protocol

Culturing pyramidal neurons from the early

postnatal mouse hippocampus and cortex
Gerard M J Beaudoin III1,3,4, Seung-Hye Lee1,3, Dipika Singh2, Yang Yuan2, Yu-Gie Ng1, Louis F Reichardt1 &
Jyothi Arikkath2,3
1
Department of Physiology, University of California–San Francisco (UCSF), San Francisco, California, USA. 2Munroe-Meyer Institute, University of Nebraska
Medical Center (UNMC), Omaha, Nebraska, USA. 3These authors contributed equally to this work. 4Present addresses: Neurosciences Institute, University of Texas
at San Antonio, San Antonio, Texas, USA (G.M.J.B.); Genentech, Inc., South San Francisco, California, USA (S.-H.L.). Correspondence should be addressed to
J.A. (jyothi.arikkath@unmc.edu).

Published online 30 August 2012; doi:10.1038/nprot.2012.099

The ability to culture and maintain postnatal mouse hippocampal and cortical neurons is highly advantageous, particularly for
studies on genetically engineered mouse models. Here we present a protocol to isolate and culture pyramidal neurons from the
early postnatal (P0-P1) mouse hippocampus and cortex. These low-density dissociated cultures are grown on poly-l-lysine–coated
glass substrates without feeder layers. Cultured neurons survive well, develop extensive axonal and dendritic arbors, express
neuronal and synaptic markers, and form functional synaptic connections. Further, they are highly amenable to low- and high-
© 2012 Nature America, Inc. All rights reserved.

efficiency transfection and time-lapse imaging. This optimized cell culture technique can be used to culture and maintain neurons
for a variety of applications including immunocytochemistry, biochemical studies, shRNA-mediated knockdown and live imaging
studies. The preparation of the glass substrate must begin 5 d before the culture. The dissection and plating out of neurons takes
3–4 h and neurons can be maintained in culture for up to 4 weeks.

INTRODUCTION
The hippocampus and cortex have fascinated neuroscientists for
years because of their precise organizations and functional roles
in cognition, learning and memory1. As with most structures in
the brain, the complexity of these structures makes it challenging
to analyze and manipulate easily in vivo. Cell lines derived from
­central nervous system precursors have limitations because the
neurons derived from these lines fail to recapitulate characteristics
of central neurons, including the ability to form well-defined axons,
dendrites and synapses. Instead, primary cell culture techniques
have been successfully developed to study these neurons in vitro.
Primary neuronal cell cultures
One of the most well-established and widely used techniques for
the study of hippocampal and cortical pyramidal neurons has been
the primary dissociated cell culture system pioneered by Kaech
and Banker2 for the culture of embryonic rat neurons. This culture
system allows neurons to be cultured in vitro in a far less complex
environment than that used in vivo, making them highly accessible
to manipulations and observations. The vast majority of cells in
these preparations are excitatory pyramidal neurons, with inhibi-
tory neurons representing about 18–20% of the population. Several
studies, including those from our group3–8, have shown that dis-
sociated embryonic neurons maintained in culture undergo dis-
tinct stages of differentiation and form well-established synaptic
connections9–11. These cells can be grown and maintained in single
layers on glass coverslips or plated at higher densities for biochemi-
cal studies. Furthermore, coupled with advances in imaging tech-
nologies and manipulation of gene expression, these cultures have
allowed unprecedented views into the intricate inner workings of
neuronal cells. Thus, this technique has been well exploited to dis-
sect out various aspects of molecular and cellular mechanisms that
underlie neuronal morphogenesis and connectivity12.
The culture of postnatal versus embryonic neurons is advanta-
geous for several reasons. It reduces the necessity of killing ­animals.
It is possible to euthanize one or a few animals from one litter and
save the rest for further breeding, which is not possible with embry-
onic tissue. Further, the mother need not be killed and can be used
for further breeding or other purposes. Moreover, the technique
makes it possible to (i) study neurons from genetically engineered
mice that are early postnatal lethal; (ii) control expression of genes
in a temporally controlled manner using RNAi-­mediated knock-
downs or tetracycline-regulated expression plasmids in wild-type
neurons; or (iii) delete genes in individual neurons using transfec-
tion of a vector expressing Cre recombinase in neurons from floxed
mice. The ability to delete genes from small subsets of neurons
makes it possible to distinguish cell-autonomous versus non-cell-
autonomous effects of individual genes and their disease counter-
parts on neuronal structure and function13,14.
Culturing postnatal mouse neurons has proved to be more chal-
lenging than culturing embryonic mouse or rat neurons. Neurons
at earlier developmental stages, as in the case of embryos, are less
susceptible to damage during the processing required for their cul-
ture. This is partly because their neuronal processes are relatively
less well developed in comparison with postnatal animals15,16. In
practical terms, the meninges are also more adherent to the tissue
in postnatal animals. Thus, postnatal cultures tend to be less robust
and have a lower yield of healthy viable cells in comparison with
embryonic rat or mouse hippocampal and cortical neurons and
have a higher proportion of non-neuronal cells.
Types of primary neuronal cultures
Two types of cultures have been described for culturing primary
neurons from rats and mice2. In one method, the neurons are
grown sandwiched on an astrocytic feeder layer17. This allows for
the neurons not to be in direct contact with the astrocytes, but to
be exposed to factors secreted by astrocytes into the medium. The
alternate protocol does not involve the use of the feeder layer, and
neurons are maintained in serum-free medium (B27) supplemented

nature protocols | VOL.7 NO.9 | 2012 | 1741

 protocol
with cofactors necessary for neuronal growth and maintenance18.
Both techniques have been widely used successfully. The genera-
tion of the astrocyte feeder layer can be cumbersome, but it can
also yield high-quality primary neurons and is advantageous for
studying neuron-glia interactions19. Some laboratories have also
pioneered the use of the NS21 supplement as an alternative to
the use of B27 supplement for both rat20 and embryonic mouse
cultures21 and postnatal mouse cultures22.
Optimized protocol for culture
Here we describe optimized protocols for the culture
(PROCEDURE) and transfection (Box 1) of postnatal mouse hip-
pocampal and cortical neurons. An overview of the procedure is
shown in Figure 1. These protocols are based on the widely used,
non-feeder layer–based, rat embryonic culture procedures with
modifications that improve the health, viability and yield of post-
natal mouse neurons4,7,8. These cultures can be prepared by pooling

  1 | Transfection of neurons ● TIMING 1 h

Box
Neurons cultured as described in the main PROCEDURE are easily amenable to transient transfections. For all transfections, we prepare
the DNA following the manufacturer’s instructions. We have had success with using the EndoFree DNA preparation kit from Qiagen.
For low-efficiency transfections, such as those required for single-neuron imaging studies, we routinely take advantage of a lipid-based
technique. With this technique, transfection efficiencies vary from greater than 10% in neurons transfected before DIV 7 to 0.5–5%
in neurons transfected at DIV 8–14. We routinely use transfections for overexpression and shRNA-mediated knockdown. For shRNA-
­mediated knockdown, we have had success with the pSUPER.gfp+neo (Oligoengine) vectors.
For high-efficiency transfections, such as those required for biochemical studies, we take advantage of an electroporation-based
technique, commonly called nucleofection. The obvious disadvantage of the electroporation-based technique that we have described
© 2012 Nature America, Inc. All rights reserved.

