Cortical Actin Flow in T Cells Quantified by Spatio-temporal Image Correlation Spectroscopy of Structured Illumination Microscopy Data

Summary

To investigate flow velocities and directionality of filamentous-actin at the T cell immunological synapse, live-cell super-resolution imaging is combined with total internal reflection fluorescence and quantified with spatio-temporal image correlation spectroscopy.

Abstract

Filamentous-actin plays a crucial role in a majority of cell processes including motility and, in immune cells, the formation of a key cell-cell interaction known as the immunological synapse. F-actin is also speculated to play a role in regulating molecular distributions at the membrane of cells including sub-membranous vesicle dynamics and protein clustering. While standard light microscope techniques allow generalized and diffraction-limited observations to be made, many cellular and molecular events including clustering and molecular flow occur in populations at length-scales far below the resolving power of standard light microscopy. By combining total internal reflection fluorescence with the super resolution imaging method structured illumination microscopy, the two-dimensional molecular flow of F-actin at the immune synapse of T cells was recorded. Spatio-temporal image correlation spectroscopy (STICS) was then applied, which generates quantifiable results in the form of velocity histograms and vector maps representing flow directionality and magnitude. This protocol describes the combination of super-resolution imaging and STICS techniques to generate flow vectors at sub-diffraction levels of detail. This technique was used to confirm an actin flow that is symmetrically retrograde and centripetal throughout the periphery of T cells upon synapse formation.

Introduction

The goal of the experiment is to image and characterize the molecular flow of filamentous- (F-) actin in living T cells, beyond the diffraction limit in order to establish flow direction and magnitude during immunological synapse (IS) formation. This technique will be carried out using a structured illumination microscope (SIM) in total internal reflection fluorescence (TIRF) mode, imaging Jurkat T-cells forming an artificial synapse on activating coverslips coated with anti-CD3- and anti-CD28 antibodies. The IS forms between a T cell and its target; an antigen presenting cell (APC) upon T cell receptor (TCR) activation^1,2. As this is the first step in determining whether the adaptive immune system generates an immune response to an infection, the consequences of incorrect responsiveness to stimuli can cause a number of diseases. The importance of the T cell-APC interaction is well documented, however, the role(s) of the mature IS after initial TCR signal internalization are yet to be fully understood.

As the cytoskeleton is thought to aid two processes linked to TCR activation (the movement of molecules at the plasma membrane and the control of signaling molecules dependent on the actin cytoskeleton^3,4 ), understanding how the cytoskeleton rearranges spatially and temporally will elucidate the steps of molecular reorganization during synapse formation. The retrograde flow of F-actin is thought to corral TCR clusters towards the center of the immunological synapse, this translocation is believed to be vital for signal cessation and recycling; aiding the fine balance of T cell activation⁵.

While standard fluorescence microscopy is a flexible tool that continues to provide insight about many biological events including cell-cell interactions, it is limited by the system's ability to accurately resolve multiple fluorophores residing closer than the diffraction limit (≈200 nm) of each other. As many molecular events within cells involve dense populations of molecules and exhibit dynamic rearrangement either in quiescence or upon specific stimulation, conventional light microscopy does not provide the full picture.

The use of super resolution imaging has allowed researchers to circumvent the diffraction limit using a variety of techniques. Single molecule localization microscopy (SMLM) such as PALM and STORM^6,7 has the ability to resolve the location of molecules up to a precision of around 20 nm, however, these techniques rely on long acquisition times, building up an image over many frames. Generally this requires samples to be fixed or for structures to exhibit low mobility so single fluorophores are not misinterpreted as multiple points. While stimulated emission depletion (STED) imaging allows super-resolution through very selective confocal excitation⁸, the required scanning approach can be slow over whole cell fields of view. STED also demonstrates significant phototoxicity for higher resolutions due to the increase in power of the depletion beam.

Structured illumination microscopy (SIM⁹) provides an alternative to these methods, capable of doubling the resolution of a standard fluorescence microscope and is more compatible with live cell imaging than current SMLM techniques. It uses lower laser powers, on a wide-field system for whole-cell fields of view; providing increased acquisition speeds, and is compatible with standard fluorophore labelling. The wide-field nature of SIM also permits the utilization of total internal reflection fluorescence (TIRF) for highly selective excitation, with penetration depths in the axial dimension of <100 nm.

Image correlation spectroscopy (ICS) is a method of correlating fluctuations in fluorescence intensity. An evolution of ICS; Spatio-temporal image correlation spectroscopy (STICS¹⁰) correlates intracellular fluorescent signals in time and space. By correlating pixel intensities with surrounding pixels in lagging frames via a correlation function, information is gained regarding flow speeds and directionality. To ensure static objects do not interfere with the STICS analysis, an immobile object filter is implemented, which works by subtracting a moving average of the pixel intensities.

