Brought to you by Portland Press Ltd.
Published on behalf of the International Federation for Cell Biology
Cancer Cell death Cell cycle Cytoskeleton Exo/endocytosis Differentiation Division Organelles Signalling Stem cells Trafficking
Cell Biology International (2008) 32, 198–209 (Printed in Great Britain)
Effect of nicotine and polyaromtic hydrocarbons on cerebral endothelial cells
Pilaiwanwadee Hutamekalinab, Attila E. Farkasa, Anna Orbóka, Imola Wilhelma, Péter Nagyőszia, Szilvia Veszelkaa, Mária A. Delia, Krisztina Buzásc, Éva Hunyadi‑Gulyásc, Katalin F. Medzihradszkycd, Duangdeun Meksuriyenb and István A. Krizbaia*
aInstitute of Biophysics, Biological Research Centre, Temesvári krt. 62, 6726 Szeged, Hungary
bDepartment of Biochemistry, Faculty of Pharmaceutical Sciences, Chulalongkorn University, Phayathai Road, Pathumwan, Bangkok 10330, Thailand
cProteomics Research Group, Biological Research Centre, Temesvári krt. 62, 6726 Szeged, Hungary
dDepartment of Pharmaceutical Chemistry, University of California San Francisco, San Francisco, CA 94143-0446, USA


Abstract

The present study was designed to investigate the effect of nicotine and polyaromatic hydrocarbon compounds on cerebral endothelial cells (CECs). Nicotine treatments from 15min to 5h did not cause any changes in the expression and localization of principal junctional proteins. One day of treatment with a relatively high concentration of nicotine induced a decrease in the expression of the tight junction protein ZO-1, occludin, and the adherens junction protein, cadherin. Treatment with 3×10−5M phenanthrene for 24h caused a redistribution of occludin from the Triton X-100 insoluble to the Triton X-100 soluble fraction. Transendothelial electrical resistance was not significantly affected by 24h treatments with nicotine, methylanthracene or phenanthrene. However, 24h nicotine treatment increased transendothelial permeability in CECs exposed to oxidative stress. Both nicotine and phenanthrene were able to regulate the expression of a large number of proteins as revealed by 2D electrophoresis. Our experiments suggest that tobacco smoking may affect the junctional complex of CECs, and that this effect is enhanced by oxidative stress.


Keywords: Nicotine, Blood–brain barrier, Cerebral endothelial cells, Polyaromatic hydrocarbons, Tight junction, Proteomics.

*Corresponding author. Tel.: +36 62 599 602; fax: +36 62 433 133.


1 Introduction

Tobacco-related diseases are among the world's leading preventable causes of death affecting not only active but passive smokers as well. Cigarette smoke is a complex mixture of different chemicals including nicotine and polyaromatic hydrocarbons (PAHs). The high number of patients affected by neurological disorders related to tobacco use suggests that one of the main targets of tobacco smoke components is the central nervous system (CNS) including the blood–brain barrier (BBB). Despite the high incidence of stroke among smokers and the role of BBB in stroke, little is known about the direct effects of smoking on the BBB. One of the most controversial issues is to which extent nicotine per se is involved in the development of cerebrovascular diseases and what the role of other components of cigarette smoke is.

The cerebral endothelium is a single-cell layer lining the blood vessels of the brain and constitutes the principal component of the BBB. By forming an active interface between blood and neuronal tissue it plays a key role in the maintenance of the homeostasis of CNS. The morphological basis of the endothelial barrier forming function is constituted by the interendothelial junctions: tight junctions (TJ) and adherens junctions (AJ). In this respect the transmembrane proteins of the junctional complexes may have a special importance. These include occludin (Furuse et al., 1993), the claudin family (Furuse et al., 1998) and the junctional adhesion molecules (Martin-Padura et al., 1998). Occludin binds to the members of the zonula occludens protein family (ZO-1, ZO-2, ZO-3) and cingulin which connect the junctional complex to the cytoskeleton. The transmembrane proteins of the adherens junctions are the cadherins, in the case of vascular endothelial cells VE-cadherin which is linked through the catenins (α, β, γ) to the cytoskeleton. A proper function of the adherens junction is needed for tight junction formation, blockade of cadherin with peptides directed against the extracellular region of cadherin causes an increase in BBB permeability (Pal et al., 1997).

