Protocol to study microcircuits in the medial entorhinal cortex in mice using multiple patch-clamp recordings and morphological reconstruction

Summary

Multiple patch-clamp recordings and morphological reconstruction are powerful approaches for neuronal microcircuitry dissection and cell type classification but are challenging due to the sophisticated expertise needed. Here, we present a protocol for applying these techniques to neurons in the medial entorhinal cortex (MEC) of mice. We detail steps to prepare brain slices containing MEC and perform simultaneous multiple whole-cell recordings, followed by procedures of histological staining and neuronal reconstruction. We then describe how we analyze morphological and electrophysiological features.

For complete details on the use and execution of this protocol, please refer to Shi et al.¹

Graphical abstract

Highlights

• •Detailed surgical procedures for obtaining horizontal brain slices containing the MEC
• •Simultaneous octuple patch-clamp recordings from superficial MEC neurons
• •Morphological reconstructions of DAB-stained neurons
• •Concise instructions for analyzing electrophysiological and morphological data

Publisher’s note: Undertaking any experimental protocol requires adherence to local institutional guidelines for laboratory safety and ethics.

Multiple patch-clamp recordings and morphological reconstruction are powerful approaches for neuronal microcircuitry dissection and cell type classification but are challenging due to the sophisticated expertise needed. Here, we present a protocol for applying these techniques to neurons in the medial entorhinal cortex (MEC) of mice. We detail steps to prepare brain slices containing MEC and perform simultaneous multiple whole-cell recordings, followed by procedures of histological staining and neuronal reconstruction. We then describe how we analyze morphological and electrophysiological features.

Before you begin

Institutional permissions

This protocol was fully executed in the study by Shi et al.,¹ where young adult male and female mice (≥P22, with ∼90% of them to be P25-65), including wild type or Vgat-Cre/Ai9, PV-Cre/Ai9, SOM-Cre/Ai9 and VIP-Cre/Ai9 mice, were used. The Cre and Ai9 reporter lines were purchased from Jackson Laboratory (Vgat-Cre, #028862; PV-Cre, #017320; SOM-Cre, #013044; VIP-Cre, #010908; Ai9, #007909). All procedures for animal surgery and maintenance were approved by the Institutional Animal Care and Use Committee (IACUC) of Harbin Institute of Technology. Other laboratories should first attain the approval from the animal ethics committee of their institution before replicating the following experiments.

Preparation for brain dissection and slicing

Timing: 30 min

• 1.Prepare artificial cerebral spinal fluid (ACSF): Place 1 L ACSF into a fridge at 4°C prior to preparation of acute slices.

Note: Add 4.5 g glucose, 2 mL 1 M CaCl₂, and 1 mL 1 M MgCl₂ to 1 L of the prepared ACSF stock solution (see materials and equipment ) and saturate with carbogen (95% O₂/5% CO₂) to fully dissolve them on the day before preparing brain slices.

• 2.
  Prepare the incubating chamber:

  • a.
    Put a custom 3D-printed incubating chamber with eight isolated compartments into a 300 mL beaker (Figure 1).
  • b.
    Attach a piece of mesh to the bottom of the incubating chamber to hold brain slices and permit solution exchange (Figures 1B and 1C).
  • c.
    Place the ventilation tube with an infusion needle in the beaker for bubbling solution with carbogen (95% O₂/5% CO₂).
  • d.
    Fill the incubating chamber with ACSF, bubble carbogen (95% O₂/5% CO₂) into the solution and place the whole set in a water bath at 37°C near the dissection area. Allow 10–15 min for the ACSF to reach 37°C (Figure 2A).
• 3.
  Prepare the vibratome:

  • a.
    Insert a single edge razor blade into the blade holder on the vibratome (Leica VT1200S), fasten with a hexagonal screwdriver, and lower the blade into the vibratome buffer tray.

    Note: Rinse and wipe the razor blade with pure ethyl alcohol.

  • b.
    Put crushed ice into the vibratome ice tray, pack the ice around the buffer tray and then fill the ice tray with cold water (Figure 2A).
• 4.
  Prepare the dissection area:

  • a.
    Lay two pieces of paper towels near the vibratome. Place dissecting instruments (large and fine scissors, forceps and scalpel), bent spatula handle (used to scoop brain out of solution), superglue, 1 mL syringe with a bent needle (used to cut and trim brain slices), transfer pipette (with a rubber bulb covering the truncated narrow end), specimen plate, custom-made metal ramps (with a magnet disc glued on the bottom side) on the paper towels (Figures 2A–2C).
  • b.
    Prepare a large box filled with crushed ice. Add 100 mL of ACSF to a small beaker (100 mL) and embed the beaker firmly in the ice (Figure 2D). Bubble the ACSF with carbogen (95% O₂/5% CO₂).

    Note: To reach 0°C for ACSF, sprinkle some salt around the beaker in ice to freeze ACSF, which usually takes 10–15 min.

  • c.
    Switch on the vibratome and place the control panel in a proper position for operation (Figure 2E).

Figure 1.

Assembly of the incubating chamber

(A) Unassembled parts of a custom 3D-printed incubating chamber with 8 compartments.

(B) Nylon mesh with a piece cut off for the incubating chamber.

(C) Top view of the assembled incubating chamber.

(D) Side view of the assembled incubating chamber in a 300 mL beaker.

Figure 2.

Instruments for brain dissection and slicing

(A) Tools and equipment for dissection and slicing were properly arranged in a dissection area.

(B) Top: specimen plate, metal ramps and filter paper; bottom: super glue, transfer pipette, 1 mL syringe with a bent needle and bent spatula handle.

(C) Fine scissors, forceps, scalpel and large scissors.

(D) Ice box with a beaker containing oxygenated ACSF.

(E) Control panel of the vibratome.