The technique presented here is believed to be the first demonstration of image correlation spectroscopy being applied to super-resolution microscopy data to quantify the molecular flow in live cells¹¹. This method improves on diffraction-limited imaging of F-actin flow via STICS analysis¹² and would suit researchers who wish to determine two-dimensional molecular flow in live cells, from data acquired at spatial resolutions beyond the diffraction limit of light.

Protocol

Note: Stages 1 and 2 take place before the day of imaging; ensure all media and supplements to be used before or on the day of imaging are equilibrated. Equilibration is achieved through warming of media and supplements to 37 °C in an incubator set to 5% CO₂. All cell culture work and coverslip plating steps take place in sterile conditions under a laminar flow hood.

1. Cell Transfection

1. Transfect the T cells to be imaged 24 hr before the experiment with a plasmid encoding a fluorescent fusion construct for F-actin GFP labelling and imaging. Note: Cells should be split 24 hr prior to transfection (48 hr prior to imaging) to ensure cells are at optimal health and in their logarithmic growth phase.
   1. Place 5 mL of RPMI supplement (RPMI 1640 plus 10% FBS and 1% Pen-Strep (optional)) in a single well of a 6-well plate into the incubator to equilibrate.
   2. Pellet 2 x 10⁷ T cells cultured at 37 °C +5 % CO₂ in RPMI supplement using a centrifuge tube at 327 x g for 5 min and wash in an equal volume of pre-warmed electroporation media. Note: Electroporation medium is a reduced serum medium containing buffers and supplements (see consumables).
   3. After re-pelleting at 327 x g for 5 min, resuspend the cells in 500 µl of electroporation media and slowly add to a cuvette, add 5 µg of the plasmid in PCR grade dH₂O for labelling F-actin.
   4. Gently tap the cuvette on the hood workspace to ensure removal of any air bubbles in the media, which can cause arcing of the electricity and to distribute the cells evenly within the media.
   5. Turn on the electroporation apparatus and set voltage to 300 V and capacitance to 290 Ω. Electroporate sample using an exponential decay for 20 msec. Note: Optimization may be required for different cells and different plasmid constructs.
   6. Once the transfection protocol is complete, store cells on ice for 1 min before continuing.
   7. Carefully transfer the cells from the cuvette to the 6-well plate containing 5 mL of equilibrated RPMI supplement from step 1.1.1.
   8. Incubate the cells O/N at 37 °C + 5% CO₂.

2. Coat Coverslips and Pre-warm Imaging Supplement

1. Coat coverslips with antibodies to induce T cell stimulation.
   1. Coat each chamber of an 8-well coverslip with 1 µg / ml αCD3 and αCD28 in 200 µl PBS.
   2. Store at 4 °C O/N. Alternatively, coat coverslips 2 hr prior to imaging and keep in an incubator at 37 °C to reduce drift during the imaging process by minimizing temperature changes to the coverslip.
2. Create a coverslip test sample with 0.1 µm sub-diffraction fluorescent microspheres.
   1. Vortex stock solution of beads to avoid clumps.
   2. On a new, clean coverslip of the same type as those used in step 2.1.1, pipette 5 µl fluorescent microsphere stock solution onto the surface (as per manufacturer's instructions).
   3. Spread microsphere solution over the coverslip with pipette tip.
   4. Allow to dry, then apply 200 µl mounting medium such as PBS.
      Note: the mounting medium should match that used for imaging cell samples.
3. Equilibrate HBSS (+ 20 mM HEPES, hereafter called 'HBSS supplement') in a centrifuge tube by placing in an incubator to pre-warm O/N at 37 °C + 5% CO₂.
   Note: the addition of HEPES is to buffer CO₂, if using an incubation chamber with CO₂ input skip this supplementation.

