Acute Mouse Brain Slicing to Investigate Spontaneous Hippocampal Network Activity

Summary

This protocol describes the preparation of horizontal hippocampal-entorhinal cortex (HEC) slices from mice exhibiting spontaneous sharp-wave ripple activity. Slices are incubated in a simplified interface holding chamber and recordings are performed under submerged conditions with fast-flowing artificial cerebrospinal fluid to promote tissue oxygenation and the spontaneous emergence of network-level activity.

Abstract

Acute rodent brain slicing offers a tractable experimental approach to gain insight into the organization and function of neural circuits with single-cell resolution using electrophysiology, microscopy, and pharmacology. However, a major consideration in the design of in vitro experiments is the extent to which different slice preparations recapitulate naturalistic patterns of neural activity as observed in vivo. In the intact brain, the hippocampal network generates highly synchronized population activity reflective of the behavioral state of the animal, as exemplified by the sharp-wave ripple complexes (SWRs) that occur during waking consummatory states or non-REM sleep. SWRs and other forms of network activity can emerge spontaneously in isolated hippocampal slices under appropriate conditions. In order to apply the powerful brain slice toolkit to the investigation of hippocampal network activity, it is necessary to utilize an approach that optimizes tissue health and the preservation of functional connectivity within the hippocampal network. Mice are transcardially perfused with cold sucrose-based artificial cerebrospinal fluid. Horizontal slices containing the hippocampus are cut at a thickness of 450 μm to preserve synaptic connectivity. Slices recover in an interface-style chamber and are transferred to a submerged chamber for recordings. The recording chamber is designed for dual surface superfusion of artificial cerebrospinal fluid at a high flow rate to improve oxygenation of the slice. This protocol yields healthy tissue suitable for the investigation of complex and spontaneous network activity in vitro.

Introduction

Electrophysiological measurement from living hippocampal slices in vitro is a powerful experimental approach with numerous advantages. The experimenter can use a microscope, micromanipulators, and a recording system to directly visualize and collect measurements from individual neurons in the tissue. Tissue slices are also very accessible to photostimulation or drug delivery for optogenetic, chemogenetic, or pharmacological experiments.

The hippocampal network generates highly synchronous population activity in vivo, visible as oscillations in the extracellular local field potential^1,2,3,4,5. Brain slice methods have been leveraged to gain insight into the cellular and circuit mechanisms underlying these neuronal network oscillations. Foundational work from Maier et al. demonstrated that sharp wave-ripple complexes (SWRs) can emerge spontaneously in slices of the ventral hippocampus^6,7. Subsequent studies from multiple investigators have gradually elucidated many aspects of SWRs, including the role of neuromodulators in regulating the network state of the hippocampus^8,9,10 and the synaptic mechanisms that drive the in vitro reactivation of neuronal ensembles previously active during behavior in vivo¹¹. Brain slice experiments have also provided insight into the gamma range oscillation (30–100 Hz), a distinct hippocampal network state believed to support memory encoding and recall^12,13. Finally, recognizing the central role of the hippocampus and associated structures in the pathophysiology of temporal lobe epilepsy^14,15, researchers have used hippocampal slice preparations to investigate the generation and propagation of epileptiform activity. Carter et al. demonstrated that combined hippocampal-entorhinal cortex slices prepared from chronically epileptic animals can spontaneously generate epileptiform discharges in vitro¹⁶. Subsequently, Karlócai et al. explored the mechanisms underlying epileptiform discharges in hippocampal slices by using modified artificial cerebrospinal fluid (ACSF) with altered ion concentrations (reduced Mg²⁺ or elevated K⁺) or added drugs (4AP or gabazine)¹⁷.

Investigators have developed numerous hippocampal slice approaches that differ in key ways: (1) the region of the hippocampus contained in the slice (dorsal, intermediate, or ventral); (2) the presence or absence of extrahippocampal tissues such as the entorhinal cortex; (3) the orientation used to cut slices (coronal, sagittal, horizontal, or oblique); and (4) the conditions under which the tissue is maintained after slicing (submerged fully in ACSF or held at the interface of ACSF and humidified, carbogen-rich air).

The choice of which slicing approach to use should be determined by the experimental objective. For example, transverse or coronal slices of the dorsal hippocampus maintained under submerged conditions have been used very effectively for the investigation of intrahippocampal circuitry and synaptic plasticity^18,19,20. However, such preparations do not spontaneously generate network oscillations as readily as slices from the ventral hippocampus^21,22,23. Although a state of persistent SWR activity can be induced by tetanic stimulation in transverse slices from the dorsal and ventral hippocampus²⁴, spontaneous SWRs are more readily observed in ventral slices^7,25.