Preparation for multiple whole-cell patch-clamp recordings

Timing: 10 min

• 5.Prepare patch recording pipettes: Pull glass patch recording pipettes with a resistance of 4–7 MΩ (about 1.6 ± 0.3 μm) using a pipette puller (P-97, Sutter Instrument) (Figure 3A).

Note: Pull the patch pipettes one day in advance and store them in a custom-made dustproof jar. A thick plate with holes is placed inside the jar to hold the patch pipettes (Figure 3B).

• 6.
  Prepare the bath solution:

  • a.
    Fill ACSF into the reservoir of the gravity-driven perfusion system and permit the solution to flow into the recording chamber (custom-made, diameter 76 mm, depth 7 mm).
  • b.
    Bubble the solution in the reservoir with carbogen (95% O₂/5% CO₂) and heat the solution in the chamber to 32°C–34°C.
  • c.
    Switch on the peristaltic pump to continuously circulate the bath solution back into the reservoir.
• 7.
  Prepare the intracellular solution:

  • a.
    Aliquot and store intracellular solution in 1.5 mL centrifuge tubes (1 mL) at −20°C prior to use.
  • b.
    On the day of recording, thaw a tube of intracellular solution and aspirate it into a 1 mL syringe.
  • c.
    Place a sterile syringe filter (Millipore, 0.22 μm) on the syringe tip and then a custom-made loading tip on the filter.

Note: Keep the intracellular solution on ice and protect it from light throughout the entire experiment.

• 8.
  Start-up equipment:

  • a.
    Switch on all equipment for whole-cell patch-clamp recording, including micromanipulators, patch-clamp amplifiers, digital camera, light source and mercury lamp of the microscope, and computer.
  • b.
    Open software for electrophysiological recording and imaging.

Figure 3.

Preparation of patch recording pipettes

(A) Patch pipettes were pulled using a micropipette puller.

(B) Patch pipettes were stored in a custom-made dustproof jar.

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Chemicals, peptides, and recombinant proteins
Isoflurane	RWD	Cat#: R510-22-10
Paraformaldehyde	Sigma-Aldrich	Cat#: V900894
3,3-Diaminobenzidine	Sigma-Aldrich	Cat#: MKCK2487
Avidin-biotin-horseradish peroxidase complex	Vector Labs	Cat#: PK-6100
Mowiol	Sigma-Aldrich	Cat#: STBH8076
Potassium D-gluconate	Sigma-Aldrich	Cat#: G4500
HEPES	Sigma-Aldrich	Cat#: V900477
MgATP	Sigma-Aldrich	Cat#: A9187
Na3GTP	Sigma-Aldrich	Cat#: V900868
Na2-phosphocreatine	Sigma-Aldrich	Cat#: V900832
Biocytin	Sigma-Aldrich	Cat#: B4261
D-(+)-glucose	Sigma-Aldrich	Cat#: V900392
H2O2	Sigma-Aldrich	Cat#: 1.08597
Triton X-100	Sigma-Aldrich	Cat#: V900502
Na2HPO4	Sigma-Aldrich	Cat#: V900061
NaH2PO4	Sigma-Aldrich	Cat#: V900060
H3PO4	Sigma-Aldrich	Cat#: 695017
NaOH	Sigma-Aldrich	Cat#: V900797
Tris base	Sigma-Aldrich	Cat#: V900483
NaCl	Sigma-Aldrich	Cat#: V900058
CaCl2·2H2O	Sigma-Aldrich	Cat#: V900269
MgCl2·6H2O	Sigma-Aldrich	Cat#: V900020
NaHCO3	Sigma-Aldrich	Cat#: V900182
KCl	Sigma-Aldrich	Cat#: V900068
Experimental models: Organisms/strains
Mouse: C57BL/6Cnc (male/female, ≥P22)	Charles River Laboratories	N/A
Mouse: Vgat-Cre: B6J.129S6(FVB)-Slc32a1tm2(cre)Lowl/MwarJ (male/female, ≥P22)	The Jackson Laboratory	JAX: 028862
Mouse: PV-Cre: B6.129P2-Pvalbtm1(cre)Arbr/J (male/female, ≥P22)	The Jackson Laboratory	JAX: 017320
Mouse: SOM-Cre: Ssttm2.1(cre)Zjh/J (male/female, ≥P22)	The Jackson Laboratory	JAX: 013044
Mouse: VIP-Cre: Viptm1(cre)Zjh/J (male/female, ≥P22)	The Jackson Laboratory	JAX: 010908
Mouse: Ai9: B6.Cg-Gt(ROSA)26Sortm9(CAG-tdTomato)Hze/J (male/female, ≥P22)	The Jackson Laboratory	JAX: 007909
Software and algorithms
Igor Pro	WaveMetrics	https://www.wavemetrics.com/products/igorpro
PEPOI; an Igor-based operation and analysis program for simultaneous electrophysiology, optogenetics and imaging experiments	Home-made; Wang et al.2	Contact UVA Patent Foundation (https://lvg.virginia.edu/) for the end user license
Neurolucida explorer	MBF Bioscience	https://www.mbfbioscience.com/products/neurolucida-explorer
GNU Octave	John W. Eaton	https://octave.org/
Python	Python Software Foundation	https://www.python.org/
RStudio	Posit	https://posit.co/download/rstudio-desktop/
Microsoft Excel	Microsoft	https://www.microsoft.com/en-us/microsoft-365/excel
Other
Borosilicate glass capillaries	Sutter Instrument	Cat#: B200-116-10
Vibratome	Leica Biosystems	Model: VT1200S
Micropipette puller	Sutter Instrument	Model: P-97
Patch-clamp amplifiers	Axon	Model: Axopatch 200B
Patch-clamp amplifiers	A-M Systems	Model: Model 2400
Data acquisition interface boards	HEKA Instruments	Model: LIH 8 + 8
Digital camera	Hamamatsu Photonics	Model: ORCA-FLASH4.0
Upright microscope	Olympus	Model: BX51WI
Micromanipulator systems	Luigs & Neumann	Model: SM-10
Vibration isolation table	TMC	Model: 63-7590M
Neurolucida system	MBF Bioscience	Model: MAC 6000
Upright microscope	Olympus	Model: BX53
Analytical balance	Sartorius	Model: BSA124S-CW

Materials and equipment

ACSF stock solution

Chemical	g/L	Concentration (mM)
NaCl	6.95	119
KCl	0.19	2.5
NaH2PO4	0.12	1
NaHCO3	2.18	26
distilled water	Fill up to 1 L	-

Note: Adjust to pH 7.4 and 305 mOsm. Store at 4°C. It is recommended to prepare fresh ACSF stock solution at least every week.

