Confocal Microscopy to Measure Three Modes of Fusion Pore Dynamics in Adrenal Chromaffin Cells

Summary

This protocol describes a confocal imaging technique to detect three fusion modes in bovine adrenal chromaffin cells. These fusion modes include 1) close-fusion (also called kiss-and-run), involving fusion pore opening and closure, 2) stay-fusion, involving fusion pore opening and maintaining the opened pore, and 3) shrink-fusion, involving fused vesicle shrinkage.

Abstract

Dynamic fusion pore opening and closure mediate exocytosis and endocytosis and determine their kinetics. Here, it is demonstrated in detail how confocal microscopy was used in combination with patch-clamp recording to detect three fusion modes in primary culture bovine adrenal chromaffin cells. The three fusion modes include 1) close-fusion (also called kiss-and-run), involving fusion pore opening and closure, 2) stay-fusion, involving fusion pore opening and maintaining the opened pore, and 3) shrink-fusion, involving shrinkage of the fusion-generated Ω-shape profile until it merges completely at the plasma membrane.

To detect these fusion modes, the plasma membrane was labeled by overexpressing mNeonGreen attached with the PH domain of phospholipase C δ (PH-mNG), which binds to phosphatidylinositol-4,5-bisphosphate (PtdIns(4,5)P₂) at the cytosol-facing leaflet of the plasma membrane; vesicles were loaded with the fluorescent false neurotransmitter FFN511 to detect vesicular content release; and Atto 655 was included in the bath solution to detect fusion pore closure. These three fluorescent probes were imaged simultaneously at ~20-90 ms per frame in live chromaffin cells to detect fusion pore opening, content release, fusion pore closure, and fusing vesicle size changes. The analysis method is described to distinguish three fusion modes from these fluorescence measurements. The method described here can, in principle, apply to many secretory cells beyond chromaffin cells.

Introduction

Membrane fusion mediates many biological functions, including synaptic transmission, blood glucose homeostasis, immune response, and viral entry^1,2,3. Exocytosis, involving vesicle fusion at the plasma membrane, releases neurotransmitters and hormones to achieve many important functions, such as neuronal network activities. Fusion opens a pore to release vesicular contents, after which the pore may close to retrieve the fusing vesicle, which is termed kiss-and-run^1,4. Both irreversible and reversible fusion pore opening can be measured with cell-attached capacitance recordings combined with fusion pore conductance recordings of single vesicle fusion.

This is often interpreted as reflecting full-collapse fusion, involving dilation of the fusion until flattening of the fusing vesicle, and kiss-and-run, involving fusion pore opening and closure, respectively^{5,6,7,8,9,10,11,12,13}. Recent confocal and stimulated emission depletion (STED) imaging studies in chromaffin cells directly observed fusion pore opening and closure (kiss-and-run, also called close-fusion), fusion pore opening that maintains an Ω-shape with an open pore for a long time, termed stay-fusion, and shrinking of the fusing vesicle until it complete merges with the plasma membrane, which replaces full-collapse fusion for merging fusing vesicles with the plasma membrane^{4,8,14,15,16,17}.

In neurons, fusion pore opening and closure have been detected with imaging showing the release of quantum dots preloaded in vesicles that are larger than the fusion pore and with fusion pore conductance measurements at the release face of nerve terminals^5,18,19. Adrenal chromaffin cells are widely used as a model for the study of exo- and endocytosis^20,21. Although chromaffin cells contain large dense-core vesicles, whereas synapses contain small synaptic vesicles, the exocytosis and endocytosis proteins in chromaffin cells and synapses are quite analogous^{10,11,12,20,21,22,23}.

Here, a method is described to measure these three fusion modes using a confocal imaging method combined with electrophysiology in bovine adrenal chromaffin cells (Figure 1). This method involves loading of fluorescent false neurotransmitters (FFN511) into vesicles to detect exocytosis; addition of Atto 655 (A655) in the bath solution to fill the fusion-generated Ω-shape profile, and labeling of the plasma membrane with the PH domain of phospholipase C δ (PH), which binds to PtdIns(4,5)P₂ at the plasma membrane^8,15,24. Fusion pore dynamics can be detected through changes in different fluorescent intensities. Although described for chromaffin cells, the principle of this method described here can be applied widely to many secretory cells well beyond chromaffin cells.

Protocol

NOTE: The animal use procedure followed NIH guidelines and was approved by the NIH Animal Care and Use Committee.