is that the neurons can only be subjected to the treatment before plating and establishment of neurites. However, new technologies
available from Lonza Biosciences claim to support the nucleofection of established neurons in a 96-well format. Coating of DNA onto
magnetic beads also permits transfection of plated neurons26. Such transfections might be very useful for library screens and high-
efficiency transfections of mature neurons. More recently, we have successfully used recombinant lentiviruses to infect rat neurons in
culture, and these could be adapted for use in mouse neurons8. In addition, it might be possible to use the nucleofection technique
to introduce plasmids in inducible vectors (e.g., the tetracycline-regulated expression vectors) at plating and induce or suppress the
expression of the gene of interest at much later stages in culture, thus allowing temporal control of gene expression in large cell
populations. Clearly, advances in such technologies will allow further exploitation of this neuronal cell culture technique to assess
various aspects of neuronal physiology, structure and function.
We present here a low-efficiency transfection protocol using Lipofectamine. For high-efficiency transfections, we have successfully
overexpressed and knocked down protein levels using the nucleofection technique (Lonza Biosciences). We routinely use low-efficiency
transfections for imaging studies. This allows the delineation of entire arbors from individual neurons. High-efficiency transfections are
suitable for biochemical studies.
Low-efficiency transfection of neurons for imaging
1. Adjust the volume of the culture medium in each well of a 12-well plate containing growing neurons to 1 ml.
2. Add 1 µl of Lipofectamine to an Eppendorf tube containing 25 µl of Neurobasal medium. Incubate for 5 min.
3. In a separate tube, add 0.1 to 1 µg of DNA for transfection to 25 µl of Neurobasal medium. We prefer using a high concentration
of DNA for transfections (1–3 mg µl − 1).
4. Combine solutions from Steps 2 and 3 with gentle pipetting.
 CRITICAL STEP It is important that the solutions be combined with gentle mixing so as to not disrupt the lipid-DNA complexes.
5. Incubate in the laminar flow hood for 20–30 min at room temperature.
6. Add the mix to each well and swirl gently.
7. Return cells to the incubator.
8. After 24 h, replace half of the medium in each well with fresh maintenance medium at 37 °C. Continue to maintain the neurons
as described in PROCEDURE Step 42.
High-efficiency transfection:
1. Prepare Nucleofector working solution by adding supplementary solution to the main reaction buffer obtained from
the manufacturer.
2. Pellet 1 to 2 million cells obtained after PROCEDURE Step 39 gently (500g for 5 min) in plating medium.
3. Aspirate the medium carefully and resuspend the cells in 100 µl of the mouse Nucleofector solution.
 CRITICAL STEP It is important to be very quick from this point onward so as to minimize the time that the neurons are in the
Nucleofector solution.
4. Add 4 µg of plasmid DNA and gently mix to resuspend the pellet.
5. Quickly transfer the cell-DNA mix to the cuvette and pulse it in the Nucleofector using the recommended program (O-005).
6. Add a small volume of maintenance medium to the cuvette and transfer the cell suspension to the dish containing maintenance
medium, and then mix gently. These cells can be plated on Petri dishes for biochemical studies or Petri dishes containing coated
coverslips for imaging studies. A substantial number of the surviving neurons should be transfected. We routinely obtain 60–80%
transfection efficiencies with this technique.
7. Incubate in a cell culture incubator at 37 °C.
8. Change the medium and maintain cells as described in PROCEDURE Step 42.
? TROUBLESHOOTING

1742 | VOL.7 NO.9 | 2012 | nature protocols

 protocol
Figure 1 | Overview of the technique. Coverslips are treated, prepared and Preparation of coverslips
coated with poly-l-lysine. After washing, they are incubated in medium.
Hippocampi or cortices are dissected; cells are trypsinized and dissociated, and
plated on coverslips. The neurons are maintained in culture for up to 28 d.

Coating with poly-L-lysine

tissue from several animals or by using individual animals to Washing and incubation in medium
generate separate cultures. Individual mouse cultures allow the
comparison of control and mutant neurons from animals of the
same litter. This culturing protocol is relatively straightforward and
does not use an astroglial feeder cell layer. The serum-free main-
tenance medium for neurons is not conducive to the growth and
Dissection of the hippocampus/cortex
survival of astrocytes18, and thus the surviving cell population is
predominantly neuronal. However, one can adapt this protocol to
include a feeder layer, if useful, for example, to study the effects of
secreted astrocyte factors on neurons. Hippocampi are dissected,
after which neurons are trypsinized, dissociated and plated and
Cell dissociation and plating
maintained in medium containing the B27 supplement. By using
this protocol, postnatal neurons can be maintained well in culture
for up to about 4 weeks. During this period, they show the
© 2012 Nature America, Inc. All rights reserved.

immunohistochemical characteristics of neurons, develop well-

differentiated axons and dendrites, establish spines and form func-
tional synaptic connections. They are also highly amenable to both
low-efficiency and high-efficiency transfections. We demonstrate
the application of these cultures to shRNA-mediated knockdowns
(Box 1), immunohistochemistry (Box 2) and time-lapse imaging
Maintenance of neurons
(Box 3). Other potential applications of these cultures include
in situ hybridization, nucleic acid and proteomic analyses, neuro- from embryos because of the advanced developmental stage of the
toxicology and electrophysiology4,23. animals. To avoid glial overgrowth, it is possible to use inhibitors,
such as araC, to inhibit the growth of dividing non-neuronal cells
Limitations. Although the neurons can be used for a variety of in long-term cultures. In addition, it is very difficult to maintain
applications, as discussed above, the culture system does have these cultures beyond 28 days in vitro (DIV); hence, studies must be
some disadvantages. Neurons prepared using this technique have a limited to this time period. Also, primary cultured neurons, either
higher proportion of non-neuronal cell types than cultures prepared embryonic or postnatal, are highly sensitive even to subtle changes in

  2 | Immunostaining and mounting of mouse neurons cultured on

Box
coverslips ● TIMING 24–36 h
1. Warm the paraformaldehyde/sucrose mixture to 37 °C.
 CRITICAL STEP It is important to ensure that the solution is warmed to 37 °C to preserve neuronal morphology.
2. Gently aspirate the medium from wells containing coverslips on which neurons are plated.
3. Add 1 ml of the warmed paraformaldehyde/sucrose mixture quickly, but carefully, to the well to ensure that neurites are not damaged.
4. Incubate on bench at room temperature for 10 min.
5. Aspirate the paraformaldehyde/sucrose mixture and add 1 ml of 0.1% (vol/vol) Triton X-100 solution in PBS to the well. Incubate on
bench for 10 min at room temperature.
6. Wash the coverslips once gently with PBS and add 1 ml of 5% (wt/vol) bovine serum albumin or 10% serum in PBS. Incubate on
bench for an hour at room temperature.
7. Add primary antibody diluted in 1% (wt/vol) BSA or 1% (wt/vol) serum in PBS to the coverslip and incubate overnight in a
­humidified chamber at 4 °C.
8. Wash the coverslips three times in PBS.
9. Add secondary fluorescent-conjugated antibody diluted in 1% (wt/vol) BSA/PBS to the coverslips and incubate on bench at room
temperature for an hour.
 CRITICAL STEP From this point onward, it is necessary to protect the coverslips from light.
10. Wash the coverslips three times with PBS.
11. On a clean slide, add a drop of ‘antifade’ mounting solution. Gently hold the edge of the coverslip with forceps and tap the edge
carefully on a Kimwipe to drain excess liquid.
12. Carefully invert the coverslip, cell side down, onto the ‘antifade’ mounting medium on the slide.
13. Gently remove excess mounting medium from the sides of the coverslip.
 CRITICAL STEP Be very gentle and careful and ensure that the coverslip does not move on the slide. This can result in considerable
damage to neuronal morphology. (continued)

nature protocols | VOL.7 NO.9 | 2012 | 1743

 protocol

  2 | (continued)
Box
14. Allow the coverslip to dry in the dark for an hour.
15. Gently seal the sides of the coverslip with clear nail polish. Let the nail polish dry completely before imaging.
16. The slides with the coverslips are now ready for imaging.
 PAUSE POINT The slides can also be stored at  − 20 or  − 80 °C for several months without substantial loss
of fluorescence.

