Tuning in the Hippocampal Theta Band In Vitro: Methodologies for Recording from the Isolated Rodent Septohippocampal Circuit

Summary

Here, we present a protocol for recording rhythmic neuronal network theta and gamma oscillations from an isolated whole hippocampal preparation. We describe the experimental steps from extraction of the hippocampus to details of field, unitary and whole-cell patch clamp recordings as well as optogenetic pacing of the theta rhythm.

Abstract

This protocol outlines the procedures for preparing and recording from the isolated whole hippocampus, of WT and transgenic mice, along with recent improvements in methodologies and applications for the study of theta oscillations. A simple characterization of the isolated hippocampal preparation is presented whereby the relationship between internal hippocampal theta oscillators is examined together with the activity of pyramidal cells, and GABAergic interneurons, of the cornu ammonis-1 (CA1) and subiculum (SUB) areas. Overall, we show that the isolated hippocampus is capable of generating intrinsic theta oscillations in vitro and that rhythmicity generated within the hippocampus can be precisely manipulated by optogenetic stimulation of parvalbumin-positive (PV) interneurons. The in vitro isolated hippocampal preparation offers a unique opportunity to use simultaneous field and intracellular patch-clamp recordings from visually-identified neurons to better understand the mechanisms underlying theta rhythm generation.

Introduction

Hippocampal theta oscillations (4 - 12 Hz) are amongst the most predominant forms of rhythmic activity in the mammalian brain and are believed to play key roles in cognitive functions such as processing of spatiotemporal information and formation of episodic memories^1,2,3. While several in vivo studies that highlight the relationship of theta-modulated place-cells with spatial navigation and lesion studies, as well as clinical evidence, support the view that hippocampal theta oscillations are involved in memory formation^4,5,6, the mechanisms associated with generation of hippocampal theta oscillations are still not fully understood. Early in vivo investigations suggested that theta activity depended mainly on extrinsic oscillators, in particular rhythmic input from afferent brain structures such as the septum and entorhinal cortex^7,8,9,10. A role for intrinsic factors - internal connectivity of hippocampal neural networks together with the properties of hippocampal neurons - was also postulated based on in vitro observations^{11,12,13,14,15,16,17,18}. However, apart from a few landmark studies^19,20,21, difficulties in developing approaches that could replicate physiologically realistic population activities in simple in vitro slice preparations have, for a long time, delayed more detailed experimental examination of the intrinsic abilities of the hippocampus and related areas to self-generate theta oscillations.

An important downside of the standard in vitro thin-slice experimental setting is that the 3D cellular and synaptic organization of brain structures is usually compromised. This means that many forms of concerted network activities based on spatially distributed cell assemblies, ranging from localized groups (≤1 mm radius) to populations of neurons spread across one or more brain areas (>1 mm), cannot be supported. Given these considerations, a different type of approach was needed to study how theta oscillations emerge in the hippocampus and propagate to related cortical and subcortical output structures.

In recent years, the initial development of the "complete septo-hippocampal" preparation to examine bidirectional interactions of the two structures²², and the ensuing evolution of the "isolated hippocampus" preparation, have revealed that intrinsic theta oscillations occur spontaneously in the hippocampus lacking external rhythmic input²³. The value of these approaches lies on the initial insight that the whole functional structure of these regions had to be preserved in order to function as a theta rhythm generator in vitro²².

Protocol

All procedures have been performed according to protocols and guidelines approved by the McGill University Animal Care Committee and the Canadian Council on Animal Care.

1. Acute Hippocampus In Vitro Preparation

NOTE: Isolating the intact hippocampal preparation involves three major steps: (1) Preparation of solutions and equipment, (2) Dissection of the hippocampus and (3) Setting up the fast perfusion rate system necessary for generation of intrinsic theta oscillations. In this protocol, the timely performance of procedures – from dissection to recording – is particularly important because the isolated hippocampus constitutes such a dense, but delicate, preparation that maintaining functional connectivity of the structure in vitro requires great care. Preparing everything beforehand ensures that an adequate level of perfusion is available as early as possible to minimize cell damage and maintain physiological function.

