AccPbFRET: An ImageJ plugin for semi-automatic, fully corrected analysis of acceptor photobleaching FRET images
- 10k Downloads
The acceptor photobleaching fluorescence resonance energy transfer (FRET) method is widely used for monitoring molecular interactions in cells. This method of FRET, while among those with the simplest mathematics, is robust, self-controlled and independent of fluorophore amounts and ratios.
AccPbFRET is a user-friendly, efficient ImageJ plugin which allows fully corrected, pixel-wise calculation and detailed, ROI (region of interest)-based analysis of FRET efficiencies in microscopic images. Furthermore, automatic registration and semi-automatic analysis of large image sets is provided, which are not available in any existing FRET evaluation software.
Despite of the widespread applicability of the acceptor photobleaching FRET technique, this is the first paper where all possible sources of major errors of the measurement and analysis are considered, and AccPbFRET is the only program which provides the complete suite of corrections – for registering image pairs, for unwanted photobleaching of the donor, for cross-talk of the acceptor and/or its photoproduct to the donor channel and for partial photobleaching of the acceptor. The program efficiently speeds up the analysis of large image sets even for novice users and is freely available.
KeywordsFluorescence Resonance Energy Transfer Fluorescence Resonance Energy Transfer Efficiency Donor Channel Fluorescence Resonance Energy Transfer Ratio Acceptor Channel
Fluorescence resonance energy transfer (FRET) is a powerful technique that can be applied to study nanoscale intra- and intermolecular events and interactions of molecules in situ in biological systems . In assessing FRET, fluorescence of a spectrally matched donor and acceptor dye pair can be measured to reveal the radiationless transfer of excitation energy from the donor to the acceptor, in the case that their dipoles are properly oriented and the two are in spatial proximity (usually at a distance of 1–10 nm) . This latter phenomenon is the basis of the popularity of FRET in biology: The distance over which FRET occurs is small enough to characterize the proximity of possibly interacting molecules, under special circumstances it even provides quantitative data on exact distances, and, additionally, information on the spatial orientation of molecules or their domains. Hence the very apropos term from Stryer, who equaled FRET to a "spectroscopic ruler" . FRET can be measured both in microscopic imaging and in flow cytometry. While flow cytometric FRET (FCET) carries the advantage of examining large cell populations in a short time, microscopic approaches have the ability to provide subcellular detail and the possibility to correlate FRET values with other biological information gained from fluorescent labeling, on a pixel by pixel basis . In a review about FRET imaging, Jares-Erijman and Jovin classified 22 different approaches to quantifying FRET in a systematic way. The techniques fall in two major groups: most of them are based on donor quenching and/or acceptor sensitization, and a few on measuring emission anisotropy of either the donor or the acceptor . In the practice of cell biology, ordinary confocal microscopy is now broadly available, and brings three quantitative FRET approaches within close reach. These are the various ratiometric approaches, donor photobleaching FRET and acceptor photobleaching FRET [4, 6]. Some other approaches based on anisotropy [7, 8], fluorescence lifetime [9, 10], imaging spectroscopy , or lifetime imaging spectroscopy  require more specialized equipment, while yet others lack the quantitative measurement of FRET efficiency and rely on various FRET parameters that are usually made unreliable by the varying amounts and ratios of donor and acceptor in each examined pixel .
A measurement that exploits this proportionality is facilely implemented in confocal microscopy, thus providing the option of distinguishing various molecular association states even at the subcellular level. The method is also applicable to the ever-spreading family of green fluorescent protein (GFP) derivatives .
Image manipulation and analysis in biological research are often performed with the free ImageJ package . In spite of the numerous plugins available, there are only three tools to help the evaluation of FRET. Two of them aid the assessment of ratiometric FRET images [22, 23], while that provided by D. Stepensky has been the only freely available tool for acceptor photobleaching FRET . This plugin allows the calculation of FRET efficiency based on average fluorescence signals (i.e. not on a pixel-by-pixel basis) from pixels above a pre-defined threshold in acceptor photobleaching images. It does not provide correction possibilities, registration, and it allows selection of rectangular shaped ROIs only. As the need for evaluating larger data sets for molecular interactions increases, we have undertaken to develop a program that addresses all the above deficiencies, and is also capable of quick semi-automatic processing of serial measurements.
