Effect of human astrocytes on the characteristics of human brain-microvascular endothelial cells in the blood–brain barrier

Colloids and Surfaces B: Biointerfaces 86 (2011) 225–231

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

Effect of human astrocytes on the characteristics of human brain-microvascular endothelial cells in the blood–brain barrier Yung-Chih Kuo ∗ , Chin-Hsun Lu Department of Chemical Engineering, National Chung Cheng University, Chia-Yi 62102, Taiwan, ROC

a r t i c l e

i n f o

Article history: Received 19 February 2011 Received in revised form 28 March 2011 Accepted 1 April 2011 Available online 8 April 2011 Keywords: Blood–brain barrier Human brain-microvascular endothelial cells Human astrocyte Transendothelial electrical resistance Permeability

a b s t r a c t A blood–brain barrier (BBB) model in vitro was established by cultivating human brain-microvascular endothelial cells (HBMECs) with the regulation of human astrocytes (HAs) (HBMEC/HA). Astrocyteconditioned medium (ACM) was employed to constitute a confluent monolayer of HBMECs without directly conjugated HAs. HBMECs exhibited an orientated multiplication on the supporting membrane; while HAs grew in an overlapping fashion. In addition, HBMECs could propagate over the membrane pore, and the end-feet of HAs extended into the membrane pore to improve the integral feature of the BBB. HBMEC/HA demonstrated a high transendothelial electrical resistance (TEER) about 230  cm2 and low permeability of propidium iodide (PI) about 4 × 10−6 cm/s. The order in TEER was HBMEC/HA > HBMECs with 100% ACM > HBMECs with 50% ACM > HBMECs. The reverse order was valid for the permeability of PI and uptake of calcein-AM by HBMECs. The tranwell culture of HBMECs and HAs displays appropriate characteristics of the BBB and can be applied to estimate the delivery efficiency of therapeutic chemicals for the brain-related disease. © 2011 Elsevier B.V. All rights reserved.

1. Introduction The transport property of the blood–brain barrier (BBB) is crucial to the delivery of antiviral and neural drugs into the central nervous system (CNS) [1–5]. In anatomical structure, the BBB comprises mainly brain-microvascular endothelial cells (BMECs) and astrocytes, locating in the frontier between blood and brain interstitial fluid [6]. Small and hydrophobic molecules can transfer across the BBB via specific channels. However, the BBB prevents the majority of pharmaceuticals and proteins from infiltration and restricts the invasion of viruses, bacteria, exogenic biomacromolecules, and toxic blood ingredients into the brain. These functions derive from the fact that BMSCs develop intercellular tight junction (TJ), express few pinocytotic vesicles, and lack of membrane fenestration [7,8]. Since the BBB is adjacent to the brain parenchyma, it can also regulate the supply of nutrients such as glucose, removal of metabolites, and exchange of gases in the brain [9,10]. The other physiological roles of the BBB are to sustain homeostasis of the CNS and to block

Abbreviations: ACM, astrocyte-conditioned medium; BBB, blood–brain barrier; HA, human astrocyte; HBMEC, human brain-microvascular endothelial cell; HBMEC/ACM, HBMECs cultured with ACM; HBMEC/HA, HA-regulated HBMECs; PET, polyethylene terephthalate; PI, propidium iodide; TJ, tight junction. ∗ Corresponding author. Tel.: +886 5 272 0411x33459; fax: +886 5 272 1206. E-mail address: [email protected] (Y.-C. Kuo). 0927-7765/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2011.04.005

the influence of hormones and peripheral neurotransmitters on the brain [11,12]. The primary culture of BMECs has been achieved from mammals including bovine, porcine, ovine, murine, canine, and human and used in subculture [13–17]. These isolated BMECs expressed specific enzyme markers, Factor VIII antigen, polarization of membrane proteins, and adenosine triphosphate-binding cassette transporters [18,19]. In addition, cultured BMECs demonstrated a high transendothelial electrical resistance (TEER) and low permeability [20]. Moreover, the interaction between BMECs and astrocytes can enhance the unique characteristics of the BBB. The essential tasks of mature astrocytes include providing structural support to contact neighboring neurons, delivering nutrients (energy sources) to neurons, and mediating neuronal homeostasis [21]. A co-culture of bovine BMECs and rat astroglia was observed to increase the length, width, and complexity of TJ [22]. A co-culture of bovine BMECs and rat glial cells was also observed to promote the genetic expression (messenger ribonucleic acid) of P-glycoprotein and multidrug resistance-associated protein (MRP) 1, MRP4, and MRP6 [23]. The aim of this study is to establish a BBB model by human BMECs (HBMECs) and human astrocytes (HAs) via a transwell culture with an insert of polyethylene terephthalate (PET) membrane. Effect of astrocyte-conditioned medium (ACM) on the BBB model was also considered. The proliferation of HBMECs and HAs, the transmission electron microscope (TEM) morphology of the

