Introduction

The Bend.3 Cells line — also known by the synonyms bEND.3, b.End3, bEnd3, BEND-3, BEND3, and brain-derived Endothelial cells.3 — is a transformed mouse brain microvascular endothelial cell line that has become one of the most widely used in vitro tools in neurovascular biology. The line was established from murine brain endothelioma tissue following transformation with the polyoma virus middle T antigen, a manipulation that confers stable, indefinite proliferative capacity while preserving key endothelial characteristics. Since its characterisation in the early 1990s, bEnd.3 has served as an accessible and reproducible surrogate for studying the blood-brain barrier (BBB), cerebral ischemia, stroke, reperfusion injury, angiogenesis, and endothelial cell biology more broadly. Its utility in modelling BBB disruption in the context of ischemic stroke, Alzheimer disease, and traumatic brain injury has made it a widely adopted tool in both basic and translational neuroscience research.

Key Takeaways

• bEnd.3 is a transformed mouse brain microvascular endothelial cell line derived from BALB/c mouse brain tissue and transformed with the polyoma virus middle T antigen (Cellosaurus accession CVCL_0170).
• It is widely used as an in vitro model of the blood-brain barrier for studying ischemic stroke, reperfusion injury, Alzheimer disease, and neuroinflammation.
• The cells express key BBB-associated markers including tight junction proteins claudin-5, occludin, and ZO-1, as well as eNOS and PECAM-1.
• bEnd.3 cells express lymphatic endothelial markers such as LYVE-1 and VEGFR-3, broadening their relevance to vascular biology research.
• The cell line has a doubling time of 26–30 hours, making it efficient for large-scale experimental work.
• Barrier tightness is limited: a maximum transendothelial electrical resistance of approximately 130 Ω·cm² was reached only in astroglial co-culture, and the line discriminates poorly between paracellular and transcellular permeability markers.
• bEnd.3 cells have been shown to support the maintenance of human pluripotent stem cells as feeder cells, expanding their applications in stem cell biology.

What is bEnd.3?

bEnd.3 was derived from brain tissue of a 6-week-old BALB/c mouse of unspecified sex (Cellosaurus accession CVCL_0170). The original endothelioma cells were generated by transforming murine endothelial cells with the polyoma virus middle T (mT) antigen. A landmark 1990 study by Montesano, Pepper, and colleagues demonstrated that mT-expressing endothelioma cells exhibit markedly increased proteolytic activity — particularly elevated urokinase-type plasminogen activator together with reduced plasminogen activator inhibitor production — and form large hemangioma-like cystic structures within fibrin gels, explaining the oncogene's profound subversion of normal vascular development [1]. This mechanistic foundation helped define the transformed phenotype that underpins bEnd.3's biological behaviour and utility in experimental settings.

Subsequent characterisation work has defined the cell line's morphological and molecular profile in detail. Omidi and colleagues reported that bEnd.3 cells display a distinctive spindle-like squamous morphology in culture and express vesicle-associated proteins including clathrin, caveolin-1, flotillin, and dynamin II, consistent with active vesicular trafficking machinery [2]. The same study confirmed transcripts for the tight junctional elements ZO-1, JAM, occludin, claudin-1 and claudin-5, as well as functional GLUT-1-mediated glucose uptake and P-glycoprotein efflux. In co-culture with C6 astroglia the monolayers reached a maximum transendothelial electrical resistance of approximately 130 Ω·cm², although the authors also found that the cells did not meaningfully discriminate between paracellular and transcellular permeability probes. Yuen and colleagues later reported that bEnd.3 cells also express lymphatic endothelial markers LYVE-1 and VEGFR-3, as well as progenitor markers Sca-1 and CD133, and form capillary-type tubes upon stimulation with TNF-alpha — a discovery that broadened the cell line's biological significance beyond classical blood vascular research [3].

Advantages of bEnd.3

One of the most significant advantages of bEnd.3 is its accessibility and practical reproducibility. Unlike primary brain microvascular endothelial cells, which require laborious isolation procedures and exhibit limited passaging capacity, bEnd.3 is an immortalised line that can be expanded reliably across laboratories. With a doubling time of 26–30 hours, the cells are well-suited to experiments requiring large numbers of uniform cells. Their stable expression of endothelial markers — including tight junction proteins claudin-5, occludin, and ZO-1, as well as eNOS and PECAM-1 — makes them a practical stand-in for brain microvascular endothelium across a broad range of experimental conditions, provided the barrier-function caveats outlined below are taken into account.

