Imaging Initial Ca2+ Microdomains in Primary T Cells

Summary

Here, we provide a comprehensive protocol to resolve initial local Ca²⁺ signals, known as Ca²⁺ microdomains, in primary murine and human T cells using fluorescence microscopy. This protocol serves as a valuable resource for researchers examining Ca²⁺ signaling pathways within immune cells and to further unravel their function.

Abstract

Local, sub-second Ca²⁺ signals, termed Ca²⁺ microdomains, are highly dynamic and short-lived Ca²⁺ signals, which result in a global [Ca²⁺]_i elevation and might already determine the fate of a T cell. Upon T cell receptor activation, NAADP is formed rapidly, binding to NAADP binding proteins (HN1L/JPT2, LSM12) and their respective receptors (RyR1, TPC2) sitting on intracellular Ca²⁺ stores, like the ER and lysosomes, and leading to subsequent release and elevation of [Ca²⁺]_i. To capture these fast and dynamically occurring Ca²⁺ signals, we developed a high-resolution imaging technique using a combination of two Ca²⁺ indicators, Fluo-4 AM and FuraRed AM. For postprocessing, an open-source, semi-automated Ca²⁺ microdomain detection approach was developed based on the programming language Python. Using this workflow, we are able to reliably detect Ca²⁺ microdomains on a subcellular level in primary murine and human T cells in high temporal and spatial resolution fluorescence videos. This method can also be applied to other cell types, like NK cells and murine neuronal cell lines.

Introduction

The presented fluorescence microscopy technique enables the visualization of local, temporal initial calcium (Ca²⁺) signals in primary mouse T cells, termed Ca²⁺ microdomains. Ca²⁺ microdomains represent highly dynamic and short-lived Ca²⁺ signaling events, posing challenges for effective live cell imaging and analysis¹.

T cells are challenging for live cell imaging due to the relative differences in central and peripheral fluorescence intensity, which can be attributed to their spherical shape and their small diameter of ~6-8 µm. Upon stimulation and immune synapse formation, T cells undergo morphological changes, further complicating the imaging of T cells¹. Therefore, employing a ratiometric analysis becomes imperative, achieved by either recording two images representing different properties of a Ca²⁺ dye or utilizing a combination of two Ca²⁺ dyes. The demanding characteristics of Ca²⁺ microdomains include their rapid, temporally, and spatially limited nature. To capture this, the Ca²⁺ dyes used must possess both a high basal brightness and a high signal-to-noise ratio (SNR) to obtain the highest temporal and spatial resolution possible. Optimal results have been achieved using a combination of the dual-wavelength dye Fura Red and the single-wavelength dye Fluo-4. Co-loading cells with Fluo-4 and Fura Red mitigate the challenges posed by strong photobleaching of dual-emission wavelength dyes and the temporal delay associated with dual-excitation dyes, ensuring suitability for fast image acquisition. This approach further facilitates the visualization of changes in shape and subtle movements. Special demands are also placed on the imaging system in terms of spatial resolution to enable the visualization of Ca²⁺ signals originating from the opening of small clusters of channels or even single channels¹.