An inherent physiological and anatomical distinction between the dorsal and ventral hippocampus is supported by studies performed both in vivo and in vitro²⁶. Recordings in rats revealed strongly coherent theta rhythms throughout the dorsal and intermediate hippocampus, yet poor coherence between the ventral region and the rest of the hippocampus²⁷. SWRs in vivo propagate readily between the dorsal and intermediate hippocampus, while SWRs that originate in the ventral hippocampus often remain local²⁸. The associational projections originating from CA3 pyramidal neurons that reside in the dorsal and intermediate hippocampus project long distances along the longitudinal axis of the hippocampus. CA3 projections originating from ventral regions remain relatively local, and thus are less likely to be severed during the slicing process^29,30. Ventral slices may, therefore, better preserve the recurrent network necessary to generate population synchrony. The propensity of ventral slices to generate spontaneous network activities in vitro may also reflect higher intrinsic excitability of pyramidal neurons or weaker GABAergic inhibition in the ventral hippocampus as compared to more dorsal regions³¹. Indeed, ventral hippocampal slices are more susceptible to epileptiform activity^32,33. Thus, many studies of spontaneous physiological^8,9,11,24 or pathological^16,34,35,36 network oscillations have traditionally used a horizontal slicing approach, sometimes with a slight angle in the fronto-occipital direction, which yields tissue slices parallel to the transverse plane of the ventral hippocampus.

Network connectivity is unavoidably impacted by the slicing procedure as many cells in the slice will be severed. The angle and thickness of the slice and the tissue retained in the preparation should be considered to optimize connectivity in the circuits of interest. Many studies have utilized horizontal combined hippocampal-entorhinal cortex slices (HEC) to explore interactions between the two structures in the context of physiological or pathological network oscillations. Roth et al. performed dual recordings from the CA1 subfield of the hippocampus and layer V of the medial entorhinal cortex to demonstrate propagation of SWR activity through the HEC slice³⁷. Many studies of epileptiform activity have used the HEC slice preparation to investigate how epileptiform discharges propagate through the corticohippocampal network^16,35,36,38. It is important to note that preservation of the intact corticohippocampal loop is not a prerequisite for spontaneous SWRs, epileptiform discharges, or gamma oscillations; network oscillations can be generated in transverse slices of the dorsal or ventral hippocampus with no attached parahippocampal tissues^{21,22,23, 25,39,40,41}. A more important factor for the spontaneous generation of network oscillations in hippocampal slices may be the thickness of each slice, as a thicker slice (400–550 μm) will preserve more connectivity in the CA2/CA3 recurrent network^21,22,25.

Although angled horizontal HEC slices (cut with an approximately 12° angle in the fronto-occipital direction) have been used to study the functional connectivity of the corticohippocampal loop^{11,16,34,35,42}, such angled preparations are not required for spontaneous network activity^43,44,45. However, the use of an angled slicing plane does allow the investigator to selectively make slices that best preserve the transversely-oriented lamellae of either the ventral or intermediate hippocampus, depending on whether a downward or an upward angle is applied (Figure 1). This approach is conceptually similar to that used by Papatheodoropoulos et al., 2002, who dissected each hippocampus free and then used a tissue chopper to create transverse slices along the entire dorsal-ventral axis²¹. In the light of the aforementioned functional distinctions between the ventral and dorsal-intermediate hippocampus, investigators should consider the anatomical origin of slices when designing experiments or interpreting results. Using an agar ramp during the slicing procedure is a simple way to preferentially produce slices from either the intermediate or ventral hippocampus.

Hippocampal slices can be maintained in either a submerged chamber (with the tissue fully immersed in ACSF), or an interface-style chamber (e.g., Oslo or Haas chamber, with slices covered only by a thin film of flowing media). Interface maintenance enhances oxygenation of the tissue, which promotes neuronal survival and allows for sustained high levels of interneuronal activity. Traditionally, submerged recording conditions use a slower ACSF flow rate that does not provide adequate tissue oxygenation for stable expression of network-level oscillations. In submerged hippocampal slices carbachol-induced gamma oscillations are only observed transiently^46,47, while they can be stably maintained in interface recording chambers^10,48,49. As such, many studies of complex spontaneous activity in vitro have relied on interface recording chambers to investigate sharp-wave ripple complexes^{6,7,8,9,10,25,37}, gamma oscillations^10,13, and epileptiform activity^16,38,45,47.