  3 | Live imaging and FM4-64 labeling to examine synaptic function

Box
FM dyes are widely used in cell biology to examine various aspects of exocytosis and endocytosis. FM4-64 is a lipophilic styryl dye that
reversely binds to the outer leaflet of the lipid bilayer, but is not membrane permeable. At neuronal synapses, recycling of synaptic
vesicles can be induced by depolarization through electrical stimulation or incubation in high K +  buffers. During the subsequent endo-
cytosis, the dye becomes trapped in endocytic vesicles and can be used as a marker of a functional synapse. Advasep-7 is used as a FM
dye scavenger to reduce the background FM dye signal. An alternate protocol (see modifications to make at Steps 2 and 6) takes
advantage of FM4-64FX to label recycling vesicles. Cells that take up this dye can be fixed and immunostained without loss of FM4-
64FX fluorescence. This offers the advantage of being able to store coverslips for later imaging or co-labeling with antibodies.
© 2012 Nature America, Inc. All rights reserved.

FM4-64 labeling of recycling vesicles in live neurons:

1. Incubate live cultured neurons in Tyrode’s solution in the cell culture incubator for 10 min.
2. Aspirate the solution carefully and incubate with the high-KCl Tyrode’s solution containing 10 mM FM4-64 for 1 min in the cell
culture incubator. If you wish to fix cells later, use 10 mM FM4-64X in place of FM4-64.
 CRITICAL STEP From Step 3 until imaging, use Tyrode’s with CNQX and AP5. CNQX is an AMPA/kainate receptor antagonist and AP-5
is an NMDA receptor antagonist. These inhibitors are used to prevent additional firing to keep FM4-64 in recycled vesicles.
3. Aspirate the solution and add fresh Tyrode’s solution; repeat twice.
4. Add Tyrode’s solution containing Advasep-7 for 1 min at 37 °C.
5. Aspirate the solution and replace it with Tyrode’s solution.
6. Wash three times with fresh Tyrode’s solution and incubate in fresh Tyrode’s solution for 10 min in a cell culture incubator. If you
wish to fix the cells, fix them using 4% (wt/vol) paraformaldehyde in HBSS without phenol red for 10 min on ice and then wash them
twice with HBSS. If desired, immunofluorescence staining and mounting can be performed on fixed cells as described in Box 2.
7. Image by using an inverted confocal imaging system after placing the cells in a CO2 and temperature-controlled chamber. FM4-64
can be excited at 510–570 nm (peak at 538 nm) and emission can be collected at 560–750 nm (peak at 654 nm) in a lipid bilayer
membrane environment. FM4-64 excitation/emission spectra do not overlap with GFP or Alexa Fluor 488 excitation/emission spectra,
and thus can be used for co-labeling with GFP or Alexa Fluor 488.
8. Recycling vesicles can be imaged as punctae at synaptic terminals (Fig. 5d).

reagents or environment. We have provided a troubleshooting guide
to help identify and overcome some of these culture problems.
Experimental design
For routine studies requiring imaging fixed cells, we plate
neurons on 18-mm coverslips in 12-well dishes. Alternatively, four
or five coverslips can also be placed in 100-mm Petri dishes for
such studies. When using coverslips, they need to be prepared
and coated with poly-l-lysine, as described in Steps 1–11 of the
PROCEDURE. For time-lapse imaging, we plate neurons on glass-
bottom dishes. For biochemical studies, we plate neurons on tissue
culture–treated Petri dishes coated with poly-l-lysine. When doing
time-lapse imaging or biochemical studies, start the PROCEDURE
at Step 7, as there is no need to prepare coverslips.

MATERIALS
REAGENTS For culture
For preparing coverslips • P0 mouse pups (8–10 pups, either sex; we routinely use C57BL/6 mice)
• Borate buffer, 0.1 M ! CAUTION All experiments must be conducted in accordance with the
• Poly-l-lysine hydrobromide powder (Sigma-Aldrich, cat. no. P-9155, 5 mg ­relevant institutional and governmental guidelines and regulations.
each; cat. no. P2636, 25 mg each) • Medium components, as described in Table 1
• Coverslips (18 mm; Fisher Scientific, cat. no. 12-545-84 18CIR-1D) • Cell culture plates (12 well)
• Concentrated nitric acid • Culture dishes (60 mm)
• Sterile double-distilled water • Sterile microcentrifuge tubes for individual mouse culture
• Boric acid • Time-lapse dishes (Mattek, cat. no. P3SG-0-14-C)
• Sodium tetraborate • Petri dishes, bacteriological grade, 100 mm

1744 | VOL.7 NO.9 | 2012 | nature protocols

 protocol
Table 1 | Medium composition.

Components Final concentration Source

Dissection/dissociation HBSS (Ca2 +  and Mg2 +  free) 97.5% Invitrogen, cat. no. 14175095
medium
11 mg ml − 1 sodium pyruvate (100×) 1× Invitrogen, cat. no. 11360070

20% (wt/vol) glucose (in Milli-Q water, filter sterilized) 0.1% Sigma-Aldrich, cat. no. G-6152

1 M HEPES (pH 7.3) (in Milli-Q water, filter sterilized) 10 mM Sigma-Aldrich, cat. no. H-4034

Plating medium MEM Eagle’s with Earle’s BSS 86.55% Invitrogen, cat. no. 21010046

FBS (re-filtered, heat-inactivated) 10% Invitrogen, cat. no. 16140071

20% (wt/vol) glucose 0.45%

100 mM sodium pyruvate (100×) 1× Invitrogen, cat. no. 11360070

200 mM glutamine (100×) 1× Invitrogen, cat. no. 25030081

Penicillin/streptomycin (100×) 1× Invitrogen, cat. no. 15140122

© 2012 Nature America, Inc. All rights reserved.