1 M CaCl₂

Chemical	g/50 mL
CaCl2·2H2O	7.3505
distilled water	Fill up to 50 mL

Note: Store at room temperature (23 ± 2°C). It is recommended to prepare fresh chemical stocks at least every 1 month.

1 M MgCl₂

Chemical	g/50 mL
MgCl2·6H2O	10.1650
distilled water	Fill up to 50 mL

Note: Store at room temperature (23 ± 2°C). It is recommended to prepare fresh chemical stocks at least every 1 month.

Intracellular solution

Chemical	g/100 mL	Concentration (mM)
Potassium D-gluconate	2.8109	120
HEPES	0.2385	10
KCl	0.0298	4
MgATP	0.2030	4
Na3GTP	0.0157	0.3
Na2-phosphocreatine	0.2552	10
Biocytin	0.5000	13.5
distilled water	Fill up to 100 mL	-

Note: Adjust to pH 7.25 and 310 mOsm. Aliquot 100 mL into 1.5 mL tubes and store at −20°C. It is recommended to prepare fresh intracellular solution every 6 months.

0.1 M PB

Chemical	g/1 L
Na2HPO4	10.9
NaH2PO4	3.2
distilled water	Fill up to 1 L

Note: Adjust to pH 7.3. Store at room temperature (23 ± 2°C). It is recommended to prepare fresh PB every 1 month.

0.2 M PB

Chemical	g/1 L
Na2HPO4	21.8
NaH2PO4	6.4
distilled water	Fill up to 1 L

Note: Adjust to pH 7.3. Store at room temperature (23 ± 2°C). It is recommended to prepare fresh PB every 1 month.

0.01 M PBS

Chemical	g/1 L
NaCl	9
0.2 M PB	50 mL
distilled water	Fill up to 1 L

Note: Adjust to pH 7.3. Store at room temperature (23 ± 2°C). It is recommended to prepare fresh PBS every 1 month.

1% H₂O₂ solution

Chemical	mL
30% H2O2 solution	2
0.01 PBS	18

Note: It is recommended to prepare fresh 1% H₂O₂ solution 30 min in advance.

Triton-PB solution

Chemical	Amount
Triton X-100	400 μL
0.01 PBS	20 mL

Note: It is recommended to prepare fresh Triton-PB solution containing 2% Trion X-100 30 min in advance.

DAB solution

Chemical	Amount
3,3-diaminobenzidine	15–30 mg
0.1 M PB	30 mL
30% H2O2 solution	10 μL

Note: It is recommended to prepare fresh DAB solution 30 min in advance.

0.2 M Tris-Cl solution

Chemical	Amount
Tris base	24.22 g
distilled water	Fill up to 1 L

Note: Adjust to pH 8.5 with concentrated HCl.

Mowiol solution

Chemical	Amount
Mowiol	4.8 g
Glycerol	12 g
distilled water	12 mL
0.2 M Tris-Cl solution	24 mL

Note: After heating and dissolving Mowiol solution at 50°C, centrifuge at 5000 × g for 15 min and store the supernatant at −20°C.

Step-by-step method details

Dissection and slice preparation

Timing: 20 min

The MEC slice preparation for electrophysiological recordings followed our previous neocortical studies^3,4 but with adaptation to the MEC.

• 1.
  Dissect brain:

  • a.
    After preparing the dissection area (Figure 2) and oxygenating and freezing the ACSF in the small beaker (Figure 2D), place the mouse in an anesthesia induction chamber and deeply anesthetize the mouse by inhaled isoflurane.

    Note: Under deep anesthesia, the mouse exhibits reduced breathing and no reflexes.

  • b.
    Rapidly decapitate mouse using a pair of large scissors (Figure 2C).
  • c.
    Rapidly cut the skin and skull along the midline using a pair of fine scissors (Figure 2C).
  • d.
    Pluck the skull using forceps (Figure 2C).
  • e.
    Gently extract the brain out of the skull using a scalpel (Figure 2C). Place the brain in the small beaker filled with ice-cold and oxygenated ACSF about 2 min before slicing (Figures 2D and 4A).

    CRITICAL: The brain dissection time after decapitation should be less than 30 s to ensure high-quality brain slices.

• 2.
  Prepare horizontal brain slices of the MEC:

  • a.
    Apply a small amount of superglue to a custom ramp with a 5°–15° tilt angle to form a thin layer (Figure 4B).

    Note: Because the cortex is not plane but convex, the relative angle of the slicing plane to the cortical surface is slightly different across slices.

  • b.
    Transfer the brain from the small beaker to filter paper using the bent spatula handle (Figure 2B) to absorb the solution off the ventral surface of the brain.
  • c.
    Place the brain on the superglue layer with the ventral surface facing down. Pay attention to the position of the brain so that the MEC is erect (Figures 4B–4D).

    Note: Rapidly perform steps b and c to prevent the superglue from drying out.

  • d.
    Place the custom ramp in the vibratome buffer tray. Position the brain with the rostral portion facing the vibratome blade (Figure 4E).

    Note: The ramp is fixed in the tray by the magnet beneath the ramp.

  • e.
    Flood the buffer tray with ice-cold ACSF in the small beaker (Figure 4E).