1. Bovine chromaffin cell culture

1. Prepare Locke's solution (Table 1) and autoclave tools 1 day before chromaffin cell culture.
2. Obtain bovine adrenal glands from a local abattoir on the culture day, and keep them submerged in ice-cold Locke's solution before dissection.
   NOTE: Adrenal glands are from 21-27-month-old, healthy, black Angus of either sex (mainly male) with body weight around 1,400 pounds (~635 kg).
3. Prepare 30 mL (for 3 adrenal glands) of fresh enzyme solution containing collagenase P, trypsin inhibitor, and bovine serum albumin (Table 1) before dissection and keep it at room temperature.
4. Choose 3 intact glands without cuts or bleeds on the surface and remove the fat tissue with scissors (Figure 1A). Wash the glands by perfusion with Locke's solution until no blood comes out. To achieve this, inflate the gland through the adrenal vein (Figure 1B) using a 30 mL syringe attached with a 0.22 µm filter as many times as needed²⁵.
   NOTE: Approximately 150 mL of Locke's solution is usually needed to wash 3 glands.
5. For digestion, inject the enzyme solution through the adrenal vein using a 30 mL syringe attached with a 0.22 µm filter until the gland starts to swell. Then, leave the glands at 37 °C for 10 min. Inject once again and leave the glands at 37 °C for another 10 min.
   NOTE: Approximately 30 mL of enzyme solution is needed to digest 3 glands.
6. After digestion, cut the gland longitudinally from the vein to the other end with scissors to unfold the gland (Figure 1C). Isolate the medullae by tweezing out the white medulla into a 10 cm Petri dish containing Locke’s solution. 
   NOTE: The comparison of the interior of the gland before and after digestion is shown in Figure 1C. The details of the digestion and medullae isolation are reported previously^23,25.
7. Cut and mince the medulla into small pieces with scissors (Figure 1D). Filter the medulla suspension with an 80-100 µm nylon mesh into a beaker. Then transfer the filtrate to a 50 mL conical tube for centrifugation at 48 × g, room temperature, for 3 min with deceleration of 3.
   NOTE: The mincing of the medullae usually takes ~10 min. To obtain a good yield of cells, the minced pieces must be very small, until they cannot be tweezed up.
8. After centrifugation, remove the supernatant and resuspend the cell pellet with Locke's solution by pipetting. Filter the cell suspension with an 80-100 µm strainer, and centrifuge at 48 x g, room temperature, for 3 min with deceleration of 3.
9. Remove the supernatant and resuspend the cell pellet with 30 ml of culture medium (Table 1). Determine the cell number using hemacytometer counting chambers²⁵.

2. Transfection with electroporation

1. Transfer 2.8 × 10⁶ cells into a 15 mL tube. Pellet the cells by centrifugation at 48 × g for 2 min with deceleration of 3. Add 100 µL of transfection buffer (see the Table of Materials) provided by the manufacturer to the cell pellet, and then add 2 µg of the PH-mNG plasmid.
   NOTE: The PH-mNG plasmid was created by replacing the enhanced green fluorescent protein (EGFP) with mNG in PH-EGFP¹⁵ (see the Table of Materials).
2. Gently mix the suspension by pipetting the solution up and down, and transfer the mixture into an electroporation cuvette without delay (Figure 1E). Immediately transfer the cuvette to the electroporator (see the Table of Materials), select the O-005 program in screen list and press Enter to perform electroporation.
   NOTE: Prepare the cell suspension for transfection in a cell culture hood. Do not introduce air bubbles into the suspension during the mixing step. Proceed to the next step without delay.
3. After electroporation, add 1.8 mL of medium immediately to the cuvette and mix gently with a micropipettor equipped with a sterile tip. Then add 300 µL of the suspension of the electroporated cells on top of the coverslip (see the Table of Materials) in each dish, plating 5-6 dishes in total for one electroporation reaction.
4. Carefully transfer the dishes to a humidified incubator at 37 °C with 9% CO₂ for 30 min, and gently add 2 mL of prewarmed medium to each dish after 30 min.
5. Keep the transfected cells in the humidified incubator at 37 °C with 9% CO₂ for 2-3 days before experiment.
   ​NOTE: The cultured cells will last for a week. It is best to use cultured cells on days 2-3.

3. Preparation for patch-clamp recording and confocal imaging

NOTE: This protocol was performed with a laser scanning confocal microscope and patch-clamp amplifier with voltage-clamp recording together with a lock-in amplifier for capacitance recording. An XY plane confocal imaging at a fixed Z-plane (XY/Z_fixed scanning) was used to image all three fluorescent signals simultaneously. The Z-plane was focused at the cell bottom where the plasma membrane was adhering to the coverslips.

