A Galvanotaxis Assay for Analysis of Neural Precursor Cell Migration Kinetics in an Externally Applied Direct Current Electric Field

All procedures involving animal handling were approved by the University of Toronto Animal Care Committee in accordance with institutional guidelines (protocol no. 20009387). The following methods should be performed using sterile tools and techniques, in a laminar flow hood where applicable.

In the protocol text below, the phrase "EFH-SFM" refers to serum free media supplemented with epidermal growth factor, basic fibroblast growth factor and heparin. EFH-SFM is used when investigating the galvanotaxis of undifferentiated NPCs because these mitogens maintain NPCs in their undifferentiated state⁴. When investigating the galvanotaxis of NPCs induced to differentiate into mature cell types, "FBS-SFM" refers to serum free media supplemented with 1% fetal bovine serum. FBS promotes the differentiation of NPCs into mature neural phenotypes⁵.

1. Isolation and Culture of Neural Precursors (Not shown in video)

1. Anaesthetize a CD1 mouse (6-8 weeks old) with isofluorane and sacrifice via cervical dislocation.
2. Douse the head in 70% ethanol and decapitate the animal with sharp dissection scissors.
3. While holding the head with surgical forceps, remove the skin on the dorsal surface to expose the skull.
4. Using a scalpel and no. 11 blade, score the skull at the frontal sinus along the mediolateral axis, and also along the sagittal suture in the rostrocaudal direction.
5. Peel the parietal bones away from the head with no. 7 curved forceps, taking care not to pierce the brain tissue.
6. Insert a thin spatula underneath the brain starting from beneath the cerebellum and advancing toward the olfactory bulbs. While holding the skull in place with forceps, gently pull the brain from the skull and immediately place it in ice-cold artificial cerebrospinal fluid (see recipes below).
7. Under a dissection microscope, using sterile dissection scissors and forceps cut the brain in half along the midline. Rotate each hemisphere so that the medial (cut) surface faces upwards.
8. Select a hemisphere, and with the medial surface facing upward, locate the splenium of the corpus callosum (posterior region of the corpus callosum).
9. Make an incision from the surface of the cortex to the splenium of the corpus callosum along the dorsoventral axis.
10. Peel the incised cortex toward the olfactory bulb to expose the medial and lateral walls of the lateral ventricle.
11. Rotate the hemisphere so that the dorsal surface faces upwards, and use curved microscissors to cut out and collect the exposed medial and lateral walls, which contain the periventricular region where NPCs reside⁶.
12. Repeat steps 1.8-1.11 for the other hemisphere.
13. Pipette the isolated tissue into 7 ml of trypsin solution (see recipe below) in a 15 cc tube, and place the tube on a rocker in a 37 °C incubator for 25 min.
14. Centrifuge the tube at 1,500 rpm for 5 min, aspirate the supernatant, and resuspend the tissue in 2 ml of trypsin inhibitor solution (see recipe below).
15. Gently triturate the tissue with a small borehole Pasteur pipette 30-50 times carefully to avoid air bubbles.
16. Centrifuge the tube at 1,500 rpm for 5 min, aspirate the supernatant, and resuspend in 1-2 ml of SFM (see recipe below) by triturating the pellet 3-5 times.
17. Centrifuge the tube at 1,500 rpm for 3 min, aspirate the supernatant and resuspend in 1 ml of SFM + EFH.
18. Count live cell density with a haemocytometer and plate the cells in a T25 culture flask at a density of 10 cells per μl in SFM + EFH.
19. Allow the culture to grow for 7 days undisturbed to yield free-floating primary neurospheres comprised of NPCs.