    CRITICAL: We typically limit the exposure time of the brain in the air to less than 15 s to prevent compromising the quality of the brain slices.

  • f.
    Move the blade to brain surface and cut 300–400 μm thick horizontal slices containing the MEC consecutively (Figures 4D and 4E).

    Note: Ideal brain slices can typically be obtained by setting the slicing speed to 0.14 mm/s and the vibrational amplitude to 0.65 mm (Figure 2E). Vibratome VT1200S (Leica) has a fixed sectioning frequency at 85 Hz. For a vibratome with adjustable sectioning frequency (Ted Pella DTK-Zero 1 for example), set it above 40 Hz.

  • g.
    Use the bent syringe needle (Figure 2B) to trim the slices and separate them from each hemisphere before transferring to the incubating chamber.

    Note: We usually discard the first two or three slices and then collect four slices containing MEC from each hemisphere sequentially.

  • h.
    Transfer the slices using the wide-mouth pipette to the incubating chamber filled with oxygenated ACSF prewarmed to 37°C and recover for 10–15 min (Figure 4G).

    Pause point: Remove the incubating chamber from the water bath and leave it at room temperature (23 ± 2°C) for an additional 0.5–1 h (Figure 4F) until slices are transferred to the recording chamber.

Figure 4.

Dissection and slice preparation

(A) An extracted brain was immerged in the small beaker filled with ice-cold and oxygenated ACSF.

(B) The brain was glued on a custom ramp with a 5°–15° tilt angle.

(C) Schematic showing the position of the MEC (green) in a mouse brain.

(D) Schematic showing that horizontal slices were cut.

(E) The brain was cut into horizontal slices containing the MEC.

(F) Trimmed slices in the incubating chamber.

(G) Slices in the incubating chamber were incubated at 37°C for 10–15 min.

Perform multiple whole-cell patch-clamp recordings

Timing: 1.5 h

• 3.
  Transfer slice to the recording chamber (Figures 5A and 5B):

  • a.
    Use a wide-mouth pipette to transfer a brain slice from the incubating chamber to the recording chamber.
  • b.
    Submerge the horizontal slice in the recording chamber.
  • c.
    Take care to ensure that its dorsal surface is kept upwards and position the slice with the caudal end toward the operator (Figure 5B).

Note: Control the ACSF flow rate (3–4 mL/min) in the recording chamber to avoid interference caused by surface fluctuations.

• 4.Check and select slices under the upright microscope with a 40× water-immersion lens (Figures 5A and 5B).

CRITICAL: When checking slices one by one in order under the microscope, we can observe that the orientation of apical dendrites of pyramidal cells (PCs) gradually changes from tending toward the upper slicing plane to tending toward the lower slicing plane. There is one intermediate slice preserving the most apical dendrites parallel to the slicing planes and therefore the most neurons. In this way, ideal slices for electrophysiological recordings are obtained.

Note: Regular Kohler illumination was used for visualization of the neurons. IR-DIC can improve visualization and is especially useful for patch-clamp recording from a dendrite or an axonal bleb, but regular Kohler illumination is enough for soma recording.

• 5.Stabilize the selected slice with fine fishing lines attached to a steel ring (Figure 5B).
• 6.Identify MEC superficial layers (i.e., layers 1–3; L1-L3 thereafter for short) based on their differentiations in cytoarchitecture as well as anatomical position (Figures 5E–5G).

Note: Compared to L1, L2 has a sharply increased soma density. L2PCs are organized in islands,^5,6 which encroach upon adjacent L1. These island PCs are not treated as L1 excitatory neurons by defining L1/2 border along the upper edge of islands. In contrast to L2, L3 contains smaller and loosely arranged PCs, and is brighter under transparent microscope. The deeper border of L3 is the lamina dissecans containing sparse cells. L1/L2 border is typically 100–160 μm from the pia, while L2/L3 border is typically 240–320 μm from the pia.

• 7.Fill multiple patch pipettes with intracellular solution without bubbles. Insert and fasten the pipettes in recording electrode holders.

Note: Ensure that Ag wires of headstage amplifiers are suitably coated with AgCl prior to any recording. Avoid overfilling the intracellular solution so that it just submerges the Ag wires.

• 8.Apply weak positive pressure to each pipette using a syringe connected to the corresponding recording electrode holder via plastic tubing.
• 9.Observe with the 40× water-immersion lens. Steer the patch pipettes using micromanipulators and park them above the slice and near the cells to be recorded.

Note: The current protocol is not subject to the mouse line. However, some transgenic mouse lines (e.g. Vgat-Cre/Ai9, PV-Cre/Ai9, SOM-Cre/Ai9 and VIP-Cre/Ai9 mice; see institutional permissions and key resources table) were used to facilitate searching interneurons. In these mice, all or subpopulations of interneurons were labeled with tdTomato. Randomly select healthy tdTomato-expressing and non-expressing cells in superficial layers of the MEC. Avoid selecting cells close to the surface of the slice.

• 10.Once a group of healthy target cells are selected, apply stronger positive pressure to one of the patch pipettes and move it near the membrane of one of the target cells until a small dimple appears on the cell membrane.
• 11.Immediately release the positive pressure by detaching the syringe from the plastic tubing. Swiftly reconnect the syringe with the tubing and apply a slow and continuous negative pressure.

Note: Negative pressure sucks the cell membrane onto the surface of the pipette tip to form a high resistance seal (“giga-seal”).

• 12.Set the holding voltage to −60 mV to approach the resting membrane potential.
• 13.Apply brief negative pressure pulses to rupture the cell membrane within the pipette.

Note: Once successful, wider membrane capacitive currents in response to the test pulses and a sharp decrease in membrane resistance can be observed.

• 14.Once whole-cell configuration is established for one cell, repeat steps 10–13 for other target cells one by one, but avoid interfering with cells that have already been recorded (Figures 5C–5G).

