Examination of Synaptic Vesicle Recycling Using FM Dyes During Evoked, Spontaneous, and Miniature Synaptic Activities

Summary

We describe the use of styryl FM dyes to image synaptic vesicle recycling in functional nerve terminals. This protocol can be applied not only to evoked, but also spontaneous and miniature synaptic activities. The protocol expands the variety of synaptic events that can be effectively evaluated.

Abstract

Synaptic vesicles in functional nerve terminals undergo exocytosis and endocytosis. This synaptic vesicle recycling can be effectively analyzed using styryl FM dyes, which reveal membrane turnover. Conventional protocols for the use of FM dyes were designed for analyzing neurons following stimulated (evoked) synaptic activity. Recently, protocols have become available for analyzing the FM signals that accompany weaker synaptic activities, such as spontaneous or miniature synaptic events. Analysis of these small changes in FM signals requires that the imaging system is sufficiently sensitive to detect small changes in intensity, yet that artifactual changes of large amplitude are suppressed. Here we describe a protocol that can be applied to evoked, spontaneous, and miniature synaptic activities, and use cultured hippocampal neurons as an example. This protocol also incorporates a means of assessing the rate of photobleaching of FM dyes, as this is a significant source of artifacts when imaging small changes in intensity.

Introduction

The functionality of synaptic vesicles is an important determinant of synaptic transmission. These vesicles release neurotransmitters when they fuse with the presynaptic plasma membrane (exocytosis), and they become ready for another cycle of release after being regenerated from the plasma membrane (endocytosis) and reloaded with neurotransmitter. Research into the dynamics of and mechanisms underlying synaptic vesicle recycling has been greatly accelerated by the introduction of styryl FM dyes¹. These amphipathic molecules, which have positively charged hydrophilic head groups and hydrophobic tails (multiple dyes in Figure 1A, stereoview of FM1-43 in Figure 1B), can reversibly enter and exit lipid membranes without permeating them. Groups of FM dyes share similar features that influence the range of light that they emit. For example, FM2-10, FM1-43, and FM1-84 have one double bond between two cyclic compounds and show green emission. The difference between them is the length of the hydrophobic tail, which determines its hydrophobicity and therefore the rate of exit from the membrane (departitioning). In the cases of FM5-95 and FM4-64, three double bonds link the cyclic compounds, and they show red emission. These dyes differ with respect to their hydrophilic parts. In all FM dyes, the fluorescence intensity increases when they are inserted into biological membranes, due to an increase in quantum yield in the hydrophobic environment relative to the hydrophilic environment. Thus the changes in FM intensity represent the changes in membrane turnover. The different colors (emission spectra) and hydrophobicities make the FM dyes a versatile research tool in synaptic vesicle recycling.

Based on these features, the FM dyes are mostly used according to the following scheme when analyzing synaptic vesicle recycling (Figure 2). Neurons are bathed in an extracellular solution containing the FM dye, enabling it to be taken up into synaptic vesicles (SVs) as they form via endocytosis (staining). The dye is then washed out by applying a dye-free extracellular solution; this reveals the functional nerve terminals, i.e. only those actively undergoing endocytosis will contain a cluster of synaptic vesicles that are loaded with the dye (Figure 2 bottom). Subsequent exocytosis leads to loss of the FM dye to the extracellular space and a concomitant loss of fluorescence (destaining; due to both the departitioning to a hydrophilic environment and diffusion away from the site of exocytosis). Therefore the changes in FM fluorescence intensity are indicators of synaptic vesicle exo- and endocytosis.

FM dyes have been used to stain and destain the synaptic vesicles in various organisms and preparations^2,3. Examples include mammalian neuronal cultures^4-9, mammalian brain slices^10,11, neuromuscular junctions^12,13, retinal bipolar neurons^14,15, and hair cells of cochlea¹⁶.