226

Y.-C. Kuo, C.-H. Lu / Colloids and Surfaces B: Biointerfaces 86 (2011) 225–231

2.4. ACM Nomenclature Pj TEER

permeability of PI across the BBB (cm/s) transendothelial electrical resistance ( cm2 )

Subscript j index of permeability across the BBB, j denotes HBMEC/HA or HBMEC

BBB (HBMECs regulated by HAs), TEER, permeability of propidium iodide (PI), and uptake of calcein-AM were investigated.

HAs with a density of 2 × 105 cells/cm2 on a human fibronectinpretreated T75 tissue culture flask were cultivated with fresh culture medium in the CO2 incubator at 37 ◦ C over 2 days. The culture medium was replaced and the cultivation continued. After cultivation over 24 h, the culture medium was collected and replaced with fresh culture medium. This step repeated and the collected culture medium was filtrated with a membrane containing pores of 0.2 ␮m to remove large aggregates and possible contaminants. The filtrate was defined as ACM and preserved at −80 ◦ C. Addition of an equal volume of fresh culture medium in ACM yielded 50% ACM. 2.5. HBMECs regulated by HAs or ACM

2. Materials and methods 2.1. Reagents and chemicals Human fibronectin, trypsin, Dulbecco’s phosphate buffered saline (DPBS), gelatin, trypan blue, formaldehyde solution, PI, and calcein-AM were purchased from Sigma (St. Louis, MO). Osmium tetraoside, LR white acrylic resin, copper grid film (carbon-coated and glow-discharged, 300 mesh), uranyl acetate, and lead citrate were obtained from Ted Pella (Redding, CA). HBMECs, endothelial cell medium (ECM), fetal bovine serum (FBS), endothelial cell growth supplement (ECGS), penicillin/streptomycin (PS) solution, HAs, astrocyte medium (AM), and astrocyte growth supplement (AGS) were purchased from Sciencell (Corte Del Cedro Carlsbad, CA). Diaminoethanetetraacetic acid (EDTA) was obtained from Riedel-de Haen (Seelze, Germany), dimethyl sulfoxide (DMSO) from J.T. Baker (Phillipsburg, NJ), PET membrane (cell culture insert, pore size 1.0 ␮m, effective growth area 0.9 cm2 ) from BD Falcon (Franklin Lakes, NJ), and ethanol from Tedia (Fairfield, OH).

HAs were seeded with a density of 4 × 105 cells/cm2 on the bottom surface of a gelatin-coated PET membrane in a reversed transwell (BD Falcon) and cultivated in the CO2 incubator at 37 ◦ C for 1 h. The transparent PET membrane can avoid the complexity in optical examination. The PET membrane is also very good in biocompatibility without inducing cell inflammation and apoptosis. HBMECs were seeded with a density of 4 × 105 cells/cm2 on the top surface of the PET membrane. The transwell system with the sandwich configuration of HBMECs/PET/HAs was placed in a 12-well microtiter plate (Corning Costar) and cultivated in the CO2 incubator at 37 ◦ C over 14 days. HBMECs never migrated to the back of PET membrane in this study. The lower and upper chambers of the transwell were filled, respectively, with fresh culture medium of HAs and HBMECs. The two media were maintained at the same level to avoid hydrostatic interference in the insert and replaced at a rate of every 2 days to form a confluent monolayer of HA-regulated HBMECs (HBMEC/HA). For the ACM-regulated HBMECs, the bottom surface was free from HAs and 50% or 100% ACM was used as the culture medium in the lower chamber of the transwell.

2.2. HBMECs

2.6. Morphology

A protocol for expanding and preserving HBMECs was described previously [24]. Briefly, HBMECs were cultured with ECM containing 1% (v/v) ECGS, 5% (v/v) FBS, and 1% (v/v) PS solution in a humidified CO2 incubator (NuAire, Plymouth, MN) at 37 ◦ C, severed from the expanded colony, purified, and stored in liquid nitrogen.