The direct effect of nicotine on endothelial cells has been investigated in several studies (Abbruscato et al., 2002; Conklin et al., 2002; Cucina et al., 2000). The majority of these studies have been performed on endothelial cells of non-cerebral origin, showing that endothelial cells play a crucial role in the development of vascular diseases associated with cigarette smoking. Respecting endothelial barrier function it is of special importance that nicotine is able to reorganize the cytoskeleton of bovine aortic endothelial cells (Cucina et al., 2000) and that nicotine induced upregulation of vascular endothelial growth factor expression in porcine aortic endothelial cells (Conklin et al., 2002). It has also been shown that nicotine increases plasminogen activator inhibitor-1 production by cerebral endothelial cells via a PKC dependent pathway (Zidovetzki et al., 1999) and that nicotine is able to downregulate alpha 2 isoform of Na,K-ATPase at the BBB in rats, which is an important mediator of ion homeostasis in the brain (Wang et al., 1994). Furthermore, an increase in permeability and decrease in ZO-1 expression has been recently demonstrated (Abbruscato et al., 2002; Hawkins et al., 2004).

Besides nicotine, polyaromatic hydrocarbons may play an important role in endothelial damage. PAHs are present at high concentrations in cigarette smoke (1-methylanthracene: 1500ng/cigarette, phenanthrene: 362ng/cigarette) (Tithof et al., 2002). Recent studies demonstrate that cigarette smoke condensate (CSC) inhibits endothelial cell migration (Snajdar et al., 2001). In addition, CSC induces surface expression of a subset of cell adhesion molecules (CAMs) like intercellular adhesion molecule 1 (ICAM-1), endothelial leukocyte adhesion molecule 1 (ELAM-1), and vascular cell adhesion molecule 1 (VCAM-1) in human umbilical vein endothelial cells (HUVEC). These studies show that CSC-induced activation of protein kinase C in endothelial cells initiates signaling pathways leading to increased binding of NF-kappa B to specific DNA sequences, which in turn increases surface expression of a subset of CAMs. Furthermore, they demonstrate a link between CSC induced phosphorylation of PECAM-1 and the migration of blood monocytes across the vascular endothelium (Shen et al., 1996).

Previous experimental data suggested that smoking may influence BBB function. The present study was designed to clarify the structural and functional alterations at interendothelial junction (TJ and AJ) sites elicited by nicotine and PAHs and to identify proteins which may be involved in the mediation of the effect of cigarette smoke components on CECs.

2 Materials and methods

2.1 Chemicals and antibodies

All chemicals, unless otherwise stated, were purchased from Sigma (Budapest, Hungary). 2,3-dimethoxy-1,4-naphthoquinone (DMNQ) was from Alexis, protein G sepharose was from Amersham-Pharmacia Biotech. The following antibodies were used: anti-occludin, claudin-5, ZO-1, ZO-2 (Zymed, San Francisco, USA), anti-pan-cadherin, α-catenin, β-catenin (Sigma), Cy3 labelled anti-rabbit (Jackson). Fine chemicals for 2D electrophoresis were purchased from BioRad.

2.2 Cell culture

Rat brain endothelial cells were isolated from 2-week old rats. Brains were cut into small pieces and digested in two steps with collagenase type II (Sigma) and collagenase/dispase (Roche), followed by centrifugation on percoll gradient. The microvessel fragments were plated on collagen type IV and fibronectin coated dishes (Orange), glass coverslips or corneal extracellular matrix coated cell culture inserts (Transwell clear, 1cm2; pore size: 0.4μm, Costar) at a density of one brain equivalent per 10cm2. The cells were cultured in DMEM/F12 (Sigma) medium containing 20% PDS (First Link, UK), supplemented with 1ng/ml bFGF (Roche), 100μg/ml heparin, physiological concentration of hydrocortisone and antibiotics (Hoheisel et al., 1998) at 37°C and 5% CO2 until forming confluent monolayer. In the first two days, 4μg/ml puromycin was added to remove contaminating cells (Perriere et al., 2005). The cell layers reached confluency in 5–7days and were used as primary cultures for the experiments.

For co-culture primary cultures of glial cells were prepared from two-day old Wistar rats. After removing the meninges, cortices were mechanically dissociated in DMEM (Sigma) containing 10% fetal bovine serum. The cells were plated on poly-l-lysin coated 12-well dishes (Costar) and used 3weeks after reaching confluency.

