Single Plane Illumination Microscopy Fluorescence (Corss-)Correlation Spectroscopy (SPIM-FCS/FCCS)

Introduction

more

Here is a recent poster also describing this setup

The mobility and reaction parameters of molecules inside living cells can be conveniently measured using fluorescent probes. Typically fluorescence correlation spectroscopy (FCS) based on confocal microscopy is used for such measurements.

This implies high time-resolution but only for a single spot at a time. So in order to use this technique to image mobility parameters, we have to scan the focus of the confocal microscope and measure the image spot by spot sequentially. In order to overcome this tedious scheme, cameras can be used to detect fluorescence and then an FCS evaluation for every pixel can be performed. This requires an illumination scheme, that excites fluorescence at all pixels of the field of view at the same time. Currently two schemes have shown to be useful: 

  • Total internal reflection microscopy (TIRFM)
  • Single plane illumination microscopy (SPIM)

Both feature good z-selectivity and exclude out-of-focus light. As TIRF relies on total internal reflection, it is restricted to measurements close (< 500nm) to a glass surface (useful e.g. for the observation of membrane processes). As we are interested in measurements inside the cytosol and also the nucleus of living cells, we implemented the second method, SPIM.

Single Plane Illumination Microscopy (SPIM)

Single Plane Illumination Microscopy (SPIM) is a fluorescence microscopy technique, where the illumination is done perpendicularly to the detection. The technique shapes the illumination laser beam into a rectangle and then focuses it down only in one direction, using a cylindrical lens. This forms a thin "sheet of light" right in the focal plane of the detection objective. As the lightsheet can be tailored to be thin (< 2μm FWHM), we achieve good sectioning of the sample and out-of-focus light suppression. The lateral resolution is given by the detection objective only. Our setup currently uses a 60x/NA1.0 objective which leads to a lateral resolution of around 0.65μm.

The sample is mounted from above (although other implementations exist) inside a water- or buffer-filled sample chamber (stainless steal) which can also be heated to 37°C in order to create near-physiological conditions. It is embedded in a low concentration agarose gel (0.5-0.7%), or on a cover slip. As the detection objective can not be moved in our setup, the sample is mounted on a computer controlled XYZ translation stage (100nm resolution). 

As an imaging detector, we either use an electron-multiplying CCD camera (EMCCD, Andor iXon X3 850), or a custom-made single-photon avalanche diode (SPAD) array (see below).

3D volume rendering of a z-stack of a single HeLa cell expressing EGFP tetramers

z-stack of HeLa cells expressing EGFP tetramers in transmission illumination and SPIM

3D volume rendering of the z-stack above

Details of our setup

  • illumination: 488nm diode laser, 561nm DPSS laser, white/multi-color LEDs
  • lightsheet projection objective: 10x, NA 0.3
  • lightsheet width: 2.2...2.4μm (1/e2 width)
  • detection objective: 60x, NA 1.0
  • lateral resolution: 0.65μm
  • dual-color detection: Photometrics DualView DV2
  • imaging device: Andor iXon X3 860 (EMCCD), 128×128 pixels, 500fps + SPAD array, 32×32 pixels, 100000fps

Images of the setup

SPIM-Fluorescence (Cross-)Correlation Spectroscopy (SPIM-FCS/SPIM-FCCS)

Image series taken with a SPIM at high frame rates (>500fps) can be used for an FCS evaluation, which then yields a set of autocorrelation functions. After a model fit to each of these functions, we get a map of the mobility parameters inside the sample. As fluorescence fluctuations were not only measured at a single spot, we can also calculate pixel-pixel-crosscorrelation functions, which e.g. yield information about flow patterns in the sample.

In addition we recently introduced two-color fluorescence crosscorrelation spectroscopy on our lightsheet microscope, which allows us to measure molecular interactions in different compartments of the cell (cytosol, nucleoplasm, membrane) in a spatially resolved manner.

Our SPIM-FCS data evaluation software QuickFit 3.0 is available for free.

SPAD arrays for SPIM-FCS

Our EMCCD camera (500fps @ 128×128 pixels, 14000fps @ 128×1 pixels) is reasonably fast, but not fast enough for FCS measurements on small proteins in cells (e.g. free eGFP). So we also use a 32×32 pixel array of single-photon avalnche diodes (SPAD array) as a secondary image detector. Our SPAD array was developed in the group of Prof. Edoardo Charbon at the TU Delft. It's features are:

  • 32×32 pixel array with passive quenching
  • around 35% quantum efficiency at 525nm (EGFP emission)
  • dark count rate is
  • pixel pitch: 30μm; SPAD diameter 4μm
  • 1 Bit memory per pixel
  • row-wise readout (rolling shutter), possibility to define region of interest (ROI)
  • full frame rate > 100kHz, faster for sub-regions
  • two field programmable gate arrays (FPGA) are used for readout, on-line data processing and data transfer to a standard PC via USB 2.0
  • realtime autocorrelator for all 1024 pixels with 10µs minimum lag time and 2.66µs SPAD array readout time (temporal binning before correlation!)

The image on the right shows a first test measurement with the SPAD array and 20nm beads in water as a sample.

In order to improve the usability of the SPAD array for SPIM-FCS, we use the two programmable logic devices to do an online-calculation of the autocorrelation functions. Our system achieves real-time processing of all 1024 pixels at a readout rate of 100000fps. The correlators span a dynamic range of 10μs up to 1.2s, which matches the range of interest for FCS in live cells (see our OptEx publication on this topic).

Publications

  • Jan Buchholz, Jan Wolfgang Krieger, Gábor Mocsár, Balázs Kreith, Edoardo Charbon, György Vámosi, Udo Kebschull, and Jörg Langowski (2012): 
    "FPGA implementation of a 32x32 autocorrelator array for analysis of fast image series," 
    Opt. Express 20, 17767-17782,
    DOI: 10.1364/OE.20.017767 [LINK]
  • Anand Pratap Singh, Jan Wolfgang Krieger, Jan Buchholz, Edoardo Charbon, Jörg Langowski, Thorsten Wohland (2013):
    "The performance of 2D array detectors for light sheet based fluorescence correlation spectroscopy",
    Opt. Express 21(7), 8652-68,
    DOI: 10.1364/OE.21.008652
  • Jan Wolfgang Krieger, Anand Pratap Singh,, Christoph S. Garbe, Thorsten Wohland and Jörg Langowski (2013):
    "Dual-Color Fluorescence Cross-Correlation Spectroscopy on a Single Plane Illumination Microscope (SPIM-FCCS)",
    Opt. Express22(3), 2358-2375, DOI: 10.1364/OE.22.002358 [PDF] supplementary information
  • Peter Brazda, Jan Krieger, Bence Daniel, David Jonas, Tibor Szekeres, Jörg Langowski, Katalin Tóth, Laszlo Nagy and György Vámosi (2014):
    "Ligand binding shifts highly mobile RXR to chromatin-bound state in a coactivator-dependent manner as revealed by single cell imaging",
    Molecular and Cellular BiologyDOI: 10.1128/MCB.01097-13

to top