Proliferated HBMECs and HAs were visualized optically under an inverted microscope (Eclipse TS-1000-F, Nikon, Tokyo, Japan). In addition, the structure of HBMEC/HA on PET membrane was analyzed by a TEM (JEM-1400, Jeol, Tokyo, Japan). After cultivation in the transwell, HBMECs and HAs on PET membrane were washed with DPBS, fixed with 10% (v/v) formaldehyde solution at 4 ◦ C for 30 min, sliced into 1 mm × 5 mm, immersed in 1% (w/v) osmium tetraoside at 4 ◦ C for 2 h, dehydrated stepwise with ethanol from 50% to 95% for 1 h, desiccated with 99.8% ethanol for 40 min, permeated with LR white acrylic resin and 99.8% ethanol in a volume ratio of 1:1 at 4 ◦ C for 24 h, penetrated with LR white acrylic resin at 4 ◦ C for 24 h, loaded into a home-made capsule containing LR white acrylic resin at 25 ◦ C for 24 h, and embedded at 60 ◦ C for 26 h. The sample was sliced by an ultra-microtome (Reichert, Depew, NY) into 60–90 nm in thickness, attached to a copper grid film, stained with 2% (w/v) uranyl acetate at 25 ◦ C for 30 min in darkness, and stained with 0.4% (w/v) lead citrate at 25 ◦ C for 10 min. The double staining of uranium and lead could enhance the contrast between the cells and PET.

2.3. HAs Unfrozen HAs with a density of 7500 cells/cm2 were seeded on T75 tissue culture flask (Corning Costar, Cambridge, MA) pretreated with human fibronectin. These HAs were cultured with AM, containing 1% (v/v) AGS, 2% (v/v) FBS, and 1% (v/v) PS solution, and multiplied in the CO2 incubator at 37 ◦ C over 7 days. The culture medium was replaced after the initial 8 h and subsequently at a rate of every 2 days. The expanded HAs were washed with 10 mL of DPBS, detached with 4 mL of 0.025% trypsin–0.5 mM EDTA, rinsed with 8 mL of fresh culture medium, harvested in 15 mL of sterile conical tube (BD Falcon, Franklin Lakes, NJ), centrifuged at 150 × g for 5 min, resuspended in 8 mL of fresh culture medium, and equally dispensed to 3 flasks pretreated with human fibronectin. The concentration of HAs was determined by trypan blue exclusion with a hemocytometer (Neubauer, Marienfeld, Germany) under a phasecontrast biological microscope (Motic, Richmond, BC, Canada). The excess HAs were collected in cryovials (BD Falcon), filled with the fresh culture medium containing 10% (v/v) DMSO and 10% (v/v) FBS, frozen in an ultralow temperature freezer (Sanyo, Osaka, Japan) at −80 ◦ C over 1 day, and preserved in liquid nitrogen.

2.7. TEER The transwell with PET membrane was filled with two fresh culture media in the lower and upper chambers at the same height. The TEER of blank PET membrane was determined by a Millicell electrical resistance system (Millipore, Bedfoed, MA). After cultivation, the two culture media were replaced with the fresh culture media. The TEER of the transwell insert containing cells was determined by