2.3 Preparation of cell extracts and western-blot

Confluent monolayers of brain endothelial cells were treated with different concentrations of nicotine, phenanthrene and methylanthracene. The highest concentrations used (nicotine: 10−5M, phenanthrene and methylanthracene: 3×10−5M) were not cytotoxic, as assessed by the XTT assay. After exposure, cells were washed with PBS and scraped into ice-cold lysis buffer (20mM Tris–HCl pH 7.5, 150mM NaCl, 0.5% Triton X-100, 0.5% NP-40, 2mM CaCl2, 5mM NaF, 1mM Na orthovanadate and 1mM Pefabloc (Roche)) and incubated on ice for 30min. Lysates were clarified by centrifugation at 10,000×g for 10min and protein concentration was determined with the BCA method (Pierce). Within each set of samples, equal amount of protein was loaded per gel and electrophoresed using standard SDS-PAGE procedures, and finally blotted on to PVDF membranes (Pall). The immunoreaction was visualized by a chemiluminescence detection kit (Pierce). Gel loading uniformity has been checked by staining the membranes with amido black.

For the preparation of Triton X-100 soluble fraction, cells were scraped into lysis buffer (20mM Tris–HCl pH 7.5, 150mM NaCl, 0.5% Triton X-100, 1mM Na orthovanadate and 1mM Pefabloc) and incubated on ice for 30min. Homogenates were centrifuged at 10,000×g for 10min and the supernatant was used as the Triton X-100 soluble fraction. The pellet was resuspended in SDS sample buffer and used as the Triton X-100 insoluble fraction.

2.4 Immunoprecipitation

For immunoprecipitation, cells were homogenized as described above in 500ml lysis buffer. The lysates were centrifuged at 10,000rpm in a microfuge and the supernatant was subjected to immunoprecipitation. Briefly, after preclearing with protein G-Sepharose supernatants were incubated with 2–5μg primary antibody (β-catenin) at 4°C for 4h. The formed immunocomplexes were precipitated by incubating the samples overnight with protein G-Sepharose beads (Amersham-Pharmacia Biotech). The precipitates were washed 4 times with lysis buffer, boiled in sample buffer, and subjected to SDS-PAGE and immunoblotting.

2.5 Immunofluorescence

For immunofluorescent studies cells were cultured on collagen/fibronectin coated coverslips. Cells were fixed using a mixture of ice cold ethanol:acetic acid (95:5) for 10min. After blocking with 3% bovine serum albumin for 30min coverslips were incubated with primary antibodies. The staining was visualized using Cy3 conjugated secondary antibodies. Coverslips were mounted in anti-fading embedding medium (Biomeda) and the distribution of the signal was studied using a Nikon Eclipse TE2000U photomicroscope with epifluorescent capabilities connected to a digital camera (Spot RT KE).

2.6 Transendothelial electric resistance (TEER)

In order to obtain a better BBB model, endothelial cells were co-cultured with astrocytes which is a widely accepted model of the BBB (Dehouck et al., 1990). Brain endothelial cells were cultured on cell culture inserts in plates containing astroglia at the bottom of the wells with endothelial culture medium in both compartments. TEER was measured using a chopstick electrode and an EVOM epithelial voltohmmeter (World Precision Instruments). The resistance of the coated filter was substracted from the measured value to obtain the resistance of the endothelial monolayer itself.

2.7 Statistical analysis

All TEER data presented are means±S.D. The values were compared using paired t-test. Changes were considered significant at P<0.05. The number of parallel filter inserts was 3 for each time-point and treatment.

2.8 2D electrophoresis

Primary cultures of rat brain endothelial cells were treated with 10−5M nicotine or 3×10−5M phenanthrene. The cells were put in lysis buffer (2% NP40, 2% CHAPS, 20mM Tris–HCl, 1mM EDTA, 0.2mM vanadate, 1mM pefabloc, 1mM NaF, 2% ampholite). After centrifugation, the protein concentration was determined using the BCA method (Pierce) from the soluble fraction. The sample-rehydration buffer was produced by adding the necessary amount of reagents and protein samples to reach the final concentration of 7M urea, 1% DTT, 2% CHAPS, 20mM Tris–HCl and a protein concentration of &007E;1mg/ml. The 2D electrophoresis was carried out using a Protean IEF Cell, a Protean II xi 2-D Cell (BioRad Laboratories Inc.) according to the manufacturer's instructions. The gels were stained with silver using a protocol appropriate for MALDI mass spectrometry, then scanned and analysed with the PDQuest 7.3 software (BioRad Laboratories Inc.). Spots were excised with a spot cutter (ProteomeWorks spot cutter, BioRad Laboratories Inc.) and were subjected to tryptic digestion and MALDI-MS analysis. Two parallel gels were run for each experiment.