In a submerged-style recording chamber, an immersion microscope objective can be used to visualize individual cells and selectively target healthy-looking cells for recordings. The submerged preparation also allows fine control over the cellular milieu, as submersion facilitates rapid diffusion of drugs or other compounds to the tissue. Thus, a modified methodology in which stable network oscillations are maintained under submerged conditions represents a powerful experimental approach. This approach is exemplified by the work of Hájos et al., in which hippocampal slices recover in a simplified interface-style holding chamber for several hours before transfer to a modified submerged recording chamber with a high flow rate of ACSF (~6 mL/min) to enhance oxygen supply to the tissue^12,48,49. Under these conditions, high levels of interneuron activity and stable spontaneous network oscillations can be maintained in a submerged recording chamber. This modified approach allows the investigators to perform visually guided whole-cell patch clamp recordings and characterize the contribution of morphologically identified cell types to carbachol-induced gamma oscillations¹². SWRs can also occur spontaneously in submerged hippocampal slices with a fast flow rate of ACSF^11,48,49. Maier et al. demonstrated that hippocampal slices that recovered in an interface chamber before transfer to a submerged recording chamber reliably exhibited spontaneous SWRs, whereas slices that recovered submerged in a beaker before transfer to a submerged recording chamber showed smaller evoked field responses, lower levels of spontaneous synaptic currents, and only very rarely exhibited spontaneous SWRs⁴³. Schlingloff et al. used this improved methodology to demonstrate the role of parvalbumin-expressing basket cells in the generation of spontaneous SWRs⁴⁴.

The following protocol presents a slicing method through which spontaneously active neurons in horizontal hippocampal slices can be recovered under interface conditions and subsequently maintained in a submerged recording chamber suitable for pharmacological or optogenetic manipulations and visually guided recordings.

Protocol

All methods described here have been approved by the Institutional Animal Care and Use Committee at Columbia University (AC-AAAU9451). 1. Prepare solutions Prepare sucrose cutting solution for slicing as described in Table 1. NOTE: After preparing 1 L of sucrose solution, freeze a small amount (approximately 100–200 mL) in an ice tray. These frozen sucrose ice cubes will be blended into an icy slurry (see step 4.3). Prepare a…

Representative Results

Presented here are representative recordings from HEC slices prepared as described in this protocol. Following recovery in an interface holding chamber (Figure 1C), slices are transferred individually to a submerged recording chamber (Figure 2B). The recording chamber is supplied with carbogen-saturated ACSF using a peristaltic pump (Figure 2A). The pump first draws ACSF from a holding beaker into a heated reservoir. Carbogen lines …

Discussion

There are several steps in this slicing protocol designed to promote tissue health and favor the emergence of spontaneous naturalistic network activity: the mouse is transcardially perfused with chilled sucrose cutting solution; horizontal-entorhinal cortex (HEC) slices are cut at a thickness of 450 μm from the intermediate or ventral hippocampus; slices recover at the interface of warmed ACSF and humidified, carbogen-rich air; during recordings slices are superfused with ACSF warmed to 32 °C and delivered at a…