2. Galvanotaxis Chamber Preparation

1. Place 3 square glass no. 1 cover slips (22 x 22 x 0.17 mm) in a bottle of 6N hydrochloric acid overnight.
2. The next day, use a diamond-tip glass-cutter to cut 6 rectangular strips (22 x 5 x 0.17 mm) of glass from square no. 1 cover slips.
3. Transfer the acid-washed square slips and rectangular slips into a laminar flow hood. Wash the rectangular and square strips first with 70% ethanol, then with tissue culture-grade autoclaved water, and allow to dry on a Kim Wipe (for added sterility, the glass may be allowed to air dry).
4. Apply vacuum grease to the perimeter of one surface of the square glass slides, and seal them to the base of 60 mm plastic Petri dishes.
5. Apply vacuum grease along the long axis of one surface of the rectangular glass strips, and seal them to opposite edges of the square glass slides (such that they are parallel to each other) in order to create a central trough.
6. UV-sterilize the chambers for at least 15 min in the laminar flow hood.
7. Pipette 250-300 μl of poly-L-lysine onto the central trough of the chambers and incubate at room temperature for 2 hr.
8. Approximately 15 min prior to the end of the incubation period, prepare the Matrigel solution (see recipe below).
9. Aspirate the poly-L-lysine, wash the central troughs with 1 ml of autoclaved water, and pipette 250-300 μl of Matrigel solution onto the central troughs.
10. Incubate the chambers at 37 °C for 1 hr.
11. Aspirate the Matrigel solution and gently wash the central troughs with 1-2 ml of SFM.
12. Pipette 100 μl of EFH-SFM or FBS-SFM onto the central troughs and transfer the galvanotaxis chambers onto the stage of a counting microscope.
13. Pipette 3-4 ml of the neurosphere-containing culture into a 60 mm Petri dish and transfer the Petri dish to the stage of the counting microscope.
14. At a viewing objective of 5x, use a P10 pipette to transfer 5-8 whole neurospheres (up to four at a time) onto the central trough of each galvanotaxis chamber without dissociating them, and carefully spread the neurospheres around the central trough without disrupting the Matrigel substrate.
15. Add an additional 150-200 μl of EFH-SFM or FBS-SFM onto the central troughs.
16. Transfer the galvanotaxis chambers into a 37 °C, 5% CO₂, 100% humidified incubator for 17-20 hr (if analyzing undifferentiated NPCs) to allow the neurospheres to adhere to the Matrigel substrate and dissociate into single cells as shown in Figure 1. If analyzing differentiated NPCs, the incubation period should be extended to 69-72 hr to allow the differentiation of the cells.

3. Live Cell Time-Lapse Imaging

1. Allow the live cell imaging system to equilibrate at 37 °C, 5% CO₂ for a minimum of 30 min prior to initiation of the time-lapse recording.
2. Cut two 12 cm pieces of 1 mm diameter Silver wire, coil them from one end, and place them in Clorox bleach for 20 min to form Ag/AgCl electrodes.
3. Transfer the galvanotaxis chambers onto the stage of a counting microscope and select which chamber will be used for live-cell imaging migration analysis based on the following criteria: i) the neurospheres should be almost completely dissociated into single cells and ii) the cells should possess round morphologies with little-to-no processes extending from the cell bodies.
4. Transfer the selected galvanotaxis chamber into a laminar flow hood, along with a separate square no. 1 glass cover slip and vacuum grease.
5. Wash the cover slip first with 70% ethanol, then with autoclaved water, and apply a strip of vacuum grease on two parallel edges of the cover slip.
6. Aspirate the culture media from the central trough of the chamber, then quickly place the cover slip (grease-side facing down) onto the chamber such that the grease strips rest on the two parallel rectangular glass strips, effectively creating a roof to the chamber.
7. Pipette 100 μl of fresh EFH-SFM or FBS-SFM into the central trough via capillary action.
8. Use vacuum grease to create borders for pools of culture media on each end of the central trough, as shown in Figure 2.
9. Cut two 15 cm pieces of PVC tubing, and use a 10 cc syringe with an 18 gauge needle to carefully inject agarose solution into the tubing, ensuring no bubbles form in the tubes, and allow the gel to solidify for 5 min.
10. Transfer the galvanotaxis chamber to the live cell imaging system, along with the agarose gel tubes, Ag/AgCl electrodes, and a pair of empty 60 mm Petri dishes that will be used as culture media reservoirs and will contain the Ag/AgCl electrodes. Allow the galvanotaxis chamber to rest within the 37 °C, 5% CO₂ environment for 20-30 min.
11. During this time, prepare the lids of the 2 empty Petri dishes and the lid of the galvanotaxis chamber's Petri dish by drilling holes into them with a Dremel or similar tool as shown in Figure 3.
12. Pipette 1-1.5 ml of EFH-SFM or FBS-SFM onto either side of the central trough, and 7-8 ml of SFM into each empty Petri dish. Place one Petri dish on each side of the galvanotaxis chamber's central trough and place one Ag/AgCl electrode into each dish. Bridge the gap between the galvanotaxis chamber and the Petri dishes to establish electrical continuity with the agarose gel bridges, as shown in Figure 4.
13. Connect the Ag/AgCl electrodes to an external power supply, with an ammeter in series to measure electrical current, and turn on the power supply. Use a voltmeter to measure the strength of the electric field directly across the central trough, and adjust the output of the power supply until the desired electric field strength is achieved (the assays performed in this lab utilize a dcEF strength of 250 mV/mm with electrical current between 1 and 1.5 mA).
14. Initiate the time-lapse module on the live cell imaging system, and allow the experiment to run for the desired amount of time. After completion of the assay, fix the cells in 4% paraformaldehyde for standard immunostaining analysis.