Maintenance medium Neurobasal medium 96% Invitrogen, cat. no. 21103049

B-27 (50×) 1× Invitrogen, cat. no. 17504044

200 mM glutamine (100×) 1× Invitrogen, cat. no. 25030081

Penicillin/streptomycin (100×) 1× Invitrogen, cat. no. 15140122

• Trypsin (Worthington, cat. no. LS003707)
• DNase (Sigma-Aldrich, cat. no. DN25)
• Sterile tissue culture plasticware
• Sterile, long, cotton-plugged Pasteur pipettes, 9 inches
• Trypan blue stain (Optional; Invitrogen, cat. no. 15250-061)
• Cell strainer (70 µm, nylon, BD Falcon, cat. no. 352350)
For transfection with Lipofectamine and nucleofection only (Box 1)
• Lipofectamine 2000 (Invitrogen, cat. no. 11668-027)
• EndoFree kit for preparation of DNA (Qiagen, cat. no. 12362)
• Neurobasal medium (Invitrogen, cat. no. 21103-049)
• Mouse Nucleofector kit (Lonza Biosciences, cat. nos. VAPG-1001,
VPG-1001, VVPG-1001, includes cuvettes)
• Maintenance medium
For immunostaining only (Box 2)
• Fixing solution (4% (wt/vol) paraformadehyde/4% (wt/vol) sucrose in 1×
PBS, see Reagent Setup for details)
• Triton X-100
• Primary antibodies. We have used the following antibodies for immuno­
fluorescence: mouse gephyrin-specific (1:2,000; Synaptic Systems), rabbit
anti-vGAT (1:1,000; Synaptic Systems), guinea pig vGluT1-specific (1:2,000;
Millipore Bioscience Research Reagents), mouse psd95-specific (1:250; clone
6G6-1C9, Thermo Scientific, cat. no. MA1-045), mouse MAP2-specific
(1:2,000, clone HM-2, Sigma-Aldrich cat. no. M4403), rabbit Tau-specific
(1:1,000, Abcam) and γ-aminobutyric acid (1:4,000, Sigma-Aldrich). We have
used the following antibodies for western blotting: mouse β-tubulin–specific
(Clone E7, University of Iowa Developmental Studies hybridoma bank) and
mouse δ-catenin–specific (BD transduction Labs, cat. no. 611537)
• DAPI (Sigma-Aldrich, cat. no. D-9564)
• Serum (species in which the secondary antibody is raised) or BSA
• Fluorophore-conjugated secondary antibodies (We have used Invitrogen’s
Alexa Fluor–conjugated secondary antibodies.)
• PBS
• ProLong Gold antifade (Invitrogen, cat. no. P36934)
• Glass slides (Fisher Scientific, cat. no. 12-544-3)
• Nail polish
For FM4-64 labeling of recycling vesicles only (Box 3)
• FM4-64 dye (Invitrogen, cat. no. T-13320) or FM4-64FX dye (Invitrogen,
cat. no. F34653)
• Tyrode’s solution (124 mM NaCl, 3 mM KCl, 2 mM CaCl2, 1 mM MgCl2,
10 mM HEPES (pH 7.3) and 5 mM d-glucose; see REAGENT SETUP)
• Tyrode’s solution with CNQX (Sigma-Aldrich, cat. no. C127) and AP5
(dl-2-amino-5-phosphonovaleric acid; Sigma-Aldrich, cat. no. A5282)
(see REAGENT SETUP)
• High-KCl Tyrode’s solution (72 mM NaCl, 50 mM KCl, 2 mM CaCl2,
1 mM MgCl2, 10 mM HEPES (pH 7.3) and 5 mM d-glucose; see
REAGENT SETUP)
• Paraformaldehyde (4% (wt/vol)) in Hank’s buffered saline solution (HBSS)
without phenol red ! CAUTION This is a hazardous material and a possible
carcinogen; use it in a chemical safety hood.
• Sodium hydroxide
• DMSO
EQUIPMENT
• Two Dumont-style (no. 5) forceps—to handle coverslips
• One small scissor, one Dumont-style forceps—to remove brain
• Two Dumont-style (no. 5) forceps, one fine scissor—to dissect hippocampus
• Dry sterilizer (Germinator 500; Roboz, cat. no. DS-401, optional)
• Dissecting microscope with illumination
• Hemocytometer
• Light microscope
• Fluorescent/confocal microscope
• 37 °C water bath
• Fire-polished glass pipette (orifice size 1.5–2 mm)
• Laminar flow cell culture hood
• High-temperature dry oven (225–270 °C).
• Bunsen burner
• Cell culture incubator—5% CO2, 95% humidity
• Lonza Nucleofector II (Lonza Biosciences, cat. no. AAD-1001S)
• CO2 and temperature-controlled chamber mounted on microscope for live
imaging
• Microcentrifuge tubes (e.g., Eppendorf tubes)
• Kimwipes
REAGENT SETUP
Borate buffer (0.1 M)  Mix 1.24 g of boric acid and 1.90 g of sodium tetraborate
in 400 ml of H2O (pH 8.5); filter-sterilize using a 0.2-µm filter. Note that several
hours of stirring may be required for these chemicals to dissolve. The solution
may be stored at 4 °C for 3–4 weeks.

nature protocols | VOL.7 NO.9 | 2012 | 1745

 protocol
Poly-l-lysine solution  Dissolve 0.5 mg ml − 1 of poly-l-lysine in borate
buffer; the solution may be stored at  − 20 °C for 2–3 weeks.
Trypsin  Dissolve 2.5% (wt/vol) trypsin in sterile double-distilled water;
aliquot and store at  − 20 °C. The solution is stable for 2–3 months.
DNase  Dissolve 1% (wt/vol) DNase in sterile double-distilled water; aliquot
and store at  − 20 °C. The solution is stable for 2–3 months.
Media  As described in Table 1.
Fixing solution  Weigh out 4 g of paraformaldehyde into a beaker. Add
10 ml of 10× PBS and 60 ml of water. Stir continuously and add a drop or
two of concentrated sodium hydroxide. This should allow the parafor-
maldehyde to go into solution immediately. Add 4 g of sucrose, allow to
dissolve completely and adjust the volume of the solution to 100 ml. Alter-
natively, this solution can be made by diluting out commercially available
concentrated paraformaldehyde solution. The solution can be stored for
up to a month at  − 20 °C. ! CAUTION Paraformaldehyde is a hazardous
material and a possible carcinogen; use it in a chemical safety hood.
Sodium hydroxide is a caustic chemical; use it in a chemical safety hood.
© 2012 Nature America, Inc. All rights reserved.
FM4-64 or FM4-64FX  Dissolve in DMSO to make a 10 mM stock solution.
Protect from light. The stock solution can be stored at  − 20 °C for up to
2 weeks.
Tyrode’s solution  Mix 124 mM NaCl, 3 mM KCl, 2 mM CaCl2, 1 mM
MgCl2, 10 mM HEPES (pH 7.3) and 5 mM d-glucose in distilled water.
Filter-sterilize using a 0.2-µm filter. This solution can be stored for up to
4 weeks at 4 °C.
Tyrode’s solution with CNQX and AP5  Add 10 mM CNQX and 50 mM AP5
to sterile Tyrode’s solution right before experiments.
High-KCl Tyrode’s solution  Dissolve 2 mM NaCl, 50 mM KCl, 2 mM CaCl2,
1 mM MgCl2, 10 mM HEPES (pH 7.3) and 5 mM d-glucose in distilled water.
Filter-sterilize through a 0.2-µm filter. This solution can be stored for up to
4 weeks at 4 °C.
EQUIPMENT SETUP
Pasteur pipette for trituration of cells  Take a sterile glass Pasteur pipette and
flame the tip carefully to obtain a smooth surface. The opening of the tip should
be comparable to that of a 1-ml pipette tip.

PROCEDURE
Preparation of coverslips ● TIMING 2 d, 5 d before culture
1| Place coverslips in a heat-resistant glass container with a large surface area. Add concentrated nitric acid (70% (wt/wt)).
Swirl coverslips in the acid, cover them well and leave them in acid for at least 12 h. The acid can be reused two or three
times, although it becomes discolored by exposure to light.
! CAUTION Nitric acid should be handled in a fume hood.

2| Remove the nitric acid and rinse the coverslips four or five times in several volumes of double-distilled water. Afterward,
for each wash, allow the coverslips to soak in water for 2 h, before replacing the wash with fresh water. At each wash, swirl
the coverslips around carefully.

3| Aspirate water carefully and ensure that there are no droplets of water sticking to the coverslips.

4| Cover the container with aluminum foil and bake in a dry-heat oven at 225–270 °C for 12–16 h.
 PAUSE POINT Baked coverslips may be stored at room temperature (20–25 °C) for 1 month if covered tightly.

Coating of coverslips, time-lapse dishes and Petri dishes with poly-L-lysine ● TIMING 1 h, 2 d before culture
5| When working in a sterile laminar hood, use sterile forceps to place individual coverslips in single wells in a
12-well dish.

6| Sterilize by placing the open dishes with the coverslips under ultraviolet light for 30 min. We do not routinely sterilize
the time-lapse dishes or Petri dishes, as they come in a sterile package.

7| Add enough poly-l-lysine solution to cover the coverslip and the glass bottom of the time-lapse dishes. Similarly, for
Petri dishes, ensure that the volume of poly-l-lysine covers the dish entirely.

8| Wrap the dish in aluminum foil to prevent evaporation and leave it overnight at room temperature.
 CRITICAL STEP It is important to make sure that the poly-l-lysine does not dry out during incubation. If you are working
in the laminar hood, be sure to turn the blower off during the incubation.