Materials

3D printer	Lulzbot	LulzBot TAZ 6
Acute brain slice incubation holder	NIH 3D Print Exchange	3DPX-001623	Designed by ChiaMing Lee, available at https://3dprint.nih.gov/discover/3dpx-001623
Adenosine 5′-triphosphate magnesium salt	Sigma Aldrich	A9187-500MG
Ag-Cl ground pellets	Warner	64-1309, (E205)
agar	Becton, Dickinson	214530-500g
ascorbic acid	Alfa Aesar	36237
beaker (250 mL)	Kimax	14000-250
beaker (400 mL)	Kimax	14000-400
biocytin	Sigma Aldrich	B4261
blender	Oster	BRLY07-B00-NP0
Bonn scissors, small	becton, Dickinson	14184-09
borosilicate glass capillaries with filament (O.D. 1.5 mm, I.D. 0.86 mm, length 10 cm)	Sutter Instruments	BF150-86-10HP	Fire polished capillaries are preferable.
calcium chloride solution (1 M)	G-Biosciences	R040
camera	Olympus	OLY-150
compressed carbogen gas (95% oxygen / 5% carbon dioxide)	Airgas	X02OX95C2003102
compressed oxygen	Airgas	OX 200
constant voltage isolated stimulator	Digitimer Ltd.	DS2A-Mk.II
coverslips (22×50 mm)	VWR	16004-314
cyanoacrylate adhesive	Krazy Glue	KG925	Ideally use the brush-on form for precision
data acquisition software	Axograph	N/A	Any equivalent software (e.g. pClamp) would work.
Dell Precision T1500 Tower Workstation Desktop	Dell	N/A	Catalog number will depend on specific computer – any computer will work as long as it can run electrophysiology acquisition software.
Digidata 1440A	Molecular Devices	1-2950-0367
digital timer	VWR	62344-641	4-channel Traceable timer
disposable absorbant pads	VWR	56616-018
dissector scissors	Fine Science Tools	14082-09
double-edge razor blades	Personna	BP9020
dual automatic temperature controller	Warner Instrument Corporation	TC-344B
dual-surface or laminar-flow optimized recording chamber	N/A	N/A	The chamber presented in this protocol is custom made. A commercial equivalent would be the RC-27L from Warner Instruments.
equipment rack	Automate Scientific	FR-EQ70"	A rack is not strictly necessary but useful for organizing electrophysiology
Ethylene glycol-bis(2-aminoethyiether)- N,N,N',N'-teetraacetic acid (EGTA)	Sigma Aldrich	324626-25GM
filter paper	Whatman	1004 070
fine scale	Mettler Toledo	XS204DR
Flaming/Brown micropipette puller	Sutter Instruments	P-97
glass petri dish (100 x 15 mm)	Corning	3160-101
glucose	Fisher Scientific	D16-1
Guanosine 5′-triphosphate sodium salt hydrate	Sigma Aldrich	G8877-250MG
ice buckets	Sigma Aldrich	BAM168072002-1EA
isoflurane vaporizer	General Anesthetic Services	Tec 3
lab tape	Fisher Scientific	15-901-10R
lens paper	Fisher Scientific	11-996
light source	Olympus	TH4-100
magnesium chloride solution (1 M)	Quality Biological	351-033-721EA
magnetic stir bars	Fisher Scientific	14-513-56	Catalog number will be dependent on the size of the stir bar.
micromanipulator	Luigs & Neumann	SM-5
micromanipulator (manual)	Scientifica	LBM-2000-00
microscope	Olympus	BX51WI
microspatula	Fine Science Tools	10089-11
monitor	Dell	2007FPb
MultiClamp 700B Microelectrode Amplifier	Molecular Devices	MULTICLAMP 700B	The MultiClamp 700B should include headstages, pipette holders, and a model cell.
N-(2-Hydroxyethyl)piperazine-N′-(2-ethanesulfonic acid), (HEPES)	Sigma Aldrich	H3375-25G
needle (20 gauge, 1.5 in length)	Becton, Dickinson	305176
nylon filament	YLI Wonder Invisible Thread	212-15-004	size 0.004. This cat. # is from Amazon.com
nylon mesh	Warner Instruments Corporation	64-0198
perstaltic pump	Harvard Apparatus	70-2027
Phosphocreatine di(tris) salt	Sigma Aldrich	P1937-1G
pipette holders	Molecular Devices	1-HL-U
platinum wire	World Precision	PT0203
polylactic acid (PLA) filament	Ultimaker	RAL 9010
potassium chloride	Sigma Aldrich	P3911-500G
potassium gluconate	Sigma Aldrich	1550001-200MG
potassium hydroxide	Sigma Aldrich	60377-1KG
razor blades	VWR	55411-050
roller clamp	World Precision Instruments	14041
scale	Mettler Toledo	PM2000
scalpel handle	Fine Science Tools	10004-13
slice harp	Warner	SHD-26GH/2
sodium bicarbonate	Fisher Chemical	S233-500
sodium chloride	Sigma Aldrich	S9888-1KG
sodium phosphate monobasic anhydrous	Fisher Chemical	S369-500
sodium pyruvate	Fisher Chemical	BP356-100
spatula	VWR	82027-520
spatula/spoon, large	VWR	470149-442
sterile scalpel blades	Feather	72044-10
stirrer / hot plate	Corning	6795-220
stopcock valves, 1-way	World Precision Instruments	14054
stopcock valves, 3-way	World Precision Instruments	14036
sucrose	Acros Organics	AC177142500
support for swivel clamps	Fisher Scientific	14-679Q
surgical scissors, sharp/blunt	Fine Science Tools	14001-12
syringe (1 mL)	Becton, Dickinson	309659
syringe (60 mL with Luer-Lok tip)	Becton, Dickinson	309653
three-pronged clamp	Fisher Scientific	05-769-8Q
tissue forceps, large	Fine Science Tools	11021-15
tissue forceps, small	Fine Science Tools	11023-10
transfer pipettes	Fisher Scientific	13-711-7M
tubing	Tygon	E-3603	ID 1/16 inch, OD 3/16 inch
tubing	Tygon	R-3603	ID 1/8 inch, OD 1/4 inch
vacuum grease	Dow Corning	14-635-5D
vibrating blade microtome	Leica	VT 1200S
vibration-dampening table with faraday cage	Micro-G / TMC-ametek	2536-516-4-30PE
volumetric flask (1 L)	Kimax	KIM-28014-1000
volumetric flask (2 L)	PYREX	65640-2000
warm water bath	VWR	1209