Ca²⁺ signaling plays a pivotal role in activating immune functions within T cells, including synapse formation and cytokine production and release^2,3. The specific fate of the cell is regulated through the differently pronounced and locally distributed Ca²⁺ signals, the Ca²⁺ microdomains³. Notably, those local Ca²⁺ signals precede a widespread rise in intracellular Ca²⁺ levels in T cells and the formation of Ca²⁺ microdomains depends on both Ca²⁺ entry and release^1,4,5. Upon T cell receptor (TCR)/CD3 stimulation, the formation of Ca²⁺ releasing second messengers, like nicotinic acid adenine dinucleotide phosphate (NAADP), D-myo-inositol 1,4,5-trisphosphate (IP₃) and cyclic ADP ribose (cADPR) is triggered, leading to rising intracellular Ca²⁺ levels up to 1 µM^6,7. Early Ca²⁺ signaling events are linked to the release of Ca²⁺ from intracellular Ca²⁺ stores such as the endoplasmic reticulum (ER), with channels such as the ryanodine receptor 1 (RyR1) and the IP₃ receptor (IP₃R) being predominantly responsible for this signaling. This subsequently triggers extracellular Ca²⁺ influx and results in a global Ca²⁺ signal via store-operated Ca²⁺ entry (SOCE)⁸. In addition, there are other channels involved in Ca²⁺ signaling during T cell activation⁹, e.g., P2X4 and P2X7 channels ensure adenosine triphosphate (ATP)-dependent cation influx, contributing to the rise in intracellular Ca²⁺. Remarkably, initial adhesion-dependent Ca²⁺ microdomains (ADCMs) are already formed before TCR stimulation but with lower Ca²⁺ amplitudes and frequencies. These initial TCR-independent Ca²⁺ signals most likely serve the migration of the T cell to the site of inflammation and prime the T cells for restimulation at the site of infection^10,11.

By developing the described method for local Ca²⁺ imaging, we have gained an additional tool for exploring the origin and significance of early Ca²⁺ signals in T-cell activation. This method enables the user to detect smaller, short-lived, and more rapidly occurring Ca²⁺ signals than previously possible. Moreover, Deconvolution, Analysis, Registration, Tracking, and Shape normalization (DARTS), the Python-based analysis pipeline, enables sharing the analysis tools with a broader audience¹².

Protocol

All animal experiments were approved and performed in accordance with the animal welfare guidelines of the Institutional Animal Care and Use Committee at University Medical Centre Hamburg-Eppendorf.

1. Isolation of primary mouse T cells from lymph nodes and spleens

1. Harvest the spleens/lymph nodes under sterile conditions according to the ethical guidelines and place them in a tube with ice-cold Clicks medium (10% Fetal Calf Serum (FCS), 2 mM L-Glutamine, Penicillin/Streptomycin (P/S) 100 U/mL, 50 nM β-Mercaptoethanol).
   NOTE: Wild type and KO mouse T cells are isolated in the same way.
2. Cool down the centrifuge to 4 °C.
3. Place the spleen and lymph nodes in a 70 µm cell strainer in a sterile Petri dish and add spleen isolation medium (RPMI medium + 7.5 % NCS + 1% P/S) to a total volume of 20 mL.
4. Thoroughly but gently disrupt the spleen using a plastic mortar. Transfer the cells afterward to a 50 mL centrifuge tube.
   NOTE: A total volume of 20 mL should now be in the 50 mL centrifuge tube. Spleen and lymph nodes should be completely disrupted. Work on ice from this point onwards. Keeping the cells on ice slows down their metabolism and prevents cell death, ensuring they remain viable and functionally intact. This helps to preserve their physiological state for a longer period of time, which is especially important for accurate downstream experiments.
5. Centrifuge the cell suspension at 300 x g for 5 min at 4 °C.
6. Discard the supernatant and ensure resuspension of the resulting pellet in 2 mL of Dulbecco`s Phosphate Buffered Saline (DPBS) (without Ca²⁺ and Mg²⁺). Transfer to a new 12 mL centrifuge tube.

2. Negative selection of CD4⁺ T cells

NOTE: For the negative selection of CD4⁺ T cells, a T cell isolation kit containing a FcR blocker, biotinylated antibodies against non-CD4⁺ T cells, and streptavidin-coated magnetic particles are used.

