A protocol for preparation and transfection of rat entorhinal cortex organotypic cultures for electrophysiological whole-cell recordings

Graphical abstract


Plasmid constructs
Plasmids carrying reporter genes for either green fluorescent protein (GFP, pGFP-V-RS) or red fluorescent protein (RFP, pRFP-C-RS) were purchased from OriGene (Rockville, MD). Plasmids were amplified in E. coli and isolated using a mega prep kit purchased from Qiagen (cat#: 12181). DNA working solutions used for cartridge preparation were diluted to 1 mg/mL in TE.

Procedure overview EC slice culture
Setup for culturing procedure (estimated time$1 h) 1. Prepare and autoclave instrument packages and foil-covered 20, 30, and 250 mL beakers a. Package #1: large surgical scissors, fine surgical scissors, fine forceps, spatula ( Fig. 1A) b. Package #2: spatula, scalpel blade, syringe with modified needle, brush, filter paper, fine forceps ( Fig. 1A) 2. Aliquot necessary solutions in BSC and pre-equilibrate media a. Aliquot $200 mL of filter sterilized cutting solution into a small sterile bottle for freezing at À80 C i. Critical step: Ensure that cutting solution is a very icy slurry prior to dissection to avoid poor slice health b. Add $1.1 mL of SM to anticipated number of petri dishes/6-well plates needed i. We typically culture 3-4 EC slices per insert and thus require between 4 and 6 inserts for one animal ii. Pre-equilibrate dishes by placing them into a 37 C humidified 5% CO 2 incubator at least 30 min prior to dissection procedure c. Aliquot $2 mL and $10 mL of SM and MEM, respectively, and place on ice for use during dissection procedure 3. Prepare and UV sterilize BSC for slicing ( Fig. 1C) a. Spray down BSC with 70% ethanol (EtOH) and lay out: vibratome equipment, both dissection packages, sterile beakers, 4-40 mm petri dishes, 1 mL transfer pipette, 1 mL transfer pipette tips, anticipated number of cell culture inserts, and cyanoacrylate glue b. UV sterilize cabinet and materials for at least 15 min 4. Place foil covered sterile beakers on ice with SM and MEM 5. Use sterile scissors from dissection package to cut and modify tip of 1 mL transfer tip a. These tips are used for transferring slice from cutting chamber, MEM washes, and placement on insert (see Graphical abstract) 6. Invert lids of 100 mm petri dish to use as brain-prepping stage (used prior to gluing brain to vibratome platform) and place filter paper atop 7. Fill vibratome trough with ice, turn on and prepare vibratome 8. Once cutting solution is sufficiently slushy and chilled, aliquot $150 mL into 250 mL beaker, $25 mL into 30 mL beaker, and $15 mL into 20 mL beaker in BSC a. Recover with foil and return to ice bucket Dissection and slice collection procedure (estimated time $30 min) 1. In a fume hood, deeply anesthetize animal in bell jar with isoflurane, spray animal with 70% EtOH, decapitate using large scissors from dissection package #1, and transfer head to 30 mL beaker of icy slurry solution a. Replace foil, change gloves, and move to BSC with remaining tools in dissection package #1 2. On surgical draping, dissect brain into 20 mL beaker of icy solution and allow to cool for at least 1 min a. Meanwhile, use EtOH to disinfect dissection region of BSC 3. Once the brain is chilled, pour off excess cutting solution and guide brain onto a piece of filter paper placed on prepping stage 4. With scalpel blade, gently invert brain so that it sits on its dorsal region and remove cerebellum and anterior portion of brain not containing hippocampus 5. Using the spatula and scalpel, carefully transfer the brain to the second piece of filter paper to remove excess cutting solution from the bottom (dorsal) part of the brain 6. Apply glue to vibratome platform, gently guide brain onto glue, transfer platform to cutting reservoir, and add $150 mL ice cold cutting solution without disturbing brain with large ice chunks 7. Collect 300 mm horizontal sections in reservoir 8. While slicing, remove an ice pack from the freezer and place in BSC; place the four petri dishes atop the pack 9. Using a paintbrush and modified syringe/needle, carefully isolate the EC by making two cuts (see Graphical abstract) a. The first cut removes a portion of the perirhinal cortex and provides a visual cue for medial vs. lateral orientation at later stages in culture b. The second cut near the hippocampal sulcus removes the hippocampus proper and leaves the EC intact 10. Fill three of