1. Sucrose-based artificial Cerebral Spinal Fluid (aCSF) solution
   1. Prepare 1X stock sucrose solution to be used for dissection and post-dissection incubation. Combine all components listed in Table 1, except for CaCl₂, into 1 L of deionized water, and store at 4 °C for up to 2 weeks.
2. Standard aCSF solution
   1. Prepare standard aCSF for high flow rate perfusion of the isolated hippocampus during electrophysiological recordings. To make the concentrated (5X) stock aCSF, dissolve all the components listed in Table 2, except CaCl₂, in 2 L of deionized water and store at 4 °C.
3. Prepare the equipment
   1. Before starting dissection, set up the bench area with an ice-filled plastic tray (30 x 20 x 5 cm), carbogen connected tubing lines and three glass Petri dishes (10 x 2 cm) (See Figure 1).
   2. Add 300 mL of sucrose solution (1X stock) to a 500 mL beaker, bubble it with carbogen (95% O₂, 5% CO₂) and add Ca²⁺ [1.2 mM].
   3. Transfer 60 mL of this sucrose solution to a glass Petri dish and leave at room temperature. Place the remainder in a 4 °C freezer for 10 - 15 min.
   4. Bubble the ice-cold sucrose solution with carbogen throughout the dissection.
   5. Fill a second Petri dish (cold holding chamber) with sucrose solution (4 °C) and keep on ice.
   6. Add 30 mL of ice-cold sucrose solution to a 50 mL beaker.

2. Whole Hippocampus Dissection

NOTE: The method for dissecting the isolated hippocampus is essentially identical to the one developed and described originally²², but with additional details and changes regarding the perfusion rate and recording techniques.

1. Brain dissection
   1. Anesthetize mouse (P20 - 35) under a fume hood with a small paper towel soaked with 1 mL of isoflurane (100%) that is added into the cage.
   2. Decapitate the anesthetized mouse and quickly submerge the head in ice-cold sucrose solution (50 mL beaker, ~ 5 s). Transfer onto a paper towel.
   3. Use a scalpel to make a medial skin incision (caudal to rostral) and push skin on the sides to expose the cranium.
   4. Cut open the skull with a small scissors and use a spatula to rapidly extract the brain and drop it into the Petri dish containing ice-cold sucrose solution. Allow 1-2 min for the brain to cool.
   5. While the brain is recovering, add 2 - 3 mL of sucrose solution onto a lens paper covered Petri (dissection) dish on ice.
2. Isolating the hippocampus
   1. Place the brain onto the dissection dish in an upright position.
   2. Remove the cerebellum and frontal cortex with a razor blade. Cut the brain in half along the midsagittal plane and return the two isolated hemispheres to the holding chamber. Allow a further 1 - 2 min for recovery.
   3. Place single hemisected brain upright onto the dissection dish.
   4. Rotate the dish until midsagittal structures face the observer and the embedded outline of the septal complex is visible as a thin pear-shaped layer of tissue anterior to the thalamus.
   5. Hold the hemi-sected brain steady with the microspatula and place the small end of the polytetrafluoroethylene (PTFE)-coated spatula immediately behind (caudal to) the septal area.
   6. Insert the coated spatula beneath the septum and move the tip down until the dissection dish is reached. Sever the fibers that connect the septal area caudally. Repeat the same operation along the anterior edge of the septum and cut the fibers that connect to the frontal part of the brain.
   7. With the spatula lightly holding the inner part of the cortex just above the hippocampus in an upright position, use the microspatula to carefully pull down the thalamic, hypothalamic and remaining brain stem nuclei. Use the spatula to cut off and remove the pulled away tissue.
   8. Turn the isolated hemisphere on its side and identify the outline of the hippocampus.
   9. Insert the coated spatula in the lateral ventricle, underneath the rostral end of the dorsal hippocampus.
   10. Hold the spatula horizontally aligned with the midsagittal plane of the hemi-sected brain and slide it through the smooth contour of ventricular walls until the tip emerges caudally.
   11. Hold the spatula beneath the hippocampus and lightly press it onto the connecting fibers, along the inside layer where the hippocampus joins with the overlying cortex. Apply the microspatula on the external side of this layer and slide against the coated spatula to slice through the connecting fibers.
   12. To complete the extraction, rotate the dissection dish and insert the coated spatula beneath the ventral hippocampus. Lightly hold the inside of the cortical fold with the spatula and cut through the entorhinal connections using a slicing motion against the microspatula.
       NOTE: The whole septohippocampal complex can be removed by placing the coated spatula under the entire hippocampus, and pulling out the remaining brain tissue from underneath.
   13. Once isolation is complete, keep the hippocampus resting on the dish and add a drop of ice-cold sucrose solution to keep it cool. Carefully trim off any remaining cortex and fibers and separate the hippocampus from the septum by gently applying a razor blade through the fornix.
       NOTE: If patch clamp recordings are to be performed from pyramidal cells (see Section 4.6 Optogenetic control), the isolated hippocampus can be transversally cut at a 45° angle to expose the pyramidal layer of CA1 ("angle-cut" preparation), without compromising the local network circuitry in the remaining portion of the hippocampus.
   14. Transfer the preparation to room temperature sucrose solution and let it recover for 15 - 30 min before transferring to the recording chamber.