3. Microscope Set-up

1. Turn on the stage-top incubator of the TIRF-SIM microscope and place enough distilled water in the reservoir of the microscope to fully cover the base of the heating element. Allow water to equilibrate to 37 °C. Add the heating collar to the objective lens, also set to 37 °C.
   Note: Due to the sensitivity of the set-up, these steps are done the night before so the optical components are at a stable temperature prior to imaging.
2. Place the TIRF 488 compatible grating block in the light path of the microscope and secure in place.
   1. Switch the channel mode and the fiber cable to the 'Single' position on the laser bed and microscope stage respectively to engage the correct optical path for the TIRF-SIM mode. Note: Skipping this step will lead to an error (Figure 1A) within the software window.
3. Turn on all microscope and computer components and open the acquisition control software.
   1. Open the OC panel, Ti Pad, N-SIM Pad and ND Acquisition windows (Figure 1B). In the 'OC Panel' select the option 'TIRF 488' (Figure 2B, yellow arrow).
   2. In the N-SIM Pad window, select the 'TIRF-SIM' and 'Moving' settings (Figure 2C, yellow arrows). Select the settings button (Figure 1C, yellow box) and select 'TIRF 488 (for 100x TIRF 488nm)' from the dropdown menu, hit 'Apply' and 'OK'.
   3. Set the camera settings within the 'DU897 settings' to: Frame Rate 50 msec, Readout mode, EM Gain 3 MHz at 14-bit, and EM Gain Multiplier 300 with a Conversion Gain of 1 (Figure 1C, green box).
      Note: see Representative Results for frame rate considerations.
   4. Calibrate the microscope's illumination for TIRF-SIM by selecting the 100x 1.49 NA CFI Apochromat TIRF objective, on the 'N-SIM Pad' turn the 488 nm laser power to 10%.
      Note: Pixel size of the raw images is 60 nm, with a 512 x 512 pixel field of view.
   5. Clean the objective lens using lens tissue to remove dust and oil.
4. Find the TIRF-SIM focal plane.
   1. Start with the objective in the lowest possible z-plane. Place a drop of lens oil on the objective and add the fluorescent microsphere sample from step 2.2 to the microscope stage.
   2. Select the Perfect Focus System (PFS) function located on the body of the microscope.
   3. Slowly move the objective up through the z-plane until the PFS button stops flashing to confirm the focal plane has been reached.
   4. Switch to 'Live' view on the control software and image the microspheres for final adjustments made by using the PFS micro-manipulator wheel.
5. Observe the uniform illumination with no out-of-focus fluorescence emission above the 75 nm evanescent wave.
   Note: Confirm structured illumination is fulfilled by observing a fluctuating illumination pattern from the fluorescent microsphere sample.
6. Measure fluorescent microsphere sample and quantify the improved imaging resolution by observing the Full Width at Half Maximum (FWHM) of the intensity profile.
   1. Open the Analyze software (v4.20.01), with the NIS-A N-SIM Analysis module installed and select 'Applications' -> 'Show N-SIM window'.
   2. With the illumination modulation contrast (IMC) and high resolution noise suppression (HRNS) set to 1 for TIRF-SIM datasets (Figure 2), reconstruct the microsphere sample image to generate a reconstructed image of 1,024 x 1,024 pixels with a pixel size of 30 nm.
   3. Using the line profile tool with a width of 1 pixel (Figure 3A, yellow box), draw a line through the center of a single microsphere.
   4. Select the FWHM button (Figure 3B, yellow box) and click at the intensity maxima of the Gaussian distribution, followed by its intensity minima.
   5. Confirm the FWHM of the microsphere is in the range of 100 nm (Figure 3C). Note: the resolution achieved when imaging biological samples will vary depending on the signal to noise ratio from labelling efficiency and background fluorescence.

4. Sample Preparation for Imaging

1. Coverslip preparation.
   1. Retrieve αCD-coated coverslip from step 2.1 from fridge or incubator and remove PBS containing αCD3 and αCD28 from each well.
   2. Add 200 µl of pre-warmed HBSS supplement from step 2.3 to each well of the coverslip. Wash and remove to ensure minimal αCD3 and αCD28 is left in solution. Add another 200 µl of the HBSS supplement to the coverslip wells.
   3. Clean the underside of the coverslip with lens tissue to remove oil and particulates.
2. Position coverslip on the microscope stage and find the TIRF focal plane of the coverslip by using the PFS z-plane as found in step 3.4.
3. Cell preparation.
   1. Pipette 200 µl of the media containing transfected cells from step 1.1.8 into a micro-centrifuge tube.
   2. Pellet cells in a micro-centrifuge at 268 x g for 20 sec and remove cell media.
   3. Resuspend cells in 200 µl pre-warmed HBSS supplement from step 2.3.

5. Imaging Cells

1. Take cell-containing HBSS supplement to the microscope and gently pipette 200 µl into one of the wells.
2. While remaining in the TIRF focal plane at the coverslip (in PFS mode), use the microscopes eyepiece and epifluorescence illumination to track fluorescent cells approaching the coverslip.
3. Once the cells land on the coverslip, switch to 'Live' mode in the N-SIM Pad to bring the image up on the monitor and check the cells are in focus on the computer screen.
4. To record time-courses, open the ND Acquisition window and select the 'Time' tab, click 'Add' and set 'Interval' to No delay, 'Duration' to 10 min, 'Loops' is set to 1.
   Note: Duration can be changed depending on the process of interest; the STICS software can extract quantitative results with image stacks of ≈10 frames when captured with 'No Delay'.
5. With identical acquisition settings to the bead sample, start recording data by selecting 'Run now' in the ND Acquisition window.