1. Add 20 µL/mL of mouse FcR blocker from the T cell isolation kit to the cell suspension.
2. Further, add 100 µL of the mouse CD4⁺ T cell isolation cocktails at a concentration of 50 µL/mL, mix well, and incubate for 10 min at room temperature (RT).
3. Vortex streptavidin-coated magnetic particles for 30 s, add 150 µL to the cell suspension, mix well, and incubate for 2 min 30 s at RT.
4. Add DPBS to the cell suspension, filling it to a total volume of 7 mL. Ensure to pipette up and down gently, then incubate in the magnet for 2 min 30 s at RT.
5. Subsequently, gently collect the enriched T cells in the supernatant and transfer them to a new 12 mL centrifuge tube.
6. Centrifuge cells at 300 x g at RT for 5 min.
7. Discard the supernatant.
8. Resuspend the isolated and enriched CD4⁺ T cells in 1 mL of DPBS, thoroughly pipetting up and down. Incubate the mixture for 2 min 30 s in the magnet at RT to ensure no bead contamination.
9. Collect the supernatant, transfer it into a new centrifuge tube and count the enriched T cells.
10. To count the cells, stain the cells with trypan blue and count using an automated cell counter.
    NOTE: It is best to directly load the isolated CD4⁺ T cells and image them on the same day of isolation. However, primary murine T cells can be cultured without any stimulation overnight in an incubator at 37 °C with 5 % CO₂ in an isolation medium (RPMI 1640 + 10% FCS + P/S).

3. Loading of primary mouse CD4⁺ T cells

NOTE: To measure free cytosolic Ca²⁺ concentrations, fluorescent and membrane-permeable dyes are employed in the Ca²⁺ imaging experiments. A combination of Fluo-4- acetoxymethyl ester (AM) and the ratiometric dye Fura Red-AM serves as an indicator for rapid detection of local Ca²⁺ signals. Ensure to work in the dark while using fluorescent dyes.

1. Centrifuge 2-5 x 10⁶ cells for 5 min, 300 x g, discard the supernatant and resuspend in 480 µL of T cell medium (RPMI 1640, 10% FCS) with 10 µM Fluo-4-AM (Stock: [1 mM]), and 20 µM Fura Red-AM (Stock: [4 mM]).
2. Incubate initially for 20 min at RT under the bench in the dark, covering the falcon with aluminum foil.
   NOTE: Do not exceed the first incubation time, as the Ca²⁺ indicators are diluted in DMSO, which could be harmful to the cells at a longer exposure time in the low total volume. Keep the cells shielded from light after loading due to the light sensitivity of the dyes.
3. At the end of the initial 20-min incubation, add 2 mL of T cell medium to the cells and continue the incubation for an additional 30 min in the dark at RT.
4. Centrifuge the cell suspension at 300 x g for 5 min at RT.
5. Discard the supernatant and cleanse the pellet by washing it with 2 mL of Ca²⁺ measuring buffer (140 mM NaCl, 5 mM KCl, 1 mM MgSO₄, 1 mM CaCl₂, 20 mM HEPES, 1 mM NaH₂PO₄, 5 mM glucose, pH 7.4) followed by centrifugation at 300 x g for 5 min at RT.
6. Finally, discard the supernatant and resuspend it with the Ca²⁺ measuring buffer. Adjust the cell number to ~100,000 cells per 10 µL.
   NOTE: A volume of 10 µL is used for one measurement. Avoid overdiluting the cells, as one might need to further adjust the cell count under the microscope to reach an optimum for the measurement conditions.
7. Allow the cells to incubate for ~20 min to ensure complete cell infiltration and de-esterification of the dyes. Store cells on ice and in the dark until measurement.
   NOTE: Loaded primary murine T cells can be used up to ~4 h after loading.