1. On the day of the patch-clamp and imaging experiment, observe the cells under a fluorescence microscope. Use brightfield to make sure the cell culture is not contaminated (Figure 1F) and epifluorescence to check for proper expression of the fluorescent-tagged protein.
   NOTE: For example, PH-mNG is expressed at the plasma membrane of ~20-30% of the cells.
2. Prepare patch pipettes from borosilicate glass capillaries. To do this, pull the pipettes with a pipette puller, coat their tips with liquid wax, and polish them with a microforge (see the Table of Materials).
3. Turn on the patch-clamp recording amplifier and start the patch-clamp recording software (see the Table of Materials). Set the appropriate parameters for calcium current and capacitance recording in the software (see the Table of Materials).
   1. Set a recording protocol of 60 s duration in total, where the stimulation starts at 10 s.
   2. Set the holding potential for voltage clamp recording to -80 mV. Set a 1 s depolarization from -80 mV to 10 mV as the stimulus to induce calcium influx and capacitance jump.
   3. For capacitance measurements, set the frequency of the sinusoidal stimulus to a range of 1,000-1,500 Hz with a peak-to-peak voltage of no more than 50 mV.
4. Save the protocol and create a new file for recording.
5. Turn on the confocal microscope system and set the appropriate parameters in the software (see the Table of Materials).
   1. Turn on the lasers, including 458 nm, 514 nm, and 633 nm, and set the emission collection range for each laser according to each fluorescence probe as in the following. FFN511: excitation wavelength (EX), 458 nm; emission wavelength (EM), 468-500 nm. mNG: EX, 514 nm; EM, 524-560 nm. A655: EX, 633 nm; EM, 650-700 nm.
   2. Use sequential imaging for FFN511 and mNG to avoid crosstalk between these two probes. Set a timelapse of 1 min duration for image recording (Figure 1G).

4. Patch-clamp recording and confocal imaging

1. Choose a dish with good cell state and proper expression and add 2 µL of fluorescent false neurotransmitter FFN511 (10 mM stock, 1:1,000 working solution) into the medium. Put the dish back in the incubator for 20 min. Alternatively, load FFN511 after step 3.2 and perform steps 3.3-3.5 while waiting to save time.
2. After FFN511 loading finished, prepare the recording chamber and add 2 µL of fluorescent dye A655 into 500 µL of the bath solution (Figure 2A and Table 1). Transfer the coverslip from the dish into the recording chamber (see the Table of Materials) with tweezers, and immediately add A655-containing bath solution (Figure 2B).
   NOTE: The A655 is kept in -20 °C with concentration of 10 mM and the working concentration is 40 µM.
   CAUTION: Wear gloves to avoid direct skin contact with FFN511 or A655.
3. Place a drop of oil (refractive index: 1.518; see the Table of Materials) on the 100x oil immersion objective (Numerical Aperture = 1.4). Mount the chamber in the microscope and use the adjustment knob to make the oil just contact the bottom of the coverslip, then immerse the ground wire tip into the bath solution (Figure 2C, D).
   NOTE: Choose an immersion oil appropriate for working at room temperature. Switching between the low-and high-magnification lens is unnecessary. The room temperature is maintained around 20-22 °C during recording.
4. Bring the cells into focus and use brightfield and confocal imaging to find a good cell with mNG expression. Zoom in on the selected cell and adjust it to the center of the view to minimize blank regions.
   NOTE: The schematic drawing of fluorescence labeling is shown in Figure 3A. A good cell usually has a smooth membrane and clean edge (Figure 3B). With epifluorescence or confocal imaging, a good cell appears to have a large and flat bottom with bright mNG expression (Figure 3C).
   The pixel size is ~50 nm, within a range of ~40-80 nm according to different cell size and zoom factor.
5. Set up parameters for XY plane confocal imaging at a fixed Z-plane (XY/Z_fixed scanning) of FFN511, PH-mNG, and A655, with a minimized time interval. Adjust the focus to the bottom of the cell with the fine adjustment knob.
   NOTE: PH-mNG signals the contour of the cell plasma membrane at the confocal XY-plane crosssection (except the cell bottom). Near the cell membrane bottom, A655 spots can be observed, which reflect Ω-shape or Λ-shape membrane invaginations. If the Z-focal plane is lower than the cell bottom, the intensity of all fluorescence will become weaker.
6. Adjust the excitation laser power in the software to find a setting to get the best signal-to-noise ratio and avoid significant fluorescence bleaching. To do so, start with an initial test value of 2.5 mW power for FFN511 excited at 458 nm and 1 mW for mNG excited at 514 nm. For A655 excited at 633 nm with an HeNe laser, use high laser power, ~12-15 mW (i.e., 70%-80% of the maximum), which, upon continuous excitation, can bleach all fluorescent A655 inside an Ω-shape profile when its pore is closed.
   NOTE: If the laser power is too high, PH-mNG and FFN511 fluorescence will be bleached quickly. If the laser power is too low, signals will be too weak or noisy to be analyzed.
7. Add 9 µL of internal solution (Table 1) into a patch pipette and attach the pipette to the holder in the patch-clamp amplifier stage (see the Table of Materials).
8. Apply a small amount of positive pressure with a syringe and move the pipette tip to touch the bath solution with a micromanipulator. Ensure the amplifier shows that the pipette resistance is ~2-4 MΩ with a voltage pulse test. Press LJ/Auto to cancel the liquid junction.
9. Move the pipette toward the selected cell with the micromanipulator. To form a cell-attached mode, move the pipette tip to touch the cell (the resistance will increase); applying negative pressure gently with the syringe (resistance will increase to more than 1 GΩ), change the holding potential from 0 to -80 mV.
10. Once resistance passes 1 GΩ, wait for ~30 s for the configuration to stabilize. Press C-fast/Auto to compensate for fast capacitance.
11. To form a whole-cell mode, apply pulsed short yet powerful negative pressure with the syringe to rupture the membrane (the shape of the current pulse changes with a charged and discharged membrane capacitor). Press C-slow/Auto to compensate for slow capacitance.
12. Adjust the imaging focus slightly to focus on the cell bottom. Start confocal timelapse imaging and patch-clamp recording simultaneously by clicking the Start buttons in the two different software applications at the same time.
    NOTE: The confocal imaging software makes a movie at a rate of ~20-90 ms/frame. The patch-clamp recording includes the resting state for 10 s, the 1 s depolarization stimulation, and an additional 49 s after stimulation. The patch-clamp configuration permits recording of calcium current, capacitance jump, and decay induced by the 1 s depolarization from -80 to +10 mV (Figure 3D).
13. Once recording is finished, make sure the data are saved. Change the holding potential back to 0 mV. Move the pipette out of the bath solution and discard it.
14. Repeat steps 4.7-4.13 to record another cell in the dish. After 1 h of recording, change to another dish for recording.
    ​NOTE: After 1 h of recording, the success rate of patch-clamp recording decreases significantly. To increase the efficiency of data collection, recording for only 1 h is recommended in each dish.

