Leica_Dye_Finder :53 Uhr Seite 6 Dye Finder LAS AF

Transcription

1 Dye Finder LAS AF

2 Dye Finder Multicolor live cell fluorescence microscopy is limited by the availability of spectrally separable fluorescent dyes. Fluorescent dyes (or spectral GFP variants) with incongruent excitation and emission maxima are commonly employed in combination with dichroic filters or beam splitters in order to separate their signals into different channels. In general, fluorescence spectra in liquid solutions tend to expand so that significant spectral overlap does occur, no matter how hard one tries to optimize a given set of fluorescent dyes and filters. The resulting multicolor images from such experiments show a mixture of colors with many color nuances in between the original selected channels. A lack of separation between the dyes often makes it difficult for users to identify which dye the different areas of a sample are derived from. This problem can be solved by the Leica Dye Finder. A Ch1 Ch2 Ovl B Ch1 Raw Baseline Median Separated Raw Baseline Median Separated Fig. 1: NIH 3T3 mouse fibroblast cells stably expressing H2A-YFP ( Ch2, red) were transiently transfected with Actin- GFP ( Ch1, green). GFP and YFP have strongly overlapping emission spectra with considerable spectral crosstalk ( Raw images). The crosstalk is most apparent in the overlay images as a yellow color ( Ovl ). Both Leica s Automatic Dye Separation tool (panel A) and Channel Dye Separation (panel B) were applied. Before dye separation the inbuilt baseline ( Baseline ) correction and a median filter ( Median ) with kernel size 3x3 were applied to reduce background and noise respectively. Both Automatic as well as Channel Dye Separation lead to efficient unmixing ( Separation ). Magnification 63x objective. Courtesy of Constantin Kappel, DKFZ, Heidelberg, Germany Ch2 Ovl

3 This software tool offers several different algorithms to separate the dyes after image acquisition and assigns the different color nuances back to the dye it originates from. As a result, the spatial distribution of Before Dye Separation After the dyes can easily be identified by the user. The Dye Finder tool can also help to separate signals from unbalanced fluorochromes. It allows an increase in acquisition speed, whilst still seeing well separated channels. Users of confocal systems can separate fluorochromes that are excited with one and the same laser line. The Dye Finder can be used on both the Leica AF6000 system series for widefield imaging and analysis and the Leica TCS SP5 confocal systems. Channel Automatic Dye Separation how it works The Leica Dye Finder offers three methods to separate fluorochromes with overlapping emission spectra from each other, i) Automatic Dye Separation, ii) Channel Dye Separation and iii) the Spectral Dye Separation. Automatic Dye Separation The Automatic Dye Separation tool employs an advanced clustering algorithm, resulting in robust unmixing of fluorescence channels (Fig. 1 A, Fig. 2). The user only needs to specify two parameters relating to separation stringency and image dynamics. The rest is done by the powerful Automatic Dye Separation function. The dyes are separated without any further user interaction. Fig. 2: Enlargement of areas from Fig. 1 in highlighted boxes. Channel Dye Separation the chosen regions of interest are shown for channel 1 in green and for channel 2 in red. Note the absence of actin fluorescence in the H2A channel, the low nuclear fluorescence in the actin channel (there is naturally cytosolic fluorescence above and below the nucleus) as well as a clearly red fluorescence signal in the nucleus of a non-transfected cell in the upper right corner. Magnification 63x objective. Courtesy of Constantin Kappel, DKFZ, Heidelberg, Germany Channel Dye Separation The Channel Dye Separation method uses reference regions to identify the distribution coefficients of the fluorochromes in the different channels. Ideally these regions are defined on separately acquired single dye images, but can also be areas within the specimen which clearly contain single dye regions. The user then defines all reference regions within the same sample (Fig. 2, red and green ROIs, Fig. 4). A color distribution box helps the user to easily identify the regions. Once these reference regions are defined for each channel, the Channel Dye Separation tool calculates an unmixing matrix and applies this matrix to the desired image set Mean Intensity x 10 3 all channels Fig 3: The color distribution box helps the user to identify single dye regions. Reference regions are characterized by a high degree of differentiation between the channels.