Typically in such experiments, both staining and destaining are triggered by extensively stimulating the neurons (evoked activity). Recently, however, synaptic vesicle recycling in response to weak stimulation has also been analyzed, as has recycling in the absence of an external stimulus (spontaneous and miniature synaptic activity)^9,17-19. Spontaneous and miniature synaptic activities are defined as those that occur in the absence of external stimuli, with the former involving the spontaneous firing of action potentials (Figure 3). These weak synaptic activities are associated with smaller changes in FM signals than those triggered by extensive stimulation. The measurement requires that the changes in FM fluorescence intensity accurately reflect synaptic vesicle exocytosis or endocytosis but not artifactual changes in intensity. One cause of the artifact is the presence of nonspecific staining of the plasma membrane by FM dyes. Gradual washout of this component will lead to a gradual change in the measured fluorescence intensity, which will be erroneously ascribed to synaptic activities. This factor can be reduced by appropriate methods (see Protocol). The most notable cause of the artifact is the photobleaching of FM dye retained within synaptic vesicles. The photobleaching-related changes in FM intensity must be small in comparison to the biological (synaptic) changes that are measured. The recent development of sensitive cameras, e.g. the electron-multiplying charge-coupled device (EMCCD) camera, makes it possible to minimize photobleaching by shortening exposure time and weakening the intensity of the light used to excite the fluorophore. Another cause of the artifact is a drift in the focusing level of light microscope. The focus drift during an imaging session can be caused by mechanical or thermal effects, and will erroneously lead to a change in the measured fluorescence intensity.

Here we describe protocols and equipment that make it possible to use FM dyes to analyze synaptic vesicle recycling even in the context of weak or no stimulation, in particular, the miniature synaptic activity. We show examples of the staining and destaining of vesicles during evoked and spontaneous synaptic events, using cultured rodent hippocampal neurons, and imaging the destaining phase. We also demonstrate how to evaluate the degree of FM dye photobleaching, in the absence of any FM dye loss due to synaptic activities.

Protocol

1. Primary Culture of Neurons from the Mammalian Brain

All animal procedures performed in this study are approved by the Institutional Animal Care and Use Committee of the University of Iowa.

1. Prepare the dissociated cell culture of the CA3-CA1 regions of hippocampus from mice or rats on postnatal days 0-1^19,20. Plate the hippocampal cells on 12-mm coverslips (thickness number 0) preseeded with the rat glial feeder layer, in 24-well dishes, and at a density of 12,000 cells/well.
2. Culture the hippocampal neurons for at least 8 days for functional nerve terminals to develop²¹. We use neurons on 11-14 days in vitro when individual nerve terminals are relatively isolated without much clustering. Neurons can be used for experiments up to about 21-28 days in vitro e.g. Willig et al.²² and Nosyreva and Kavalali²³.
3. For functional studies, assess coverslips (before staining) for indicators of healthy neurons using transmitted light microscopy (e.g. phase contrast optics at a magnification of 10-20X). The indicators of good health include: a uniform glial cell layer, a clear cellular margin around the neurons, extended dendrites without beaded structures, a lack of clustered somata, and a lack of bundled neurites.