5. Patch-clamp data analysis

1. Open appropriate software (see the Table of Materials) for data analysis. Click the PPT | Load PULSE file | File buttons to load the .dat file. Click Do it and four traces will be plotted in one graph automatically including the calcium current and capacitance traces.
2. Click the Windows | New Graph buttons, choose the Pulse_1_1_1, the first wave in Y Wave(s), and click Do it to plot the calcium current graph. Click the Windows | New Table buttons, choose the Pulse_1_1_1, and click Do it to show the raw data of calcium current, which could be used to plot the averaged trace of multiple cells.
3. Draw a square accordingly in the calcium current graph, right click and expand the current signal to include the baseline and the peak of calcium current. Click Graph | Show Info to show cursors A and B and move them to the baseline and peak respectively. The graph information will be shown in the bottom and parameters can be estimated.
4. Repeat step 5.2, but choose the Pulse_1_1_1_Cm, the second wave in Y Wave(s), to plot the capacitance trace. Repeat step 5.3 and place A and B cursors in the appropriate position to measure the capacitance parameters, such as baseline, amplitude, and decay rate.
5. Copy raw data in step 5.2 into a spreadsheet, calculate the mean and standard error of mean for a group of recorded cells, and plot the averaged traces of calcium current and membrane capacitance (Figure 3E) in an appropriate software.

