Original research article

The authors used this protocol in:
Sep 2016

Navigate this Article


Isolation and Cultivation of Primary Brain Endothelial Cells from Adult Mice    

How to cite Favorites Q&A Share your feedback Cited by


Brain endothelial cells are the major building block of the blood-brain barrier. To study the role of brain endothelial cells in vitro, the isolation of primary cells is of critical value. Here, we describe a protocol in which vessel fragments are isolated from adult mice. After density centrifugation and mild digestion of the fragments, outgrowing endothelial cells are selected by puromycin treatment and grown to confluence within one week.

Keywords: Primary culture, Blood-brain barrier, Tight junctions, CD31, Occludin, Claudin-5, ZO-1, VE-cadherin


The blood-brain barrier protects the brain from uncontrolled entry of cells and substances. This is mainly achieved by brain endothelial cells that form a barrier composed of tight and adherens junctions to restrict paracellular transport.

This protocol was developed to overcome the limited availability of mouse brain endothelial cell lines that maintain their key characteristics, e.g., the expression of sufficient amounts of tight junction proteins such as occludin, ZO-1 or claudin-5 to induce a high transendothelial resistance.

In addition, the isolation of brain endothelial cells from genetically modified mice allows investigating of gene-specific functions in vitro.

Using this method, we previously complemented in vivo studies demonstrating the importance of NF-κB signaling in brain endothelial cells for maintaining normal blood-brain barrier function (Ridder et al., 2015).

Materials and Reagents

  1. Materials
    1. Multiwell plate (cell culture grade) (Greiner Bio One International, catalog number: 662160 )
    2. Cellulose chromatography paper (sterilize at 180 °C) (Whatman, catalog number: 3030-931 )
    3. 50 ml centrifuge tubes (cell culture grade) (Greiner Bio One International, catalog number: 210261 )
    4. 10 ml disposable pipette (Greiner Bio One International, catalog number: 607160 )
    5. Mice (C57BL/6, age 6 weeks up to 1 year from Charles River, Germany or the in-house breeding facility)
    6. Ice

  2. Reagents

    Catalog number
    Stock concentration
    Working concentration

    Hydrochloric acid (HCl) 1 N (sterile filtered)
    Carl Roth 

    1 L

    0.05 N 
    1 ml HCl 1 N + 19 ml H2O, sterile
    Dulbecco’s PBS (DPBS)

    500 ml 

    4 °C 
    70% EtOH (denatured)
    Th. Geyer 




    (sterile filtered) 
    Roche Diagnostics 

    500 mg/5 ml H2O, sterile 
    200 µl 
    -20 °C 
    100 mg/ml 
    1 mg/ml 
    100 µl collagenase/dispase in 10 ml medium

    DNase I 
    Roche Diagnostics

    100 mg/10 ml H2O, sterile
    100 µl 
    -20 °C

    10 mg/ml 
    4 µg/ml 
    40 µl DNase I in 10 ml medium

    Nα-Tosyl-L-lysine chloromethyl ketone hydrochloride (TLCK)

    14.7 mg/10 ml H2O, sterile
    200 µl 
    -20 °C 
    1.47 mg/ml diluted to 14.7 µg/ml
    0.147 µg/ml 
    100 µl TLCK in 10 ml medium 

    2.5 mg/10 ml H2O, sterile
    500 µl 
    -20 °C 
    0.25 mg/ml 
    8 µg/ml 
    32 µl puromycin in 1 ml medium
    Trypsin-EDTA 0.25% 
    Thermo Fisher Scientific
    100 ml 

    -20 °C 
    4% paraformaldehyde 
    Merck Millipore 

    4% paraformaldehyde in DPBS

    -20 °C 

    BD Pharmingen

    α-smooth muscle Actin (α-SMA)
    Acris Antibodies

    Wako Pure Chemical Industries

    Glial Fibrillary Acidic Protein (GFAP)
    EMD Millipore 

    Zona occludens-1
    Thermo Fisher Scientific

    Santa Cruz Biotechnology

    Claudin-5 (Cldn-5) 
    Thermo Fisher Scientific

    Occludin (Ocln)

    Mouse collagen, type IV 

    890 mg/ml 0.05 N HCl
    100 µl
    -80 °C 
    890 mg/ml 
    50 µg/ml 
    56 µl collagen IV + 944 µl 0.05 N HCl 
    Dextran MW 60,000-90,000
    Alfa Aesar 

    1 kg


    5.4 g dextran in 30 ml DPBS
    Penicillin/streptomycin (100x)