3. Set Up the Fast Perfusion for Recording the Isolated Hippocampus

1. Set up the gravity-fed perfusion system to enable continuous high-speed inflow of oxygenated aCSF at 20 - 25 mL/min.
   1. Prepare aCSF (1X) flasks in advance and immerse them in a warming bath (~30 °C) at least 15 min before start of experiment. Turn on the in-line solution heater system (30 °C).
   2. Use aquarium bubblers and tubing lines connected to a carbogen tank to oxygenate aCSF throughout the experiment, and add CaCl₂ [2 mM] prior to the start of perfusion.
   3. Stop aCSF flow and transfer the hippocampus to the recording chamber using the wide end of a glass pipette. Allow the sucrose saturated preparation to sink and settle at the bottom.
   4. Place the preparation in the center of the recording chamber (in line with the inlet-outlet axis) with the smooth surface of CA1 and SUB on top. Stabilize the hippocampus with small weights at the septal and temporal extremities and restart aCSF flow.

4. Electrophysiology in the Isolated Hippocampus

1. Extracellular recording of in vitro hippocampal theta oscillations
   1. For extracellular monitoring of Local Field Potential (LFP) or single-unit activity from the in vitro isolated hippocampus, use glass micropipettes (1 - 3 MΩ) filled with aCSF. Record low-noise AC-coupled field potential signals with a patch clamp amplifier of gain x1000, online band-pass filtering 0.1 - 500 Hz and a sampling rate of 5-20 kHz.
      NOTE: The custom-built microscope used in this protocol is equipped with low- (2.5X) and high-magnification (40X) objectives for low-magnification view of the hippocampus, as well as infrared video microscopy and epifluorescence for single cell recognition. The components of this system include the upright fluorescence microscope, fluorescence filter cubes, an analogue video camera and digital frame grabber with imaging software, an on-stage custom perfusion chamber (Figure 2A, inset i) and high-precision micromanipulators for neuronal recordings and optic fiber placement.
   2. Use the camera mounted on the microscope and connected to a computer display to inspect the hippocampal preparation under low magnification (2.5X) and rapidly check that there are no undercuts or damage to the external surface. The diagonally oriented arrangement of alveus fibers along the smooth surface of CA1 should be visible (Figure 2A, inset ii) when shining transmitted light through the hippocampus.
   3. Use the low-magnification wide-field objective to view the isolated hippocampus and the placement of LFP electrodes in specific recording locations.
      NOTE: A layout of the isolated hippocampal preparation with multiple electrodes placed in different recording locations is illustrated in Figure 2A.
   4. Lower an LFP electrode into the bath until it touches the surface (alveus) of the CA1 area.
      NOTE: This usually produces a fast transient in the AC-coupled recording similar to what happens when the electrode comes in contact with the recording media. An audio monitor can be useful to listen to the LFP channel signal after this step.
      NOTE: Oftentimes, early signs of activity will only appear after 10 - 15 min, therefore it is important to not advance quickly into the preparation. If after this period the surface LFP electrode is still not picking up activity, advance a little further down through the stratum oriens and pause periodically to see if any form of activity arises while approaching the principal cell layer. If nothing is detected after a further 5 - 10 min, carefully retract the electrode and place it elsewhere along the same region.
   5. Advance the LFP electrode through the pyramidal layer. Observe that extracellular spiking activity initially increases as single-unit discharge from individual neurons (mostly pyramidal and inhibitory basket cells) is detected. Lower the electrode further and note that spiking begins to fade again as the tip crosses into the radiatum. Observe that a clearly visible network oscillation in the theta-frequency range (4 - 12 Hz) becomes apparent and reaches maximum amplitude (~100 - 300 µV) as the recording location is lowered through the radiatum.
2. Theta oscillations across a single region
   1. To test the spatial properties of spontaneous theta oscillations across the CA1 region, place the tip of a reference LFP electrode in a medial CA1 site (medially located along the longitudinal septo-temporal axis) and lower it into the radiatum as described previously.
   2. Place a second LFP electrode into a CA1 site (200 - 800 µm septal or temporal from the first LFP) and observe that CA1 theta oscillations can synchronize over relatively large distances.
3. Theta oscillations across layers
   1. To test the properties of theta oscillations across hippocampal layers, leave a reference LFP electrode in a CA1 radiatum site. Starting from just above the stratum oriens, lower a second electrode into the pyramidal cell layer, and through the radiatum. Observe a gradual inversion of the LFP signal (relative phase reversal) across the pyramidal layer.
      NOTE: For representative examples of these types of activity, see Figure 2B. Examples of laminar/depth profiles and Current Source Density (CSD) analysis illustrating the site of phase reversal in isolated hippocampi have been published previously^23,24,25.