Note: The process of steps 10–14 should be fast enough. For example, the time spent in establishing whole-cell configuration for eight cells is usually less than 20 min to avoid any possible deterioration in the initially recorded cells. Less experienced experimenters are suggested to take training step by step, i.e. performing double recordings first, then quadruple, and finally octuple recordings. Principal neurons (stellate cells and pyramidal cells) can be patched first, and then interneurons, as the former typically remain healthy for a relatively longer time after recording.

• 15.
  Measure electrophysiological properties and identify connections between neurons in current-clamp mode (Figure 6):

  • a.
    Inject depolarizing and hyperpolarizing step currents of different amplitudes (40 steps, −200–580 pA for 500 ms with a step interval of 20 pA) into the recorded cells simultaneously. Sub- and supra-threshold electrophysiological properties can be measured and calculated from the recorded responses (Figures 6A and 6B).
  • b.
    Hold neurons at −55 ± 3 mV in current-clamp mode. Inject depolarizing currents (2 nA for 2 ms at 0.01–0.1 Hz) into each member of a set of recording neurons in turn to evoke action potentials (APs; Figures 6C and 6D). Monitor the unitary excitatory and inhibitory postsynaptic potentials (uEPSPs and uIPSPs) produced by other neurons (Figures 6C and 6D).
  • c.
    Inject hyperpolarizing currents (−200 pA for 200 ms at 0.01–0.1 Hz) into the presynaptic neuron to evoke postsynaptic responses to identify electrical coupling (Figures 6C and 6E).

Note: Recorded traces of postsynaptic potentials (PSPs) shown in Figure 6 are averages of 20–100 consecutive episodes and these averages are also used to calculate amplitudes and other kinetic properties of PSPs as well as electrical coupling coefficient.

• 16.Record neurons for a long time (usually ≥ 1 h) to allow biocytin to diffuse into distal axons. Once multiple whole-cell recordings are finished, carefully remove the electrodes from all the recorded neurons.

CRITICAL: Recording pipettes should be withdrawn very slowly for the initial ∼10 μm to maintain the seal and then retracted rapidly. It is ideal that a patch of the cell membrane adheres to the pipette and is stretched out of the soma. After the patch detaches from the soma, the membrane can reseal, leaving the cell intact.

• 17.Quickly transfer the brain slice using a pipette to 2–3 mL of 4% paraformaldehyde in 0.1 M phosphate buffer (PB). Place only one brain slice in a well of the tissue culture plate. Label each brain slice with a unique ID number on its corresponding well.

Note: Once the slice is removed from the perfusion ACSF, it should be immediately fixed. Gently shake to ensure that the slice does not adhere to the well wall.

CRITICAL: Avoid dilution of 4% paraformaldehyde solution by ACSF in the pipette during brain slice transfer.

• 18.Transfer the slices to 4°C for storage for at least 48 h.

CRITICAL: To ensure the health of the brain slices, we usually record only one or two slices per mouse.

Pause point: Slices can be stored in paraformaldehyde for up to one week. Longer storage time may require cryoprotectant.

Figure 5.

Multiple patch-clamp recording system

(A) The setup for octuple whole-cell recordings.

(B) Recording chamber with a brain slice stabilized by a steel ring with fishing lines.

(C) Simultaneous octuple whole-cell recordings.

(D) Images of the recorded neurons and the acquired response traces were displayed on two computer screens.

(E) Image of octuple whole-cell recordings from a MEC slice under the 10× objective.

(F) Region within the black box in (E) was imaged under the 40× objective. Cell recorded by electrode 3 was focused (red arrow).

(G) Same with (F) but with a different focal plane. Cell recorded by electrode 2 was focused (red arrow).

Figure 6.

Example of simultaneous multiple patch-clamp recordings

(A) Image showing simultaneous octuple whole-cell recordings under a 10× objective.

(B) Traces of simultaneously recorded neurons in response to hyperpolarizing, near-threshold, and suprathreshold current injections.

(C) Connection diagram of the eight recorded neurons. Bullet-headed lines show inhibitory chemical connections symbolically. Arrow lines show excitatory chemical connections symbolically. Zigzag lines show electrical coupling symbolically.

(D) APs elicited by current injection (I_inj) in presynaptic neurons and uIPSPs/uEPSPs evoked in postsynaptic neurons. Scale bars apply to all recording traces. The vertical scale bar shows amplitudes of I_inj (nA), APs (mV), and uIPSPs/uEPSPs (mV).

(E) Hyperpolarization elicited by negative I_inj in presynaptic neurons and voltage responses evoked in postsynaptic neurons. Scale bars apply to all recording traces. The vertical scale bar shows amplitudes of I_inj (nA), hyperpolarization (mV), and voltage responses (mV).

Histological staining

Timing: 1–2 days

To reveal cell morphology, the avidin-biotin-peroxidase method was employed following the procedures of our previous studies.^3,4 Conduct all the following steps at room temperature (23 ± 2°C).

• 19.Carefully extract the slices fixed in 4% paraformaldehyde solution using a pipette.
• 20.Wash the slices in a six-well tissue culture plate for 10 min, with each well containing fresh 0.01 M PBS (3–5 mL). Repeat this step three times.

Note: At this and the following steps, always place only one slice in each well and retain numbering order. Before step 28, always place the tissue culture plate containing slices on a shaker when waiting for time.

• 21.Transfer the slices into a 24-well tissue culture plate using a pipette and incubate for 30 min, with each well containing 1% H₂O₂ solution (1 mL).

Note: To reduce the background color, the concentration can be increased up to 3%.

• 22.Use a pipette to transfer the slices again to a six-well tissue culture plate and wash for 10 min, with each well containing 0.01 M PBS (3–5 mL). Repeat this step three times.

Note: Ensure that no bubbles remain on the slices after washing.