2. Loading the Synaptic Vesicles with the FM Dye (Staining)

1. Staining due to synaptic activity evoked by electrical field stimulation (involves action potentials)
   1. Prepare staining solution 1 by adding the FM dye to a HEPES-based dye-free solution, e.g. Tyrode's solution (see SOLUTIONS for details). The concentration of FM dye is 10 μM FM1-43 and 2.5 μM FM4-64. Typical concentration ranges are: 2-15 μM FM1-43, or 2.5-20 μM FM4-64^3,6. For experiments requiring suppression of recurrent glutamatergic synaptic activity and the induction of synaptic plasticity, further add antagonists of ionotropic glutamate receptors: AMPA receptors (e.g. 10 μM CNQX) and NMDA receptors (e.g. 50 μM AP5).
   2. Transfer a coverslip with cells from the culture medium to the imaging chamber prefilled with plain Tyrode's solution. We use an imaging chamber with the stimulation electrodes attached to the side walls (platinum wires separated by 6.3 mm, RC-21BRFS). This chamber is designed as a closed imaging chamber, but we use it as an open chamber (i.e. without a top coverslip), with a bath volume of ~200 μl. During transfer, do not expose the cultured cells to air, and avoid pinching the cultured cells with the tweezers (pick the coverslip up from the perimeter rather than near the center).
   3. Perfuse the imaging chamber with plain Tyrode's solution at room temperature (23-25 °C).
   4. Apply staining solution 1 by constant perfusion.
   5. Evoke synaptic activity by applying electrical pulses. For maximal staining of the total recycling pool of synaptic vesicles in hippocampal neurons, stimulate the cultures at 10 Hz for 60-120 sec ²⁴. When trying to attain maximal efficiency of field stimulation, keep several features in mind. 1) The height of the bath solution should be as low as possible, while keeping the neurons and stimulating electrodes entirely immersed in a thin layer of Tyrode's solution (in order to keep the neurons alive and to obtain a homogeneous electrical field). 2) Current intensity and duration should be optimized using an indicator of neuronal excitability, e.g. fluorescent Ca²⁺ imaging; in our case, a constant current of 30 mA intensity and 1 msec duration gives reliable results. 3) The bath solution should be applied at constant flow during stimulation; the electrical stimulation leads to electrolysis, causing protons (H⁺) to accumulate at the positively charged anode and the pH near the anode to decrease rapidly in a static bath. Check the acidification in a separate experiment by passing more than a few hundred pulses through a solution that contains a pH-indicator (e.g. Hanks’ balanced salt solution with phenol red).
   6. Leave the neurons in staining solution 1 for an additional 60 sec after field stimulation is terminated, so that post-stimulus endocytosis will be completed²⁵.
   7. Wash the staining solution out of the imaging chamber by perfusing it with rinsing solution 1, a dye-free Tyrode's solution plus antagonists of AMPA receptors and NMDA receptors (see section 2.1.1), and a blocker of voltage-dependent Na⁺ channels (e.g. 0.5-1 μM tetrodotoxin, TTX). The combination of blockers will suppress the loss of FM dye through spontaneous synaptic activity. To maximize the washing efficiency, increase the perfusion rate (e.g. up to 5-6 ml/min) for 1-2 min, and add a bolus (e.g. 1 ml) of rinsing solution 1 directly to the open imaging chamber manually, using a pipette. Control the strength and direction of manual infusion, so as not to disturb the cultured neurons.
2. Staining due to synaptic activity evoked by high-K⁺ solution (does not involve action potentials).
   1. Prepare staining solution 2 by adding the FM dye to high-K⁺ Tyrode's solution. Specifically, prepare a modified Tyrode's solution with 45 mM KCl, by equimolar substitution of KCl for NaCl. If required, add antagonists of AMPA and NMDA receptors. This high-K⁺ solution will depolarize neurons continuously, enable Ca²⁺ influx into nerve terminals, and initiate synaptic vesicle exocytosis to the maximal extent²⁶. Higher concentrations (70-90 mM) were also used in other reports e.g. Klingauf et al.²⁷ and Richards et al.²⁸
   2. Transfer a coverslip with cultured neurons from the culture medium to the imaging chamber prefilled with plain Tyrode's solution.
   3. Stain the neurons at room temperature for 1-2 min by applying staining solution 2 at constant perfusion or using a pipette^26,28. In the latter case, the final dye concentration should be controlled by adding a known volume of staining solution 2 to a known volume of dye-free solution.
   4. Wash the staining solution out of the imaging chamber, as described in section 2.1.7.
3. Staining due to spontaneous and miniature synaptic activity.­
   1. Prewarm bicarbonate-based solution (e.g. Minimum Essential Medium, MEM) to 37 °C by placing in the culture incubator for more than 60 min.
   2. For staining based on spontaneous synaptic activity, prepare staining solution 3 by adding the FM dye to MEM. For staining based on miniature synaptic activity, prepare staining solution 4 by adding TTX to the staining solution 3.
   3. Stain neurons by transferring a coverslip with cultured neurons from culture medium to staining solution 3 or 4, and leaving them in the same staining solution at 37 °C for 10 min.
   4. Rinse the coverslip briefly in rinsing solution 1.
   5. Transfer the coverslip to rinsing solution 1 in an imaging chamber. Keep in mind the following. 1) Cells should not be exposed to air; if necessary, transfer the coverslip directly to the imaging chamber without rinsing. 2) Neurons can be stained at room temperature by replacing staining solution 3 or 4 with a HEPES-based solution. 3) Neurons can be stained for evoked synaptic activity in the context of high-K⁺ solution using this method, by replacing staining solution 3 or 4 with staining solution 2 and carrying out the remaining steps at room temperature.
   6. Wash the staining solution out of the imaging chamber as described in section 2.1.7.