6. Confocal imaging data analysis

1. Open the raw imaging files with any manufacturer-supplied software (see the Table of Materials).
   NOTE: Some other free programs, such as ImageJ or Fiji, could be utilized for data processing and quantifying the images obtained in section 4.
2. Click Process | ProcessTools and use the tools to generate rolling average (e.g., rolling average for every 4 images) files for each timelapse image and save those files.
   NOTE: After rolling average, appropriate adjustment of the brightness and background could be done to show the fluorescence changes of three channels better, which may help to identify fusion events. Check the fluorescent intensity in some regions without fusion event may help to distinguish the fluorescent signal of fusion events from background signals.
3. Click the buttons Quantify | Tools | Stack Profile, check timepoints before and after stimulation to identify fluorescence changes in each channel. Click the Draw ellipse button to circle the regions of interest (ROIs) for fusion events. Right click on the image and click Save ROIs to save the ROIs.
   NOTE: The size of ROI, which covers all the three fluorescent signals, was similar with vesicle size in chromaffin cells²⁴. Raw data were used for the analysis, while the rolling averaging data which increase the signal-to-noise ratio were provided to show clearly the three signals.
4. Click Open projects to locate the raw file, right click on the image and click Load ROIs to load the ROI file in the raw imaging files to measure the fluorescence intensities.
   NOTE: The raw fluorescent signal will be used to plot traces of different channels. For each ROI, the baseline is defined as the intensity before stimulation, and the intensity traces can be normalized to baseline.
5. Click Tools | Sort ROIs in the software and plot the traces for all three channels of each ROI. Click Report to save the ROI data, including both digital data and imaging traces for each ROI, into a file folder.
   1. Identify a fusion event by the simultaneous increase in the intensity of PH-mNG (F_PH) and A655 (F₆₅₅) spot fluorescence (within a single frame), accompanied by a decrease in FFN511 spot fluorescence (F_FFN).
   2. Look for the appearance of F_PH and F₆₅₅, which indicates PH-mNG/A655 spot formation due to an Ω-profile generated by fusion, allowing the diffusion of PH-mNG from the plasma membrane and persistent diffusion of A655 from the bath.
   3. Look for F_FFN decrease, which indicates its release from a vesicle due to fusion pore opening and excludes the possibility that F_PH and F₆₅₅ appearance may be from membrane invagination caused by endocytosis.
   4. Inspect the timelapse XY/Z_fixed images that show fusion events happening on the cell bottom. Analyze three modes of fusion events: 1) close-fusion, 2) stay-fusion, and 3) shrink-fusion.
      NOTE: Fusion events were rarely observed before depolarization, whereas tens of fusion events could be observed after depolarization. Following fusion, the Ω-profile may close its pore, maintain its open pore, or shrink to merge into the plasma membrane.
      1. To identify close-fusion, look for dimming F₆₅₅ because fusion pore closure prevents the exchange of bath A655, while F_PH sustains, reflecting the continuing conversion of PtdIns(4,5)P₂ into PtdIns(4)P, or F_PH decays with a delay, reflecting vesicle pinch-off (close-fusion, Figure 4A,B).
      2. Look for sustained F_PH and F₆₅₅ to identify stay-fusion (Figure 4C).
      3. Look for parallel decreases of F_PH and F₆₅₅, indicating Ω-profile shrinkage to identify shrink fusion (Figure 4D).
         NOTE: Check the original imaging files to inspect the imaging traces if uncertain of the event type. Checking the fluorescent intensity in some regions without fusion event may help to distinguish the fluorescent signal of fusion events from background signals.
6. Plot representative traces of intensity changes for these three channels: FFN511, PH-mNG, and A655. Quantify the percentage of different fusion modes in each cell and plot figures.

Representative Results

Following the experimental procedures shown in Figure 1 and Figure 2, chromaffin cells from bovine adrenal glands were transfected with PH-mNG to label the plasma membrane; A655 was added to the bath solution to detect fusion pore closure; and fluorescent false neurotransmitter FFN511 was loaded in vesicles for detection of release. Next, XY-plane confocal timelapse imaging of FFN511, PH-mNG, and A655 was performed every 20-90 ms at the cell bottom (Z-focal plane ~100-200 nm above the cell membrane). Whole-cell patch-clamp recording and application of a 1 s depolarization from -80 to +10 mV was performed to evoke exo- and endocytosis (Figure 3A-C). This depolarization induced an inward calcium current, a capacitance jump that indicates exocytosis, and a capacitance decay after the jump that indicates endocytosis (Figure 3D, E).

With timelapse XY/Z_fixed imaging at the cell bottom, many individual fusion events were observed^8,24 after depolarization (Figure 4A), whereas rare fusion events were observed before depolarization. Fusion events induced by the 1 s depolarization protocol were identified as FFN511 spot fluorescence (F_FFN) decrease reflecting FFN511 release, accompanied by F_PH and A655 spot fluorescence (F₆₅₅) increase reflecting PH-mNG and A655 diffusion from the plasma membrane and the bath solution into the fusing vesicle (the fusion-generated Ω-profile)¹⁵.

After fusion, the Ω-profile generated by vesicle fusion with the plasma membrane may 1) close its pore, termed close-fusion, 2) maintain an open fusion pore, termed stay-fusion, or 3) shrink to merge into the plasma membrane, termed shrink-fusion. Close-fusion was identified as F₆₅₅ dimming while F_PH sustained or decayed with a delay (Figure 4B). Stay-fusion was detected as the sustained presence of both PH-mNG and A655 spots (Figure 4C). Shrink-fusion was detected as parallel F_PH and F₆₅₅ decay accompanied with a parallel size reduction of the PH-mNG spot and the A655 spot (Figure 4D)^8,15,16,24.

Figure 1: Schematic representation of the experimental protocol. (A, B) Bovine adrenal glands are trimmed with scissors to remove fat tissue (A), flushed with Locke's solution, and digested through adrenal vein (B). (C) The interior of an adrenal gland without digestion (left) or after proper digestion (right). (D) After being washed and digested, the medullae are isolated and minced into small pieces, and chromaffin cells are separated from minced medullae after filtering and centrifuging. (E) Chromaffin cells are electroporated and plated on coverslips for incubation. (F) On days 2-3, check the chromaffin cells under a microscope before experimentation. (G) The cell sample is embedded in the chamber for patch-clamp recording and confocal imaging. Calcium current and capacitance changes are recorded, amplified, and displayed on the monitor. Fluorescence changes upon stimulation are detected and displayed on the monitor. Scale bars = 40 mm (A), 20 mm (B-D), 20 µm (F). Please click here to view a larger version of this figure.