Kinematic analysis reveals that in the presence of a 250 mV/mm dcEF, undifferentiated NPCs exhibit highly directed and rapid galvanotaxis toward the cathode (Figure 5A, Movie 1). In the absence of a dcEF, random movement of the cells is observed (Figure 5B, Movie 2). At this field strength, > 98% of undifferentiated NPCs migrate for the entire 6-8 hr for which they are imaged, and since dead cells do not migrate this suggests that they remain viable during this period in the absence or presence of a dcEF.

Differentiated phenotypes undergo negligible migration both in the presence and the absence of a dcEF (Movies 3, 4). Subsets of differentiated cells extend processes that tend to align perpendicular to the direction of the dcEF, but no noticeable cell body translocation is observed.

Immunostaining verifies that NPCs maintain positive expression of the neural precursor marker nestin after 6 hr of dcEF exposure (Figure 6A). In the assays involving NPCs induced to differentiate into mature phenotypes, the majority of cells express the mature astrocyte marker glial fibrillary acidic protein (GFAP) after 6 hr of dcEF exposure (Figure 6B).

The primary antibodies used in these analyses were as follows: mouse monoclonal anti-nestin (1:400, Millipore, Canada), and rabbit polyclonal anti-GFAP (1:500, Sigma, Canada). The secondary antibodies used in these analyses were as follows: goat-anti-mouse conjugated with Alexafluor 568 (1:400, Invitrogen-Gibco, Canada), and goat-anti-rabbit conjugated with Alexafluor 488 (1:400, Invitrogen-Gibco, Canada).

Figure 1. Neurospheres adhere to the Matrigel substrate and dissociate into single cells following 17 hr of incubation at 37 °C/5% CO₂, 100% humidity in EFH-SFM.

Figure 2. Illustration of galvanotaxis chamber. Strips of vacuum grease are used to create a pool of culture media on either side of the central trough.

Figure 3. Schematic of holes that should be drilled into the lids the galvanotaxis chamber and the culture medium reservoirs. The 7 mm diameter holes in both lids are used to insert the agarose gel tubes. The 4 mm diameter holes in the lid of the galvanotaxis chamber is used to measure the electric potential directly across the central trough using a voltmeter. The 4 mm diameter hole in the lid of the culture medium reservoir is to allow the Ag/AgCl electrode to protrude from the lid of the dish for connection to the power supply.

Figure 4. Illustration of galvanotaxis chamber assembly for time-lapse imaging. The galvanotaxis chamber is the central Petri dish, in which the cells are plated. The media inside the central Petri dish housing the galvanotaxis chamber is either SFM + EFH or SFM + 1% FBS. The Petri dishes on either side of the galvanotaxis chamber are filled with SFM, and also contain the Ag/AgCl electrodes. These electrodes are connected to an external power supply, and bridging the three Petri dishes with agarose-gel bridges forms a complete circuit.

Figure 5. In the presence of a dcEF undifferentiated NPCs undergo rapid galvanotactic migration toward the cathode (A), whereas in the absence of a dcEF the cells undergo random radial migration (B).