Washing of coverslips after coating ● TIMING 1 h, 1 d before culture

9| When working in a sterile laminar hood, aspirate the poly-l-lysine carefully.

10| Add 1 ml of sterile water into each well and aspirate; repeat this step three times.
 CRITICAL STEP Ensure that the glass surface does not dry out during the washing procedure.

1746 | VOL.7 NO.9 | 2012 | nature protocols

 protocol

11| Aspirate water, add maintenance a b c d

medium and leave the coverslips in the
tissue culture incubator until you are
ready to plate cells, but preferably not
for more than 24 h. If it is necessary
to leave them for more than 24 h,
change the medium to fresh main- e f g h
tenance medium a few hours before
plating the cells.

Removal of brains from P0 mouse

pups ● TIMING 3–6 min for each
animal
! CAUTION All experiments must be
performed in accordance with relevant
institutional and governmental guide-
lines and regulations.
© 2012 Nature America, Inc. All rights reserved.
Figure 2 | Illustration of the technique to remove brains from P0 mice. (a,b) Decapitate pups and
remove the head. By using a fine scissors, make a midline incision at the skin surface close to the
hindbrain region. (c) Follow the incision to the extreme rostral region. (d) Make a small incision at
the base of the skull and follow along the midline. (e) Separate the two halves of the skull to reveal
the brain. (f) Gently place the forceps underneath the brain and separate it from the underlying
tissue. (g,h) Gently remove the intact brain (g) and quickly place it into dissection medium (h).

12| Sterilize instruments by heating
them in a dry sterilizer or washing
them with 70% (vol/vol) ethanol. Dry
thoroughly if ethanol is used.
It is important to be extremely fast and careful to avoid contamination. All animal procedures
were carried out in accordance with approved protocols from the Institutional Animal Care and Use
Committees at UNMC and UCSF.

13| Prepare 60-mm dishes with dissection medium. If you are culturing from individual mice, prepare microcentrifuge tubes
with 1 ml of dissection medium (Table 1).

14| Euthanize the mouse pup by decapitation and separate the head from the body (Fig. 2).

15| Place the head on a dish and hold down the sides with forceps.

16| Gently dissect the skin on the top of the head and hold down the skin on either side with the forceps.

17| By using fine scissors, cut open the skull by making an incision at the base of the brain. Separate the two halves of the
skull and remove carefully. Take care to not cut through the brain tissue when removing the skull bone.

18| By using forceps, pinch off the brain from the base and transfer it to a 60-mm dish containing dissection/dissociation
medium (Table 1).

19| To culture neurons from individual pups, put each brain in a separate dish. For pooled cultures, put two or three brains
into one 60-mm dish.
 CRITICAL STEP Ensure that the brains are submerged in the medium, and do not let them dry out at any point.

Dissection of the hippocampus and cortex ● TIMING 4–6 min for each mouse
20| Separate the two halves of the brain by making a sagittal cut along the midline. Discard the cerebellum (Fig. 3).

21| Place the brain such that the outer surface of the hemisphere faces the bottom of the dish.

22| Under a dissecting microscope, gently remove the midbrain and thalamic tissue to leave an intact hemisphere containing
the cortex and hippocampus.

23| Turn the tissue over so that the hippocampus faces the bottom of the dish.

24| By using forceps, pin down the anterior cortex, taking care not to damage the hippocampus.

nature protocols | VOL.7 NO.9 | 2012 | 1747

 protocol
Figure 3 | Steps for dissection of the
hippocampus from the intact brain. a b c
(a) Place the brain dorsal side up in dissection
medium. (b) Separate the hindbrain region and
make a sagittal incision to separate the two
hemispheres. (c) Place each hemisphere’s cortex
side down and remove any noncortical forebrain
tissue. (d) Hold the hemisphere in place using
forceps; use caution so as not to damage the
hippocampus. (e) Remove the meningeal tissue
with another pair of forceps. Dissect out the
hippocampus (f) and the cortex. All animal
procedures were carried out in accordance with
d e f
approved protocols from the Institutional Animal
Care and Use Committees at UNMC and UCSF.

25| Use another pair of forceps to

pick and grab the meninges carefully
and gently peel them off, ideally as a
single piece. Check for the remaining
© 2012 Nature America, Inc. All rights reserved.

pieces of meninges and remove them

completely.
 CRITICAL STEP At this stage during mouse development, the meninges can be sticky and difficult to remove, compared
with the embryonic stage. However, it is important to ensure that the meninges are completely removed, so that they do not
contribute any non-neuronal cells to the culture.

26| Orient the tissue so that the hippocampus is on the top. At this stage, the hippocampus can be identified by its
C-shaped structure and opacity, which differ from the neighboring cortical tissue. The cortex can also be dissected out and
processed similarly for cortical cultures.

27| By using forceps or fine scissors, separate out the hippocampus carefully. Collect it in a 60-mm dish containing fresh
dissection/dissociation medium. For culture of individual hippocampi or cortices, collect each pair of hippocampi or cortices
in a 1.5-ml Eppendorf tube containing 1 ml of dissection/dissociation medium.
 CRITICAL STEP It is important that the dissection be done as quickly as possible to ensure cell viability and health.

Cell dissociation and plating ● TIMING ~1–1.5 h

28| Place all the hippocampi or cortices in a 15-ml centrifuge tube for a pooled mouse culture. For culturing individual
mouse cultures, leave hippocampi or cortices in Eppendorf tubes. Wait until all tissues settle to the bottom and remove all
but 5–10% of the medium.

29| Add 10 ml of fresh dissection/dissociation medium, wait for the tissue to settle to the bottom of the tube, aspirate and
repeat twice. For tissue in Eppendorf tubes, wash it twice with 1 ml each of dissection medium.
 CRITICAL STEP It is important to be very gentle with the washing, so as to not damage the hippocampi or break them
apart. Smaller pieces of tissue are more difficult to handle during the washing procedure.

30| Resuspend the tissue in 4.5 ml of fresh dissection medium and add 0.5 ml of trypsin solution; swirl the tube around
gently to mix. For tissue in individual Eppendorf tubes, scale the volumes down to 450 and 50 µl, respectively.

31| Incubate at 37 °C in a water bath for 20 min.

 CRITICAL STEP It is important to ensure that this incubation does not proceed for longer than 20 min.

32| Add 0.5 ml of DNase solution and incubate in a laminar flow hood at room temperature for 5 min. For tissue in
Eppendorf tubes, add 20 µl of DNase and incubate for 2–5 min.

33| Aspirate the medium and wash the tissue twice gently with 10 ml of fresh dissection medium for each wash. For tissue
in Eppendorf tubes, use 1 ml of dissection medium for each wash.
 CRITICAL STEP Be extremely gentle while washing the hippocampi and cortices because their tissue structures have now
been loosened by trypsin incubation.

1748 | VOL.7 NO.9 | 2012 | nature protocols

 protocol

34| Wash twice, each with 10 ml a b c

(or 1 ml, for tissue in Eppendorf tubes) Polished
Pipette tip
Unpolished
of plating medium.
 CRITICAL STEP Use temperature-
equilibrated plating medium from
this point forward, as the sera will
inactivate the trypsin.

35| Resuspend hippocampi or cortices

in 2.5 ml (for tissue from six to ten
Figure 4 | Trituration of isolated hippocampi/cortex. (a) Comparison of unpolished and polished
animals) or 0.5 ml (for tissue from tips of glass Pasteur pipettes. A plastic 1,000-µl pipette tip is shown for comparison. Note that the
individual animals in Eppendorf tubes) polished tip has smooth edges and an orifice size comparable to that of the pipette tip. (b) Intact
of plating medium. dissected hippocampi in dissection medium at the start of trypsinization. (c) Setup of Petri dishes for
triturating cells. All animal procedures were carried out in accordance with approved protocols from the
36| Place the lid of a 100-mm Institutional Animal Care and Use Committees at UNMC and UCSF.
­bacteriological-grade Petri dish face
down on the floor of the hood. Place
the base of the dish on the top of the inverted lid such that the base is at about 30° resting both on the top of the lid and
© 2012 Nature America, Inc. All rights reserved.

on the floor of the hood. This is to ensure the use of minimal surface area for dissociating the hippocampi. Bacteriological-
grade dishes do not support the adherence of neurons to the dish and hence result in greater yield.