The versatility of bEnd.3 extends across multiple research disciplines. Studies have used this line to model BBB disruption during ischemic stroke and reperfusion injury, investigate neuroinflammation triggered by adhesion molecules, profile angiogenic gene expression, and explore connexin-mediated intercellular communication. Furthermore, Joubin and colleagues demonstrated that bEnd.3 cells can sustain human pluripotent stem cells in an undifferentiated state, functioning as feeder cells that can be genetically engineered — a capability not easily replicated with short-lived primary cultures [4]. This breadth of application speaks to the cell line's utility across neuroscience, vascular biology, pharmacology, and stem cell research.

Limitations of bEnd.3

Despite its widespread use, bEnd.3 presents important limitations when employed as a BBB model. Omidi and colleagues measured a maximum transendothelial electrical resistance of only about 130 Ω·cm², and that value required co-culture with C6 astroglia; more critically, the monolayers showed little discrimination between transcellular and paracellular probes, with permeability coefficients for propranolol and sucrose differing only marginally [2]. The authors concluded that the line is unsuitable for transendothelial permeability assessments, while remaining valuable for studies of BBB transport mechanisms. Permeability data generated in this system should therefore be interpreted with caution and, where possible, supplemented with co-culture systems or in vivo validation.

The organisation of the tight junctions themselves is a related weakness. Song and Pachter compared bEnd.3 with freshly purified primary murine brain microvascular endothelial cells and found that, whereas the primary cells concentrated claudin-5, occludin, and ZO-1 along intercellular borders and partitioned occludin and ZO-1 with the detergent-resistant cytoskeletal framework, bEnd.3 cells displayed diffuse cytoplasmic localisation of occludin and ZO-1 and retained these proteins predominantly in the soluble fraction [5]. Expression of a tight junction protein in bEnd.3 therefore does not by itself imply correct junctional assembly.

The transformed nature of bEnd.3, arising from mT oncogene expression, may also introduce phenotypic artefacts not present in normal cerebral endothelium. Elevated proteolytic activity — a defining feature of mT-expressing endothelioma cells [1] — and the expression of lymphatic markers such as LYVE-1 and VEGFR-3 suggest the cell line does not fully recapitulate normal blood vascular endothelium [3]. Additionally, donor sex is unspecified for this line, which limits the applicability of findings to sex-stratified research questions. These considerations do not diminish the cell line's value, but they do underscore the importance of understanding its specific biological context.

View full specifications and order

Applications

Blood-Brain Barrier Modelling and Ischemic Stroke

The most prominent application of bEnd.3 is as an in vitro model of the blood-brain barrier under ischemic and post-ischemic conditions. Koto and colleagues demonstrated that hypoxia (1% O₂) altered the localisation and protein level of claudin-5 in bEnd.3 monolayers, causing a measurable decrease in transendothelial electrical resistance, and confirmed the corresponding loss of claudin-5 in retinal microvascular endothelium in vivo [6]. This established a mechanistic link between tight junction remodelling and barrier breakdown during neural hypoxia, and positioned bEnd.3 as a tractable system for studying the molecular events underlying BBB failure in stroke.

A frequently cited study by Kim and colleagues used the cell line as an in vitro BBB model exposed to oxygen-glucose deprivation (OGD) to simulate ischemia. Eighteen hours of OGD induced cellular dysfunction, increased permeability, degradation of occludin, and activation of autophagy pathways, with degraded occludin co-localising with ischemic autophagosomes; both the occludin loss and the permeability change were markedly reduced by the autophagy inhibitor 3-methyladenine. The same molecular changes and their reversal by 3-methyladenine were subsequently reproduced in brain capillaries from rats subjected to middle cerebral artery occlusion [7]. This work established a mechanistic framework linking autophagic flux to tight junction protein loss and provided important translational insights into stroke pathology.