Figure 2: Patch-clamp and confocal setup. (A) On the day of experimentation, the chromaffin cells grown on coverslips are incubated with FFN511 for 20 min. The dye A655 is added to the bath solution. (B) Chromaffin cells on the coverslip are transferred to a recording chamber, and the bath solution with A655 is added to the chamber. (C) After adding a drop of oil on the 100x oil immersion objective, the chamber is mounted onto the stage of an inverted confocal microscope. The tip of the ground wire is immersed in bath solution. The pipette is brought into position after loading with pipette solution and held by a pipette holder, which is attached to a headstage. The headstage is controlled by a motorized micromanipulator. (D) After finding a good cell, move the pipette tip into the bath solution with the micromanipulator to start whole-cell patch-clamp recording and confocal imaging. Scale bars = 10 mm (A), 5 mm (B). Please click here to view a larger version of this figure.

Figure 3: Whole-cell voltage-clamp recordings of calcium currents and capacitance changes induced by depolarization. (A) Setup drawing of chromaffin cells during whole-cell voltage-clamp recording. The chromaffin cell is immersed in A655-containing bath solution (red), and the cell membrane and vesicles are labeled with PH-mNG (green) and FFN511 (cyan), respectively. This image has been modified with permission from ²⁴. (B) A representative image of a patch-clamped chromaffin cell observed by brightfield. (C) Representative images of PH-mNG (green), A655 (red), and FFN511 (cyan) in a cell with good focus at the cell footprint. (D) An example of calcium current and capacitance changes induced by 1 s depolarization from -80 to +10 mV. (E) The averaged traces of calcium currents (ICa) and capacitance (Cm) changes collected from 20 chromaffin cells. This image has been modified with permission from ²⁶. Scale bars = 5 µm (B, C). Abbreviations: ICa = calcium current; Cm = capacitance; Depol = depolarization. Please click here to view a larger version of this figure.

Figure 4: Visualization of fusion events under the confocal microscope. (A) Many fusion spots can be detected with confocal XY-plane imaging of PH-mNG (green), A655 (red), and FFN511 (cyan) at the cell bottom. Fusion was evoked by a 1 s depolarization from -80 to +10 mV (depol_1s). FFN spots underwent release and close-fusion (circles). Images before (-1 s) and after (+1 s and +10 s) depolarization are shown. (B) An example of close-fusion. (C) An example of stay-fusion. (D) An example of shrink-fusion. This image has been modified with permission from ²⁴. Scale bars = 1 µm (A), 0.5 µm (B-D). Abbreviations: Depol = depolarization; F = fluorescent intensity. Please click here to view a larger version of this figure.

Medium/Solution	Description
Locke's solution	145 mM NaCl, 5.4 mM KCl, 2.2 mM Na2HPO4, 0.9 mM NaH2PO4, 5.6 mM glucose, and 10 mM HEPES, pH 7.3, adjusted with NaOH
Enzyme solution	1.5 mg/mL collagenase P, 0.325 mg/mL trypsin inhibitor, and 5 mg/mL bovine serum albumin in Locke's solution
Culture medium	DMEM medium supplemented with 10% fetal bovine serum
Internal solution	130 mM Cs-glutamate, 0.5 mM Cs-EGTA, 12 mM NaCl, 30 mM HEPES, 1 mM MgCl2, 2 mM ATP, and 0.5 mM GTP, pH 7.2, adjusted with CsOH
Bath solution	125 mM NaCl, 10 mM glucose, 10 mM HEPES, 5 mM CaCl2, 1 mM MgCl2, 4.5 mM KCl, and 20 mM TEA, pH 7.3, adjusted with NaOH

Table 1: Details regarding the composition of culture medium and solutions.

Discussion

A confocal microscopic imaging method is described to detect the dynamics of fusion pore and transmitter release, as well as three fusion modes, close-fusion, stay-fusion, and shrink-fusion in bovine adrenal chromaffin cells^4,24. An electrophysiological method to depolarize the cell and thereby evoke exo- and endocytosis is described. Systematic confocal image processing provides information about different modes of pore behaviors for fusion and fission events.

Simultaneous monitoring of calcium current and capacitance changes at the same cell with the whole-cell configuration provides additional information about exo- and endocytosis, implying the ratio of pore opening and closure at a whole-cell level at any given time. These methods are, in principle, applicable to other excitable cells and nonelectrically excitable cells in primary culture or in cell lines containing vesicles of ~200-1,600 nm¹⁵. In addition to chromaffin cells, the method described here has been applied to a rat pancreatic beta-cell line, INS-1 cells⁴.