The plugin was written in Java v1.6, and tested with ImageJ version 1.38×.
where FD 1(i, j)and FD 2(i, j)are the donor fluorescence values of the pixel (i, j) before (1) and after (2) photobleaching the acceptor, and FA 1(i, j)the acceptor fluorescence for the same pixel before photobleaching. All F values are background corrected throughout. α, γ, δ and ε are correction factors that are described below.
In some cases, photobleaching of the acceptor is not complete. As shown by van Munster et al. , the average FRET efficiency is directly proportional to the amount of available acceptor molecules assuming that photobleaching occurs indiscriminately to all acceptor molecules, and there is not more than one acceptor per donor molecule present. To correct for incomplete acceptor bleaching, the correction factor
α is calculated asα = ⟨FA 2(i, j)/FA 1(i, j)⟩
where FA 2(i, j)and FA 1(i, j)are intensities in the acceptor channel in pixels above threshold of the donor and acceptor labeled sample, before (1) and after (2) photobleaching.
γ = ⟨FDd 1(i, j)⟩/⟨FDd 2(i, j)⟩
γ = ⟨FDd 1(i, j)/FDd 2(i, j)⟩
where FDd 1(i, j)and FDd 2(i, j)are donor fluorescence intensities of donor only (Dd) samples in pixels above threshold before (1) and after (2) photobleaching the acceptor, and the ⟨⟩ signs denote mean value. Since FRET protects the donor from photobleaching , this factor calculated based on a sample labeled with donor only is not exactly accurate. However, the difference in practice is 10–20% of 1–2%, which may not cause great errors in determining FRET. Nevertheless, caution needs to be taken to minimize photobleaching of the donor during the measurement.
The program offers (also as for the correction factors δ and ε), the possibility to calculate the factor on a pixel-by-pixel basis and then average pixels above threshold for raw data, or, alternatively, to average the raw data for these pixels and then calculate an average correction factor.
In the case that the acceptor dye also fluoresces in the donor channel, FRET would be underestimated without correcting for this cross-talk . The appropriate correction factor is calculated as
δδ = ⟨FDa 1(i, j)⟩/⟨FAa 1(i, j)⟩
δ = ⟨FDa 1(i, j)/FAa 1(i, j)⟩
where FDa 1(i, j)and FAa 1(i, j)are signals in the donor and acceptor channels in pixels above threshold of an acceptor only labeled sample, before (1) photobleaching the acceptor.
In some cases, photobleaching the acceptor can yield a photoproduct with distinct absorption and emission properties, which can contribute to the post-bleach donor signal, resulting in the overestimation of FRET efficiency . The correction factor ε for such acceptor-photoproduct is calculated as
ε = ⟨FDa 2(i, j)⟩/⟨FAa 1(i, j)⟩
ε = ⟨FDa 2(i, j)/FAa 1(i, j)⟩
where FDa 2(i, j)and FAa 1(i, j)are intensities in the donor and acceptor channels in pixels above threshold of an acceptor only labeled sample, before (1) and after (2) photobleaching.
The calculation of the constants γ, δ and ε requires taking images with the same photobleaching protocol on samples labeled with donor only and acceptor only, which usually need to be taken anyway. To correct for shifts in the x-y plane, the images are registered using the Fast Hartley Transform algorithm  implemented in the ImageJ package. All corrections are optional and can be activated/inactivated in the "Corrections" menu of the plugin.