2.9 Protein identification by mass spectrometry

Proteins were in-gel digested by side-chain protected porcine trypsin according to a published protocol (http://donatello.ucsf.edu/ingel.html). The unfractionated digests were analyzed on a Bruker Reflex III MALDI-TOF mass spectrometer, using 2,5-dihydroxy-benzoic acid as the matrix. Database searches were performed with the monoisotopic MH+ lists against the NCBI protein database using the MS-Fit program of ProteinProspector. (http://prospector.ucsf.edu). Protein IDs were confirmed by MS/MS (Post-source decay) analysis.

2.10 Induction of oxidative stress

To investigate the additional effects of oxidative stress, nicotine treatment was combined with DMNQ (10−5M) for 1–24h. DMNQ (2,3-Dimethoxy-1,4-naphthoquinone) is a cyclical-quinone widely applied to study the effect of reactive oxygen species.

3 Results

3.1 Expression of junctional proteins

Treatment of cerebral endothelial cells with nicotine (10−8, 10−7, 10−6 and 10−5M) for 15min, 60min and 5h did not cause changes in occludin expression even when the highest nicotine concentration was used (Fig. 1A). Treatment of the cells for 24h with 10−5M nicotine caused a slight decrease in the expression of occludin in the Triton X-100 insoluble fraction (Fig. 1B). The junctional plaque protein ZO-1 showed similar changes to occludin. Only high concentrations of nicotine (10−5M) for 5h and 24h caused decrease in its expression (Fig. 2A), but the decrease was more pronounced than in the case of occludin. A 5h treatment of CECs with nicotine did not change the expression of cadherin. However, 24h treatment with high concentrations of nicotine led to a decrease in its expression (Fig. 2B). No changes in the expression of β-catenin were observed (Fig. 2C).


Fig. 1

Expression of occludin in response to nicotine. Cells were treated with 10−8–10−4M nicotine for 15min, 60min, 5h (A) and 24h (B). Occludin expression was unaffected by the applied concentrations of nicotine after 15min to 5h, while treatment of the cells for 24h with 10−5M nicotine caused a slight decrease of expression in the Triton X-100 insoluble fraction. Densitometrical analysis has been performed for the samples treated with nicotine for 24h (C). One representative of 3 independent experiments is shown.


Fig. 2

Expressional changes of ZO-1 and cadherin in response to nicotine. Cells were treated with 10−6–10−5M nicotine for 5h and 24h. The level of ZO-1 (A) decreased when treated with 10−5M nicotine, cadherin (B) was affected only after 24h treatment while beta-catenin (C) was unaffected. Representative blots of three independent experiments are shown.



Treatment with 3×10−5M phenanthrene for 24h (Fig. 3) caused a redistribution of occludin from the Triton X-100 insoluble to the Triton X-100 soluble fraction. A similar but less pronounced effect was seen in the case of claudin-5. No changes were observed in the expression of α- and β-catenin and ZO-1. Methylanthracene (3×10−5M) was not able to induce changes in junctional protein expression, except a slight decrease in claudin-5 expression in the Triton X-100 insoluble fraction. The interaction of adherens junction proteins was studied by coimmunoprecipitation. Our results show that phenanthrene treatment did not affect the amount of cadherin or α-catenin in samples coimmunoprecipitated with β-catenin, indicating that phenanthrene did not cause changes in the interactions between these components of the adherens junction (Fig. 4).


Fig. 3

Expression of junctional proteins in response to 3×10−5M methylanthracene (MA) or 3×10−5M phenanthrene (Ph) in the Triton X-100 soluble and Triton X-100 insoluble fraction. Phenanthrene caused a decrease of occludin in the insoluble fraction while, increasing the amount of soluble occludin. Moreover phenanthrene caused a similar but less pronounced change in the expression of claudin-5. Methilanthracene induced a slight decrease in claudin-5 expression in the Triton X-100 insoluble fraction. Representative blots of 3 independent experiments are shown.


Fig. 4

Effect of phenanthrene (Ph) on the interaction of β-catenin with α-catenin and cadherin. CECs were treated with 3×10−5M phenanthrene for 24h. Immunoprecipitation was performed using anti β-catenin antibodies and the blots were stained with β-catenin, α-catenin and pan-cadherin antibodies. The treatment did not cause changes in the β-, α-catenin and cadherin content of the precipitated samples, indicating that the interaction of these proteins remained unchanged.