37| Carefully place the tissue and the solution in the base of the 100-mm dish such that it occupies the volume at the
bottom of the dish (Fig. 4).

38| By using a fire-polished glass pipette (Fig. 4), carefully triturate the tissue eight to ten times to dissociate the cells
gently and obtain a homogenous cell suspension. At this point, there should be no or minimal chunks of tissue.
 CRITICAL STEP The trituration of cells should be done slowly and carefully to minimize damage to cells. It is best to avoid
any bubbling during the procedure and to try to keep the pipette within the cell suspension at all times. The tip diameter of
the glass pipette is crucial. Too small an orifice can result in cell damage, whereas too large an orifice can cause inadequate
trituration. It is also necessary to be careful with the length of trituration. Too much can damage cells and too little can
cause large chunks of tissue to be retained, with either case resulting in a loss of yield. Use a cell strainer if it is challenging
to remove tissue chunks, especially for preparing cortical cultures or hippocampal cultures from more than ten mice.
? TROUBLESHOOTING

39| Take a 10-µl aliquot of the dissociated cells to estimate viable cell density on a hemocytometer. Viable cells can be
identified by their glossy appearance, whereas dead cells appear dark. Typical yields are 400,000–600,000 hippocampal
cells and 600,000–800,000 cortical cells per animal. An optional alternative is to use trypan blue staining to estimate cell
viability. Add 10 µl of cells to 40 µl of maintenance medium with 50 µl of trypan blue. Determine the density of cells on a
hemocytometer. In this case, viable cells are colorless, whereas dead and damaged cells are blue. If you wish to perform a
high-efficiency transfection, proceed to Box 2.
? TROUBLESHOOTING

40| At this point, the desired number of cells can be plated out or used for nucleofection (Box 2). We routinely add
90,000 cells to an 18-mm coverslip in a 12-well dish (65 cells per mm2) and 45,000 cells per time-lapse dish that contains
maintenance medium (65 cells per mm2). Gently mix to ensure even plating on the coverslip. Cells that do not settle on
the coverslip will not survive. For biochemical studies, we plate 500,000 to 1 million cells on a 60-mm-diameter Petri dish
(44–88 cells per mm2) and 1–2 million cells on a 100-mm-diameter Petri dish (32–64 cells per mm2).
 CRITICAL STEP It is important to plate the neurons, particularly the cortical neurons, quickly to avoid any clumping.
If clumping is an issue, dilute the cell suspension with plating medium and triturate gently again, as described
in Step 38.

41| Incubate in the cell culture incubator at 37 °C for 2–6 h.

42| Four hours after plating cells, examine under the microscope to ensure that the cells have settled on the substrate.
? TROUBLESHOOTING

nature protocols | VOL.7 NO.9 | 2012 | 1749

 protocol

43| Gently aspirate the medium from each well and add 1.5 ml of fresh maintenance medium warmed to 37 °C to each well
in the 12-well dish, 1 ml to each time-lapse dish and 10 ml in the 100-mm Petri dish.
 CRITICAL STEP It is necessary to ensure that the medium has been warmed to 37 °C and that the manipulation is gentle
so as not to damage the cells. We prefer to aspirate the medium using a pipette and not a vacuum-based method. Avoid
aspiration or addition of medium directly above cells; instead, allow the medium to trickle down gently through the side wall
of the plate or well.

44| Incubate in a cell culture incubator at 37 °C.

45| If desired, 2 d after plating, add cytosine arabinoside (araC; 1-b-d-arabinofuranosylcytosine) to a final concentration of
1–5 µM to inhibit the proliferation of dividing non-neuronal cells. Several investigators continue to maintain araC in the
medium through the life of the culture. However, in our hands, we have noticed that occasionally the addition of the araC
causes the death of neurons. To avoid this, we replace the araC-containing medium with fresh maintenance medium 24–48 h
after adding the araC.

Maintenance of neurons ● TIMING Up to 28 d

46| Twice a week, aspirate half the medium from each well/time-lapse dish and replace it with fresh maintenance medium
warmed to 37 °C. These neurons can be maintained in culture for up to DIV 28. The neurons may be used for low-efficiency
© 2012 Nature America, Inc. All rights reserved.

transfections (Box 1), immunostaining (Box 2) or live imaging (Box 3) anytime during this period (DIV 0–28). The cultures
can be used for examining synaptic function by FM4-64 dye uptake between days 11 and 28 (Box 3).
 CRITICAL STEP It is important to not replace the entire medium. Neurons secrete factors that promote growth and
survival.
? TROUBLESHOOTING

? TROUBLESHOOTING
Troubleshooting advice can be found in Table 2.

Table 2 | Troubleshooting table.

Step Problem Possible reason Solution

38 Difficulty in triturating Insufficient trypsinization Ensure the correct concentration and incubation time of
hippocampi trypsin, use fresh trypsin

DNA from damaged cells makes cell Consider using DNase at the right concentration
suspension viscous

Insufficient incubation with DNase Increase incubation time with DNase

Opening of Pasteur pipette tip for Ensure that the opening of the Pasteur pipette tip is
trituration is too large adequate

39 Excessive number of dead Opening of Pasteur pipette orifice is Ensure that the opening of the Pasteur pipette is the
cells (more than 25% at the too small or not smooth enough correct size and smooth
time of counting)

Harsh trituration Triturate the cells gently, ensuring little or no bubbling

of liquid

Insufficient trypsin activity, strain on Ensure the right concentration of trypsin or replace the
cells during trituration batch of trypsin

Wrong pH of the dissection buffer Check the pH of the dissection buffer

Extended time period between remov- Reduce the time for the entire procedure by practicing
ing tissue and trituration of tissue techniques

Low yield of cells Clumping of cells Ensure adequate and gentle trituration of cells
(continued)

1750 | VOL.7 NO.9 | 2012 | nature protocols


Table 2 | Troubleshooting table (continued).
Step Problem Possible reason
39 Cells adhere to Petri dish during the
trituration process
Insufficient starting material
42 Neurons do not adhere to Incorrect coverslip glass
coverslips when plated
Insufficient etching of coverslips
Insufficient volume of poly-L-lysine
on coverslip
Poly-L-lysine–coated coverslips are
stored for too long
© 2012 Nature America, Inc. All rights reserved.
Incorrect composition of borate
buffer
Poor quality poly-L-lysine
46 Contamination Nonsterile conditions
Cells do not appear viable or Inappropriate medium composition
healthy 4–5 d after plating
Inappropriate coating of coverslips or
coated coverslips were too old
Medium change is inappropriate
Medium changing procedure takes
too long
Box 1 Variability in transfection— Neurons are not healthy
low-efficiency transfection
Transfection reagents are not fresh or
have changed
protocol
Solution
Use bacteriological-grade dishes that are not tissue
culture coated
Ensure enough number of hippocampi used
Ensure that the right coverslip is being used; neurons are
very sensitive to the glass surface
Follow the nitric acid soaking and baking protocol; bake
coverslips in small batches to ensure adequate etching
Use enough poly-L-lysine to coat the coverslip entirely
Discard and coat fresh coverslips
Check the concentration and pH of borate buffer
Try a different batch of poly-L-lysine
Sterilize tools by autoclaving or using a dry sterilizer
(e.g., Germinator 500 from Cell Point Scientific)
Clean working surface areas with 70% (vol/vol) ethanol
Observe sterile working conditions
Separate tools for culture from tools for other
experiments
Use fresh components for all media. Do not store the
maintenance medium for more than 10–12 d at 4 °C.
Do not freeze thaw serum for plating medium more than
once. Do not warm aliquots of maintenance medium more
than two times
Ensure coverslips are coated with the correct concentra-
tion of poly-L-lysine. Do not store coated coverslips at
4 °C for more than a week
Replace half the medium with fresh medium warmed to
37 °C every 3–4 d
In changing the medium, work with cultures in the hood
for as short a period of time as possible to avoid creat-
ing an alkaline medium and killing cells. When changing
medium from multiple dishes, it is better to only take out
and feed one dish at a time
Optimize neuronal culture to obtain healthy cells that can
be maintained in culture
Try a new vial
(continued)
nature protocols | VOL.7 NO.9 | 2012 | 1751
 protocol
Table 2 | Troubleshooting table (continued).