• 23.Transfer the slices to a 24-well tissue culture plate and permeate for 30 min to prevent non-specific binding, with each well containing Triton-PB solution (1 mL).
• 24.
  Make avidin-biotin-peroxidase complex (ABC) solution using an ABC-HRP Kit (Vector, PK-6100):

  • a.
    Add 8 drops of solution A (avidin) and 8 drops of solution B (biotinylated HRP) in 20 mL of 0.01 M PBS solution and shake by hand, and then set for at least 15 min to form a complex.
  • b.
    Then add 200 μL Triton X-100 and mix thoroughly, and set for at least 15 min to form ABC solution.

CRITICAL: Freshly prepare ABC solution and keep it in the dark before use.

• 25.Transfer the slices to a 24-well tissue culture plate containing ABC solution and incubate for 3 h in the dark, with each well containing 1 mL ABC solution.

Pause point: To further increase permeability, the incubation time of slices can be extended to 12 h.

• 26.Use a pipette to transfer the slices again to a six-well tissue culture plate and wash for 10 min, with each well containing 0.01 M PBS (3–5 mL). Repeat this step three times.
• 27.Rinse with 0.01 M PBS solution for 1 h for sufficient cleaning.
• 28.Preparation of diaminobenzidine (DAB) solution at least 30 min in advance due to the difficulty of DAB dissolution. Avoid light during the preparation process.
• 29.Transfer the slices to a 24-well tissue culture plate containing DAB solution and react for 5–10 min to observe the staining of cells in the recorded area.

Note: Observe the staining degree of the slices under a stereo microscope and stop the DAB reaction when the required staining intensity is reached (Figures 7A–7C).

• 30.Wash the slices in a six-well tissue culture plate for 10 min, with each well containing fresh 0.1 M PB (3–5 mL). Repeat this step three times.
• 31.Transfer the slices using a pipette and mount them on a glass slide.
• 32.Use a micropipette and filter paper to remove excess 0.1 M PB solution and permit the slices to dry in air for a while, and then seal the slices with Mowiol solution.
• 33.Cover a coverslip and air dry at room temperature (23 ± 2°C).

CRITICAL: We usually mount 6 slices on each slide and label each slice with an ID number to avoid confusion in order (Figure 7E). Avoid generating bubbles during the sealing process.

Pause point: Air dry the mounted slices for at least 7 days before morphological reconstruction.

Figure 7.

Morphological reconstruction of neurons

(A) Example of slice with ideal intensity of DAB staining.

(B) Example of slice with insufficient DAB staining.

(C) Example of slice with excessive DAB staining.

(D) Neurolucida system for morphological reconstruction.

(E) Stained slices mounted on a slide.

(F) The slide with slices was placed on the motorized stage and observed under a 60× oil-immersion lens.

(G) Neurolucida software user interface showing a MEC slice overlaid with reconstructed neurons.

(H) Schematic showing the key steps of axonal density analysis. Reconstructed neurons are adapted with permission from Shi et al.¹

Morphological reconstruction

Timing: 1–2 days

After Mowiol got dry, the morphologically recovered cells were examined and reconstructed using the Neurolucida on an upright microscope (Figure 7D).

• 34.Switch on the computer, the microscope and the control box of the motorized stage. Allow 15–20 s for an initial positioning calibration, and then open the Neurolucida software.

Note: Avoid touching the motorized stage while waiting for the initial positioning calibration.

• 35.Place a glass slide with embedded brain slices (Figure 7E) on the motorized stage and locate the target brain area under a low magnification objective (4× or 10×).

CRITICAL: If it is impossible to complete the depiction of a slice in one session, a deviation of the slide position may happen at the next time of depiction, leading to a mismatch in the cell morphology. Therefore, it is necessary to ensure that the slide is perfectly mounted in the slot every time to avoid possible positional deviation.

• 36.Drop a small amount of oil onto the slide and use a 60× oil-immersion lens to observe neurons (Figure 7F).
• 37.
  After clicking the mouse on the screen to place the reference point for drawing, move the joystick and adjust the fine adjustment knob to trace the axonal and dendritic branches at different fields of view and depths, and then manually draw the cell body, dendrites and axon of the neuron by clicking the mouse (Figure 7G).

  CRITICAL: We usually place the reference point of the depiction at a unique point like the intersection of the cell body surface and one of the dendrites to ensure rapid alignment when resuming depiction.

  • a.
    Reconstruct cell body:

    • i.
      Determine the range of the cell body on the Z-axis by adjusting the fine adjustment knob.
    • ii.
      First outline the cell body by depicting the contour of the largest cross-section of the cell body.
    • iii.
      Then sequentially depict the contours of cross-sections of the cell body at difference focal planes (with a 0.2 μm interval on Z-axis), until cross-sectional contours of the complete cell body are depicted.
  • b.
    Reconstruct dendrites:

    • i.
      Given that a neuron typically has multiple dendrites, trace them in order.
    • ii.
      Place the cursor over the cell body surface at the location where a dendrite originates.
    • iii.
      Click to place the first point. Record the diameter of the dendrite by adjusting the mouse scroll wheel to match the circular cursor to the diameter.
    • iv.
      Continue to place points in short increments to depict the dendrite.

      Note: Adjust the focus and cursor diameter before placing each point. When reaching a branch point, mark it so that the system will automatically return to the point to begin the next one after completing one dendritic branch. When reaching the end of the branch, mark the end.

  • c.
    Reconstructing axon: Place the starting point at the axonal hillock and then follow a similar procedure to step b to depict the axon.

    Note: Neurons typically have only one axon, arising from the cell body or one dendrite. Axons and dendrites can be distinguished by their morphological features. Axons are usually longer than dendrites. The diameter of axons is usually small and relatively constant, while the diameter of dendrites gradually decreases from the cell body to the branch ends. In addition, dendrites of principal neurons usually have dendritic spines, making them distinct from axons.

• 38.After completing the reconstruction of one neuron (2 h typically needed for one neuron), repeat step 37 for other stained neurons in the slice.