Complementary research demonstrated that agmatine attenuates matrix metalloproteinase-2 and -9 expression in cerebral endothelial cells — including bEnd.3 — subjected to OGD-reperfusion injury, with the downregulation of MMP-9 running parallel to upregulation of eNOS and maintained nitric oxide release, offering a potential neuroprotective mechanism relevant to brain injuries and reperfusion injury [8]. Studies on Ang II-mediated apoptosis further showed that hyperoside protects bEnd.3 cells and preserves the integrity of an in vitro BBB monolayer, acting on JAM-A, claudin-5, ZO-1, caveolin-1, and Mfsd2a and suggesting therapeutic avenues for cerebrovascular protection [9]. Research on circLphn3 overexpression in bEnd.3 cells also highlighted the role of non-coding RNA in maintaining tight junction protein expression — including ZO-1, ZO-2, and occludin — following traumatic brain injury, with circLphn3 acting as a molecular sponge for miR-185-5p [10].

Neuroinflammation, Endothelial Toxicology, and Alzheimer Disease

bEnd.3 cells have been extensively used to study inflammatory signalling at the cerebrovascular endothelium. Jeong and colleagues showed that cadmium — a ubiquitous heavy metal that accumulates outside the BBB — stimulated ICAM-1 expression in bEnd.3 cells through NF-κB activation driven by tyrosine phosphorylation of IκBα rather than its degradation, providing a mechanistic basis for cadmium-induced cerebrovascular toxicity [11]. Similarly, Zhang and colleagues reported that C-reactive protein at concentrations of 1–75 mg/L increased ICAM-1 and VCAM-1 expression and LDH leakage in bEnd.3 cells, suggesting a direct role for this acute-phase protein in BBB injury during ischemic stroke [12].

Pesticide-induced endothelial toxicity has also been investigated in this cell line. Jung and colleagues demonstrated that N-nitrosocarbofuran, the nitroso metabolite of the carbamate pesticide carbofuran, induced apoptosis in bEnd.3 cells via MAP kinase pathways, with the ERK pathway playing a predominant role in cell death [13]. Lysophosphatidylcholine-induced cytotoxicity in bEnd.3 cells was examined by Tsai and colleagues, who traced cell death to cytosolic calcium overload, loss of mitochondrial membrane potential, p38 activation, and caspase-3 activation — notably without reactive oxygen species generation — and identified heparin as a protective agent acting downstream of p38 [14]. More recent work on florfenicol treatment documented ROCK/cofilin-mediated F-actin remodelling contributing to reduced proliferation and migration in bEnd.3 endothelial cells [15].

In the realm of Alzheimer disease research, bEnd.3 cells have served as a model for amyloid beta peptide-mediated BBB disruption. Chan and colleagues showed that Aβ oligomer-induced alterations of autophagy in bEnd.3 cells were blocked by RAGE small interfering RNA, linking RAGE signalling through the CaMKKβ/AMPK/mTOR axis to autophagy-dependent injury of the tight junction scaffold proteins ZO-1, occludin, and claudin-5 [16]. Separately, lychee seed polyphenol was shown to protect BBB integrity by inhibiting Aβ(25–35)-induced NLRP3 inflammasome activation through the AMPK/mTOR/ULK1-mediated autophagy pathway in bEnd.3 cells, with results corroborated in APP/PS1 transgenic mice — a study that advanced understanding of autophagy-inflammasome crosstalk at the neurovascular interface [17]. Photobiomodulation has been examined in the same context: Ma and colleagues used LPS-stimulated bEnd.3 cells alongside APP/PS1 mice to show that near-infrared treatment improves mitochondrial energy metabolism, upregulates occludin, claudin-5, and ZO-1, and protects barrier integrity in an AMPK-dependent manner [18].

Angiogenesis and Vascular Signalling

bEnd.3 has proven highly informative in studies of angiogenic signalling. Yue and colleagues employed cDNA microarray to profile angiogenic gene expression in bEnd.3 cells treated with basic fibroblast growth factor (bFGF), identifying upregulation of pro-angiogenic genes including ADAMTS1, MMP-9, angiopoietin-1, PDGF-B, and G-CSF, downregulation of anti-angiogenic genes such as TIMP-2 and TSP-3, and induction of cyclooxygenase-2 (COX-2) at both mRNA and protein level [19]. This high-throughput approach established bEnd.3 as a convenient platform for genome-scale angiogenesis profiling.