In principle, the method can also be applied to secretory cells containing vesicles smaller than 200 nm. However, as the vesicle size is decreased, the signal-to-noise-ratio may be too low for reliable detection. Super-resolution imaging instead of confocal imaging may become necessary. So far, the methods described here have not been applied to synapses containing ~40 nm synaptic vesicles.

Successful implementation of the method described here depends critically on several general steps, such as the culture cell quality, plasmid transfection efficiency, and electrophysiological recording success rate. First, healthy cells are vital for both patch-clamp recording and confocal imaging. It is not advised to use antibacterial or antifungal agents during culture because these chemicals may affect the excitability of chromaffin cells. Thus, diligent exercise of sterile technique and use of freshly prepared media are important. Although primary culture chromaffin cells can survive in dishes for at least one week²⁵, it is important for new users to use cells on days 2 to 3 for the experiments. As fibroblasts grow gradually, it becomes more difficult to establish the whole-cell patch-clamp configuration. Second, the electroporation efficiency of plasmids into chromaffin cells is approximately 20-30%. The electroporation efficiency needs to be optimized if it is too low or PH-mNG fluorescence is too dim. Third, cells were imaged at their bottoms with a 100x oil objective to obtain substantial events indicated by changes in three fluorescent probes: FFN511, PH-mNG, and A655. Although DIC provides a clearer view than bright field or naked-eye imaging when establishing the whole-cell configuration, any visualization technique that allows the user to establish the configuration is sufficient.

Appropriate laser power strength is vital for confocal imaging in live cells. As FFN511 and mNG may be bleached by high laser power, it is better to look for a good cell with epifluorescence before visualization with confocal lasers. For A655, high-power laser excitation is needed to bleach the dye inside the vesicles. In addition, the most common reason this experiment is unsuccessful is a failure of patch-clamp recording. A clean and polished pipette tip of the proper size and practice generating the whole-cell configuration on chromaffin cells are key factors. Although it may take some time and effort to succeed, once established, these experiments provide substantial data regarding both fusion and fission events, indicating membrane pore opening and closure at a single event level.

The protocol described here can be used with some modifications to further different experimental goals. However, regardless of experimental goals, it is advisable to first use high potassium solution (such as 70 mM KCl) as a stimulus to see the changes the membrane undergoes upon being depolarized, ensuring the intensity changes in these three dyes, FFN, PH, and A655, can be visualized. The plasma membrane can be labeled with other membrane binding fluorophores, such as mCLING (membrane-binding fluorophore-cysteine-lysine-palmitoyl group)^4,27 or CAAX (a protein motif that targets cytosol-faced plasma membrane via cysteine residue isoprenylation)^4,28. A655 can be replaced with other dyes with different spectra, such as Atto 532 (A532), depending on the combination of fluorescent molecules used in certain experiments¹⁵. Neuropeptide Y can be used instead of fluorescent false neurotransmitters¹⁶.

FFN511 was excited at 458 nm instead of 405 nm in this experiment even though the peak excitation spectrum of FFN511 is about 405 nm. Study with capacitance recording showed no significant difference in chromaffin cells with or without PH-mNG, indicating that overexpression of PH-mNG does not affect exocytosis and endocytosis^15,24. Similar percentages for fusion events were verified in chromaffin cells labeled with PH-mNG and A532 only or with FFN511, demonstrating that loading of FFN511 into chromaffin vesicles does not affect the content release and different fusion modes¹⁶. The 1 s depolarization stimulation can be replaced by high potassium solution or other pattens of depolarization. These modifications can be used to adapt to different experimental aims.

Despite the advantages of this powerful method, it has some limitations. The biggest limitation is related to the diffraction-limited resolution (~230 nm) of confocal microscopy, which can be overcome by more advanced super-resolution microscopy techniques²⁹. The granule diameter in bovine chromaffin cells is ~300 nm⁶, making it possible to observe membrane pore opening and closure within the spatial resolution of confocal microscopy. However, synaptic vesicles are ~30-60 nm³⁰, which is beyond confocal resolution. Extending these live-cell imaging measurements to small synaptic vesicles proves to be difficult with confocal imaging.

Live-cell imaging techniques with better temporal and/or spatial resolution, such as STED microscopy, stochastic optical reconstruction microscopy (STORM), photoactivated localization microscopy (PALM), total internal reflection fluorescence (TIRF) microscopy, and minimal photon fluxes (MINFLUX) nanoscopy have been developed and may be applied to this method^31,32,33,34. Indeed, STED microscopy has been used to reveal fusion pore dynamics in chromaffin cells. The different modes of fusion events and preformed Ω-profile closure can be further verified using super-resolution microscopy such as STED microscopy²⁴. Other limitations include photobleaching and cytotoxicity of fluorescent proteins and dyes.

This combination of electrophysiology and confocal microscopy can be applied widely to many secretory cells. The combination of super-resolution microscopy with the methods described here would be a valuable tool for measuring fusion pore dynamics in neural circuits in the future, provided that super-resolution microscopy will develop to such an extent that it can resolve the fusion pore at nerve terminals.