Pause point: Morphological reconstruction is a time-consuming work. One can take a short break at any time or decide to pause depiction and resume on another day.

• 39.After all neurons in the slice are reconstructed, outline the entire brain slice and delineate boundaries between different areas/layers.
• 40.Set neurons to different colors, add the scale bar, and export data files.
• 41.Turn off the device switches, remove the glass slide, and wipe the lens with ethyl alcohol.

Expected outcomes

This protocol is specifically designed for simultaneous multiple patch-clamp recordings and morphological reconstruction of MEC neurons, as validated in our previous study.¹ However, with slight modification, it should also be used in research of other brain areas like the neocortex.^3,4 With the current protocol, it is expected to obtain sub- and supra-threshold electrophysiological properties of diverse MEC neurons and evaluate uEPSP/uIPSP and electrical coupling between these neurons. Moreover, this protocol also enables researchers to obtain morphological features of the neurons. With these data, different cell types of neurons can be identified based on their morphological and electrophysiological features and cell-type-specific chemical and electrical microcircuits can be demonstrated.¹

Quantification and statistical analysis

Morphological analysis

• 1.Morphological features, including number of axonal nodes, axonal length, number of primary dendrites, dendritic length and soma volume, were measured using Neurolucida Explorer (MBF Bioscience).
• 2.Tangential and vertical ranges of axons and dendrites were measured manually. Varicosity density was obtained by counting varicosities on a segment of axon (e.g., initial 100 μm from the cell body) and then calculating the linear density. Dendritic diameter was calculated as the average of diameters of all the primary dendrites on each cell, which were measured in Neurolucida software.
• 3.Axonal length density was analyzed using GNU Octave. To obtain 2D axonal length density maps, we first flattened the axonal arborization of a neuron by projecting axonal segments to tangential and vertical coordinates as necessary and then divided XY plane into pixels of 20 μm × 20 μm. Axonal length density in each pixel was calculated to form a 2D axonal length density map (Figure 7H).
• 4.To represent cell-type-specific axonal density maps, all maps of the same cell type were spatially aligned by setting their somata at the origin of coordinates and added together, followed by Gaussian smoothing (standard deviation σ = 20 μm; Figure 7H). To analyze (sub)laminar distribution of axons, all maps of the same cell type were spatially aligned by setting their somata at the origin of X-axis and the vertical position of pia or laminar boundary at the origin of Y-axis (Figure 7H).
• 5.To calculate 1D axonal length density curves, values in pixels of the 2D density map were added horizontally or vertically. After the mean density curve was calculated, the full width at half maximum (FWHM) of the curve was measured when the density was halfway between the peak and zero.

Electrophysiology analysis

Various electrophysiological properties are measured from the recorded traces responding to depolarizing and hyperpolarizing step currents in Igor Pro. These properties together with the above morphological properties can be further used in other sophisticated analysis like clustering and classification.

• 6.Resting membrane potential (RMP) can be determined from the trace without injecting current, but we usually recorded it soon after gaining whole-cell access during experiments. Membrane time constant and input resistance were determined from the average membrane response to small hyperpolarizing currents (−20 pA, 500 ms, 20 sweeps). Membrane time constant was obtained by fitting an exponential function to the average trace, and input resistance was calculated using Ohm’s law. Sag ratio and time constant to sag minimum was determined from the membrane response to a large hyperpolarizing current (−200 pA, 500 ms). Sag ratio was calculated by dividing the maximum voltage deflection (the sag minimum) from the steady state by the steady state hyperpolarization voltage during the pulse. Time constant to sag minimum was obtained by fitting an exponential function to the response trace.
• 7.Rheobase current was determined as the amplitude of the 500 ms depolarizing current step that evoked the first AP. All AP properties were measured on the first AP at rheobase current. AP threshold was determined when the slope of the spike increases suddenly (e.g., the moment when the slope exceeds 10 V/s and will also exceed 20 V/s and 30 V/s at two sequential time points). AP amplitude was calculated as the difference between the spike peak and the AP threshold. AP FWHM was measured when the voltage is halfway between the AP threshold and the AP peak. AP latency was determined as the delay of the first spike peak relative to the onset of the rheobase current step. AP rise time was calculated as the duration when the spike rises from 20% to 80% of its total height. AHP amplitude was measured as the difference between the AP threshold and the largest hyperpolarization voltage immediately following the AP. AHP latency was measured as the time difference between the AP threshold and the AHP minimum. AHP FWHM was measured when the voltage is halfway between the AP threshold and the minimum AHP value.
• 8.Properties of firing pattern were obtained from responses to depolarizing step currents of different amplitudes (0–580 pA, 500 ms, step interval 20 pA). Maximum frequency, coefficient of variation (C.V.) of interspike interval (ISI) and adaptation index were measured from response traces between rheobase current and the current evoking the most spikes. Maximum frequency was determined as the maximum of firing frequencies among these traces while C.V. of ISI and adaptation index were represented by the median across the traces. Adaptation index was defined as the ratio of the sum of the first three ISIs to the sum of the last three ISIs.

Connection analysis

• 9.Connection probability was calculated as the ratio between the total number of observed connections and the total number of tested connections. For each pair of neurons, there is one possible electrical connection and two possible chemical connections, so the number of chemical connections tested is twice the number of electrical connections tested.
• 10.PSP properties were measured from average PSP traces (averages of 20–100 consecutive episodes). PSP Amplitude was calculated as the difference between the PSP peak/trough and the baseline. PSP Latency was measured as the time difference between the onset of the PSP and the moment when the inward current of the presynaptic neuron reaches the maximum (i.e., the slope of the AP is the maximum). Rise time of PSP was measured as the difference between the time points when the response reaches 10% and 90% of the peak during the rising phase, respectively. FWHM of PSP was measured when the response is halfway between the baseline and the PSP peak. Decay time of PSP was obtained by fitting an exponential function to the response trace during the decay phase.
• 11.Coupling coefficient of an electrical synapse was calculated as the ratio of the voltage change in the postsynaptic neuron to the voltage change in the presynaptic neuron.

