Localization of Angulin-1/LSR and Tricellulin at Tricellular Contacts of Brain and Retinal Endothelial Cells in vivo

Abstract

The paracellular pathway of an epithelial cellular sheet can be divided into two parts: one between two adjacent cells sealed by tight junctions (TJs) and one at tricellular contacts (TCs), where the corners of three cells meet. At TCs of epithelial cells, there is a specialized mode of TJs, namely tricellular TJs (tTJs), required for full barrier function of the cellular sheet. However, tTJs have not been described in endothelial cells to date. Here, we investigated whether tTJs occur in endothelial cells by analyzing the TC localizations of tTJ markers, tricellulin and angulin family proteins (angulin-1/LSR, angulin-2/ILDR1, and angulin-3/ILDR2), by immunofluorescence staining of frozen sections of various tissues from adult mice. Endothelial TCs in most tissues revealed no detectable staining of tricellulin or angulins. However, tricellulin and angulin-1/LSR were specifically concentrated in TCs of brain and retinal endothelial cells, which form the blood–brain barrier (BBB) and inner blood–retinal barrier (BRB), respectively. Even in the brain, endothelial cells in the choroid plexus and the median eminence, one of the circumventricular organs, did not show concentration of tricellulin or angulins at TCs. These findings indicate the existence of tTJs in endothelial cells in vivo and suggest that tTJs impart important characteristics to the BBB and inner BRB.

Introduction

Endothelial cellular sheets are generally fairly leaky compared with epithelial cellular sheets, consistent with the basic functions of blood vessels, i.e. delivery of oxygen, nutrients, and water throughout the body (Bendayan, 2002; Komarova and Malik, 2010). However, the endothelia in the central nervous system and retina, which form the blood–brain barrier (BBB) and inner blood–retinal barrier (BRB), respectively, are much less permeable. The endothelial cells in such organs have a strong paracellular seal, which is established by the continuous plasma membrane and developed tight junctions (TJs) (Abbott et al., 2010; Engelhardt and Sorokin, 2009). TJs are one mode of epithelial and endothelial cell–cell junctions and circumscribe the cells as belts (Farquhar and Palade, 1963). TJs prevent leakage of solutes through the paracellular space, thereby contributing to the barrier function of epithelial and endothelial cellular sheets (Anderson and Van Itallie, 2009; Tsukita et al., 2001). The structural core of TJs consists of a network of fibril-like structures that form complete contacts between the adjacent plasma membranes, namely TJ strands (Staehelin et al., 1969). The claudin family membrane proteins are the major components of TJ strands and determine the barrier property of TJs (Anderson and Van Itallie, 2009; Angelow et al., 2008; Furuse, 2010).

To be more precise, the paracellular pathway within a cellular sheet is divided into two parts: one between two adjacent cells and one at tricellular contacts (TCs), where the corners of three cells meet. To maintain the integrity of cellular sheets, it is essential to tightly seal TCs in addition to bicellular contacts with TJs. Epithelial cells have a specialized mode of TJs at TCs, named tricellular tight junctions (tTJs) (Ikenouchi et al., 2005). In freeze-fracture replica electron micrographs of TCs, the most apical elements of the TJ strands from both sides join and turn to extend in the basal direction together to form tTJs (Friend and Gilula, 1972; Staehelin, 1973; Wade and Karnovsky, 1974; Walker et al., 1985). Thus, there are three sealing elements closely attached to each other to form a narrow tube in the extracellular space at TCs (Staehelin, 1973). Two types of integral membrane proteins, tricellulin (Ikenouchi et al., 2005) and angulin family proteins, which include angulin-1/LSR, angulin-2/ILDR1, and angulin-3/ILDR2 (Higashi et al., 2013; Masuda et al., 2011), have been identified as specific molecular components of tTJs. Tricellulin is a member of the TAMP family containing occludin with four transmembrane domains (Ikenouchi et al., 2005; Raleigh et al., 2010; Steed et al., 2009), while angulins are single-pass membrane proteins with an Ig-like domain (Masuda et al., 2011; Higashi et al., 2013). Notably, angulins recruit tricellulin to TCs and at least one of the angulin family members is expressed in each epithelium with tricellulin (Higashi et al., 2013; Masuda et al., 2011). shRNA-mediated depletion of tricellulin or angulin-1/LSR in cultured epithelial cells revealed that these molecules are required for normal tTJ formation as well as the full barrier function of epithelial cellular sheets (Higashi et al., 2013; Ikenouchi et al., 2005; Masuda et al., 2011). Therefore, it is reasonable to consider that the concentrated localization of a combination of tricellulin and angulins at TCs is a good indicator of tTJs, despite their original definition by freeze-fracture replica electron microscopy.