Step Problem Possible reason Solution

Box 1 DNA concentration is not right or DNA Measure DNA concentration and absorbance ratio at
quality is not good 260/280 nm (A260/A280) ratio again. These should be
above 1 mg ml − 1 and ~1.8, respectively. Prepare fresh
DNA

Too many or too few cells are Adjust the amount of DNA for transfection
transfected

Lipid-DNA complex incubation time Make sure the incubation time is between 20–30 min
is either too long or too short

Low-quality DNA Use high-quality DNA. We have had success with using
the EndoFree maxi prep kit from Qiagen (cat. no. 12362)

Variability in transfection— Inappropriate solution for Use the manufacturer’s solution for electroporation
high-efficiency transfection electroporation
© 2012 Nature America, Inc. All rights reserved.

Too few cells are transfected Adjust the amount of DNA for transfection

Low-quality DNA Use high-quality endotoxin-free DNA

DNA concentration is not right or Measure DNA concentration and A260/A280 ratio again.
DNA quality is not good These should be above 1 mg ml − 1 and ~1.8, respectively.
Prepare fresh DNA

Too many cells die after nucleofection Ensure that the right number of cells and program are
being used. Leave cells in the nucleofection solution for
the minimum time possible

● TIMING
Preparation of coverslips should be started 5 d before the day of culturing, but it requires only about an hour each day.
Preparation of the hippocampal cultures will take 4–8 h, depending on the number of mice. Subsequent maintenance of the
cultures requires an hour twice a week to warm and replace the medium, and an additional hour for performing low-efficiency
transfections.

ANTICIPATED RESULTS
The culture of hippocampal and cortical pyramidal neurons a DIV 0 DIV 1 DIV 2

is widely used and well characterized. Neurons in culture

undergo distinct stages of development, starting with
extension of lamellopodia, followed by axon and dendrite
specification and extension and synapse formation and
maturation. All of these stages can be distinguished in
these cultures. Note, however, that the cell population DIV 5 DIV 10 DIV 14

is not completely synchronized and hence all the neurons

may not simultaneously be at the exact same stage of
development2. Consequently, it is possible to detect neurons
with different levels of MAP2 staining (Supplementary
Fig. 1a). Moreover, we detect some DAPI-positive

Figure 5 | Images of cultured hippocampal neurons. (a) DIC images of

b MAP2 Tau Merge

neurons at different stages during culture—DIV 0, 1, 2, 5, 10 and 14. Note

how neuronal morphology and arborization develop over time. Scale bar,
10 µm. (b) Immunostaining of DIV 7 neurons with axonal (Tau, green) and
dendritic (MAP2, red) markers. Note that axons and dendrites can be clearly
distinguished by Tau-specific and MAP2-specific labeling and morphology.
Scale bar, 20 µm.

1752 | VOL.7 NO.9 | 2012 | nature protocols

 protocol
Figure 6 | Morphology and immunostaining a
of neurons for synaptic markers. (a) Confocal
microscope images of a DIV 17 neuron
expressing GFP. The dendrites and axons can be
distinguished clearly by their morphology.
Dendrite
Note the presence of spines on the dendrites
and the thin morphology of the axon. Scale
bar, 20 µm. (b) Immunostaining of DIV 14 GFP-
transfected neurons with antibodies to excitatory
and inhibitory pre- and postsynaptic markers. The
majority of inhibitory synapses (presynaptic— Dendrite
vGAT, blue; postsynaptic—Gephyrin, red) form
on dendritic shafts (*, left image), whereas the
majority of excitatory synapses (presynaptic—
vGlut1, blue; postsynaptic—psd95, red) form on
spine heads (*, right image) and can be clearly
Axon
distinguished. Scale bars, 5 µm.
b GFP GFP

structures that are smaller than neu- Gephyrin psd95

ronal nuclei. We suspect that these

© 2012 Nature America, Inc. All rights reserved.

represent non-neuronal dividing cells vGAT vGlut1

that are undergoing cell death after

Merge Merge
araC treatment or fragmented cell * *
*
debris from the preparation proce- * * *
dure. Note that these structures do
not usually have a distinct cell body
associated with them (Supplementary Fig. 1b). Hence, DAPI staining may not be a reliable indicator of total cell number
if not counted carefully, and differential interference contrast (DIC) imaging might be a more reliable indicator of total cell
counts. In these cultures, we detect about 6–8% astrocytes, as identified by glial fibrillary acidic protein (GFAP) reactivity
(­Supplementary Fig. 2), and about 20% of the neuron population comprises inhibitory neurons, as identified by immuno­
staining to γ-aminobutyric acid (GABA) (Supplementary Fig. 3).
Mouse postnatal neurons cultured as described here can be successfully maintained in culture for several days. These neurons
extend neurites that have characteristic morphologies (Fig. 5a) and express appropriate dendritic (MAP2) and axonal (Tau) mark-
ers (Fig. 5b). These neurons are amenable to transfections. In neurons transfected with a GFP-expressing vector, it is possible to
distinguish axons and dendrites by virtue of their morphology after about 10–12 d of culture (Fig. 6a; axons are thin and lack
spines; dendrites are thicker and mature dendrites typically contain spines). Neurons develop excitatory synapses as indicated
by staining for psd95 (excitatory postsynaptic marker) and vGlut1 (excitatory presynaptic marker) and inhibitory synapses as
indicated by gephyrin (inhibitory postsynaptic marker) and vGAT (inhibitory presynaptic marker) labeling (Fig. 6b). Synaptic
function can be assessed using the FM4-64 dye uptake assay (Fig. 7).
In neurons that have been transfected with the high-efficiency nucleofection with vector and shRNA-expressing constructs,
it is possible to detect the knockdown of the protein of interest by western blot analysis (Supplementary Fig. 4). Moreover,
cells are quite amenable to time-lapse imaging. Figure 8 shows an example of a dendritic segment with spines in a GFP-
expressing neuron imaged over time. Such imaging allows for clear examination of spine morphology over time. It is also
possible to couple such studies with co-transfections with fluorescently labeled synaptic markers, such as psd-95, to simulta-
neously examine spine and synapse dynamics over time.
The ability to culture mouse postnatal neurons offers the potential to investigate multiple aspects of neuronal structure
and function. However, the great utility of cultured neurons is in the ability to be manipulated with low- and high-efficiency
transfections. Transfected neurons are
ideal for studying various aspects of
GFP FM4-64 Merge
neuronal structure and function by
imaging or by electrophysiology. In
addition, these cells can be further
manipulated by overexpression or
knockdown of proteins. By using genet-
ics to manipulate gene expression and
* *
high-efficiency transfections to mark
the populations, one can generate Figure 7 | Synaptic functional assay on cultured neurons. A GFP-transfected neuron at DIV 17 subjected
mixed cultures of mouse neurons from to FM4-64 dye uptake protocol to examine functional synapses (*). Scale bar, 5 µm.