Limitations

In this protocol, we described how to record a group of MEC neurons using a multiple patch-clamp recording system. Simultaneous multiple whole-cell recordings enable one to test more connections from a relative smaller population of neurons (e.g., 56 from one set of octuple recordings vs. 8 from four sets of double recordings). However, to achieve multiple whole-cell recordings, experimenters need more training to get used to the setup and gain necessary skills. The recorded neurons need be carefully selected before recording so that all electrodes can reach the target neurons without interfering with others. Since each electrode need penetrate brain tissue to reach the target neuron, it is expected that multiple whole-cell recordings may induce more damage to the slice. Therefore, excessive movements of electrodes in tissue should be avoided.

One main limitation of in vitro whole-cell recordings is that slicing inevitably causes partial damage on neurites. Therefore, slicing procedure is critical for preservation of neuronal morphology. To obtain ideal MEC slices for whole-cell recordings, we tilted the brain with ramps (5°–15°) before slicing and selected brain slices with the most intact apical dendrites of pyramidal cells. A similar slicing angle has also been used by other leading laboratories of the MEC research.^7,8,9 Furthermore, during electrophysiological experiments, we only recorded healthy cells below the surface, which were less damaged. As we know, excitatory neurons have axonal collaterals that are usually sparse locally and project to long distance. Slicing may influence their appearance seriously. Thus, pyramidal cells are usually not classified by their axonal morphologies. Instead, dendritic morphologies, especially morphologies of apical dendrites, are used.¹⁰ However, interneurons are different. Because of their dense axonal arborizations, morphological features can be preserved after slicing.¹¹ Moreover, axonal morphologies are very distinct between different types of interneurons. Thus, interneurons are usually classified by their axonal morphologies.^{8,11,12,13,14} In addition, distinct electrophysiological features usually provide further evidence for the classification of the morphological types. Nonetheless, one important part uncovered by the current protocol is genetic identification of recorded neurons. Single-cell RNA sequencing techniques including Patch-seq have recently been used to define transcriptomic cell types across the neocortex and hippocampal formation.^{11,15,16,17,18,19,20} It will be more powerful to adopt Patch-seq in the current protocol.

Troubleshooting

Problem 1

Unhealthy brain slices (steps 1, 2).

Potential solution

• •Thoroughly clean all tools and containers during the preparation phase to ensure that the brain slices are in a clean environment throughout the entire experimental process.
• •Check the water quality, pH, and osmolarity of ACSF before the experiment.
• •The key step in obtaining healthy slices is to dissect as quickly as possible. Extract the brain after decapitation within 30 s and avoid damage to the brain caused by dissecting instruments.
• •Ensure that the extracted brain is quickly placed in ice-cold and oxygenated ACSF. Both conditions are crucial for slice health.
• •The action of gluing the brain on a ramp should be as fast as possible. Avoid prolonged exposure of the brain to the air.
• •Ensure that oil on the vibratome blade is cleaned with pure ethyl alcohol, maintain the appropriate angle of the blade, and set the slicing speed and the vibrational amplitude of the blade to proper values.
• •Adjust the incubation time of slices at 37°C and room temperature (23 ± 2°C).
• •Inspect the slices sequentially under a microscope and select the best one to record as described in this protocol.

Problem 2

Unable to achieve high-quality simultaneous multiple whole-cell recordings (steps 3–14).

Potential solution

• •Avoid selecting cells close to the surface of the slice. We typically record multiple healthy cells that are more than 20 μm beneath the surface of the slice.
• •When selecting the target cells, pay attention to their locations and avoid interference between electrodes.
• •Among the target cells, first patch those that are deeper beneath the surface than others, which should help in reducing interference between electrodes.
• •Patch principal cells prior to interneurons.
• •Avoid excessive perturbations in the tissue as an electrode approaches the cell.

Problem 3

Fail to induce any PSPs between neurons (step 15).

Potential solution

• •Since the connection probability between neurons decreases sharply with distance, first try to record neurons not too far away from each other.
• •Record cells that are deep enough beneath the surface of the slice to avoid loss of connections caused by severe axonal damage.

Problem 4

Unsuccessful slice staining (steps 19–33).

Potential solution

• •Pull recording electrodes with a proper tip size. Small electrode tip may cause insufficient perfusion of intracellular solution containing biocytin after breaking the cell membrane.
• •Record neurons for a long time (typically ≥ 1 h) to allow biocytin to diffuse into distal axons.
• •The preparation of ABC solution requires first mixing A and B solutions to form a complex, and then dissolving in a solution containing Triton X-100 to form an ABC mixed solution.
• •Enough diaminobenzidine must be well dissolved in 0.1 M PB solution.

Problem 5

Positional deviation during neuronal reconstruction (steps 34–41).

Potential solution

• •Ensure that the Neurolucida system performs initial positioning calibration properly before use. Calibrate the system if necessary.
• •Ensure that the stained slices are thoroughly air dried before starting reconstruction.
• •Pay attention to mount the glass slide perfectly in the slot of the stage when resuming depiction.
• •Conduct manual tracing as precisely as possible.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Guangfu Wang (wangguangfu@hit.edu.cn).

Technical contact

Technical questions on executing this protocol should be directed to and will be answered by the technical contact, Guangfu Wang (wangguangfu@hit.edu.cn).

Materials availability

This study did not generate new unique reagents.

Data and code availability

This study did not generate new datasets or code. Source data for figures in the paper will be shared by the lead contact upon request. Our morphological files of diverse MEC neurons are available at NeuroMorpho.Org (“Wang_G″ archive). 3D CAD files for custom-made parts used in this protocol will be shared by the lead contact upon request.