To date, tTJs have been observed and analyzed in epithelial cells, but have not been described in endothelial cells. Based on intensive analyses by freeze-fracture replica electron microscopy, Walker et al. (1994) reported that TJs are discontinuous at TC regions of pulmonary capillary endothelial cells, indicating a lack of tTJs. It is reasonable that tTJs are not required for leaky endothelia. On the other hand, specific endothelial cells, such as those in brain capillaries that form the BBB, have strong barrier properties (Abbott et al., 2010; Engelhardt and Sorokin, 2009). It is intriguing to clarify whether tTJs exist in such tight endothelia. However, it is very difficult to identify tTJs in tissues by freeze-fracture replica electron microscopy, because of the extremely low probability of encountering a fractured plane that contains endothelial TC regions.

In this study, we examined whether endothelial cells possess tTJs based on the concentrated localization of two tTJ markers, tricellulin and angulin family proteins, at TCs evaluated by immunofluorescence staining of frozen sections of various mouse tissues. We found that these tTJ markers were localized at TCs of endothelial cells in the brain and retina, but not in other tissues, thereby suggesting the presence of tTJs in endothelial cells that form the BBB and inner BRB.

Materials and Methods

Antibodies

Rat anti-tricellulin monoclonal antibody (mAb), rat anti-angulin-1/LSR mAb, rat anti-occludin mAb, rabbit anti-occludin polyclonal antibody (pAb), rabbit anti-angulin-2/ILDR1 pAb and rabbit antiangulin-3/ILDR2 pAb were raised and characterized as described previously (Higashi et al., 2013; Ikenouchi et al., 2005; Saitou et al., 1997). Goat anti-VE-cadherin pAb (sc-6458; Santa Cruz Biotechnology Inc. Dallas, TX, US) was obtained commercially. As secondary antibodies, Alexa 488-labeled donkey anti-goat IgG (Invitrogen Molecular Probes, Carlsbad, CA, US), Cy3-labeled donkey anti-rat IgG (Jackson ImmunoResearch Laboratories, Inc. West Grove, PA, US), and Alexa 647-labeled donkey anti-rabbit IgG (Invitrogen Molecular Probes) were used.

Immunofluorescence microscopy

For immunofluorescence microscopy of frozen tissue sections, dissected tissues from adult C57BL/6J mice were frozen in liquid nitrogen, and cut into 10-μm-thick sections using a cryostat (Leica CM1850). The sections were mounted on coverslips, air dried, fixed in 95% ethanol at 4°C for 30 minutes and treated with 100% acetone at room temperature for 1 minute. After washing with PBS, the sections were blocked with 1% bovine serum albumin in PBS, and incubated with primary antibodies for 30 minutes at room temperature, followed by incubation with secondary antibodies for 30 minutes. After rinsing with PBS, the sections were embedded in Fluorsave (Merck KGaA, Darmstat, Germany). Images were obtained using an IX70 microscope (Olympus, Tokyo, Japan) equipped with a UPlanApo ×40 (NA 0.85) objective lens.