    100 ml
    1 ml
    -20 °C 
    100 µl pen/strep in 10 ml medium/dextran
    Thermo Fisher Scientific
    100 ml
    1 ml
    -20 °C 
    200 mM (100x) 
    2 mM (1x) 
    100 µl L-glutamine in 10 ml medium 
    DMEM-F12 w/o glutamine 
    Thermo Fisher Scientific
    500 ml

    4 °C 
    DMEM w/o glucose 
    Thermo Fisher Scientific
    500 ml

    4 °C 
    Plasma-derived bovine serum (PDS)
    First Link 

    500 ml
    10 ml 
    -20 °C 
    10 ml PDS in 50 ml medium
    Antibiotic/antimycotic (100x) 
    Thermo Fisher Scientific
    100 ml 
    1 ml 
    -20 °C 
    100 µl AA in 10 ml medium/dextran 

    PZN 003029843
    1 ampule 

    4 °C 
    5,000 I.E./ml 
    750 I.E./50 ml 
    150 µl heparin in 50 ml medium
    Endothelial Cell Growth Supplement (ECGS)

    15 mg/5 ml DPBS 
    500 µl 
    -20 °C 
    3 mg/ml 
    30 µg/ml 
    500 µl ECGS in 50 ml medium 
    18% dextran solution (see Recipes)

    Working medium (see Recipes)

    Digestion medium (see Recipes)

    Full medium (see Recipes)

    Note: Mouse collagen, type IV: Defrost stock vial slowly on ice at 4 °C overnight. Vortex thoroughly. Aliquot and store at -80 °C. The collagen concentration varies from lot to lot. Therefore, the amount of HCl added has to be adjusted for every new lot.


  1. Refrigerator (4 °C)
  2. Shaker
  3. Sterile beakers 100-150 ml (sterilize at 180 °C)
  4. Laminar flow work bench
  5. Dounce tissue grinder, 15 ml, autoclave (Sigma-Aldrich, catalog number: D9938 )
  6. Scalpel
  7. Tweezers (sterilize at 180 °C)
  8. Centrifuge (Hettich Lab Technology, model: UNIVERSAL 320 R ),
  9. Fixed-angle rotor (Hettich Lab Technology, catalog number: 1620A )
  10. Big scissors (sterilize at 180 °C)
  11. Small scissors (sterilize at 180 °C)
  12. Pipette or vacuum pump
  13. Water bath
  14. Microwave oven


  1. Preparations (Day 1)–Coating of wells with collagen
    1. Defrost one collagen aliquot for 2 wells of a 6-well plate or an according volume for other well sizes (see Table 1) slowly (2-3 h) on ice in a refrigerator (4 °C)
      If necessary, defrosting aliquots in the refrigerator without ice is possible.

      Table 1. Volume adjustment according to well size

    2. Dilute collagen to 50 µg/ml with 0.05 N HCl.
      Note: 0.05 N HCl aliquots can be stored at -20 °C.
    3. Vortex thoroughly. At least 10 sec until small bubbles form (turn the tube to ensure that the viscous collagen stock solution does not continuously stick to the bottom).
    4. Coat wells evenly with collagen solution (volume see Table 1).
    5. Put the plate on a shaker, 1 h at room temperature, 25 rpm.
    6. Move the plate to 4 °C for storage (overnight possible).
      Note: It is also possible to coat the plates on the day of endothelial cell preparation.