Y.-C. Kuo, C.-H. Lu / Colloids and Surfaces B: Biointerfaces 86 (2011) 225–231

227

a Millicell electrical resistance system. The cell-induced TEER was evaluated by subtracting the TEER of blank PET membrane from the TEER of the transwell insert containing cells. 2.8. Permeability of PI PET membrane with the cultured HBMEC/HA or HBMECs was fixed between donor and receiver chambers. HBMECs faced the donor chamber. The two chambers were maintained at 37 ◦ C by circulating water in the external jacket. The donor chamber was filled with 14 mL of fresh culture medium containing PI with a concentration of 0.25 mg/mL at 150 rpm. The receiver chamber was filled with 14 mL of fresh culture medium at 150 rpm. The concentration of PI in the donor chamber was controlled between 98% and 102% of its initial value. An independent experiment was sustained for 1 h in darkness. 50 ␮L of the fluid in the receiver chamber were sampled every 10 min and analyzed by a microplate fluorescent reader (Synergy HT, BioTek, Winooski, VT) using excitation at 485 nm and emission at 590 nm. The receiver chamber was immediately compensated with 50 ␮L of fresh culture medium. The permeability coefficient of PI across HA-regulated HBMECs (PHBMEC/HA ) or across HBMECs (PHBMEC ) was calculated by 1/Pj = 1/Pe − 1/Pm , where j is HBMEC/HA or HBMEC, Pe is the permeability of PI across PET membrane with cells, and Pm is the permeability of PI across cell-free PET membrane. The permeability could be estimated by J/C = Vr (dCr /dt)/(A·C), where J, C, Vr , Cr , t and A are, respectively, the flux of PI from the donor to receiver chamber, the difference in the concentration of PI between the donor and receiver chamber, the volume of the receiver chamber, the concentration of PI in the receiver chamber, the time, and the transport area. 2.9. Staining of calcein-AM Cultured HBMEC/HA and HBMECs were washed with DPBS, incubated with fresh culture medium in the transwell, and reacted with calcein-AM of 0.5 ␮M in the upper chamber at 37 ◦ C for 30 min in darkness. After uptake of calcein-AM, HBMECs were fixed with 10% (v/v) formaldehyde solution at 4 ◦ C for 30 min. The fluorescent images were obtained by a fluoromicroscope (Axioskop 2 plus, Zeiss, Munchen-Hallbergmoos, Germany) using excitation at 483 nm and emission at 530 nm. 3. Results and discussion 3.1. Morphology Fig. 1 illustrates the typical structure of the BBB, the most important tissue site protecting the CNS from xenobiotic invasion. The basic cell components of the BBB in human include HBMECs and HAs. Since, an intact human brain is rarely obtained as an experimental source, establishing an HBMEC/HA-based model becomes inevitable. This model should reconstruct the physiological function and molecular mechanism of the BBB in vivo. In the brain, HBMECs could form two intercellular junctions: TJ and adherens junction [25,26]. TJ between HBMECs develops a passive diffusion barrier for restraining paracellular transport across the BBB [27,28]. Therefore, the characteristics of TJ are crucial to the validation of an in vitro model [29,30]. The junction complex between HBMECs can also maintain the apical-basolaternal polar property of the BBB [31,32]. In addition, the neuronal signals activate HAs through specific receptors on the cell surface [33]. The ending parts of HAs inflate and touch the abluminal surface of HBMECs [34]. Therefore, the feature of these perivascular end-feet is another morphological trait of the BBB.

Fig. 1. Schematic representation of the BBB structure.

Fig. 2 shows the morphology of HBMECs and HAs during proliferation. As indicated in Fig. 2(a), a spindle-shaped appearance of HBMECs emerged after cultivation over 3 days. Fig. 2(b) revealed a compact growth of HBMECs without cell pile after cultivation over 7 days. In addition, HBMECs exhibited geometrical direction of propagation, suggesting an expansion toward the endothelial monolayer [35]. As displayed in Fig. 2(c), HAs demonstrated slim peripheral branches emanating from the main cell body. Fig. 2(d) presented a dense cell population with overlapping manner when HAs were cultivated over 7 days. Moreover, the electrical resistance of the layer of HAs cultured over 7 days was in the range of 0–10  cm2 . Fig. 3 shows the TEM morphology of HBMECs and HAs on PET membrane after the transwell culture. As indicated in Fig. 3(a), the mechanical stress resulting from the process might distort the drilled pore in the membrane. As revealed in Fig. 3(b), HBMECs formed a confluent monolayer over the membrane pore. In addition, HAs evolved into a planar layer on the other side of the membrane. However, this cell layer could not be tight because the electrical resistance of the cultured HAs was low (described in the previous paragraph). As displayed in Fig. 3(c), the end-feet of HAs expanded into the membrane pore. This suggested that HBMECs and HAs in the co-culture system could interact with each other to a certain extent. If a longer culture period is applied, the endfeet of HAs may direct attach HBMECs via the pores [36]. In a rat cell-based BBB model, astrocytes could migrate to the endothelium layer in the case of the membrane containing pores with diameter of 3 ␮m; and the pores with diameter of 0.45 ␮m revealed a TEER about 130  cm2 [37]. This TEER was relatively low (see Fig. 4). Therefore, the pores with diameter of 1 ␮m in this study could be appropriate. In normal human brain, about 90% surface of HBMECs was encompassed by the end-feet of HAs, yielding a layered structure [38]. 3.2. TEER and permeability Fig. 4 shows the TEER of the in vitro BBB models and the corresponding permeability of PI. As indicated in this figure, the model using only HBMECs demonstrated a TEER around 180  cm2 and the permeability of PI around 8 × 10−6 cm/s. These values suggested that the proliferated HBMECs on PET membrane generated an acceptable model of the BBB [39]. An ideal model of the BBB displays fully interconnected TJ, which can be represented by a high

228

Y.-C. Kuo, C.-H. Lu / Colloids and Surfaces B: Biointerfaces 86 (2011) 225–231

Fig. 2. Morphology of HBMECs and HAs. (a) HBMECs cultured for 3 days; (b) HBMECs cultured for 7 days, arrow: orientated multiplication; (c) HAs cultured for 3 days; (d) HAs cultured for 7 days, ellipse: cell overlapping region.