4. Gamma oscillations and theta-gamma coupling in the intact hippocampus
   1. Place a field electrode at the CA1/SUB border and lower it until it sits at the interface between the pyramidal and molecular layers. At this level, a field potential displaying clear gamma oscillations with changes in amplitude that phase-lock to the local theta rhythm can be recorded²⁴. Adjust scaling to observe the slow time scale of theta-gamma coupling. Note that gamma bursts occur in two distinct frequency bands, which can be revealed by band-pass filtering the ongoing LFP signal in the slow (25 - 50 Hz) and fast gamma (150 - 250 Hz) range (See Figure 2C for representative filtered traces).
5. Whole-cell patch clamp recording during in vitro hippocampal theta oscillations
   NOTE: In this part of the protocol, the goal is to obtain visually guided whole-cell patch clamp recordings from identified interneurons in isolated hippocampi from transgenic PV-TOM mice expressing the fluorescent protein, tdTomato, under the control of the PV promoter (for more details see materials and Ref ²⁶).
   1. Use fluorescence video-microscopy with low (2.5X) and high power (40X) magnification to visualize tdTomato-positive interneurons located near the surface of a hippocampal preparation from a PV-TOM mouse (Figure 3). Note that while PV-TOM neurons can be successfully visualized and targeted for patch through the alveus (up to a 100 µm), non-fluorescent pyramidal cells can also be reached relatively easily when the hippocampus is prepared with the angle-cut technique (see Section 2.2.14 note).
   2. Under low magnification view of the hippocampus, place an LFP electrode in CA1/SUB to monitor theta oscillations while preparing for patch clamp experiments.
   3. Switch to 40X magnification and immerse the objective over the target region; Lower it until the top layers become visible. Manipulate the position of the microscope to ensure sufficient open access for patch pipette approach from the opposite side.
   4. Under fluorescence microscopy, select a fluorescent PV-TOM cell and approach it with a patch pipette (2 - 3 MΩ) filled with standard intracellular solution.
   5. Use a mouth piece to apply light positive pressure to the pipette and monitor the resistance as the electrode is lowered onto the cell. When the tip encounters the target cell, the pipette resistance increases and a clear dimple appears as positive pressure pushes on the membrane. Release the pressure and check that the current pulse flattens as a seal starts to form. Immediately clamp to a hyperpolarized potential (-70 mV), and once a GΩ seal is achieved, rupture the cell membrane with a small amount of negative pressure applied through the pipette holder.
   6. Once in whole-cell configuration, examine the physiological properties of identified PV cell during spontaneous hippocampal oscillations (Figure 3B).
   7. Observe that intracellular membrane potential recording from PV cells are characterized by fast-spiking behavior and bursts of action potentials synchronized to the ongoing CA1/SUB theta rhythm.
   8. Switch to voltage-clamp and hold the cell at different membrane potentials to characterize the ratio of excitatory versus inhibitory synaptic currents generated by the intrinsic rhythmic activity. An example of patch clamp recording from a pyramidal cell is shown in Figure 3A for comparison.
6. Optogenetic control in the isolated hippocampus
   NOTE: For optogenetic control of hippocampal network activity, a cell-type specific strategy involving rhythmic activation of PV-containing GABAergic cells is used here as these cells were shown to play a major role in synchronizing principal cell populations in the isolated hippocampus²⁵. Selective expression of the excitatory channelrhodopsin-2 (ChR2) in PV cells is achieved by crossing the PV-Cre mouse line with mice constitutively expressing ChR2 Cre-dependently (Ai32), thereby generating PV-ChY mice. Alternatively, viral vector constructs (e.g., AAVdj-ChETA-eYFP) can be injected into the hippocampus of PV-Cre or PV-TOM mice (P15-16) to drive the expression of the excitatory opsin in CA1/SUB interneurons (Figure 4A).
   NOTE: This strategy has been described previously²⁵.
   1. Place an LFP electrode in the CA1/SUB area and patch a nearby pyramidal cell in the isolated hippocampus of a mouse expressing the blue-light sensitive excitatory opsin ChR2 in PV interneurons.
   2. Place an optic fiber light guide (0.2 - 1 mm diameter) above the hippocampal preparation and center it on the recorded region. Use blue light (473 nm) from a LED source for optogenetic stimulation, consisting of light pulses (10 - 20 ms, triggered by a transistor-transistor logic signal) or sine wave voltage commands delivered at theta frequencies (4 - 12 Hz).
      NOTE: The custom LED-based light stimulation assembly has been described elsewhere²⁵.
   3. Switch to current clamp and characterize the activity of the pyramidal during spontaneous theta oscillations.
   4. Start the stimulation protocol and record the light responses.Observe that field oscillations and synaptic activity in the recorded neuron become increasingly synchronized during optogenetic stimulation and that rhythmic pacing of PV cells results in a robust control of both frequency and power of theta oscillations (Figure 4).

