Introduction

The method of Förster Resonance Energy Transfer (FRET) (1,2) is a powerful tool for quantifying intermolecular interactions as well as probing supramolecular assembly of, e.g., proteins into dimeric/oligomeric complexes. However, numerous nuances exist in FRET-based measurements which need to be addressed by the measuring system before the full capabilities of the method can be realized. Firstly, in order to maximize the amount of FRET signal occurring, the emission spectrum of the donor molecule must overlap appreciably with the excitation spectrum of the acceptor for efficient energy transfer. While good coupling between donor and acceptor is desirable, this situation can also lead to overlapping emission profiles, making it difficult to separate the contributions of each molecule to the detected emission. Therefore, the spectral resolution of the measurement system needs to be high enough to separate FRET pairs of closely spaced emission spectra; this typically means a spectral resolution of five nanometers or better. Secondly, the measuring system needs to acquire all spectrally resolved emission in a single scan of the sample. When multiple excitation scans are used to detect FRET, the resulting pixel-level information is compromised due to mixing caused by diffusion of molecular complexes moving during the image acquisition period. Nevertheless, detecting the presence of permutations of oligomer constructs inside the cell is vital to understanding the complexity of an oligomeric construct of interest.

An exquisitely suitable tool to performing measurements of this nature is the Optical Micro-Spectroscopy System (OptiMiS™, Aurora Spectral Technologies), which is used in conjunction with standard multiphoton microscope (MPM) setups. OptiMiS either may be added to an existing MPM setup or a whole new system may be set up around it, incorporating the customer's choice of laser, microscope platform (from any major microscope manufacturer) and EMCCD (or CMOS) camera. In this article, we will provide several examples of OptiMiS uses in FRET-based imaging of living cells.

Spectrally resolved fluorescence emission collection

Single cells measured using OptiMiS were typically scanned with a pulsed (femtosecond) infrared laser. The fluorescence emitted from excited portions of the sample is separated into its spectral components before striking the detection hardware. The software controlling the laser scanning mirrors of OptiMiS also collects the data from the detector and reconstructs images of the samples’ fluorescence intensity at various wavelengths. The wavelength channels are separated by as little as 1.25 nm and range in wavelength from 400 nm to 650 nm. Displayed in Figure 1 is a small selection of fluorescence spatial maps resulting from a single scan of yeast cells expressing GFP₂ (3) in their cytoplasm.

Figure 1. A selection of 6 fluorescence spatial maps at various emission wavelengths resulting from a scan of yeast cells expressing the fluorophore GFP₂ in their cytoplasm. (Click to enlarge)

The fluorescence spectrum of each pixel of the spectrally resolved images was unmixed (4) to reveal the pixel-level contributions to the detected emission from each fluorophore species inherent to the scanned sample. For fluorophore combinations participating in FRET, the contributions to the detected emission arise from the fluorescence of the donor in the presence of the acceptor (FDA) and the fluorescence of the acceptor in the presence of the donor (FAD). By a careful selection of the FRET pairs (5), it is possible to directly excite only the donor fluorophore with the laser; in this way, any emission emanating from the acceptor molecule is occurring solely due to FRET. Thus, the fluorescence intensity maps of the donor and acceptor emission, obtained from a single scan of the sample, are sufficient for calculating pixel-level FRET efficiencies, with no photobleaching steps required (5). The pixel-level FRET efficiencies were further analyzed by binning them according to their values to build a histogram.

Efficiency determination for some FRET standards

Chinese hamster ovary (CHO) cells were transiently transfected with two FRET oligomer standards comprised of different combinations of a donor (Cerulean, ‘C’) and acceptors (Venus, ‘V’); the plasmids for the FRET standards were a jpgt from Dr. Steven S. Vogel (NIAAA, NIH) (6). One of the oligomers was a tetramer consisting of a single donor and three acceptors, denoted by the symbol VCVV, while the other construct was a trimer, denoted by the symbol VCV.