  2. Isolation (Day 2)
    1. Preparations:
      1. Ice
      2. Sterilization of instruments
      3. Approx. 50 ml DPBS in a 150 ml sterile beaker on ice (for each sample).
      4. Disinfectant (70% EtOH) in a sterile beaker on ice (approx. 50 ml in 150 ml sterile beaker)
      5. Switch on laminar flow and prepare cellulose chromatography paper, tissue grinder, scalpel, tweezers and 50 ml centrifuge tubes (one for each sample).
      6. Precool centrifuge to 4 °C.
    2. Anesthetize mice according to your local animal regulations. We use an overdose of isoflurane, which leads to breathing arrest within one minute. Decapitate the mouse with a big scissor and dip the head in ethanol (on ice). Remove the brain swiftly (Figure 1A) and store it in DPBS on ice. Repeat for all brains.
      1. Cut off cerebellum and olfactory bulb.
      2. Remove meninges by rolling the brains on cellulose chromatography paper using blunt tweezers.
      3. Cut cerebrum in 2 to 4 pieces and put the pieces in 5 ml working medium (4 °C). Repeat for all brains.
      4. Transfer brains with 5 ml working medium (4 °C) into a tissue grinder (Figure 1E) and homogenize (30 strokes with pistil A, 25 strokes with pistil B, Figure 1F). Use a maximum of 10 brains in one tissue grinder.
      5. Transfer homogenate into a 50 ml centrifuge tube. Rinse tissue grinder with 5 ml working medium (4 °C) and add to the homogenate (10 ml altogether).
      6. Centrifuge homogenate at 1,350 x g, 5 min, 4 °C. Remove supernatant carefully using a pipette or vacuum pump.
      7. Resuspend the pellet in 15 ml dextran solution and vortex extensively (2 min). The result is a white, cloudy, homogenous suspension (Figure 1G).
      8. Centrifuge at 6,080 x g, 10 min, 4 °C. In the meantime, supplement digestion medium with 100 µl collagenase/dispase, 40 µl DNase I and 100 µl TLCK each per 10 ml digestion medium. Pre-warm digestion medium to 37 °C.
      9. After centrifugation, remove the fluffy myelin layer (top, black arrows in Figures 1H and 1I) and the dextran as completely as possible. Use a 10 ml disposable pipette. Remove the filter of the pipette first if necessary.
      10. Resuspend the pellet (white arrows in Figures 1H and 1I) in 10 ml digestion medium (37 °C).
      11. Digest the tissue for 1 h 15 min in a 37 °C water bath (shake from time to time for 2 to 3 sec–approx. every 15 min).
      12. Centrifuge cell suspension at 1,350 x g, 5 min, room temperature. In the meantime, get the pre-coated plate from the refrigerator, fill sterile DPBS (10 ml per sample) in a centrifuge tube and heat it to 37 °C. Optionally, supplement full medium with puromycin and pre-warm to 37 °C (see step B2q).
      13. Remove digestion medium.
      14. Resuspend pellet in 10 ml warm DPBS.
      15. Centrifuge at 1,350 x g, 5 min, room temperature. In the meantime, remove collagen from the coated wells and wash twice with DPBS. DPBS from the second wash is left in the wells until cells are ready for seeding.
      16. Remove DPBS and resuspend the pellet in full medium. 2.5 ml full medium per well for a 6-well plate. Use 4-6 brains per culture plate.
      17. Mix cell suspension carefully before seeding to ensure even distribution.
      18. Add puromycin as indicated in Table 2. (Can be added directly to the full medium, see step B2l).

        Table 2. Culture volume according to well size

        Figure 1. Typical images for the preparation of the brain, removal of meninges, homogenization and subsequent dextran gradient centrifugation. The images depict the first steps of the isolation procedure showing the brain in situ after removal of the skullcap (A), before (B. cut planes indicated by dashed line) and after removal of cerebellum, olfactory bulb (C) and the meninges (D). Then, collect the brains in a Dounce tissue grinder (E), homogenize them (F), and centrifuge the tissue homogenate. Next, resuspend the cells in the dextran solution and vortex extensively (G). Following the centrifugation, the resulting myelin layer is at the top while the vessel fragments collect around the edge of the tube bottom (H + I, black arrows: myelin layer, white arrows: pellet location). The size of the vessel fragment pellet depends on the number of brains used. In E-I, two brains were used for the preparation.

    3. Day 3
      1. Wash cells twice with DPBS.
      2. Change full medium.
      3. Add puromycin (alternatively, puromycin can be added in advance to the full medium).
    4. Day 4
      1. Change full medium.
        Note: No puromycin needed anymore.

  3. Cultivation
    1. Change medium 1-2 times per week, first time approx. 4-6 days after isolation.
    2. Split the culture 1:2 (or 1:3) if the cells are confluent. Use trypsin 5-10 min and inactivate with full medium.
    3. Plate cells and change medium the next day.

  4. Purity of the cell culture (Figures 2 and 3)
    Endothelial cells (CD31+) > 95%
    Pericytes (α-SMA+) < 5%
    No astrocytes (GFAP+), microglia (Iba1+), neurons (NeuN+)

    Figure 2. Representative immunofluorescence images of primary mouse brain endothelial cells. Cells were fixed with 4% paraformaldehyde 14 days after isolation and subsequently stained for CD31 (BD, 1:500) as an endothelial cell specific marker in combination with α-SMA (pericytes and smooth muscle cells, Acris, 1:200, upper row), Iba1 (microglia, Wako Pure Chemical Industries, 1:100, middle row) and GFAP (astrocytes, Millipore, 1:400, lower row). Scale bars represent 50 µm.

    Figure 3. Primary mouse brain endothelial cells maintain expression of tight and adherens junction proteins. Cells were fixed 6-8 days after isolation with ice-cold methanol and subsequently stained for the tight junction proteins Ocln (Sigma-Aldrich, 1:500), ZO-1 (Thermo Fisher Scientific, 1:500), Cldn-5 (Thermo Fisher Scientific, 1:500) and the adherens junction protein VE-Cadherin (Santa Cruz Biotechnology, 1:500). Scale bars represent 50 µm.