Hypoxia-driven angiogenic responses have also been mapped using bEnd.3 cells. Zhu and colleagues demonstrated that hypoxia increased endoglin (CD105) mRNA and protein expression in bEnd.3 cells through activation of p38 MAPK, with inhibitors of this pathway — and dominant negative MKK3 — suppressing endoglin induction [20]. Since endoglin is mutated in hereditary haemorrhagic telangiectasia and is upregulated during active angiogenesis, these findings linked hypoxia-sensitive MAPK signalling to pathological vascular remodelling. Li and colleagues further showed that hypoxia upregulates CD73 (ecto-5'-nucleotidase) activity, mRNA, and protein in bEnd.3 cells in a time-dependent manner, while reoxygenation initially reversed this induction before CD73 rose again with prolonged reoxygenation — implicating adenosine generation as a neuroprotective mechanism during ischaemia-reperfusion [21].

Angiopoietin signalling has been studied in depth using bEnd.3 as a model system. Shi and colleagues established a serum-free apoptosis model in bEnd.3 cells and demonstrated that angiopoietin-1 inhibits microvascular endothelial cell apoptosis through a mechanism involving decreased Bax, increased Bcl-2, and reduced release of cytochrome C from mitochondria into the cytoplasm [22]. Perry and colleagues extended angiopoietin biology to hemangioma research, demonstrating constitutive activation of the Tie-2 receptor in human hemangioma of infancy and using bEnd.3 cells as a murine model hemangioma system in which pharmacological inhibition of angiopoietin-2 production nearly abolished tumour growth in vivo [23].

eNOS, Connexins, and Endothelial Function

Endothelial nitric oxide synthase (eNOS) biology has been characterised in bEnd.3 cells with considerable precision. Govers and colleagues used immunofluorescence microscopy to show that eNOS localisation at the plasma membrane depends critically on cell-cell contact, with strong enrichment at intercellular contact sites resembling the PECAM-1 distribution pattern rather than that of VE-cadherin or plakoglobin, and with eNOS activity higher in confluent than in subconfluent monolayers despite unchanged expression [24]. This spatial coupling between eNOS localisation and activity highlighted bEnd.3 as a model where the physiological context of confluent endothelium can be meaningfully interrogated.

Bevers and colleagues investigated eNOS uncoupling in bEnd.3 cells, which express eNOS at levels more than two orders of magnitude above those of a comparison endothelial line while lacking iNOS and nNOS. They found that tetrahydrobiopterin (BH4) supplementation increased NO production and decreased reactive oxygen species generation, whereas adding extra L-arginine increased VEGF-induced ROS further — indicating that relative BH4 deficiency, not arginine availability, drives uncoupling in this setting [25]. This result has direct relevance to the vascular complications of conditions such as hypertension and atherosclerosis, where sustained eNOS overexpression can paradoxically promote oxidative stress.

Connexin-mediated gap junctional communication has also been examined in bEnd.3 cells. Kwak and colleagues compared connexin expression and cell-to-cell coupling before and after mechanical wounding of bEnd.3 monolayers, observing wound-induced upregulation of Cx43 and downregulation of Cx37 at the injury site, with Cx40 unaffected [26]. Stable transfection with dominant negative connexin inhibitors prolonged the time required for complete wound closure by roughly 50% without altering proliferation, single-cell migration, or proteolytic activity, establishing a specific role for gap junctional communication in coordinated endothelial wound repair.

Neurotrophic Factor Secretion and Stem Cell Applications

bEnd.3 cells secrete neurotrophic factors that act on the surrounding neural microenvironment. Wang and colleagues demonstrated that brain-derived neurotrophic factor (BDNF) is expressed and released by bEnd.3 cells, with both sustained and intermittent hypoxia stimulating BDNF secretion — intermittent hypoxia being the more potent stimulus [27]. Hypoxia-induced BDNF release was partially inhibited by N-acetyl-L-cysteine and by a superoxide dismutase mimetic, implicating reactive oxygen species in hypoxia-driven neurotrophic signalling, and required calcium mobilisation from internal stores. These findings established brain microvascular endothelium as an active participant in neural survival, not merely a passive barrier.