Figure 6. Immunostaining verifies that NPCs remain nestin-positive undifferentiated precursors after 6 hr of dcEF exposure (A), and that NPCs induced to differentiate mostly express the mature astrocyte marker GFAP after 6 hr of dcEF exposure (B).

Movie 1. Time-lapse video of undifferentiated NPCs exposed to a dcEF. NPCs plated onto galvanotaxis chambers for 17 hr in SFM + EFH, and then exposed to a 250 mV/mm dcEF exhibit rapid and directed migration toward the cathode. 1 second of video = 15 minutes real time. Click here to view movie.

Movie 2. Time-lapse video of undifferentiated NPCs in the absence of a dcEF. NPCs plated onto galvanotaxis chambers for 17 hr in SFM + EFH, and then imaged in the absence of an applied dcEF undergo random migration. 1 second of video = 15 minutes real time. Click here to view movie.

Movie 3. Time-lapse video of differentiated NPCs exposed to a dcEF. NPCs plated onto galvanotaxis chambers for 69-72 hr in SFM + FBS, and then exposed to a 250 mV/mm dcEF exhibit very little migration in any direction. 1 second of video = 15 minutes real time. Click here to view movie.

Movie 4. Time-lapse video of differentiated NPCs in the absence to a dcEF. NPCs plated onto galvanotaxis chambers for 69-72 hr in SFM + FBS, and then imaged in the absence of an applied dcEF exhibit very little migration in any direction, similar to differentiated NPCs that are exposed to a dcEF. 1 second of video = 15 minutes real time. Click here to view movie.

This protocol has been adapted from the well-established methods of previous studies^7-9. Galvanotactic chambers can be constructed using a variety of different techniques, including the construction of a separate glass well for confinement of cell seeding, or using CO₂ laser ablation for microfabrication of the central trough^10,11. Some techniques may be more laborious or costly than others. We have described a simple and cost-effective assay for constructing a NPC galvanotaxis chamber using materials commonly found in most cell biology laboratories. Our protocol includes a heated, humidified, CO₂ regulated incubator that surrounds the live cell imaging system to maintain the cells under optimal conditions for continuous long-term imaging. The lack of such apparatus is a limitation of some previously published techniques^9,12.

Recent work by another group demonstrated similar galvanotactic behavior of cells from a hippocampal cell line¹³, using a different protocol and a more labor-intensive galvanotaxis chamber design. Our assay is particularly elegant in that it permits an analysis of the identical starting population of cells - primary cell cultures of undifferentiated neurosphere derived NPCs - that have been exposed to distinct conditions thereby permitting a comparison of the migratory properties of both undifferentiated and differentiated cells, simply by modifying the factors used to supplement the culture media within the galvanotaxis chamber.

The assay presented in this protocol is a powerful tool for long-term tracking and analysis of NPC galvanotaxis, and possibly other cell types that undergo galvanotactic migration^14,15. As with any protocol, nuances in the preparation and execution of certain steps in this protocol may lead to unsuccessful results. The following guidelines and suggestions should assist in the successful execution of this protocol:

• To prevent the cells from undergoing electrical current-associated heat death, the dimensions of the galvanotaxis chamber have been calculated to minimize the effects of Joule heating¹⁶. The heat generated in the chamber is proportional to the square of the current flowing through it, which is in turn proportional to the cross-sectional area of the galvanotaxis chamber. We advise maintaining the cross-sectional area of the chamber as 2.04 mm² (12 mm x 0.17 mm).
• After 7 days of culture neurospheres can be collected, dissociated into single cells and replated in growth-factor conditions (a process referred to as 'passaging') to yield secondary neurospheres. We obtain our best results using neurospheres that have undergone less than three passages (maximum 21 days in culture), with a visible decline in migratory capabilities when using neurospheres that have been passaged a greater number of times. This is consistent with our previous work demonstrating that longer-term cultures alter cell kinetics¹⁷.
• When selecting a galvanotaxis chamber containing NPCs for observation, it is important to select a chamber in which the neurosphere derived cells have dissociated well in order to permit accurate kinematic analyses. If the neurospheres have not dissociated well (NPCs are adhered to each other and cannot be delineated), analysis of individual cell migratory behaviour will become near impossible. The cultures should be allowed more time to dissociate prior to observation. In addition, cells that are in close proximity to each other can physically obstruct the migration of neighboring cells. Our analyses have shown that cells positioned at least one cell body away from their nearest neighboring cell migrate at significantly greater velocities than cells that are clustered together near the center of the dissociated neurosphere, albeit with equal directedness.
• If there is an abundance of processes extending from the NPCs, this is an indicator that the cells have begun to differentiate and therefore they are not a true representation of an undifferentiated population. A benefit of preparing the galvanotaxis chambers in triplicate is that the remaining two chambers that are not selected for time-lapse imaging can be fixed immediately following their incubated plating period (Step 2.16), and immunostained to verify the differentiation profile of the cells.
• For optimal results the culture media should be freshly prepared and supplemented on each day that it is required.
• The time-lapse images and subsequent kinematic analyses are highly sensitive to perturbation of the sample. The live-cell imaging microscope should be placed on an air table to minimize vibrations, and the system should not be perturbed during experimentation.