3. Washing out the FM Dye

1. After neuronal staining and initial washing by any of the above-described methods (section 2.1.7), continue to perfuse the imaging chamber with rinsing solution 1 for 5-10 min at room temperature. This will remove the FM dye from the plasma membrane and the extracellular solution. In the case of our imaging chamber, the bath perfusion rate is 600-1,200 μl/min. Washing can also be carried out in the absence of extracellular Ca²⁺ to suppress FM dye loss due to synaptic activities¹⁷. The washes can be enhanced by a brief application of Advasep-7, a modified cyclodextrin that serves as a scavenger of FM dyes from the plasma membrane^29-31. It is crucial to keep any Advasep-7 exposure short (e.g. 5 sec in the case of monolayer culture) to avoid both reducing the intensity of FM puncta³¹ and compromising the health of the cells. The fluorescence from FM1-43 in the plasma membrane can also be suppressed by an application of a quencher sulforhodamine 101 (50-100 μM) that does not enter synaptic vesicles³².

4. Searching for an Optimal Image Field During Washing

1. Verify that the cells in the field to be imaged are healthy, using transmitted light microscopy with a high-magnification and high-numerical-aperture objective lens (e.g. 40X) . From this section, we continue to use an inverted microscope (Eclipse Ti, Nikon) equipped with phase contrast or differential interference contrast optics. This microscope is also equipped with Perfect Focus System (PFS) that enables continuous, real-time focus correction, which overcomes the microscope focus drift. This feature is essential for time-lapse imaging during destaining with the same focusing plane.
2. Verify that staining was effective, by using fluorescence optics. Avoid aggregates of FM puncta unless they are the target of research, because individual boutons will be difficult to discern. Pay attention to the shapes of the FM puncta: if most of the stained boutons appear as strings of circular beads, they could represent unhealthy nerve terminals. Minimize the exposure of the neurons to strong fluorescence excitation in order to avoid photobleaching.

5. Unloading FM Dye from the Synaptic Vesicles (Destaining)

1. Destaining due to synaptic activity evoked by electrical field stimulation.
   1. Wash TTX out by perfusing the neurons extensively with a rinsing solution 2 (same as rinsing solution 1 but without TTX). Confirm the effectiveness of TTX washout in separate experiments, using the same washing parameters (perfusion rate and duration), e.g. by fluorescent Ca²⁺ imaging of cytoplasmic Ca²⁺ transients induced by field stimulation, or by patch-clamp recording of voltage-dependent Na⁺ currents.
   2. Start imaging the FM dyes. For imaging FM1-43 or FM4-64, use 520 nm or 650 nm long pass emission filters, respectively and a 490 nm excitation filter. Acquire images every 1-2 sec, using a short exposure time (e.g. 10-20 msec), weak excitation intensity (e.g. 5-10% of LED power), and high sensitivity (e.g. with EM gain). Minimize photobleaching of the FM dyes by reducing exposure time and the intensity of fluorescence excitation. Minimize the microscope focus drift by turning on the Perfect Focus System.
   3. Stimulate the neurons in rinsing solution 2, using field-stimulation (e.g. at 10 Hz for 120 sec)²⁰.
   4. Apply multiple rounds of stimulation, with intervening rest periods of 1-2 min, to deplete all the recycling synaptic vesicles of FM dye. This is important when assessing the size of total recycling pool of synaptic vesicles that were stained with FM dyes in the prior state^20,33.
2. Destaining due to synaptic activities evoked in the absence of action potentials.
   1. Start imaging the FM dye.
   2. Apply the stimulating reagents dissolved in rinsing solution 1 or 2. Such reagents include: KCl (high-K⁺ solution, e.g. 45 mM)²⁶, ionomycin (a Ca²⁺ ionophore that raises the intracellular Ca²⁺ concentration by allowing Ca²⁺ influx into the cell from the extracellular solution, used e.g. at 5 μM)^20,34, and sucrose (a hypertonic solution that stimulates the release of synaptic vesicles from readily releasable pool, used e.g. at 500 mM)^35,36. Use a fast, local perfusion system for a reliable temporal control in applying the reagents, e.g. the SF-77B system from of Warner Instruments³⁷, or Y-tube system^38-40.
3. Destaining due to spontaneous and miniature synaptic activity.
   1. For destaining due to spontaneous synaptic activity, wash out TTX as above (section 5.1.1), using rinsing solution 2. Start imaging the FM dyes while perfusing neurons with rinsing solution 2. For destaining due to miniature synaptic activities, start imaging the FM dyes, while continuing to perfuse the neurons with rinsing solution 1 (with TTX).
   2. After destaining based on either spontaneous or miniature activity, identify the functional nerve terminals by destaining via evoked synaptic activity. Activity can be evoked by any of the procedures described above. This identification process is necessary because the stained structures can be those other than the functional nerve terminals (e.g. slowly recycling endosomes or astrocytic vesicles) and a large amount of destaining by the evoked activity aids in this process.