Materials

Name	Company	Catalog Number	Comments
Adenosine 5'-triphosphate magnesium salt	Sigma	A9187-500MG	ATP for preparing internal solution
Atto 655	ATTO-TEC GmbH	AD 655-21	Atto dye to label bath solution
Basic Nucleofector for Primary Neurons	Lonza	VSPI-1003	Electroporation transfection buffer along with kit
Boroscilicate capillary glass pipette	Warner Instruments	64-0795	Standard wall with filament OD=2.0 mm ID=1.16 mm Length=7.5 cm
Bovine serum albumin	Sigma	A2153-50G	Reagent for gland digestion
Calcium Chloride 2 M	Quality Biological	351-130-721	Reagent for preparing bath solution
Cell Strainers, 100 µm	Falcon	352360	Material for filtering chromaffin cell suspension
Cesium hydroxide solution	Sigma	232041	Reagent for preparing internal solution and Cs-glutamate/Cs-EGTA stock buffer
Collagenase P	Sigma	11213873001	Enzyme for gland digestion
Coverslip	Neuvitro	GG-14-Laminin	GG-14-Laminin, 14 mm dia.#1 thick 60 pieces Laminin coated German coverslips
D-(+)-Glucose	Sigma	G8270-1KG	Reagent for preparing Locke’s solution and bath solution
DMEM	ThermoFisher Scientific	11885092	Reagent for preparing culture medium
EGTA	Sigma	324626-25GM	Reagent for preparing Cs-EGTA stock buffer for bath solution
Electroporation and Nucleofector	Amaxa Biosystems	Nucleofector II	Transfect plasmids into cells
Fetal bovine serum	ThermoFisher Scientific	10082147	Reagent for preparing culture medium
FFN511	Abcam	ab120331	Fluorescent false neurotransmitter to label vesicles
Guanosine 5'-triphosphate sodium salt hydrate	Sigma	G8877-250MG	GTP for preparing internal solution
HEPES	Sigma	H3375-500G	Reagent for preparing Locke’s solution
Igor Pro	WaveMetrics	Igor pro	Software for patch-clamp analysis and imaging data presentation
Leica Application Suite X software	Leica	LAS X software	Confocal software for imaging data collection and analysis
Leica TCS SP5 Confocal Laser Scanning Microscope	Leica	Leica TCS SP5	Confocal microscope for imaging data collection
L-Glutamic acid	Sigma	49449-100G	Reagent for preparing Cs-glutamate stock buffer for bath solution
Lock-in amplifier	Heka	Lock-in	Software for capacitance recording
Magnesium Chloride 1 M	Quality Biological	351-033-721EA	Reagent for preparing internal solution and bath solution
Metallized Hemacytometer Hausser Bright-Line	Hausser Scientific	3120	Counting chamber
Microforge	Narishige	MF-830	Polish pipettes to enhance the formation and stability of giga-ohm seals
Millex-GP Syringe Filter Unit, 0.22 µm	Millipore	SLGPR33RB	Material for glands wash and digestion
mNG(mNeonGreen)	Allele Biotechnology	ABP-FP-MNEONSB	Template for PH-mNeonGreen construction
Nylon mesh filtering screen 100 micron	EIKO filtering co	03-100/32	Material for filtering medulla suspension
Patch clamp EPC-10	Heka	EPC-10	Amplifier for patch-clamp data collection
PH-EGFP	Addgene	Plasmid #51407	Backbone for PH-mNeonGreen construction
Pipette puller	Sutter Instrument	P-97	Make pipettes for patch-clamp recording
Potassium Chloride	Sigma	P5404-500G	Reagent for preparing Locke’s solution and bath solution
Pulse software	Heka	Pulse	Software for patch-clamp data collection
Recording chamber	Warner Instruments	64-1943/QR-40LP	coverslip chamber, apply patch-clamp pipette on live cells
Sodium chloride	Sigma	S7653-1KG	Reagent for preparing Locke’s solution, bath solution and internal solution
Sodium hydroxide	Sigma	S5881-500G	Reagent for preparing Locke’s solution
Sodium phosphate dibasic	Sigma	S0876-500G	Reagent for preparing Locke’s solution
Sodium phosphate monobasic	Sigma	S8282-500G	Reagent for preparing Locke’s solution
Stirring hot plate	Barnsted/Thermolyne	type 10100	Heater for pipette coating with wax
Syringe, 30 mL	Becton Dickinson	302832	Material for glands wash and digestion
Tetraethylammonium chloride	Sigma	T2265-100G	TEA for preparing bath solution
Trypsin inhibitor	Sigma	T9253-5G	Reagent for gland digestion
Type F Immersion liquid	Leica	195371-10-9	Leica confocal mounting oil