Data analysis

Primary mouse brain endothelial cells isolated by this method can be used for a variety of methods that include protein and gene expression analysis, assessment of transendothelial resistance using transwell inserts and transmigration or adhesion assays. In addition, these cells can also be grown on glass coverslips coated with collagen IV for live imaging, e.g., to monitor intracellular calcium dynamics.


  1. This protocol was developed to isolate brain endothelial cells from adult mice. We successfully isolated and cultured cells from young mice (6-8 weeks) as well as old mice (up to 1 year) in our laboratory without any modifications to the protocol. Using brains from other mouse strains has not been tested in our laboratory.
  2. Cells usually reach confluence within 6-8 days (see Figure 4). They can be maintained in culture but are eventually overgrown by pericytes after several weeks.
  3. After approximately 10 days cells do not adhere as strongly as before and are more likely to detach during staining procedures.

    Figure 4. Bright field images of primary murine brain endothelial cells 2 days (left) or 7 days (right) after isolation. Note the attached vessel fragment and its radially outgrowing endothelial cells after 2 days in culture. Scale bars represent 100 µm.


  1. 18% dextran solution (for 2 preparations)
    5.4 g dextran dissolved in 30 ml DPBS by heating (microwave)
    300 µl penicillin/streptomycin (100x)
    300 µl L-glutamine (200 mM)
    The solution can be stored at -20 °C and defrosted before usage, but penicillin/streptomycin and L-glutamine should be added after thawing
  2. Working medium (for 2 preparations)
    20 ml DMEM-F12
    200 µl penicillin/streptomycin (100x)
    200 µl L-glutamine (200 mM)
  3. Digestion medium (for 2 preparations)
    20 ml DMEM
    200 µl penicillin/streptomycin (100x)
    200 µl collagenase/dispase–add right before digestion
    200 µl TLCK–add right before digestion
    80 µl DNase I–add right before digestion
  4. Full medium (max. storage time 3-4 weeks at 4 °C)
    40 ml DMEM-F12
    10 ml PDS
    500 µl antibiotic/antimycotic (100x)
    500 µl L-glutamine (200 mM)
    150 µl heparin (5,000 U/ml)
    500 µl ECGS


The protocol described here has been modified based on the method published by Song and Pachter (2003). We would like to thank Beate Lembrich for expert technical assistance. This work was funded by the Deutsche Forschungsgemeinschaft (SCHW416/5-2, 416/9-1).


  1. Ridder, D. A., Wenzel, J., Müller, K., Töllner, K., Tong, X. K., Assmann, J. C., Stroobants, S., Weber, T., Niturad, C., Fischer, L., Lembrich, B., Wolburg, H., Grand'Maison, M., Papadopoulos, P., Korpos, E., Truchetet, F., Rades, D., Sorokin, L. M., Schmidt-Supprian, M., Bedell, B. J., Pasparakis, M., Balschun, D., D’Hooge, R., Löscher, W., Hamel, E. and Schwaninger, M. (2015). Brain endothelial TAK1 and NEMO safeguard the neurovascular unit. J Exp Med 212(10): 1529-1549.
  2. Song, L. and Pachter, J. S. (2003). Culture of murine brain microvascular endothelial cells that maintain expression and cytoskeletal association of tight junction-associated proteins. In Vitro Cell Dev Biol Anim 39(7): 313-320.
Please login or register for free to view full text
Copyright: © 2017 The Authors; exclusive licensee Bio-protocol LLC.
How to cite:  Readers should cite both the Bio-protocol article and the original research article where this protocol was used:
  1. Assmann, J. C., Müller, K., Wenzel, J., Walther, T., Brands, J., Thornton, P., Allan, S. M. and Schwaninger, M. (2017). Isolation and Cultivation of Primary Brain Endothelial Cells from Adult Mice. Bio-protocol 7(10): e2294. DOI: 10.21769/BioProtoc.2294.
  2. Khan, M. A., Schultz, S., Othman, A., Fleming, T., Lebron-Galan, R., Rades, D., Clemente, D., Nawroth, P. P. and Schwaninger, M. (2016). Hyperglycemia in Stroke Impairs Polarization of Monocytes/Macrophages to a Protective Noninflammatory Cell Type. J Neurosci 36(36): 9313-9325.

If you have any questions/comments about this protocol, you are highly recommended to post here. We will invite the authors of this protocol as well as some of its users to address your questions/comments. To make it easier for them to help you, you are encouraged to post your data including images for the troubleshooting.

If you have any questions/comments about this protocol, you are highly recommended to post here. We will invite the authors of this protocol as well as some of its users to address your questions/comments. To make it easier for them to help you, you are encouraged to post your data including images for the troubleshooting.

We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.