4. Local Ca²⁺ imaging

1. Slide preparation
   1. Coat microscope coverslips (24 mm x 46 mm) with bovine serum albumin (BSA, 5 mg/mL) and poly-L-lysine (PLL, 0.1 mg/mL) (Figure 1A). Allow BSA to sit for ~20 min before applying PLL.
      NOTE: Coating facilitates cell adhesion. For proper cell adhesion, accurately spread both BSA and PLL until no more streaks are visible.
   2. To create reaction chambers, glue reusable rubber O-rings to the slides using a silicone paste (Figure 1B,C).
      NOTE: Other slides/chambers or plates such as 35 mm slides, reusable metal chambers, or 8-, 24- or 48-well plates can be used for image acquisition.
2. Coating of Protein G magnetic beads (10 µm in diameter) with antibodies
   1. Gently mix protein magnetic bead suspension and transfer 12.5 µL to a new tube.
   2. Place the tube on a magnetic stand to remove the storage buffer and let the beads migrate to the magnet. Gently remove the storage buffer by pipetting it on the opposite side of the magnet.
   3. To remove any remaining storage buffer, wash the beads by adding 500 µL of PBS-T (PBS, 0.1% Tween) and vortexing for 10 s. Place the tube in the magnetic stand again and remove the buffer.
   4. To coat the beads with antibodies, resuspend them in 7.5 µL of PBS-T and add 5 µL of anti-CD3 (0.5 mg/mL) and anti-CD28 (0.5 mg/mL), respectively. Incubate for 30-60 min with continuous mixing at RT.
   5. Wash the coated beads three times with 500 µL of PBS-T, followed by washing once with 500 µL of Ca²⁺ measuring buffer using the magnetic stand.
   6. Resuspend beads in 200-400 µL of Ca²⁺ measuring buffer.
      NOTE: Check bead density during measurements, further dilute if necessary.
3. Local Ca²⁺ imaging microscopy
   NOTE: Imaging is performed using a brightfield light microscope with a 100-fold magnification, equipped with a xenon arc lamp as a light source. Frames are captured in 14-bit mode with two-fold binning using an electron-multiplying charge-coupled camera. To record and split the emission wavelengths of both dyes, a dual-view module is employed, featuring the following filters in nm (ex, 480/40; bs, 495; em1, 542/50; em2, 650/57). The imaging setup includes an acquisition hub and imaging software for image acquisition.
   1. Place 10 µL of the loaded cells on the prepared slide and let them attach to the slide for 3-5 min.
   2. Gently add 80 µL of Ca²⁺ measuring buffer to the slide.
   3. Select the 100x oil immersion lens and apply a small drop of immersion oil. Place the slide on the microscope table.
   4. Adjust the focus in the brightfield modus, carefully select a field of view with up to 10 cells that are not in contact with each other, and acquire an image.
      NOTE: Overlapping and touching cells will be hard to analyze later.
      1. Turn on the lamp, check the fluorescence and the loading of the T cells, and compare cells in both channels to verify that they are not preactivated.
      2. Ensure to take pictures of the field of view for both brightfield and fluorescence channels before and after the measurement to check for cell movement and loading.
   5. Start measurement and capture the basal activity for 1 min with an acquisition rate of 1 frame per 5 s.
   6. Add 10 µL of compound/stimulant (bead or stimulator/inhibitor) after 1 min and measure for a total of 3 min using 40 frames per s or the maximum frame rate possible.
      NOTE: The addition of beads is the crucial step; ensure that the beads are added close to the excitation light on the slide while not moving the slide. Representative examples of Ca²⁺ microdomains in primary CD4⁺ T cells upon bead contact from WT and P2x4-/- and P2x7-/- mice are shown in Figure 2.

5. Postprocessing/ data analysis

NOTE: For image processing and data analysis, the Python-based open-source pipeline DARTS is used. It has been developed by Woelk et al.¹² based on work by Diercks et al.¹³.