6. Reconstructing SIM Images

1. After imaging is complete, open the Analyze software (v4.20.01), with the NIS-A N-SIM Analysis module installed and select 'Applications' -> 'Show N-SIM window'.
2. With the illumination modulation contrast (IMC) and high resolution noise suppression (HRNS) set to 1 for TIRF-SIM datasets (Figure 2), reconstruct one (or multiple using the Batch tool) raw SIM file to produce a reconstructed SIM image.
3. Confirm the reconstructed file has a pixel size of 30 nm, indicated in the image status bar.
   Note: Pixel size does not affect STICS analysis but it should be at least 2x smaller in dimension than the PSF spatial resolution in order to ensure spatial correlation between pixels.
4. Export the data as .tiff files, ready for STICS analysis.

7. STICS Analysis

1. Download and extract the STIC(C)S Matlab software package (Wiseman lab; available as supplementary information). The script stics_vectormapping.m is the core of a single channel STICS analysis and the first 30 lines of code are used to define the input parameters as defined in section 2.1 of the manual. Open this script and define the input parameters for TIRF-SIM data analysis. It is suggested to also add the STIC(C)S folder with subfolders to the local machine Matlab path, as detailed in the appendix of the manual.
   1. Define the input parameters for the data analysis by editing the corresponding lines of the code
   2. To perform analysis of the TIRF-SIM data to generate 1 vector map, set the temporal parameters to temporal subregion in line 31 to the maximum number of frames and the temporal shift in line 32 to 1. To obtain a time series of vector maps vary these parameters.
   3. For the Immobile removal filter, set the parameter 'filtering' (line 23) to 'Moving average' and set the 'MoveAverage' parameter (line 24) to 21.
      Note: This setting has been shown to work for slower moving and large fluorescence signals, as it is based on subtracting serial frames from the frame being analyzed.
   4. Change lines 33-38 to set the name of the output files and titles of the output vector maps, as well as the output directory path and define in what format the vector map images should be saved.

Pixel size (µm)	0.03 µm	Line 19
Frame rate (sec)	1 sec	Line 20
Maximum lag time (frames)	20 frames	Line 21
Subregion size (pixels)	8 pixels	Line 29
Subregion spacing (pixels)	1 pixel	Line 30

2. Open the script run1ChannelSTICS in the Matlab Editor window. Load the image series by running lines 1-23 of this script.
   Note: This script can be used to load and pre-process image series, plot the vector maps and speed histograms and run a batch processing for many files. If required, crop the image to a region of interest by using the line 24-26 of this script.
   1. Run lines 29-35 of run1ChannelSTICS to start STICS analysis. When prompted, select a polygon to indicate a specific region of interest (ROI), if desired. Ensure that the ROI does not overlap with the cell edge, as boundary artefacts occur when STICS analyzes these regions.
   2. Run lines 38-52 to display the vector maps. To plot the velocity magnitude histogram, run lines 53-72.

Results

Microscope settings

After successfully achieving all steps the researcher will have accomplished structured illumination (SIM) imaging of F-actin flow in a T cell synapse (Video 1), and quantified this molecular flow by spatio-temporal image correlation spectroscopy (Figure 4¹¹). Before recording images, ensure the sample illumination is uniform across the nine raw images and that structured illuminat...

Discussion

This technique will allow researchers to image and quantify previously unresolvable details in cells expressing fluorescence. In our results we show that F-actin flows in a symmetrical fashion from the cell periphery towards the cell center in the T cell IS. These findings are consistent with previous results from 1D line profile kymograph data¹³ and extend the capacity of analysis to 2D and super-resolution data.

The presented technique is a progression from kymograph imaging which...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Consumables:
Jurkat E6.1 cells	APCC	ATTC TIB-152
RPMI	Life Technologies	21875-091
FBS	Sigma	F7524-500ML
Penicillin-Strepromycin	Life Technologies	15140-122
HBSS	Life Technologies	14025-050
HEPES	Life Technologies	15630-056
#1.5 8-well Labtek coverslips	NUNC, Labtek	155409
LifeAct GFP construct	Ibidi	60101
OptiMem	Life Technologies	51985-042
anti-CD3	Cambridge Bioscience	317315
anti-CD28	BD Bioscience / Insight Biotechnology	555725
PBS	Life Technologies	20012-019
Lens oil
Name	Company	Catalog number	Comments
Equipment:
Gene Pulsar Xcell System	BioRad	165-2660
Cuvette (0.4cm)	BioRad	165-2088
Centrifuge (5810R)	Eppendorf	5805 000.327
Centrifuge (Heraeus)	Biofuge pico	75003235
6 well plate	Corning	3516
Stage-top incubator	Tokai Hit	INUH-TIZSH
Nikon N-SIM microscope	Nikon	Contact company
Nikon Analyse software	Nikon	Contact company
Incubator (190D)	LEEC	Contact company