Most strikingly, Joubin and colleagues showed that bEnd.3 cells maintain human pluripotent stem cells in an undifferentiated state, secreting a soluble activity that mimics the niche provided by mouse embryonic fibroblasts [4]. The authors further demonstrated that bEnd.3 cells can be genetically engineered to express fluorescent reporters or drug-resistance genes, creating designer feeder layers for stem cell culture — an application that significantly expands the cell line's utility beyond classical endothelial research.

Conclusion

bEnd.3 remains one of the most versatile and productive cell lines in neurovascular and BBB research. From modelling ischemic stroke and reperfusion injury to dissecting eNOS signalling, connexin-mediated wound repair, angiogenic gene expression, and autophagy-driven tight junction disruption, this mouse brain microvascular endothelial cell line has contributed insights across multiple biomedical disciplines. Its applicability to questions in Alzheimer disease, traumatic brain injuries, and neuroprotective agent screening further cements its place in the neuroscience research toolkit — provided its limited barrier tightness and transformed phenotype are accounted for in the experimental design. Researchers seeking a well-characterised, reproducible, and experimentally tractable in vitro endothelial and BBB-transport model are encouraged to explore Bend.3 Cells available at cytion.com, where full product specifications and ordering information are provided.

Key Publications

1. Montesano R, Pepper MS, Möhle-Steinlein U, et al. (1990) Increased proteolytic activity is responsible for the aberrant morphogenetic behavior of endothelial cells expressing the middle T oncogene. Cell 62:435–445. PMID: 2379237
2. Omidi Y, Campbell L, Barar J, et al. (2003) Evaluation of the immortalised mouse brain capillary endothelial cell line, b.End3, as an in vitro blood-brain barrier model for drug uptake and transport studies. Brain Research 990:95–112. PMID: 14568334
3. Yuen D, Leu R, Tse J, et al. (2014) Novel characterization of bEnd.3 cells that express lymphatic vessel endothelial hyaluronan receptor-1. Lymphology 47:73–81. PMID: 25282873
4. Joubin K, Richardson A, Novoa N, et al. (2012) The endothelial cell line bEnd.3 maintains human pluripotent stem cells. Stem Cells and Development 21:2312–2321. PMID: 22224974
5. Song L, Pachter JS (2003) Culture of murine brain microvascular endothelial cells that maintain expression and cytoskeletal association of tight junction-associated proteins. In Vitro Cellular & Developmental Biology. Animal 39:313–320. PMID: 14613336
6. Koto T, Takubo K, Ishida S, et al. (2007) Hypoxia disrupts the barrier function of neural blood vessels through changes in the expression of claudin-5 in endothelial cells. The American Journal of Pathology 170:1389–1397. PMID: 17392177
7. Kim KA, Kim D, Kim JH, et al. (2020) Autophagy-mediated occludin degradation contributes to blood-brain barrier disruption during ischemia in bEnd.3 brain endothelial cells and rat ischemic stroke models. Fluids and Barriers of the CNS 17:21. PMID: 32169114
8. Yang MZ, Mun CH, Choi YJ, et al. (2007) Agmatine inhibits matrix metalloproteinase-9 via endothelial nitric oxide synthase in cerebral endothelial cells. Neurological Research 29:749–754. PMID: 17588309
9. Xie YY, Lu YW, Yu GR (2022) The protective effects of hyperoside on Ang II-mediated apoptosis of bEnd.3 cells and injury of blood-brain barrier model in vitro. BMC Complementary Medicine and Therapies 22:157. PMID: 35698113
10. Cheng YQ, Wu CR, Du MR, et al. (2022) CircLphn3 protects the blood-brain barrier in traumatic brain injury. Neural Regeneration Research 17:812–818. PMID: 34472480