Our lab routinely images NPC galvanotaxis for 6-8 hr, although up to 15-hr analyses have been performed. Over the 15-hr periods, the dcEF across the chamber declined from 250 mV/mm to 227 mV/mm (9.2% decrease). Longer imaging periods inevitably modify the pH of the media in the chamber, further modifying the electrical field. Thus it is recommended to replace the media approximately every 6 hr. However, we have performed 8-hr NPC galvanotaxis assays without media replacement and have not observed any noticeable cell death (dead cells do not migrate) or decline in cell motility; the cells maintain their velocity of migration throughout the experiment. The investigator may also reduce the depth of the chamber to increase the stability of the dcEF¹⁰.

Following imaging, single-cell kinematic tracking analysis is performed on a subset of the imaged cells using Zeiss Axiovision software's tracking module. Cells are selected for analysis if they are localized closer to the edge of the dissociated neurosphere and at least one cell body apart from its nearest cell. The rationale behind this is to minimize the likelihood of cells overlapping each other during tracking, which would make tracking individual cells near impossible. We analyze the following four parameters of migration:

1. X-axis displacement - the distance a cell migrates in the direction of the axis, which is parallel to the direction of the dcEF.
2. Velocity - the straight-line distance between initial and final cell positions divided by elapsed time.
3. Directedness - x-axis displacement divided by the straight-line distance between the initial and final cell positions.
4. Tortuosity - the total path distance travelled by the cell divided by the straight-line distance between initial and final cell positions.

The latter two parameters are indicators of how straight the migration pathway is in a particular direction.

Upon grasping the methods in this protocol, the investigator may wish to modify some aspects for broader applications. For example, the assay may also be modified to investigate the migratory behavior of other cell types. Genetic knockout models or siRNA transfection techniques could be used to target genes of interest that may play a role in cell migration or the transduction of a dcEF into cell motility. In our lab, we have modified the galvanotaxis chamber to permit continuous cross-perfusion of fresh culture media within the central trough during experimentation¹⁸. An appropriate substrate and culture media would need to be selected for other cell types, and the duration of cell seeding prior to imaging should be adjusted as necessary. Notably, the dimensions of the chamber may need to be modified depending on the thickness of the tissue being examined (for instance, if a tissue slice is placed in the chamber). The investigator should bear in mind that for a set electrical field, the current flow through the chamber is directly proportional to the cross-sectional area of the chamber, and therefore significant enlargements to the chamber's dimensions may result in increased Joule heating-related cell death.

The techniques described in this report provide a powerful tool for investigating the important, but not widely studied, phenomenon of galvanotaxis. The ability of cells to respond to dcEFs has direct therapeutic implications. Electrical stimulation (deep brain stimulation) has already proven useful in the treatment of Parkinson's disease, and has been shown to promote hippocampal neurogenesis in a mouse model^19,20. A complete understanding of the cellular mechanisms responsible for dcEF transduction into cellular motility may eventually lead to the use of electrical stimulation as a clinical means for enhancing and directing endogenous precursor migration to sites of injury or disease to facilitate the repair process. The methods described above provide a simple and robust way of investigating these processes.