TEER [40]. In addition, PI is a fluorescent agent, which never permeates normal cell membrane in tissue. PI can only penetrate cell and bind to nucleic acids when a living cell contains impaired membrane. Since the molecular weight of PI is low, PI may transport

across the monolayer of HBMECs via paracellular pathway. Therefore, a low permeability of PI denotes a high level of developed TJ [41]. As revealed in Fig. 4, an increase in the percentage of ACM enhanced the TEER and reduced the permeability of PI. This sug-

Fig. 3. TEM images of the HBMEC/HA-based model. (a) Transverse pore in PET membrane, arrow: pore; (b) HBMEC/HA on PET membrane, E: HBMECs, A: HAs, P: pore, M: membrane; (c) HBMEC/HA on PET membrane, A: HAs, M: membrane, arrow: end-feet of HAs.

Y.-C. Kuo, C.-H. Lu / Colloids and Surfaces B: Biointerfaces 86 (2011) 225–231

Y1

Y2

250

8

150

6

100

4

50

2

0

0

ACM 50%

(a)

(b)

Pj × 106 (cm/s)

TEER (Ω×cm2)

200

10

ACM 100%

(c)

(d)

Fig. 4. TEER and permeability of PI across the BBB. Key: (a) HBMECs; (b) HBMECs cultured with 50% ACM; (c) HBMECs cultured with 100% ACM; (d) HBMEC/HA. n = 3.

gested that ACM could improve the confluent density of the HBMEC monolayer [42]. This improvement in the compactness of HBMECs could be attributed to the HA-secreted factors, such as transforming growth factor-␤, glial-derived neurotrophic factor, interleukin-6, and glial fibrillary acidic protein, or other water-soluble molecules in ACM [43,44]. In addition, the regulation of ACM has been used to reduce the permeability of sucrose and fluorescein isothiocyanteconjugated bovine serum albumin across bovine and human BBB

229

[45]. It was also observed that ACM reduced 70% permeability and actively stimulated the expression of ZO-1, an important TJ protein [41]. As displayed in Fig. 4, the co-culture of HBMECs and HAs further increased the TEER (around 230  cm2 , the highest value in the four models) and reduced the permeability of PI (around 4 × 10−6 cm/s, the lowest value in the four models). The density of the HA layer contributed negligible enhancement in TEER because the electrical resistance of the astrocyte layer was low, in general [46]. In fact, the regulation of HAs improved the compact property of the HBMEC monolayer. The main difference between ACM and HA is that the former stimulates HBMECs by the HA-secreted biomolecules in the medium and the latter can direct contact, regulate and transmit signals to HBMECs. The TEER in the in vivo BBB is expected to reach 2000  cm2 . However, the sophisticated physiological conditions cannot be completely reproduced in the present stage. The anatomical site of BBB includes pericytes, which may affect a particular function on TJ. The other possible reason of the difference between in vivo and in vitro around HBMECs can be the shear stress of local blood flow in normal physiology. The majority of the in vitro TEER reported ranged between 100 and 150  cm2 , and the permeability between 1 and 30 × 10−6 cm/s [47]. In the study on the BBB motif, it was concluded that astrocytes could up-regulate the expression of TJ promoters, including HT7, lectin binding sites of Ulex europaeus agglutinin I, and angiotensin receptors, on endothelial cells [48,49]. 3.3. Endocytosis of calcein-AM Fig. 5 shows the fluorescent staining of calcein-AM accumulated in HBMECs of the four BBB models. As revealed in this figure, the order of the fluorescent intensity was HBMECs > HBMECs with 50% ACM > HBMECs with 100% ACM > HBMEC/HA. This was because the direct regulation of HA produced the largest quantity of P-glycoprotein (P-gp) and multi-drug resistance proteins (MRPs) on HBMECs. ACM could also stimulate the expression of the efflux transporter to restrict transcytosis across HBMECs. The

Fig. 5. Fluorescent images of the uptake of calcein-AM by HBMECs. Key: same as Fig. 4.