the four dishes with chilled MEM and rinse slices by serial transfer, transferring as many slices as possible at once (see Graphical abstract) 11 3. Measure out 12.5-15 mg of 1.6 mm gold particles in a 1.5 mL microcentrifuge tube 4. Measure out 20 mg PVP into a 1.5 mL microcentrifuge tube and dissolve with 1 mL 100% EtOH 5. Aliquot 3.1 mL of 100% EtOH into a 15 mL tube and prepare a 50 mg/mL PVP dilution using the 20 mg/mL stock (7.75 mL of stock solution to 3.1 mL) 6. Prepare DNA working solution; we typically prepare 1 mg/mL dilutions of cDNA to be transfected a. If more than one construct is to be transfected, we combine all constructs in a separate tube and mix by aspirating b. Ratios will need to be determined for each condition and by individual investigators, but our results have indicated a high success rate of near equal co-transfection rates by simply using a 1:1 ratio of two constructs with the same promoter (see below) 7. To the gold, add 100 mL 0.5 M spermidine (filter sterilized); tap to mix and sonicate briefly 8. Add 50 mg of DNA to gold mixture; tap well to mix and sonicate very briefly In a BSC, place the EtOH disinfected Biorad Helios Gene-Gun, autoclaved gene-gun transfection package (Fig. 1B), vials of gene gun cartridges, a sterile 40 mm petri dish, warmed SM, a 2 mL serological pipette, and a pipette-aid a. We also include a shooting platform in the BSC to assist in consistent shooting distances (see below under Additional information) b. A simple shooting platform can be constructed using a microcentrifuge tube rack that holds a rod upright (e.g. seriological pipette) marked at 1.25 in. distance from slice surface (Fig. 1D) 2. Open gene gun kits and load appropriate number of cartridges into cartridge holder using forceps 3. Connect gene gun to helium supply, as per manufacturer instructions, and adjust helium supply to 150 psi 4. Prime the gun by discharging twice with the cartridge holder positioned on an empty barrel 5. Remove EC slice dishes/plates from incubator and transfect one insert at a time a. Transfer insert to a new empty sterile 40 mm petri dish and change the media on the dish/plate b. Place dish containing only the insert on shooting platform, remove cover, position gun directly above insert at a distance of 1.25 in. away from slice surface and discharge cartridge i. Critical Step: Slice health is important for obtaining quality electrophysiological recordings from slice cultures and we observed that shooting distance can drastically influence the quality of transfected, patchable cells ii. Thus, using a shooting platform with a guide to improve consistency is strongly recommended (Fig. 1D) c. Return insert to original dish containing fresh media d. Repeat steps 5a-5c until all slices are transfected i. Note: if transfecting with multiple constructs, be sure to use a different gene-gun instrument package with a clean mesh diffuser screen for each construct 6. Turn off helium source, return slices to incubator, clean up BSC, and return sealed cartridge constructs to 4 C.

Method validation
We first ensured that cultured slices ( Fig. 2A) maintained a laminar distribution and that we had healthy neurons in our preparation. Slices were fixed in 4% paraformaldehyde for 24 h, after which slices were individually transferred to frosted microscope slides (VWR #12-550-15). Immunostaining was performed using methods adapted from previous work [2]. Immunofluorescent staining for the neuronal marker NeuN (1:200, Milipore, #ABN78) demonstrated that EC layers were well-preserved in slice cultures and that neurons were detected after being in culture for several days (Fig. 2B). We next assessed slice health using electrophysiological methods described in previous work [2]. Under high-magnification, differential interference contrast microscopy (DIC) images revealed numerous healthy principal neurons for patching (Fig. 2C). Direct positive current injections to cultured layer II principal neurons (DIV 4) displayed action potential firing that was graded proportionally to current steps and negative current injections displayed early sag responses, typical of superficial layer II EC neurons. Input-output curves for action potential firing in response to current injections showed no significant difference between DIV 4 cultured slices and postnatal day 16 (p16) acute slices, suggesting no substantial changes in EC principal neuron excitability were present early on in culture.