6. Assessing the Photobleaching Rate of FM Dyes

1. Stain the nerve terminals with aldehyde-fixable FM dye (e.g. FM1-43FX, FM4-64FX). This can done using the evoked, spontaneous, or miniature staining method described above, and by replacing FM dye with a fixable FM dye.
2. Wash out the fixable FM dye as described in section 2.1.7.
3. Chemically fix the neurons by transferring the coverslip to fixative solution: 4% paraformaldehyde and 4% sucrose in Tyrode's solution for 30 min at 4 °C. The fluorescence intensity of the fixable FM dye will be retained after chemical fixation.
4. Wash out the fixative for 10 min in Tyrode's solution, by transferring the coverslip to fresh Tyrode's solution twice.
5. Transfer the coverslip into fresh Tyrode's solution in the imaging chamber.
6. Start imaging the fixable FM dye using the same set of imaging parameters as for live neurons. It is important to rinse the imaging chamber very well after using any chemically fixed specimen, because any remaining fixative can affect subsequent live-cell imaging experiments carried out in the same chamber. Alternatively to the fixed cell imaging, the stained live cells might be imaged for assessing the photobleaching rate. However, it is worth keeping in mind that it is difficult to block the miniature synaptic activity acutely; for example, the frequency of miniature activity can be reduced to ~50% by removing the extracellular Ca^{2+ 23}, but it is not a complete blockade.  
7. At the end of each imaging session, acquire a background image for assessing the absolute value of the FM fluorescence intensity. The intensity of FM puncta represents not only the true signal from the FM dyes in nerve terminals, but also the background noise from 1) glial cells that underlie the neuronal monolayer culture, 2) the coverslip and optical components (e.g. objective lens, filters and mirrors), and 3) the detector (EMCCD camera). Background noise is assessed in nonstained regions of the imaged field. When such regions are limited in space or heterogeneous in intensity, substitute it with an image with the camera shutter closed and acquire noise from the detector, although it is not an ideal method. Acquire approximately 10 images using the same imaging parameters.

7. Image Analysis

1. Import the acquired data into the image-analysis software as a stack of time-series images. We use ImageJ (W. S. Rasband, NIH) and the associated plug-ins, e.g. Image Stabilizer plug-in (Kang Li) for correcting a small degree of movement among FM puncta, and Time Series Analyzer V2.0 (Balaji Jayaprakash) for analyzing the time series.
2. Calculate the changes in FM intensity at the start and end of the series (ΔFM). To this end, average the images before the start of destaining and after the end of destaining (e.g. 5 image frames each), and generate a difference image by subtracting them.
3. Identify the potentially functional nerve terminals by applying an intensity threshold to the difference image, and detect pixels whose intensities are higher than the threshold. One method of setting the threshold is to choose the standard deviation of the background intensity obtained from the bare coverslip area.
4. Identify the isolated, functional nerve terminals as the contiguous pixels that were detected and that satisfy a size criterion (minimal and maximal numbers of pixels in contiguity). This process eliminates background noise and aggregated nerve terminals from analysis.
5. Assign regions-of-interest (ROIs) on the detected pixel clusters (with individual clusters corresponding to individual nerve terminals). The total change in intensity during destaining (ΔFM) is a typical parameter of analysis. This represents the cumulative amount of evoked, spontaneous or miniature synaptic activity. Exclude ROIs if they exhibit any of the following changes in intensity depending on purposes of experiments: an increase during recording, a sudden decrease before stimulation, or a long latency after stimulation.
6. Assess the rate of photobleaching by importing the acquired time-series data into the image-analysis software. Average the background images (e.g. with a closed shutter), and subtract it from the individual images. Assign ROIs to FM puncta that putatively correspond to nerve terminals, and measure changes in ROI intensity over time.