11. Jeong EM, Moon CH, Kim CS, et al. (2004) Cadmium stimulates the expression of ICAM-1 via NF-kappaB activation in cerebrovascular endothelial cells. Biochemical and Biophysical Research Communications 320:887–892. PMID: 15240131
12. Zhang J, Rui YC, Yang PY, et al. (2006) C-reactive protein induced expression of adhesion molecules in cultured cerebral microvascular endothelial cells. Life Sciences 78:2983–2988. PMID: 16412476
13. Jung YS, Kim CS, Park HS, et al. (2003) N-nitrosocarbofuran induces apoptosis in mouse brain microvascular endothelial cells (bEnd.3). Journal of Pharmacological Sciences 93:489–495. PMID: 14737022
14. Tsai TY, Leong IL, Cheng KS, et al. (2019) Lysophosphatidylcholine-induced cytotoxicity and protection by heparin in mouse brain bEND.3 endothelial cells. Fundamental & Clinical Pharmacology 33:52–62. PMID: 29974515
15. Hu D, Shen Z, Hou X, et al. (2025) ROCK-mediated actin remodeling contributes to the reduced proliferation and migration of bEnd.3 endothelial cells treated with florfenicol. Journal of Biochemical and Molecular Toxicology 39:e70453. PMID: 40826991
16. Chan Y, Chen W, Chen Y, et al. (2020) Yi-Zhi-Fang-Dai formula exerts a protective effect on the injury of tight junction scaffold proteins in vitro and in vivo by mediating autophagy through regulation of the RAGE/CaMKKβ/AMPK/mTOR pathway. Biological & Pharmaceutical Bulletin 43:1847–1858. PMID: 33268702
17. Xiong R, Zhou XG, Tang Y, et al. (2021) Lychee seed polyphenol protects the blood-brain barrier through inhibiting Aβ(25-35)-induced NLRP3 inflammasome activation via the AMPK/mTOR/ULK1-mediated autophagy in bEnd.3 cells and APP/PS1 mice. Phytotherapy Research 35:954–973. PMID: 32893437
18. Ma C, Ye Y, Shi X, et al. (2025) Photobiomodulation mitigates blood-brain barrier disruption in APP/PS1 mouse model of Alzheimer's disease by activating the AMPK pathway. Alzheimer's Research & Therapy 17:141. PMID: 40551167
19. Yue F, Zhang GP, Jin HM (2006) Effects of basic fibroblast growth factor on the expressions of angiogenic gene profile and cyclooxygenase-2 in brain microvascular endothelial cells. Sheng Li Xue Bao 58:124–128. PMID: 16628358
20. Zhu Y, Sun Y, Xie L, et al. (2003) Hypoxic induction of endoglin via mitogen-activated protein kinases in mouse brain microvascular endothelial cells. Stroke 34:2483–2488. PMID: 12947156
21. Li X, Zhou T, Zhi X, et al. (2006) Effect of hypoxia/reoxygenation on CD73 (ecto-5'-nucleotidase) in mouse microvessel endothelial cell lines. Microvascular Research 72:48–53. PMID: 16828810
22. Shi LG, Zhang GP, Jin HM (2006) Inhibition of microvascular endothelial cell apoptosis by angiopoietin-1 and the involvement of cytochrome C. Chinese Medical Journal 119:725–730. PMID: 16701012
23. Perry BN, Govindarajan B, Bhandarkar SS, et al. (2006) Pharmacologic blockade of angiopoietin-2 is efficacious against model hemangiomas in mice. The Journal of Investigative Dermatology 126:2316–2322. PMID: 16741507
24. Govers R, Bevers L, de Bree P, et al. (2002) Endothelial nitric oxide synthase activity is linked to its presence at cell-cell contacts. The Biochemical Journal 361:193–201. PMID: 11772391
25. Bevers LM, Braam B, Post JA, et al. (2006) Tetrahydrobiopterin, but not L-arginine, decreases NO synthase uncoupling in cells expressing high levels of endothelial NO synthase. Hypertension 47:87–94. PMID: 16344367
26. Kwak BR, Pepper MS, Gros DB, et al. (2001) Inhibition of endothelial wound repair by dominant negative connexin inhibitors. Molecular Biology of the Cell 12:831–845. PMID: 11294890
27. Wang H, Ward N, Boswell M, et al. (2006) Secretion of brain-derived neurotrophic factor from brain microvascular endothelial cells. The European Journal of Neuroscience 23:1665–1670. PMID: 16553631