Having validated that cultured EC slices were healthy and contained patchable neurons, we next wanted to validate our biolistic transfection procedures. Similar to other methods [3], we initially transfected slices within a few hours following the slicing procedure. Transfecting at two hours on DIV 0 resulted in several slices being lost from the insert following the pressure burst, likely owing to insufficient time for adherence to the membrane. Furthermore, we observed consistently higher transfection rates when slices were permitted to acclimate to culture conditions for a few days. The additional time in culture likely allowed the slice to reach a new physiological steady-state and permitted intrinsic processes to clear away unhealthy cells and other debris, which may have otherwise obstructed the penetrating transfection particles from reaching healthy neurons. For these reasons, we adopted a transfection scheme where slices were transfected on DIV 2 and experiments were conducted 36-72 h later (Fig. 3A). It is important to point out that because the culturing procedure results in massive denervation of the EC and because denervation can result in the occurrence of homeostatic changes in the target cell [4], careful attention should be given to time points of culturing, transfecting, and recording to ensure consistency across experiments. This method provided high transfection rates per slice in EC regions of interest (Fig. 3B). Recordings from layer III EC neurons transfected with GFP remained healthy and continued to be electrically active (Fig. 3C and D). Because some experimental conditions may require co-transfection of multiple plasmids to permit either expression of reporter molecules or because conditions may require introduction of multiple protein components, we assessed transfection ratios of 1:1 plasmid DNA constructs (i.e. 25 mg of each plasmid added during cartridge preparation) using GFP and RFP reporters to evaluate levels of coexpression. Under these conditions, we consistently observed a prominent level of co-expression across many experiments spanning different slice culture batches, although minor variations in intensity were occasionally seen in a few cells (Fig. 3E, see white arrows). These results suggest that a 1:1 ratio may be sufficient when introducing multiple gene products. However, for experiments where one of the constructs is simply a reporter (e.g. GFP), we adjust our ratios to 0.75:0.25 for gene of [ ( F i g . _ 2 ) T D $ F I G ] interest to reporter. Taken together, our results demonstrate that this method provides several candidate cells for patching following transfection and that transfected neurons remain healthy.

Troubleshooting
The following is a list of potential problems and suggestions to improve outcomes based on our experiences: 1. Problem: Slices dry out and die a. Potential Solutions: i. The best way to reduce this problem was by reducing time spent with the petri dish cover off in BSC while placing slices on inserts ii. Also, a breathable humidity chamber within the incubator was also useful as high humidity levels were crucial to slice survival 2. Problem: DNA-gold mixture is not dissolving a. Potential Solutions: i. The mixture is tough to dissolve but this process is aided by constant brief sonication ii. If it is impossible to dissolve, double-check the DNA stock solution concentration to ensure its accuracy; if more than 50 mg is added, it is nearly impossible to dissolve the clump, in which case it is best to start over 3. Problem: Transfected slices are uneven, in that some slices are transfected while others are blank a. Potential Solutions: i. This problem is likely due to poor quality bullets and is likely the result of poor spreading of gold mixture during the coating stage; there should not be a visible gold line in the cartridge ii. Better bullets can be generated by ensuring a well dissolved mixture and spinning the tubing for a longer duration after drawing off the EtOH, but before turning on the N 2 4. Problem: Both transfected and non-transfected cells from a shot culture dish are visible but very difficult to patch and seem to be encased in some difficult-to-penetrate matrix a. Potential Solutions: i. Use a control non-transfected (i.e. non-shot) plate to ensure cell viability; if cells are visible and patchable, the problem is almost certainly due to pressure intensity or distance of shot from the top of the slice ii. We strongly recommend using at least 1.25 in. shooting distance from the top of the slice to the barrel liner

Background and discussion
Organotypic cultures are a well-established in vitro model to investigate the cytoarchitecture and neurophysiology of brain tissue [5,6]. Whereas such cultures comprised of hippocampal regions have been utilized by many labs, relatively fewer reports involve organotypic cultures consisting of entorhinal tissue. Horizontal organotypic slices consisting of both entorhinal and hippocampal regions of P0-P5 animals have been employed to study axonal regeneration and synaptic scaling at the dentate gyrus following lesions to the perforant pathway in vitro [4,7]. Our method is unique in that it specifically details the culturing and transfection protocols used for the electrophysiological investigations of EC neurons and will be helpful for mechanism studies involving EC neuromodulation.