Results and Discussion

To examine whether endothelial cells possess tTJs, we analyzed the expressions and subcellular localizations of tTJ markers, namely angulin family proteins (angulin-1/LSR, angulin-2/ILDR1, and angulin-3/ILDR2) and tricellulin, in frozen sections of various tissues from adult mice by immunofluorescence staining. The sections were counterstained with antibodies against occludin and VE-cadherin to delineate the epithelial and endothelial cell–cell junctions, respectively. Claudin-5 is known to be an endothelial claudin type, but is not suitable for a unbiquitous endothelial TJ marker since its expression is limited in a part of blood vessels (Morita et al., 1999). Since TJs and adherens junctions are intermingled with each other in endothelial cells (Leach et al., 1993; Vorbrodt and Dobrogowska, 2003), these two types of cell-cell junctions cannot be distinguished in immunofluorescence staining (Morita et al., 1999). Therefore, we used VE-cadherin, which is a good marker for endothelial adherens junctions (Giannotta et al., 2013), to visualize endotheliual TCs. First, we examined non-neural tissues. The results for the small intestine, pancreas, and aorta are shown as representatives. In the small intestine, tricellulin and angulin-1/LSR were clearly concentrated at TCs of epithelial cells (Fig. 1A, B, D, E). However, these tTJ markers were negative at TCs of endothelial cells in small vessels running inside of the intestinal villi (Fig. 1A, C, D, F). In the pancreas, which also contains abundant small vessels, tricellulin and angulin-1/LSR were not localized at TCs in endothelia (Fig. 2A). In the aorta, which is a large vessel, no concentration of angulin-1/LSR and tricellulin in endothelial TCs was observed (Fig. 2B). In addition to these three tissues, TC localization of angulin-1/LSR and tricellulin was not observed in at least the lung, colon, liver, and kidney, providing no evidence for the presence of endothelial tTJs in these tissues (data not shown). Neither angulin-2/ILDR1 nor angulin-3/ILDR2 was detected at TCs in endothelia as far as we examined in this study. As representatives, the results of the kidney (Fig. 2C) and retina (Fig. 2D) are shown for angulin-2/ILDR1 and angulin-3/ILDR2, respectively.

Fig. 1

Immunofluorescence staining of cryosections of the mouse small intestine with antibodies against tTJ markers (tricellulin and angulin-1/LSR), an endothelial cell–cell junction marker (VE-cadherin), and an epithelial TJ marker (occludin). (A) Triple labeling of the mouse small intestine with rat anti-tricellulin mAb (red), goat anti-VE-cadherin pAb (green), and rabbit anti-occludin pAb (blue). Insets b and c are presented in (B) and (C) at higher magnifications, respectively. (B) Tricellulin and occludin staining in magnified images of inset b in (A). The arrows indicate TCs of intestinal epithelial cells, where tricellulin shows a dot-like concentration. (C) Tricellulin and VE-cadherin staining in magnified images of inset c in (A). The arrowheads indicate TCs of endothelial cells within an intestinal villus. Concentration of tricellulin is not detected. (D) Triple labeling of the mouse small intestine with rat anti-angulin-1/LSR mAb (red), goat anti-VE-cadherin pAb (green), and rabbit anti-occludin pAb (blue). Insets e and f are presented in (E) and (F) at higher magnifications, respectively. (E) Angulin-1/LSR and occludin staining in magnified images of inset e in (D). The arrows indicate TCs of intestinal epithelial cells, where angulin-1/LSR is highly concentrated. (F) Angulin-1/LSR and VE-cadherin staining in magnified images of inset f in (D). The arrowheads indicate TCs of endothelial cells within an intestinal villus. Concentration of angulin-1/LSR is not detected. Bars: 10 μm in (A) and (D); 5 μm in (B), (C), (E), and (F).

Fig. 2

Immunofluorescence staining of cryosections of the mouse pancreas (A), aorta (B), kidney (C) and retina (D) with antibodies against tTJ markers, VE-cadherin and occludin. In (A) and (B), the sections were stained with a combination of anti-tricellulin mAb (magenta) and anti-VE-cadherin pAb (green) or anti-angulin-1/LSR mAb (magenta) and anti-VE-cadherin pAb. The arrowheads indicate TCs of endothelial cells, where tricellulin and angulin-1/LSR concentrations are not detected. In the aorta, the anti-tricellulin and anti-angulin-1/LSR mAbs exhibit faint background staining of the internal elastic membrane. In (C), the section was stained with a combination of rabbit anti-angulin-2/ILDR1 pAb (red), anti-VE-cadherin pAb (green) and anti-occludin mAb (blue). Occludin staining is shown in the merged image. Angulin-2/ILDR1 signal is positive at TCs of renal epithelial cells (arrows) but negative at TCs of endothelial cells (arrowheads). In (D), the section that contains retinal pigment epithelial cells and endothelial cells tangentially in the choroid was stained with a combination of rabbit anti-angulin-3/ILDR2 pAb (red), anti-VE-cadherin pAb (green) and anti-occludin mAb (blue). Occludin staining is shown in the merged image. Angulin-3/ILDR2 signal is positive at TCs of retinal pigment epithelial cells delineated by occludin staining (arrows), but negative at TCs of endothelial cells (arrowheads). Bars: 10 μm.