230

Y.-C. Kuo, C.-H. Lu / Colloids and Surfaces B: Biointerfaces 86 (2011) 225–231

rationale behind the application of calcein-AM in this uptake study is explained as follows. Original calcein-AM is not fluorescent. Esterase in HBMECs degraded calcein-AM and yielded a green fluorescent product, which could not be pumped out of HBMECs by the efflux system. Therefore, calcein-AM is a perfect indicator for judging the ability of endocytosis. In addition, HAs could rarely express drug resistance proteins [50]. However, astrocytes could enhance the transcript level of P-gp, MRP-3, MRP-4, and breast cancer resistance protein on endothelial cells in rats [51]. Therefore, the direct regulation of HA and addition of ACM could promote the expression and polar activity of the efflux proteins on HBMECs and reduce the uptake of calcein-AM. 4. Conclusions The confluent and endocytotic properties of the BBB models based on HBMECs, HAs, and ACM were assessed. The TEER of HBMEC/HA increased about 28% when compared with the monolayer of HBMECs. In addition, the monolayer of HBMECs was about 2 times the permeability of PI across HBMEC/HA. The uptake of fluorescent calcein-AM was in the order: HBMECs > HBMECs with 50% ACM > HBMECs with 100% ACM > HBMEC/HA. This suggested that HAs and ACM could effectively induce the efflux transporters on HBMECs to restrain endocytosis. The BBB model with the two human cells can serve as a standard system for estimating the transport parameters related to drug delivery into the CNS. Acknowledgements This work is supported by the National Science Council of the Republic of China. References [1] J.F. Deeken, W. Löscher, The blood–brain barrier and cancer: transporters, treatment, and trojan horses, Clin. Cancer Res. 13 (2007) 1663–1674. [2] N. Strazielle, J. Francois, G. Egea, Factors affecting delivery of antiviral drugs to the brain, Rev. Med. Virol. 15 (2005) 105–133. [3] Y.C. Kuo, J.F. Chung, Physicochemical properties of nevirapine-loaded solid lipid nanoparticles and nanostructured lipid carriers, Colloids Surf. B 83 (2011) 299–306. [4] Y.C. Kuo, H.H. Chen, Effect of nanoparticulate polybutylcyanoacrylate and methylmethacrylate-sulfopropylmethacrylate on the permeability of zidovudine and lamivudine across the in vitro blood–brain barrier, Int. J. Pharm. 327 (2006) 160–169. [5] Y.C. Kuo, Loading efficiency of stavudine on polybutylcyanoacrylate and methylmethacrylate-sulfopropylmethacrylate copolymer nanoparticles, Int. J. Pharm. 290 (2005) 161–172. [6] N.J. Abbott, Evidence for bulk flow of brain interstitial fluid: significance for physiology and pathology, Neurochem. Int. 45 (2004) 545–552. [7] A.S. Yu, K.M. McCarthy, S.A. Francis, J.M. McCormack, J. Lai, R.A. Rogers, R.D. Lynch, E.E. Schneeberger, Knockdown of occludin expression leads to diverse phenotypic alterations in epithelial cells, Am. J. Physiol. Cell Physiol. 288 (2005) 1231–1241. [8] H. Wolburg, A. Lippoldt, Tight junctions of the blood–brain barrier: development, composition and regulation, Vasc. Pharmacol. 38 (2002) 323–337. [9] I. Tamai, A. Tusji, Transporter–mediated permeation of drugs across the blood–brain barrier, J. Pharm. Sci. 89 (2000) 1371–1388. [10] Y.C. Kuo, P.I. Lin, C.C. Wang, Targeting nevirapine delivery across human brain microvesscular endothelial cells using transferrin-grafted poly(lactideco-glycolide) nanoparticles, Nanomedicine (2011), in press. [11] A.H. Schinkel, P-Glycoprotein, a gatekeeper in the blood–brain barrier, Adv. Drug Deliv. Rev. 36 (1999) 179–194. [12] H. Sun, H. Dai, N. Shaik, W.F. Elmquist, Drug efflux transporters in the CNS, Adv. Drug Deliv. Rev. 55 (2003) 83–105. [13] P.D. Bowman, S.R. Ennis, K.E. Rarey, A.L. Betz, G.W. Goldstein, Brain microvessel endothelial cells in tissue culture: a model for study of blood–brain barrier permeability, Ann. Neurol. 14 (1983) 396–402. [14] L.E. Craig, J.P. Spelman, J.D. Strandberg, M.C. Zink, Endothelial cells from diverse tissues exhibit differences in growth and morphology, Microvasc. Res. 55 (1998) 65–76. [15] N.J. Abbott, C.C. Hughes, P.A. Revest, J. Greenwood, Development and characterisation of a rat brain capillary endothelial culture: towards an in vitro blood–brain barrier, J. Cell Sci. 103 (1992) 23–37.