Results

This section illustrates examples of results that can be obtained by studying theta oscillations in the mouse isolated hippocampal preparation in vitro. The dissection procedure for extracting the isolated hippocampus is illustrated in Figure 1. Using this preparation, intrinsic theta oscillations can be examined during placement of multiple field electrodes, recording overall activity and synchronized synaptic inputs to neuronal populations in diffe...

Discussion

While electrophysiological recordings from acute hippocampal slices constitute a standard in vitro technique, the methods presented here differ substantially from the classic approach. Unlike the thin slice preparations where specific cell layers are visible at the surface and can be examined directly, the intact hippocampal preparations are more akin to in vivo configurations where electrodes are lowered into targeted brain regions while crossing through individual layers. The integrity of the hippocam...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Reagents
Sodium Chloride	Sigma Aldrich	S9625
Sucrose	Sigma Aldrich	S9378
Sodium Bicarbonate	Sigma Aldrich	S5761
NaH2PO4 - sodium phosphate monobasic	Sigma Aldrich	S8282
Magnesium sulfate	Sigma Aldrich	M7506
Potassium Chloride	Sigma Aldrich	P3911
D-(+)-Glucose	Sigma Aldrich	G7528
Calcium chloride dihydrate	Sigma Aldrich	C5080
Sodium Ascorbate	Sigma Aldrich	A7631-25G
Name	Company	Catalog Number	Comments
Equipment
Standard Dissecting Scissors	Fisher Scientific	08-951-25	brain extraction
Scalpel Handle #4, 14 cm	WPI	500237	brain extraction
Filter forceps, flat jaws, straight (11 cm)	WPI	500456	brain extraction
Paragon Stainless Steel Scalpel Blades #20	Ultident	02-90010-20	brain extraction
Fine Point Curved Dissecting Scissors	Thermo Fisher Scientific	711999	brain extraction
Teflon (PTFE) -coated thin spatula	VWR	82027-534	hippocampal preparation
Hayman Style Microspatula	Fisher Scientific	21-401-25A	hippocampal preparation
Lab spoon	Fisher Scientific	14-375-20	hippocampal preparation
Borosilicate Glass Pasteur Pipets	Fisher Scientific	13-678-20A	hippocampal preparation
Droper	Fisher Scientific		hippocampal preparation
Razor blades Single edged	VWR	55411-055	hippocampal preparation
Lens paper (4 x 6")	VWR	52846-001	hippocampal preparation
Glass petri dishes (100 x 20 mm)	VWR	25354-080	hippocampal preparation