3.2 Localization of junctional proteins by immunofluorescence

Immunofluorescent staining of primary RBEC cultures confirmed the results of western-blot analysis. The most pronounced decrease in response to 24h nicotine treatment (10−5M) was observed in case of ZO-1, accompanied by a disruption of the continuous membrane staining. A similar, but less pronounced, effect of nicotine was observed on the localization of occludin, ZO-2 and cadherin. No changes could be seen in the subcellular localization of claudin-5 and beta-catenin (Fig. 5).


Fig. 5

Immunofluorescent localization of junctional proteins in response to nicotine treatment. Cells were treated with 10−5M nicotine for 24h and stained for occludin, claudin-5, ZO-1, ZO-2, cadherin and β-catenin. A remarkable disruption was seen in continuous membrane staining of ZO-1. A similar but less pronounced effect on the localization of occludin, ZO-2 and cadherin was observed. No changes were seen in the subcellular localization of claudin-5 and beta-catenin. Arrows mark disruptions in staining. Scale bar=100μm.


Treatment with 3×10−5M phenanthrene for 24h did not cause significant changes in ZO-1 or claudin-5 localization. In case of occludin and ZO-2, a slight decrease in the continuity of the staining was detected (Fig. 6).


Fig. 6

Distribution of junctional proteins in response to phenanthrene. Cells were treated with 3×10−5M phenanthrene and stained for occludin, claudin-5, ZO-1 and ZO-2. ZO-2 and occludin showed a slight decrease in staining continuity, but no changes were observed in ZO-1 and claudin-5 distribution. Arrows mark disruptions in staining. Scale bar=100μm.


Localization of junctional proteins after a combined nicotine and phenanthrene treatment was comparable to that seen after nicotine treatment alone. After the combined treatment we observed a marked decrease in ZO-1 staining intensity and continuity, while ZO-2 and occludin staining showed a slight disruption of continuity (Fig. 7).


Fig. 7

Redistribution of junctional proteins in response to combined nicotine and phenantrene treatment. Cells were treated with 3×10−5M phenanthrene and 10−5M nicotine for 24h and stained for occludin, ZO-1 and ZO-2. Combined nicotine and phenanthrene treatment caused a pronounced disruption in the continuity of the ZO-1 staining, and a slight disruption in case of ZO-2 and occludin. Scale bar=100μm.


3.3 Assessment of the barrier function by the measurement of the transendothelial electrical resistance

For TEER measurements, CECs were co-cultured with astrocytes on Transwell filters coated with collagen IV/fibronectin and supplemented with physiological concentrations of hydrocortisone and CPT-cAMP. With this protocol we could achieve average TEER values of 200–300Ohm×cm2 with peak values of 600Ohm×cm2.

The CECs were treated with 10−5M nicotine added to the luminal side of the monolayer. No significant changes in TEER were observed compared to the control (Fig. 8A). No significant decrease in TEER was observed either after phenanthrene treatment (Fig. 8B), and similar results were obtained with methylanthracene and combined nicotine and phenanthrene treatment (not shown).


Fig. 8

Effect of nicotine (A, 10−5M) and phenanthrene (B, 3×10−5M) on the transendothelial electrical resistance. Neither nicotine nor phenanthrene caused significant changes in TEER.


3.4 Cumulative effect of nicotine treatment and oxidative stress

Cigarette smoke contains oxidants that remain stable in aqueous solution and can penetrate into the blood through the lung alveolar wall (Yamaguchi et al., 2007). Thus the effect of nicotine is often combined with oxidative stress in smokers. Since oxidative stress is known to impair BBB function (Krizbai et al., 2005), a study of the combined effect of nicotine and oxidative stress could lead to a better understanding of the damaging effect of smoking on the BBB. Combination of 24h nicotine treatment with oxidative stress induced by DMNQ (10−5M) potentiated the transendothelial resistance decrease induced by oxidative stress (Fig. 9). Furthermore, alteration in the ZO-1 staining induced by combined treatment was more pronounced compared to DMNQ or nicotine treatment alone (Fig. 10).


Fig. 9

Effect of nicotine (10−5M) and DMNQ (10−5M) on the transendothelial electrical resistance. Nicotine treatment combined with DMNQ caused a significant decrease in TEER. *P<0.05, paired t-test.