With our program AccPbFRET, which can be found in the additional file [see Additional file 1] or can be downloaded from its homepage , FRET efficiencies are calculated pixel-by-pixel, and their distribution is determined for any user defined rectangular, polygonal, or freehand type ROI or subcellular location. Accurate selection of the examined cellular components is furthered by the provision to interactively set threshold values of donor and, optionally, acceptor fluorescence intensities, and to also gate using images with relevant independent fluorescent labels in the same sample. In addition, our plugin provides automatic registration of the images, an absolute necessity for perfect alignment of donor images taken before and after photobleaching. An example analysis with and without registration (along with other examples) can be found in the additional file [see Additional file 1]. We compared the results of the same images obtained with FRETcalc  and AccPbFRET, and we obtained similar FRET efficiencies, 15.1% and 14.7%, respectively. However, when we used images that needed registration because of a few pixels shift, the results changed to 5.7% versus 14.6%.
Other important issues with acceptor photobleaching FRET is correcting for bleaching of the donor, for the cross-talk of the acceptor and/or its provisional photoproduct to the donor channel, and partial photobleaching of the acceptor, which are also solved by AccPbFRET.
When evaluating large amounts of molecular interaction data on several hundreds of image sets, the optional semi-automatic mode allows nearly three times faster processing relative to the single-image mode. In semi-automatic processing mode (tested with Zeiss LSM 510 Version 4.0), the program opens images sequentially in the chosen directory, and only threshold setting(s) and creation of the FRET image need to be done manually. The upper left corner (1/6 × 1/6 of the image) is considered automatically as background.
The AccPbFRET plugin provides an easy to use graphical interface, which leads the user through the evaluation process, and does not require cumbersome pre-setting of various parameters. It allows correcting for bleaching of the donor, for the cross-talk of the acceptor and its photoproduct to the donor channel and for partial acceptor photobleaching. Furthermore, automatic registration and semi-automatic analysis of large image sets is provided, which are not available in any existing evaluation software.
Availability and requirements
Project name: AccPbFRET
Project home page: http://www.biophys.dote.hu/accpbfret
Operating system(s): platform independent
Programming language: Java
Other requirements: ImageJ 1.38× (bundled with Java 1.6.0_02) or higher, screen resolution 1280 × 900 or higher
License: free software
Any restrictions to use by non-academics: none
This work was partially supported by OTKA K62648, K68763, K75752, MRTN-CT-2005-019481, EU FP6 LSHB-CT-2004-503467 and LSHC-CT-2005-018914.
- 4.Bastiaens PIH, Jovin TM: Fluorescence resonance energy transfer microscopy. In Cell Biology: A Laboratory Handbook. Edited by: Celis JE. New York: Academic Press; 1998:136–146.Google Scholar
- 6.Szöllösi J, Damjanovich S, Nagy P, Vereb G, Mátyus L: Principles of Resonance Energy Transfer. Current Protocols in Cytometry 2006, 1.12.1–1.12.16.Google Scholar
- 14.Jovin TM, Arndt-Jovin DJ: FRET microscopy: digital imaging of fluorescence resonance energy transfer. Application in cell biology. In "Cell Structure and Function by Microspectrofluorimetry.". Edited by: Kohen E, Hirschberg, JG. Academic Press, San Diego, CA; 1989:99–117.Google Scholar
- 15.Damjanovich S, Vereb G, Schaper A, Jenei A, Matkó J, Starink JP, Fox GQ, Arndt-Jovin DJ, Jovin TM: Structural hierarchy in the clustering of HLA class I molecules in the plasma membrane of human lymphoblastoid cells. Proc Natl Acad Sci USA 1995, 92: 1122–1126.PubMedCentralCrossRefPubMedGoogle Scholar
- 18.Vereb G, Meyer CK, Jovin TM: Novel microscope-based approaches for the investigation of protein – protein interactions in signal transduction. In "Interacting protein domains, their role in signal and energy transduction. NATO ASI series". Volume H102. Edited by: Heilmeyer LMG Jr. Springer-Verlag, New York; 1997:49–52.Google Scholar
- 23.Hachet-Haas M, Converset N, Marchal O, Matthes H, Gioria S, Galzi JL, Lecat S: FRET and colocalization analyzer – a method to validate measurements of sensitized emission FRET acquired by confocal microscopy and available as an ImageJ Plug-in. Microsc Res Tech 2006, 69: 941–956.CrossRefPubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.