Next, we investigated the expressions of angulins and tricellulin in the brain. It is intriguing to clarify whether brain endothelial cells possess tTJs, since they have an exceptionally strong endothelial barrier function that establishes the BBB (Abbott et al., 2010; Engelhardt and Sorokin, 2009). By immunofluorescence staining, we detected notable concentration of angulin-1/LSR and tricellulin at endothelial TCs delineated by VE-cadherin staining in the brain parenchyma, including the cerebrum (Fig. 3A) and cerebellum (Fig. 3B). These tTJ markers were localized at TCs of endothelial cells in most parts of the brain, regardless of the thickness of the vessels. In addition to the brain, there is a strong barrier of endothelial cells in the retina, constituting the inner BRB. We found that angulin-1/LSR and tricellulin were localized at TCs of retinal endothelial cells (Fig. 3C). On the other hand, it is widely accepted that the BBB characteristics are not maintained in limited brain regions, such as the choroid plexus, which generates cerebrospinal fluid, and circumventricular organs (CVOs), where the blood environment is monitored for internal secretion (Wolburg and Paulus, 2010; Ganong, 2000). Consistently, these endothelial cells are known to have fenestrae (Wolburg and Paulus, 2010; Ganong, 2000). By immunofluorescence staining, we did not detect any concentration of tricellulin at endothelial TCs in frozen sections of the choroid plexus, although it was concentrated at TCs of ependymal cells delineated by occludin staining (Fig. 4A). Tricellulin was also not detected at TCs of endothelial cells in the median eminence, one of the CVOs (Fig. 4B). Angulin-1/LSR expression was not detected in endothelial TCs in these brain regions (Fig. 4A, B).

Fig. 3

Double immunofluorescence staining of cryosections of the mouse cerebrum (A), cerebellum (B) and retina (C) with a combination of anti-tricellulin mAb (magenta) and anti-VE-cadherin pAb (green) or anti-angulin-1/LSR mAb (magenta) and anti-VE-cadherin pAb. The arrows indicate endothelial TCs, where concentrated localizations of tricellulin and angulin-1/LSR are detected. Bars: 10 μm.

Fig. 4

Immunofluorescence staining of cryosections of the choroid plexus and median eminence in the mouse brain. (A) Triple staining of the choroid plexus with a combination of anti-tricellulin mAb (red), anti-VE-cadherin pAb (green) and anti-occludin pAb (blue) or a combination of anti-angulin-1/LSR mAb (red), anti-VE-cadherin pAb (green) and anti-occludin pAb (blue). Occludin staining is shown in the merged images. Tricellulin and angulin-1/LSR signals are positive at TCs of ependymal cells (arrows) but negative at TCs of endothelial cells (arrowheads). (B) Double staining of the median eminence with a combination of anti-tricellulin mAb (magenta) and anti-VE-cadherin pAb (green) or anti-angulin-1/LSR mAb (magenta) and anti-VE-cadherin pAb (green). The arrowheads indicate endothelial TCs, where no concentrated localizations of tricellulin and angulin-1/LSR are detected. The asterisks indicate the lumen of the third ventricle. Bars: 10 μm.

Considering that the concentration of both angulin family proteins and tricellulin at TCs is a good indication for tTJs, our observations strongly suggest that endothelial cells in the brain and retina, which constitute the BBB and inner BRB, respectively, have tTJs, while those in many other tissues do not. Although we cannot exclude the possibility that tricellulin and angulins are expressed in endothelial cells and diffusely distributed throughout the plasma membrane in various tissues, a recent transcriptome analysis identified at least lsr as a gene that is highly expressed in brain endothelial cells compared with lung endothelial cells (Daneman et al., 2010). Tricellulin localization at TCs has been reported very recently in cultured brain endothelial cells (Mariano et al., 2013). Here, we have described clear concentration of two tTJ markers, triellulin and angulin-1/LSR, at TCs in endothelial cells for the first time in vivo. Furthermore, this study suggests the presence of variations in endothelial cells in terms of the equipment for tTJs depending on their barrier properties. To finally demonstrate the existence of tTJs in endothelial cells, analyses by freeze-fracture replica electron microscopy are required. Thus far, our intensive efforts have not yet provided a definitive conclusion, probably because of the extremely low frequency of obtaining images of TJs at TC regions in brain endothelial cells using this method.

