STED Project at Bangalore Microscopy Course 2017

  • [Credits:  
    Anwesha Guru, Tata Insitute of Fundamental Research (TIFR), Mumbai, India
    Kritarth Singh, M.S University Baroda, India
    Rituraj Marwaha, Indian Insitute of Science Education and Research (IISER), Mohali, India
    Venkata Ramaiah, University of Tokyo, Japan
    Amit Cherian, National Centre for Biological Sciences (NCBS), Bangalore, India
    Nishan B. Shettigar, Institute for Stem Cell Biology and Regenerative Medicine (inSTEM), Bangalore, India 
    Thomas van Zanten, National Centre for Biological Sciences (NCBS), Bangalore, India
     Manoj V Mathew, Centre for Cellular and Molecular Platforms (CCAMP), Bangalore, India
    Experiments were conducted at the Central Imaging and Flow Cytometry Facility (CIFF), National Center for Biological Sciences (NCBS), Bangalore, India as part of Bangalore Microscopy Course 2017]

The IXth Bangalore Microscopy Course (BMC) was organized at the National Centre for Biological Sciences (NCBS), Bangalore, India during 17-24 September 2017. This course was jointly organized by NCBS and Centre for Cellular and Molecular Platforms (CCAMP). BMC is organized annually and provides didactic and hands-on training in state-of-the-art optical microscopy techniques. A range of topics from basic microscopy to super-resolution imaging is covered. Research seminars detailing applications of light microscopy to address specific biological questions are also part of the program. Lectures are combined with hands-on training in dedicated teaching laboratories equipped with state-of-the-art microscopes.

The final two days of the course give the participants an opportunity to do independent work or a mini-project. One of these projects during BMC2017 was titled "Characterization of STED imaging and STED-FCS". The project was carried out by Anwesha, Kritarth, Rituraj, and Venkata, supported by Nishan and Amit and guided by Manoj with a inputs and help from Thomas. In this write-up, we give a short description of this project.

In STED microscopy a doughnut-shaped depletion beam is overlapped with an excitation beam. This effectively reduces the excitation volume and improves the resolution in a point scanning confocal microscope. The improvement in resolution depends on a number of factors including the intensity of depletion beam, alignment and overlap of depletion and excitation beams and the degree of aberrations or distortions in the excitation and depletion beams.

This project aimed to characterize some of the parameters that affect STED performance and help a STED user achieve optical STED imaging conditions. The participants also learned how to perform STED Fluorescence Correlation Spectroscopy (FCS) measurements. The accuracy of an FCS measurement is inversely proportional to the volume over which the FCS measurements are made. STED with its reduced excitation volumes improves FCS accuracy.

The project was carried out on the Abberior Expert Line 775nm STED which is one of the Super-Resolution imaging stations available at CIFF. An image of the system is shown in Figure 1628/1.

Figure 1628/1: Abberior Expert Line 775nm STED system at CIFF, NCBS

The system uses a pulsed depletion laser at 775nm laser and two pulsed excitation lasers at 561nm and 64onm. The system has a Spatial Light Modulator (SLM) to generate the vortex and phase ring for resolution enhancements in XY and Z respectively. The laser beams (excitation and depletion) are reflected and made colinear by dichroic mirrors and directed onto a quad scanning system that scans the focused laser spot on the sample plane. The emission light is de-scanned by passing them back through the scanning mirrors. They then pass through the dichroic mirrors onto a confocal pinhole. A secondary dichroic mirror downstream of the pinhole splits the emission light onto two (corresponding to the two excitation wavelengths) Single Photon-counting Avalanche Photo Detectors (SPAPD). The system is controlled, images generated, visualized and processed using the Imspector Software. 

Objective 1: Aligning the STED beam path

The 2D STED beam path on the Abberior STED system can be very easily aligned using the following protocol. The protocol involves imaging 100nm Tetrspeck beads and uses the 640nm excitation beam as the reference line. All other beams are aligned to the 640nm laser beam.

Step 1: Aligning the STED beam

  1. Image the beads using the 640nm excitation channel with pinhole at 4AU (open pinhole) and depletion beam off. Call this Image A.
  2. Image the beads using the 640nm excitation channel with pinhole at 4AU (open pinhole) and STED beam ON. Call this Image B.
  3. Overlay Images A and B. Image B, owing to the STED effect will show bead images with smaller diameters. If the centers of Image B and Image A are not aligned the depletion beam is misaligned.
  4. To align the depletion beam adjust the SLM. The X grating and/or Y grating can be changed to deflect the depletion beam in X and Y directions respectively and align it with the 640nm excitation beam. 