[16] C.A. Szabo, M.A. Deli, T.K. Ngo, F. Joo, Production of pure primary rat cerebral endothelial cell culture: a comparison of different methods, Neurobiology 5 (1997) 1–16. [17] M.A. Deli, C.S. Abraham, Y. Kataoka, M. Niwa, Permeability studies on in vitro blood–brain barrier models: physiology, pathology and pharmacology, Cell. Mol. Neurobiol. 25 (2005) 59–127. [18] N.J. Abbott, L. Ronnback, E. Hannson, Astrocyte–endothelial interactions at the blood–brain barrier, Nat. Rev. Neurosci. 7 (2006) 41–53. [19] R. Cecchelli, V. Berezowski, S. Lundquist, M. Culot, M. Renftel, M.P. Dehouck, L. Fenart, Modelling of the blood–brain barrier in drug discovery and development, Nat. Rev. Drug Discov. 6 (2007) 650–661. [20] Y.C. Kuo, H.H. Chen, Effect of electromagnetic field on endocytosis of cationic solid lipid nanoparticles by human brain-microvascular endothelial cells, J. Drug Target. 18 (2010) 447–456. [21] N.J. Abbott, Dynamics of CNS barriers: evolution, differentiation and modulation, Cell. Mol. Neurobiol. 25 (2005) 5–23. [22] J.H. Tao-Cheng, Z. Nagy, M.W. Brightman, Tight junctions of brain endothelium in vitro are enhanced by astroglia, J. Neurosci. 7 (1987) 3293–3299. [23] V. Berezowski, C. Landry, M.P. Dehouck, R. Cecchelli, L. Fenart, Contribution of glial cells and pericytes to the mRNA profiles of P-glycoprotein and multidrug resistance-associated proteins in an in vitro model of the blood–brain barrier, Brain Res. 1018 (2004) 1–9. [24] Y.C. Kuo, I.C. Chen, Evaluation of surface charge density and surface potential by electrophoretic mobility for solid lipid nanoparticles and human brainmicrovascular endothelial cells, J. Phys. Chem. B 111 (2007) 11228–11236. [25] E. Dejana, Endothelial cell-cell junctions: happy together, Nat. Rev., Mol. Cell. Biol. 5 (2004) 261–270. [26] J.M. Staddon, L.L. Rubin, Cell adhesion, cell junction and the blood brain barrier, Curr. Opin. Neurobiol. 6 (1996) 622–627. [27] T.S. Reese, M.T. Karnovsky, Fine structural localization of a blood–brain barrier to exogenous peroxidase, J. Cell Biol. 34 (1967) 207–217. [28] M.W. Brightman, T.S. Reese, Junctions between intimately apposed cell membranes in the vertebrate brain, J. Cell Biol. 40 (1969) 648–677. [29] L.L. Rubin, J.M. Staddon, The cell biology of blood–brain barrier, Annu. Rev. Neurosci. 22 (1999) 11–28. [30] Y.C. Kuo, C.Y. Chung, Transport of zidovudine- and lamivudine-loaded polybutylcyanoacrylate and methylmethacrylate-sulfopropylmethacrylate nanoparticles across the in vitro blood–brain barrier: characteristics of the drug-delivery system, J. Chin. Inst. Chem. Eng. 36 (2005) 627–638. [31] E.E. Schneeberger, R.D. Lynch, Structure, function, and regulation of cellular tight junctions, Am. J. Physiol. 262 (1992) 647–661. [32] S. Tsukita, M. Furuse, M. Itoh, Molecular dissection of tight junctions, Cell Struct. Funct. 21 (1996) 381–385. [33] R.C. Janzer, M.C. Raff, Astrocytes induce blood–brain barrier properties in endothelial cells, Nature 325 (1987) 253–257. [34] V.V. Jeliazkova-Mecheva, D.J. Bobilya, A porcine astrocyte/endothelial cell coculture model of the blood–brain barrier, Brain Res. Protoc. 12 (2003) 91–98. [35] Y.C. Kuo, C.W. Lin, Impact of arginine-modified solid lipid nanoparticles on the membrane charge of human brain-microvascular endothelial cells, Colloids Surf. B 72 (2009) 201–207. [36] Y. Hayashi, M. Nomura, S. Yamagishi, S. Harada, J. Yamashita, H. Yamamoto, Induction of various blood–brain barrier properties in non-neural endothelial cells by close apposition to co-cultured astrocytes, Glia 19 (1997) 13–26. [37] P. Demeuse, A. Kerkhofs, C. Struys-Ponsar, B. Knoops, C. Remacle, P. Van den Bosch de Aguilar, Compartmentalized coculture of rat brain endothelial cells and astrocytes: a syngenic model to study the blood–brain barrier, J. Neurosci. Methods 121 (2002) 21–31. [38] A. Prat, K. Biernacki, K. Wosik, J.P. Antel, Glial cell influence on the human blood–brain barrier, Glia 36 (2001) 145–155. [39] Y.C. Kuo, F.L. Su, Transport of stavudine, delavirdine, and saquinavir across the blood–brain barrier by polybutylcyanoacrylate, methylmethacrylatesulfopropylmethacrylate, and solid lipid nanoparticles, Int. J. Pharm. 340 (2007) 143–152. [40] Y.C. Kuo, C.Y. Kuo, Electromagnetic interference in the permeability of saquinavir across the blood–brain barrier using nanoparticulate carriers, Int. J. Pharm. 351 (2008) 271–281. [41] V. Siddharthan, Y.V. Kim, S. Liu, K.S. Kim, Human astrocytes/astrocyteconditioned medium and shear stress enhance the barrier properties of human brain microvascular endothelial cells, Brain Res. 1147 (2007) 39–50. [42] Y. Zhang, C.S.W. Li, Y. Ye, K. Johnson, J. Poe, S. Johnson, W. Bobrowski, R. Garrido, C. Madhu, Porcine brain microvessel endothelial cells as an in vitro model to predict in vivo blood–brain barrier permeability, Drug Metab. Dispos. 34 (2006) 1935–1943. [43] S. Hamm, B. Dehouck, J. Kraus, K. Wolburg-Buchholz, H. Wolburg, W. Risau, R. Cecchelli, B. Engelhardt, M.P. Dehouck, Astrocyte mediated modulation of blood–brain barrier permeability does not correlate with a loss of tight junction proteins from the cellular contacts, Cell Tissue Res. 315 (2004) 157–166. [44] J. Kraus, A.K. Ling, S. Hamm, K. Voigt, P. Oschmann, B. Engelhardt, Interferonbeta stabilizes barrier characteristics of brain endothelial cells in vitro, Ann. Neurol. 56 (2004) 192–205. [45] L.L. Rubin, E. Hall, S. Porter, K. Barbu, C. Cannon, H.C. Horner, M. Janatpour, C.W. Liaw, K. Manning, J. Morales, L.I. Tanner, K.J. Tomaselli, F. Bard, A cell culture model of the blood–brain barrier, J. Cell Biol. 115 (1991) 1725–1735. [46] N.F. Flecher, D.J. Brayden, B. Brankin, S. Worrall, J.J. Callanan, Growth and characterisation of a cell culture model of the feline blood–brain barrier, Vet. Immunol. Immunopathol. 109 (2006) 233–244.