1. Installation of the DARTS pipeline
   1. Install Python 3.10.0¹⁴, anaconda¹⁵, and git¹⁶, and clone the GitHub repository using the terminal command git clone: https://github.com/IPMI-ICNS-UKE/DARTS.git
   2. Create a conda environment with conda create --name DARTS and install all necessary Python packages with pip install <package>.
      NOTE: Follow the installation instructions in the DARTS GitHub repository^17,18 for more detailed information.
   3. Before using Bioformats, ensure that a Java Runtime Environment is correctly installed¹⁹.
2. Usage of the DARTS pipeline
   NOTE: Once installed, DARTS can be started from the terminal window.
   1. Navigate to the local copy of the DARTS repository (cd path/to/DARTS) that contains the main.py file. Make sure that the conda environment is activated (conda activate DARTS).
   2. Execute DARTS by typing python main.py in the terminal window and pressing Enter.
   3. In the graphical user interface (GUI), specify the source and results directory as well as the image format in the upper left section (see Figure 3).
   4. Define the properties of measurement, such as the scale (pixels per micrometer), frame rate, cell type, etc. The processing pipeline can be assembled based on the specific image data and research question.
      NOTE: Read more about the deconvolution parameters, for instance, in the corresponding documentation²⁰.
   5. Save the settings on the computer before clicking on Start. A more detailed description of the GUI can be found in the DARTS documentation.
3. Ca²⁺ microdomain imaging and bead contacts
   NOTE: For local imaging and stimulation with beads and subsequent image analysis, the cell-specific bead contacts need to be defined by the user to identify the starting point (t = 0) of the measurement period of interest (see Figure 4). A bead contact is defined as the contact site of a (stimulatory) bead with a cell of interest. It consists of the specification of (1) the bead contact time, e.g., frame 300, and (2) the bead contact position relative to the center of the cell, e.g., 1 o'clock or 12 o'clock. With the current DARTS version, the user needs to manually define and enter the information for each bead contact. Use the following procedure to define bead contacts:
   1. Use the slider to pinpoint the time of contact between a bead and a cell of interest.
   2. In the options menu on the right-hand side, select bead contact: x, y, t.
   3. Select the spot on the left side of the image where the cell and bead make contact.
   4. Select Choose cell by clicking a point inside. Click on the cell stimulated by this bead, preferably in the center.
   5. Click on ADD bead contact.
   6. Repeat steps 5.3.1- 5.3.5 for each additional bead contact in this file. Once all bead contacts are defined, click on the Continue button.
   7. Proceed with additional files. After the last file is reached, the script will automatically initiate analysis of all files.
4. Data analysis
   1. Find the processed ratio images, microdomain data for each cell over time (incl. localization, amplitude, size), source data for the dartboard projection and further resources in the results folder.
   2. To create dartboards, navigate to the folder /src/analysis/ inside the DARTS folder (pattern: cd path/to/DARTS/src/analysis) containing the script DartboardPlotGUI.py. Then, type python DartboardPlotGUI.py and press enter.
   3. Ensure that the necessary information is provided and that the spreadsheet files from the results folder are accurately selected as the source files for dartboard generation.

Discussion

We described an extensive protocol for high-resolution live cell imaging of local Ca²⁺ microdomains in primary murine and human T cells triggered by TCR/CD3 stimulation through antibody-coated beads. Moreover, we implemented a user-friendly and open-source Python-based algorithm to identify and analyze local Ca²⁺ signals. Notably, the protocol is not limited to the detection of Ca²⁺ microdomains in the context of TCR/CD3 stimulation but is adaptable to other (immune) cell types such as NK cell lines (KHYG-1)¹² or TCR-independent Ca²⁺ microdomains^10,11.

A critical step within the protocol is the size and number of the stimulating beads. To mimic an immune-synapse, the beads should be similar in size to the cells. Hence, for primary murine and human T cells as well as cell lines (Jurkat and KHYG1), we use magnetic beads with a diameter of 10 µm. Furthermore, each cell should be just stimulated by one single bead. Therefore, the number of beads added to each slide should be sufficient on the one hand, but if there are too many beads in the field of view, the background increases, and it is not possible to detect a single activation time point and contact side.

The protocol utilizes the fluorescent Ca²⁺ dyes Fluo-4 AM and FuraRed AM in a ratiometric manner, therefore allowing for calibration of the data¹³. Additionally, the protocol could be adapted to other Ca²⁺ indicator pairs, but caution must be taken in the selection process in terms of Ca²⁺ binding kinetics, subcellular distribution, and photobleaching¹. Moreover, loading conditions must be developed and optimized for each individual cell type, but the concentrations indicated here are a good starting point. To visualize Ca²⁺ microdomains, the Kd of the Ca²⁺ dyes should be in the range of 300-1200 nM and the acquisition time per frame should be ≤60 ms. If the fluorescence intensity is too low, the filter set has to be checked, but it is also possible to load a double amount of Ca²⁺ dye into T cells. However, the Ca²⁺ dye could mislocate to other organelles or sequester to vesicles, but it might also act as a Ca²⁺ buffer and affect the Ca²⁺ responses.