Solutions

1. Tyrode's solution
   • Composition (in mM): 125 NaCl, 2 KCl, 2 CaCl₂, 2 MgCl₂, 30 glucose, 25 HEPES, 310 mOsm, pH 7.4.
2. Staining solution 1
   • Tyrode's solution plus the FM dye.
   • This solution is used for the staining, based on evoked synaptic activity by electrical stimulation.
   • Add the antagonists of AMPA receptors and NMDA receptors if necessary.
3. Staining solution 2
   • High-K⁺ Tyrode's solution plus the FM dye.
   • This modified Tyrode's solution is prepared by increasing the KCl concentration to 45 mM, by equimolar substitution of KCl for NaCl.
   • This solution is used for the staining, based on evoked synaptic activity by continuous depolarization.
   • Add the antagonists of AMPA receptors and NMDA receptors if necessary.
4. Staining solution 3
   • MEM solution plus the FM dye.
   • This solution is used for the staining, based on spontaneous synaptic activity.
5. Staining solution 4
   • MEM solution plus the FM dye and TTX.
   • This solution is used for the staining, based on miniature synaptic activity.
6. Rinsing solution 1
   • Tyrode's solution plus the antagonists of AMPA receptors and NMDA receptors, and TTX.
7. Rinsing solution 2
   • Tyrode's solution plus the antagonists of AMPA receptors and NMDA receptors, but without TTX.

Results

As an example, we show representative results for the destaining time course of synaptic vesicles (Figure 4). Cultured hippocampal neurons were stained with FM4-64 using the spontaneous synaptic activity (step 2.3) and washed with dye-free solution (rinsing solution 2). The imaging shows the initial destaining time course using spontaneous activity (step 5.3) (initial part of continuous line, Figure 4A). This is followed by the destaining time course using three rounds of evoked activity...

Discussion

We have described protocols for staining and destaining synaptic vesicles in response to evoked, spontaneous and miniature synaptic activity, and for imaging during the destaining phase. In addition to the existing protocols, we have included a new protocol of observing the FM destaining based on miniature synaptic activity. Using these protocols, we previously identified abnormalities in cultured neurons from a mouse model of the movement disorder dystonia. In comparison to their counterparts in wild-type mice, those ne...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Pulse generator	AMPI	Master-8
Isolated stimulator	Digitimer	DS3
Inverted microscope	Nikon	Eclipse TS100	This is used for assessing the cell morphology at low magnification.
Inverted microscope	Nikon	Eclipse TiE	This is used for high-resolution fluorescence and transmitted light imaging, with minimal focus drift.
Objective lens	Nikon		Water-immersion lens is recommended. Oil-immersion lens is usable unless an imaged structure is deep from the coverslip surface (e.g. >10 μm).
Filter cube	Nikon	77032509	490/20-nm ex, 510-nm dclp, 520-nm-LP em for FM1-43
Filter cube	Nikon	77032809	490/420 nm ex, 510 nm dclp, 650 nm LP em for FM4-64.
EMCCD camera	Andor Technology	iXon EM+ DU-860	This EMCCD camera is used for high-sensitivity detection of fluorescence.
Liquid recirculating chiller	Solid State Cooling Systems	Oasis 160	This is used for continuously perfusing the camera with chilled water for maintaining a temperature of -80°C, and thereby reducing noise.
LED	CoolLED-Custom Interconnect	490 nm	This light source is used for rapid on/off control of fluorescence excitation.
Image acquisition software	Andor Technology	Solis
Imaging chamber	Warner Instruments	RC-21BRFS
Fast perfusion system	Warner Instruments	SF-77B
CNQX	Tocris Bioscience	1045
D,L-AP5	Tocris Bioscience	0106
Tetrodotoxin	Tocris Bioscience	1069	Caution: toxic reagent. Handle with care.
FM1-43	Invitrogen	T35356
Aldehyde-fixable FM1-43 (FM1-43FX)	Invitrogen	F35355
FM4-64	Invitrogen	T13320
Aldehyde-fixable FM4-64 (FM4-64FX)	Invitrogen	F34653
Ionomycin	Sigma-Aldrich	I0634
Hanks’ balanced salt	Sigma-Aldrich	H2387
Minimum Essential Medium	Invitrogen	51200-038	This solution does not contain phenol red that will interfere with fluorescence imaging.
Paraformaldehyde	Electron Microscopy Sciences	15710	Caution: toxic reagent. Handle with care.
Sucrose	Sigma-Aldrich	S7903