251.
[Figure 1 omitted]
|Printer friendly Cite/link Email Feedback|
|Author:||Inoue, Shinya; Goda, Makoto|
|Publication:||The Biological Bulletin|
|Article Type:||Brief Article|
|Date:||Oct 1, 2001|
|Previous Article:||Introduction to the featured report: Green Fluorescent Protein: Enhanced Optical Signals from Native Crystals.|
|Next Article:||Centrifuge polarizing microscope with dual specimen chambers and injection ports.|