Y.-C. Kuo, C.-H. Lu / Colloids and Surfaces B: Biointerfaces 86 (2011) 225–231 [47] M. Gumbleton, K.L. Audus, Progress and limitations in the use of in vitro cell cultures to serve as a permeability screen for the blood–brain barrier, J. Pharm. Sci. 90 (2001) 1681–1698. [48] N.J. Abbott, Astrocyte–endothelial interactions and blood–brain barrier permeability, J. Anat. 200 (2002) 629–638. [49] B. Reuss, R. Dono, K. Unsicker, Functions of fibroblast growth factor (FGF)-2 and FGF-5 in astroglial differentiation and blood–brain barrier permeability: evidence from mouse mutants, J. Neurosci. 23 (2003) 6404–6412.

231

[50] I. Megard, A. Garrigues, S. Orlowski, S. Jorajuria, P. Clayette, E. Ezan, A. Mabondzo, A co-culture-based model of human blood–brain barrier: application to active transport of indinavir and in vivo-in vitro correlation, Brain Res. 927 (2002) 153–167. [51] N. Perrière, S. Yousif, S. Cazaubon, N. Chaverot, F. Bourasset, S. Cisternino, X. Declèves, S. Hori, T. Terasaki, M.A. Deli, J.M. Scherrmann, J. Temsamani, F. Roux, P.O. Couraud, A functional in vitro model of rat blood–brain barrier for molecular analysis of efflux transporters, Brain Res. 1150 (2007) 1–13.