One limitation of the analysis algorithm is that a spherical shape of the cell is assumed; hence, cell types with different morphologies might need adaption of the analysis toolbox. The algorithm has been used to analyze local Ca²⁺ microdomains in primary murine T cells, as well as Jurkat T cells and an NK cell line (KHYG-1)¹² and was successful in the analysis of Ca²⁺ microdomains for a murine neuronal cell line (N2a, unpublished data). In principle, the protocol and analysis toolbox could be used to analyze non-spherical cell types such as HEK293 or HeLa cells, but for these cell types, the dartboard projection cannot be adapted because it is based on a round structure and shape normalization of the cells. In addition, the protocol to detect localized initial Ca²⁺ microdomains upon bead stimulation can be adapted to analyze local Ca²⁺ signals derived from other stimuli, such as soluble activating or inhibiting compounds, as well as adhesion-dependent and TCR/CD3 independent Ca²⁺ microdomains^10,11. Of note, it is easier to define a single bead contact in terms of time and location than to determine the starting point of activation after soluble compounds.

A general limitation for the detection of Ca²⁺ microdomain formation lies in the required high temporal-spatial resolution and necessary high signal-to-noise ratio (SNR). Currently, the resolution derived from our setup reaches a calculated spatial resolution of ~0.368 µm and temporal resolution of ~40 frames per second (fps)¹. Recent advances in camera and detector development, as well as improvement of fluorescent dyes, might lead to the possibility of reaching optical single channel recordings as they have been described for ORAI-GECI (genetically expressed Ca²⁺ indicators) constructs²⁴ for live cell imaging using Ca²⁺ indicators with higher temporal and spatial resolution in the future.

Taken together, the protocol and analysis tool for high-resolution Ca²⁺ microdomain imaging described here can be used not only to analyze initial local Ca²⁺ signals in T cells but can also be adapted to other cell types to decipher the significance of local Ca²⁺ signaling in these.

Materials

Name	Company	Catalog Number	Comments
α-CD3	BD Pharmingen	553058
α-CD28	BD Pharmingen	553295
Anaconda	Anaconda	https://docs.anaconda.com/free/anaconda/install/index.html
Bovine serum albumin	Sigma-Aldrich	A2058
Cell strainer	Biofil	CSS-013-070	70 µm diameter
Countess Automated Cell Counter	Invitrogen	A49865
Cover slips	Paul Marienfeld GmbH	101202
DARTS	GitHub	https://github.com/IPMI-ICNS-UKE/DARTS
Dulbecco’s Phosphate buffered saline	Gibco, Thermo Fisher	14190144
EasySep CD4+ T cell Isolation Kit	Stemcell	19852
EasySep Magnetic Multistand	Stemcell	18010
Fluo 4-AM	Invitrogen	F14201
Fura Red-AM	Invitrogen	F3020
Gibco RPMI 1640 medium	Gibco, Thermo Fisher	11875093
Git	GitHub	https://git-scm.com/book/en/v2/Getting-Started-Installing-Git
immersion oil	Leica	11513859
Newborn calf serum	Sigma-Aldrich	N4637
Oracle	Oracle	https://www.oracle.com/de/java/technologies/downloads/
Penicillin/Streptamycin	Gibco, Thermo Fisher	15240062
Poly-L-lysine	Sigma-Aldrich	P4707
Protein G magnetic beads	Merck	LSKMAGG02	10 µM diameter
Python	Python	https://www.python.org/downloads/
Silicon grease (basylone)	Bayer	291-1220
Time-Dependent Entropy Deconvolution	GitHub	https://ipmi-icns-uke.github.io/TDEntropyDeconvolution/General/2-usage.html#input-parameters
Tween	q-biogene	TWEEN201