Brain endothelial cells form a primary element of the BBB and show a strong barrier function with high transendothelial electrical resistance (Abbott et al., 2010; Engelhardt and Sorokin, 2009). Retinal endothelial cells, which form the inner BRB, are known to exhibit similar characteristics to brain endothelial cells in terms of the endothelial barrier function (Hosoya and Tachikawa, 2009). The BBB/BRB-forming endothelial cells have been characterized by several features, including expression of drug-discharging pumps, lack of fenestrae, low transcytosis activity, and developed TJs (Abbott et al., 2010; Engelhardt and Sorokin, 2009). Given that leakage of solutes should be strictly restricted in the BBB/BRB and that the strong paracellular barrier is developed by the sealing not only at bicellular contacts but also at TCs, the existence of tTJs in brain and retinal endothelial cells would be reasonable.

Disruptions of strong endothelial barriers are associated with various disorders, such as multiple sclerosis for the BBB and diabetic retinopathy for the inner BRB (Abbott et al., 2010; Daneman, 2012; Klaassen et al., 2013). In addition, the BBB has been regarded as a problem for drug delivery from the blood to the central nervous system (Abbott et al., 2010). Therefore, it is of interest to clarify whether tricellulin and angulin-1/LSR localizing at TCs in endothelial cells are involved in the BBB and inner BRB, and whether they are possible targets that could be artificially modified for drug delivery into the brain. Future studies involving conditional knockout of angulin-1/LSR and tricellulin in endothelial cells will lead to better understanding of these issues.

Acknowledgments

We thank Kyoko Furuse for technical assistance, and Tetsutaki Hirase, Toyoshi Fujimoto, Hartwig Wolburg, and all the members of the Furuse laboratory for helpful discussions. This work was supported by the “Funding Program for Next Generation World Leading Researchers (NEXT Program)” from the Japan Society for the Promotion of Science.