nature protocols | VOL.7 NO.9 | 2012 | 1753

 protocol
Figure 8 | Neurons in culture are amenable
to time-lapse imaging. Time-lapse imaging of
two dendritic shafts with associated spines in
GFP-transfected DIV 14 neurons. Note that the # # # #
dendritic shaft shows no evidence of damage even
* 0 min * 6 min * 12 min * 18 min
upon prolonged imaging. Note how some spines
are relatively stable (#), whereas others show
alterations in morphology (*) over time. Scale
bar, 2 µm.
# # # #

* 24 min * 30 min * 36 min * 42 min

two different strains of mice to dissect their abilities to form functional connections with each other24. In addition, transfec-
tion allows for temporal control of gene deletion; for instance, cultures from floxed mice can be transfected with a plasmid
encoding Cre recombinase to abolish the expression of the gene of interest at various time points. These cultures could also
be successfully used for routine biochemical studies and proteomics25. With the use of high-efficiency transfections, these
cells are amenable to biochemical studies of a large cohort of neurons.
In summary, primary mouse cultures generated successfully following the protocol described can be used for a variety of
cell biological and biochemical studies. These cultures, prepared with a modest level of difficulty and care, are quite robust.
© 2012 Nature America, Inc. All rights reserved.

They can offer great insights into neuronal cellular architecture and function.

Note: Supplementary information is available in the online version of the paper.
Acknowledgments We thank the past and present members of the Reichardt
laboratory for their support and insight. This work was supported by US National
Institutes of Health Grant F32-MH079661 (G.M.J.B.), the Simons Foundation
(L.F.R.) and start-up funds from the Munroe-Meyer Institute, University of
Nebraska Medical Center (J.A.).
AUTHOR CONTRIBUTIONS G.M.J.B., S.-H.L. and J.A. designed experiments.
G.M.J.B., S.-H.L., D.S., Y.Y., Y.-G.N. and J.A. performed the experiments.
G.M.J.B., S.-H.L., D.S., Y.Y. and J.A. collected, analyzed and interpreted data.
J.A. and L.F.R. supervised the experiments. G.M.J.B., S.-H.L., L.F.R. and J.A.
wrote the manuscript.
COMPETING FINANCIAL INTERESTS The authors declare competing financial
interests: details are available in the online version of the paper.
Published online at http://www.nature.com/doifinder/10.1038/nprot.2012.099. 17. Shimizu, S., Abt, A. & Meucci, O. Bilaminar co-culture of primary rat
Reprints and permissions information is available online at http://www.nature.
com/reprints/index.html.
1. Frey, U., Huang, Y.Y. & Kandel, E.R. Effects of cAMP simulate a late stage
of LTP in hippocampal CA1 neurons. Science 260, 1661–1664 (1993).
2. Kaech, S. & Banker, G. Culturing hippocampal neurons. Nat. Protoc. 1,
2406–2415 (2006).
3. Arikkath, J. et al. Erbin controls dendritic morphogenesis by regulating
localization of δ-catenin. J. Neurosci. 28, 7047–7056 (2008).
4. Arikkath, J. et al. δ-catenin regulates spine and synapse morphogenesis
and function in hippocampal neurons during development. J. Neurosci. 29,
5435–5442 (2009).
5. Bamji, S.X., Rico, B., Kimes, N. & Reichardt, L.F. BDNF mobilizes synaptic
vesicles and enhances synapse formation by disrupting cadherin-β-catenin
interactions. J. Cell Biol. 174, 289–299 (2006).
6. Bamji, S.X. et al. Role of β-catenin in synaptic vesicle localization and
presynaptic assembly. Neuron 40, 719–731 (2003).
7. Elia, L.P., Yamamoto, M., Zang, K. & Reichardt, L.F. p120 catenin
regulates dendritic spine and synapse development through Rho-family
GTPases and cadherins. Neuron 51, 43–56 (2006).
8. Lee, S.H. et al. Synapses are regulated by the cytoplasmic tyrosine kinase
Fer in a pathway mediated by p120catenin, Fer, SHP-2, and β-catenin.
J. Cell Biol. 183, 893–908 (2008).
9. Li, Z. et al. Caspase-3 activation via mitochondria is required for long-
term depression and AMPA receptor internalization. Cell 141, 859–871
(2010).
10. Woo, J. et al. Trans-synaptic adhesion between NGL-3 and LAR regulates
the formation of excitatory synapses. Nat. Neurosci. 12, 428–437
(2009).
11. Kennedy, M.J., Davison, I.G., Robinson, C.G. & Ehlers, M.D. Syntaxin-4
defines a domain for activity-dependent exocytosis in dendritic spines. Cell
141, 524–535 (2010).
12. Williams, M.E. et al. Cadherin-9 regulates synapse-specific differentiation
in the developing hippocampus. Neuron 71, 640–655 (2011).
13. Tampellini, D., Rahman, N., Lin, M.T., Capetillo-Zarate, E. & Gouras, G.K.
Impaired β-amyloid secretion in Alzheimer’s disease pathogenesis.
J. Neurosci. 31, 15384–15390 (2011).
14. Volpicelli-Daley, L.A. et al. Exogenous α-synuclein fibrils induce Lewy body
pathology leading to synaptic dysfunction and neuron death. Neuron 72,
57–71 (2011).
15. Banker, G.A. & Cowan, W.M. Rat hippocampal neurons in dispersed cell
culture. Brain Res. 126, 397–342 (1977).
16. Banker, G.A. & Cowan, W.M. Further observations on hippocampal neurons
in dispersed cell culture. J. Comp. Neurol. 187, 469–493 (1979).
cortical neurons and glia. J. Vis. Exp. doi:10.3791/3257 3257 [pii] (2011).
18. Brewer, G.J., Torricelli, J.R., Evege, E.K. & Price, P.J. Optimized survival of
hippocampal neurons in B27-supplemented Neurobasal, a new serum-free
medium combination. J. Neurosci. Res. 35, 567–576 (1993).
19. Boraso, M. & Viviani, B. Glia-neuron sandwich cocultures: an in vitro
approach to evaluate cell-to-cell communication in neuroinflammation and
neurotoxicity. Methods Mol. Biol. 758, 135–152 (2011).
20. Chen, Y. et al. NS21: re-defined and modified supplement B27 for neuronal
cultures. J. Neurosci. Methods 171, 239–247 (2008).
21. Welshhans, K. & Bassell, G.J. Netrin-1-induced local β-actin synthesis and
growth cone guidance requires zipcode binding protein 1. J. Neurosci. 31,
9800–9813 (2011).
22. Chen, Y. et al. Murine hippocampal neurons expressing Fmr1 gene
premutations show early developmental deficits and late degeneration.
Hum. Mol. Genet. 19, 196–208 (2010).
23. Beaudoin, G.M. III. et al. Afadin, a Ras/Rap effector that controls
cadherin function, promotes spine and excitatory synapse density in the
hippocampus. J. Neurosci. 32, 99–110 (2012).
24. Tarsa, L. & Goda, Y. Synaptophysin regulates activity-dependent synapse
formation in cultured hippocampal neurons. Proc. Natl. Acad. Sci. USA 99,
1012–1016 (2002).
25. Zhang, G., Neubert, T.A. & Jordan, B.A. RNA binding proteins accumulate
at the postsynaptic density with synaptic activity. J. Neurosci. 32,
599–609 (2012).
26. Buerli, T. et al. Efficient transfection of DNA or shRNA vectors into
neurons using magnetofection. Nat. Protoc. 2, 3090–3101 (2007).

1754 | VOL.7 NO.9 | 2012 | nature protocols