4 channel 1 channel 2 channel 3 channel 4 channel 5 channel 6 DAPI FITC/Alexa488 Cy3/TAMRA Texas Red Alexa633 Cy5 HeNe-633 HeNe-594 HeNe-543 Argon 488 Diodo-405 DAPI FITC/Alexa488 Cy3/TAMRA Texas Red Alexa633 Cy5 Spectral Dye Separation (available on TCS SP5 only) Provided the emission spectrum of each dye of a sample is known, the Spectral Dye Separation method can identify the spectral overlap to eliminate crosstalk between the fluorochromes (Fig. 5). The user only needs to assign the appropriate reference spectra to the dyes and apply the unmixing algorithm to the desired image set. Reference emission spectra can be measured with either a Lambdascan or extracted from literature and stored in the Spectra Database, a standard feature which is of the LAS AF core software application. Fig. 4: 5-Color FISH of a metaphase spread obtained from PHA stimulated human lymphocytes (XY); the image shows the signals in channels 1-6 (columns 1-6) upon excitation with corresponding laser lines before (lines 1-5) and after Channel Dye Separation (line 6). Color assignment after using the Channel Dye Separation Tool: grey, DAPI (entire DNA); green, Alexa488; yellow, TAMRA; red, Texas Red; magenta, Alexa633; blue, Cy5. Image courtesy of Dr. I. Solovei and Dr. M. Cremer, Bio Center, Department of Biology II, University of Munich (LMU).

5 Dye separation for fast cell dynamics and 4D live cell experiments For confocal systems, sequential scanning is the method of choice to avoid cross-talk of fluorochromes when imaging multi-labeled specimens. However, this method slows down acquisition speed and is therefore detrimental to life cell experiments. If acquisition speed is important, e.g. when dealing with fast cell dynamics or performing 4D live cell experiments, simultaneous acquisition of the signals is favourable. Any resulting cross-talk due to overlapping emission spectra of the fluorochromes, can easily be removed by adding a subsequent dye separation step with the Dye Finder. A A similar effect is seen when using multi bandpass filter cubes for widefield systems such as the AF6000 system series. The use of multi bandpass cubes avoids time consuming filter cube changes during the sequential image acquisition, by changing only the excitation wavelength and therefore increases acquisition speed. Consequently, the individual channel images in an overlay image cannot be clearly seen. The dye separation step can also serve as useful tool in this scenario. B Separation of fluorochromes excited with one laser line What if your sample is labeled with dye combinations like Alexa 488 and GFP or Alexa 633 and Cy5? These fluorochrome combinations all have one thing in common: highly overlapping emission spectra and an excitation with identical laser lines on confocal systems, i.e. sequential scanning is not possible! Previously impossible to separate, it is now an easy task to segregate such dye combinations using the Dye Finder. This tool offers much more flexibility in dye selection, offering better differentiation between dyes and convenient use of multi labeled samples. Separation of non-balanced fluorochromes Ideally, the labels of a specimen have the same emission intensity level. Unfortunately this is not always the case, especially when dealing with live cells labeled with several fluorescent proteins. If the emission intensities of the different fluorochromes are not well balanced and their emission spectra are highly overlapping, the weak signal can be dominated or even covered by the strong signal. If this misbalance cannot be compensated for during image acquisition via hardware parameters, the visualization and/or quantification of such weakly expressed signals is difficult. With the Dye Finder this problem is easily solved. Fig. 5: Drosophila melanogaster (Oocyte), Grey: Alexa 488, nuclei of terminal follicle cells; Green: Alexa 546, filamentous actin; Red: Alexa 568 adducin-like hts; Blue: TOTO-3, DNA A) overlay, simultaneous acquisition B) Color restoration after employing the Spectral Dye Separation Tool on a Lambda-Stack Courtesy of Dr. R. Pflanz, Max-Planck Institute for Biophysical Chemistry, Göttingen, Germany Author: Constantin Kappel, Div. Theoretical Bioinformatics, German Cancer Research Center (DKFZ), Heidelberg, Germany B in cooperation with Leica Microsystems

6 Copyright Leica Microsystems CMS GmbH Am Friedensplatz 3 D Mannheim, Germany Tel / Fax / LEICA and the Leica Logo are registered trademarks of Leica IR GmbH. Order no: September