 Step 2: Aligning the 561nm beam

  1. Image the beads using the 561nm excitation channel with pinhole at 4AU (open pinhole) and depletion beam off. Call this Image C.
  2. Image the beads using the 561nm excitation channel with pinhole at 4AU (open pinhole) and STED beam ON. Call this Image D.
  3. Overlay Images C and D. If the centers of Images C and D are not aligned then the 561nm excitation is misaligned with respect to the depletion beam and hence with respect to the 640nm excitation.
  4. To align the 561nm excitation move the steering mirror in the excitation path of 561nm in the STED box so as to overlay centers of images C and D

 Step 3: Aligning the pinhole

  1. Image the beads using the 640nm excitation channel with the pinhole at 4AU (open pinhole) and depletion beam ON. We called this Image B.
  2. Image the beads using the 640nm excitation channel with the pinhole at 1AU (optimal pinhole) and depletion beam ON. We called this Image E.
  3. Overlay Images B and E. If the centers of Images B and E are not aligned the pinhole misaligned with respect to the depletion beam and hence with respect to the 640nm excitation.
  4. To align the pinhole move the steering mirror in the pinhole path so as to overlay centers of images C and D. As the centers get aligned the image intensity of image D will also increase owing to better alignment of the pinhole.
Note that in this project we only explored 2D STED so did perform 3D STED alignment. 
Objective 2: Measuring the effect of pinhole on resolution 
A basic tool that brings about enhancement in resolution is the confocal pinhole. So we tried measuring the confocal XY resolution (depletion OFF) with varying pinhole sizes. This was done by imaging 50nm red fluorescent beads, plotting the line profiles of individual beads and measuring their Full Width at Half Maximum (FWHM) using the ImageJ FWHM plugin. Figure 1628/2 shows the steps involved in a typical FWHM measurement.

Figure 1628/2: Measuring the FWHM using 50nm fluorescence bead image and Image J FWHM plugin

The results of this experiment are shown in Figure 1628/3.

Figure 1628/3: Variation of FWHM as a function of pinhole size

We see a small enhancement in resolution by decreasing the pinhole size in the confocal imaging mode. But this enhancement is not significant. 
Objective 3: Measuring the effect depletion power on resolution
By keeping the pinhole at 0.8AU we observed the FWHM of 50nm Red fluorescent beads as a function of increasing depletion laser power. The results are shown in  Figure 1628/4.

Figure 1628/4: Left-Experimentally observed variation in FWHM as a function of STED depletion power. Right-Theoretically predicated curve.

We observed resolution enhancement consistent with theoretical predictions.
Objective 4: Generating super-resolution images
We imaged nuclear pore complex in confocal and STED modes with 40% depletion laser power. The result is shown in Figure 1628/5.

Figure 1628/5: Image of nuclear pore complex. Left confocal image. Right STED image

Resolution enhancement in the STED image is obvious. We obtained XY resolution better than 90nm.
Objective 5: STED FCS correlating in space and time
In a typical Fluorescence Correlation Spectroscopy (FCS) experiment, fluorescence fluctuations from the confocal volume are collected from a stationary or parked laser spot using a confocal microscope. By autocorrelating the fluorescence fluctuations, parameters like sample concentration, diffusion time and diffusion coefficient can be measured. The accuracy of FCS measurements depends on the measurement volume. Smaller the measurement volume better the accuracy. In a confocal microscope, the pinhole helps limit the detection volume to femtoliters. A STED microscope decreases the effective excitation volume as well, making it even more effective for FCS measurements.
To measure diffusion coefficients the measurement volume needs to be measured very accurately. This is usually done through a calibration step by using a solution of the known diffusion coefficient. This step can be avoided if correlations could be performed both in space and time. The Abberior STED system offers the possibility of scanning the laser beam in an ellipse or a circle in one plane. For spatio-temporal correlation, a circular scan is better than raster scan as it ensures a constant dwell time across all pixels and avoids artifacts due to flyback.
We performed spatial and temporal correlations using this technique on 50nm fluorescent beads suspended in water as well as glycerine. The scan parameters were as follows:
  •  No. of pixels = 400
  • Size of pixel = 5nm
  • Circumference of the circle= No: of pixels x Size of pixel=2000nm
  • Dwell Time per pixel = 800ns
  • Time for for 1 cycle =No: Pixels x Dwell Time = 400*800 ns = 32*10^4 ns
  • Total time of measurement = 60s
  • No. of cycles= Total Time/Time for one Cycle=60s/32*10^4 ns = 187.5*10^3
  • Sampling Rate = 1 pixel scanned every 32*10^4 ns
The collected spatio-temporal data was processed using a MatLab code and analyzed using the open source program QuickFit. We measured a diffusion rate of around 6um^2/s for the 50nm beads in water and around 3um^2/s for the same beads in glycerine. We also measured the size of beads from the spatially correlated data as a function of depletion laser power. The result is shown in Figure 1628/6.

Figure 1628/6: Size of beads measured from spatially correlated data as a function of depletion laser power

The trend in the graph is indicative of STED effect.