Plastic tray for ice; size 30 x 20 x 5 cm	n.a.	n.a.	hippocampal preparation
Single Inline Solution Heater	Warner Instruments	SH-27B	perfusion system
Aquarium air stones for bubbling	n.a.	n.a.	perfusion system
Tygon E-3603 tubing (ID 1/16 OD 1/8)	Fisherbrand	14-171-129	perfusion system
Electric Skillet	Black & Decker	n.a.	perfusion system
95% O2/5% CO2 gas mixture (carbogen)	Vitalaire	SG466204A	perfusion system
Glass bottles/flasks (4 x 1 L)	n.a.	n.a.	perfusion system
Submerged recording Chamber	custom design (FM)	n.a.	Commercial alternative may be used
Glass pipettes (1.5/0.84 OD/ID (mm) )	WPI	1B150F-4	electrophysiology
Hum Bug 50/60 Hz Noise Eliminator	Quest Scientific	Q-Humbug	electrophysiology
Multiclamp 700B patch-clamp amplifier	Molecular devices	MULTICLAMP	electrophysiology
Multiclamp 700B Commander Program	Molecular devices	MULTICLAMP	electrophysiology
Digital/Analogue converter	Molecular devices	DDI440	electrophysiology
PCLAMP10	Molecular devices	PCLAMP10	electrophysiology
Vibration isolation table	Newport	n.a.	electrophysiology
Micromanipulators (manually operated )	Siskiyou	MX130	electrophysiology (LFP)
Micromanipulators (automated)	Siskiyou	MC1000e	electrophysiology (patch)
Audio monitor	A-M Systems	Model 3300	electrophysiology
Micropipette/Patch pipette puller	Sutter	P-97	electrophysiology
Custom-built upright fluorescence microscope	Siskiyou	n.a.	Imaging
Analogue video camera	COHU	4912-2000/0000	Imaging
Digital frame grabber with imaging software	EPIX, Inc	PIXCI-SV7	Imaging
Olympus 2.5X objective	Olympus	MPLFLN	Imaging
Olympus 40X water immersion objective	Olympus	UIS2 LUMPLFLN	Imaging
Custom-made Light-Emitting Diode (LED) system	custom	n.a.	optogenetic stimulation (Amhilon et al., 2015)
Name	Company	Catalog Number	Comments
Animals
PV::Cre (KI) mice	Jackson Laboratory	stock number 008069	Allow  Cre-directed gene expression in PV interneurons
Constitutive-conditional Ai9 mice (R26-lox-stop-lox-tdTomato (KI))	Jackson Laboratory	stock number 007905	Express TdTomato following Cre-mediated recombination
Ai32 mice (R26-lox-stop-lox-ChR2(H134R)-EYFP	Jackson Laboratory	stock number 012569	Express the improved channelrhodopsin-2/EYFP fusion protein following exposure to Cre recombinase
PVChY mice	In house breeding	n.a.	Offspring obtained from cross-breeding the PV-Cre line with Ai32 mice (R26-lox-stop-lox-ChR2(H134R)-EYFP