There are several steps we employ that distinguish our method from other methods of organotypic slice culture. First, we have chosen to use a vibratome as opposed to either a tissue chopper or slicer that others have described [6,8,9]. Although use of a vibratome requires additional time for collecting tissue, we find that it produces healthy slices that are well-suited for electrophysiological recordings as previously demonstrated by others, e.g. in spinal cord organotypic cultures [10]. Second, we choose to utilize a horizontal preparation because this approach provides slices that are similar in orientation to our acute model. Third, unlike most organotypic slice culture methods, we use a cutting solution where Na + has been largely replaced with NMDG, in order to minimize excitotoxicity that may occur during the traumatic slicing process. Alternative methods to minimize excitotoxicity involve supplementing the cutting solution with antioxidants, such as pyruvate and ascorbic acid [8,11]. NMDG-containing cutting solutions have become widely adopted in electrophysiological laboratories working with acute slices because these steps are neuroprotective, especially toward the more vulnerable interneuron populations [12]. Overall, our EC slice model attempts to yield a similar quality of initial tissue preparation for longer-term culture that resembles our acute slice model and we have had success culturing slices using animal ages ranging from p6 to P18.
Our optimized transfection conditions are informed by both the instruction manual provided with the Biorad Helios Gene-Gun, as well as other detailed published protocols describing biolistic cartridge preparation and transfection [1,9,13]. The reader is referred to those references for additional detail, however, we obtained excellent results in our method by employing a few modifications. First, it is essential that the gold mixture be homogenously distributed about the cartridge to increase transfection efficiency for all available slices on a plate. We found that including a more thorough sonication periods of the DNA gold mixture was necessary to obtain a better suspension for coating the Tefzel Tubing. Additionally, a longer period of tubing rotation following EtOH removal, but prior to turning on N 2, achieved an improved mixture distribution. Secondly, neurons are very susceptible to the pressure burst during the transfection procedure and the quality of cells available for electrophysiological recordings are greatly influenced by the shooting distance used during the transfection. Initially, we held the gun and barrel liner at a distance similar to what other methods reported. On recording days, we observed an unknown matrix that seemed to encase the neurons and impede our ability to form tight gigaohm seals with transfected neurons. However, when the shooting distance was increased, this matrix was no longer a problem and whole-cell success rates were substantially improved. We therefore addressed this problem by optimizing a shooting a distance for our 11 mm petri dishes, which we empirically found to be at 1.25 in. or greater. To improve consistency across experiments, we improvised a shooting platform that was clearly demarcated with the ideal shooting distance for an easy reference (see Fig. 1D). Thirdly, although not rigorously tested, we also found that the media composition seemed to influence the quality of transfected cell-health for recordings. Initially, we were able to obtain healthy slice cultures using a more simplistic media recipe described by other labs [3]. However, transfection efficiency seemed to be less than ideal when following the biolistic transfection process. We observed a consistent improvement in efficiency after modifying the SM with the addition of glucose and glutamine, similar to the Stoppini method [6] (see under Recipes above). Finally, we report that transfecting with GFP and RFP constructs in a 1:1 ratio provided a high degree of co-transfection, in agreement with a recent report involving biolistic double-transfections [8]. Ideally, co-transfection experiments should use plasmids containing the same promoter, as was the case in our experiment. Results are expected to vary if different and/or less efficient promotor-carrying plasmids are to be used, in which case ratios will need to be determined by individual investigators.
We expect this method to be useful for other investigators interested in addressing molecular and cellular mechanisms of entorhinal neurophysiology. The details described herein will assist others in quickly adopting a similar EC slice culture method for their lab. Additionally, this information may be useful to those troubleshooting biolistic transfection methods in other brain regions.