References

• Abbott, N.J., Patabendige, A.A., Dolman, D.E., Yusof, S.R., and Begley, D.J. 2010. Structure and function of the blood-brain barrier. Neurobiol. Dis., 37: 13–25.
• Anderson, J.M. and Van Itallie, C.M. 2009. Physiology and function of the tight junction. Cold Spring Harb. Perspect. Biol., 1: a002584.
• Angelow, S., Ahlstrom, R., and Yu, A.S. 2008. Biology of claudins. Am. J. Physiol. Renal. Physiol., 295: F867–876.
• Bendayan, M. 2002. Vascular permeability in blood capillary. Microsc. Res. Tech., 57: 263–268.
• Daneman, R., Zhou, L., Agalliu, D., Cahoy, J.D., Kaushal, A., and Barres, B.A. 2010. The mouse blood-brain barrier transcriptome: a new resource for understanding the development and function of brain endothelial cells. PLoS One, 5: e13741.
• Daneman, R. 2012. The blood-brain barrier in health and disease. Ann. Neurol., 72: 648–672.
• Engelhardt, B. and Sorokin, L. 2009. The blood-brain and the blood-cerebrospinal fluid barriers: function and dysfunction. Semin. Immunopathol., 31: 497–511.
• Farquhar, M.G. and Palade, G.E. 1963. Junctional complexes in various epithelia. J. Cell Biol., 17: 375–412.
• Friend, D.S. and Gilula, N.B. 1972. Variations in tight and gap junctions in mammalian tissues. J. Cell Biol., 53: 758–776.
• Furuse, M. 2010. Molecular basis of the core structure of tight junctions. Cold Spring Harb. Perspect. Biol., 2: a002907.
• Ganong, W.F. 2000. Circumventricular organs: definition and role in the regulation of endocrine and autonomic function. Clin. Exp. Pharmacol. Physiol., 27: 422–427.
• Giannotta, M., Trani, M., and Dejana, E. 2013. VE-Cadherin and Endothelial Adherens Junctions: Active Guardians of Vascular Integrity. Dev. Cell, 26: 441–454.
• Higashi, T., Tokuda, S., Kitajiri, S., Masuda, S., Nakamura, H., Oda, Y., and Furuse, M. 2013. Analysis of the ‘angulin’ proteins LSR, ILDR1 and ILDR2—tricellulin recruitment, epithelial barrier function and implication in deafness pathogenesis. J. Cell Sci., 126: 966–977.
• Hosoya, K. and Tachikawa, M. 2009. Inner blood-retinal barrier transporters: role of retinal drug delivery. Pharm. Res., 26: 2055–2065.
• Ikenouchi, J., Furuse, M.M., Furuse, K., Sasaki, H., and Tsukita, S. 2005. Tricellulin constitutes a novel barrier at tricellular contacts of epithelial cells. J. Cell Biol., 171: 939–945.
• Klaassen, I., Van Noorden, C.J., and Schlingemann, R.O. 2013. Molecular basis of the inner blood-retinal barrier and its breakdown in diabetic macular edema and other pathological conditions. Prog. Retin. Eye Res., 34: 19–48.
• Komarova, Y. and Malik, A.B. 2010. Regulation of endothelial permeability via paracellular and transcellular transport pathways. Annu. Rev. Physiol., 72: 463–493.
• Leach, L., Clark, P., Lampugnani, M.G., Arroyo, A.G., Dejana, E. and Firth, J.A. 1993. Immunoelectron characterisation of the inter-endothelial junctions of human term placenta. J. Cell Sci., 104: 1073–1081.
• Mariano, C., Palmela, I., Pereira, P., Fernandes, A., Falcao, A.S., Cardoso, F.L., Vaz, A.R., Campos, A.R., Goncalves-Ferreira, A., Kim, K.S., Brites, D., and Brito, M.A. 2013. Tricellulin expression in brain endothelial and neural cells. Cell Tissue Res., 351: 397–407.
• Masuda, S., Oda, Y., Sasaki, H., Ikenouchi, J., Higashi, T., Akashi, M., Nishi, E., and Furuse, M. 2011. LSR defines cell corners for tricellular tight junction formation in epithelial cells. J. Cell Sci., 124: 548–555.
• Morita, K., Sasaki, H., Furuse, M., and Tsukita, S. 1999. Endothelial claudin: claudin-5/TMVCF constitutes tight junction strands in endothelial cells. J. Cell Biol., 147: 185–194.
• Raleigh, D.R., Marchiando, A.M., Zhang, Y., Shen, L., Sasaki, H., Wang, Y., Long, M., and Turner, J.R. 2010. Tight junction-associated MARVEL proteins marveld3, tricellulin, and occludin have distinct but overlapping functions. Mol. Biol. Cell, 21: 1200–1213.
• Saitou, M., Ando-Akatsuka, Y., Itoh, M., Furuse, M., Inazawa, J., Fujimoto, K., and Tsukita, S. 1997. Mammalian occludin in epithelial cells: its expression and subcellular distribution. Eur. J. Cell Biol., 73: 222–231.
• Staehelin, L.A., Mukherjee, T.M., and Williams, A.W. 1969. Freeze-etch appearance of the tight junctions in the epithelium of small and large intestine of mice. Protoplasma, 67: 165–184.
• Staehelin, L.A. 1973. Further observations on the fine structure of freezecleaved tight junctions. J. Cell Sci., 13: 763–786.
• Steed, E., Rodrigues, N.T., Balda, M.S., and Matter, K. 2009. Identification of MarvelD3 as a tight junction-associated transmembrane protein of the occludin family. BMC Cell Biol., 10: 95.
• Tsukita, S., Furuse, M., and Itoh, M. 2001. Multifunctional strands in tight junctions. Nat. Rev. Mol. Cell Biol., 2: 285–293.
• Vorbrodt, A.W. and Dobrogowska, D.H. 2003. Molecular anatomy of intercellular junctions in brain endothelial and epithelial barriers: electron microscopist’s view. Brain Res. Brain Res. Rev., 42: 221–242.
• Wade, J.B. and Karnovsky, M.J. 1974. The structure of the zonula occludens. A single fibril model based on freeze-fracture. J. Cell Biol., 60: 168–180.
• Walker, D.C., MacKenzie, A., Hulbert, W.C., and Hogg, J.C. 1985. A reassessment of the tricellular region of epithelial cell tight junctions in trachea of guinea pig. Acta Anat. (Basel), 122: 35–38.
• Walker, D.C., MacKenzie, A., and Hosford, S. 1994. The structure of the tricellular region of endothelial tight junctions of pulmonary capillaries analyzed by freeze-fracture. Microvasc. Res., 48: 259–281.
• Wolburg, H. and Paulus, W. 2010. Choroid plexus: biology and pathology. Acta Neuropathol., 119: 75–88.