Fig. 10

Effect of nicotine and oxidative stress on the localization of ZO-1. Cells were treated with 10−5M nicotine and 10−5M DMNQ for 24h. Nicotine treatment combined with DMNQ caused a substantially greater disruption in the ZO-1 staining than either nicotine or DMNQ alone. Scale bar=100μm.



3.5 Proteomic analysis

Primary cultures of rat brain endothelial cells treated with 10−5M nicotine or 3×10−5M phenanthrene were subjected to 2D electrophoresis and proteins that were differentially expressed compared to control cell cultures were identified by MALDI-MS. Tables 1 and 2 list proteins which showed altered levels after the respective treatment.


Table 1.

Differentially expressed proteins in cerebral endothelial cells following nicotine treatment


Table 2.

Differentially expressed proteins in cerebral endothelial cells following phenanthrene treatment

NCBI accession #NameRatioaMatchesbCoveragecPSDd
38181888Proteasome (prosome, macropain) 26S subunit, ATPase 2, PSMC20.1729/40 (72%)60%Yes
35–40
488838, 631812, 2501206Protein disulfide-isomerase A6 precursor (protein disulfide isomerase P5) (calcium-binding protein 1) (CaBP1)0.2014/20 (70%)32%Yes
110–113
6978501, 38197394, 56566, 71758, 113947, 203252Annexin A1 (Annexin I) (Lipocortin I) (Calpactin II) (Chromobindin 9) (P35) (Phospholipase A2 inhibitory protein)0.5523/32 (71%)59%Yes
205–212
220838Dihydrolipoamide acetyltransferase2.1311/46 (23%)22%Yes
48–55
gi|62647139Zyxin2.5115/24 (62%)25%Yes
348–359
51948378, 50925575Minichromosome maintenance protein 74.6810/30 (33%)17%
34856991Similar to GMP synthase [glutamine-hydrolyzing] (glutamine amidotransferase) (GMP synthetase)5.6110/26 (38%)22%Yes
529–539
1363329Transcription elongation factor elongin A6.026/21 (28%)9%
51859516Heat shock 90 kDa protein 1, beta7.059/19 (47%)11%
a Ratio shows protein expression change compared to control estimated from silver staining intensity.
b Matches show how many of the mass spectrum peaks match the database data.
c Coverage shows how much of the amino acid sequence is covered by the peptides of the detected peaks.
d PSD shows if the post source decay analysis successfully confirms the MS data, which improves the reliability of the protein identification. The numbers show the position of the analysed peptide in the protein.

NCBI accession #NameRatioaMatchesbCoveragecPSDd
40786469, 38303871Dihydrolipoamide dehydrogenase (E3 component of pyruvate dehydrogenase complex)0.7912/18 (66%)33%Yes
133–143
40254781, 38197560Guanosine diphosphate dissociation inhibitor 3; rab GDI beta; guanosine diphosphate (GDP) dissociation inhibitor 30.8430/37 (81%)69%Yes
380–390
34864883Similar to chaperonin containing TCP-1 beta subunit0.8925/28 (89%)57%Yes
1–13
40018616, 38969850Chaperonin containing TCP1, subunit 3 (gamma)1.1124/29 (82%)51%Yes
434–449
34872057Similar to CCT (chaperonin containing TCP-1) zeta subunit1.1315/17 (88%)41%Yes
130–138
34874349Similar to Dihydropyrimidinase related protein-2 (DRP-2) (turned on after division, 64 kDa protein) (TOAD-64)1.1524/25 (96%)46%Yes
564–570
204197Glucose-6-phosphate dehydrogenase1.1721/26 (80%)50%Yes
127–135
20302113, 38181876, 2511703Stress-induced-phosphoprotein 1, Hsp70/Hsp90-organizing protein, Hop, p60 protein1.2223/37 (62%)49%Yes
1–10
316–325
26350443Similar to Tardbp protein1.2811/18 (61%)45%Yes
103–114
18266700, 18104446Similar to hypothetical protein FLJ10849 (SEPTIN) and heterogeneous nuclear ribonucleoprotein H1, Ratsg11.296/30 (20%)19%
6981146, 37590241, 2117457, 473577Lactate dehydrogenase B1.414/15 (93%)45%Yes
120–127
40786436, 39793992Eukaryotic translation initiation factor 4A, isoform 11.6822/24 (91%)55%Yes
325–334
34868946Similar to Vinculin (Metavinculin)2.3327/34 (79%)32%No
488838, 2501206, 631812Protein disulfide-isomerase A6 precursor (protein disulfide isomerase P5) (Calcium-binding protein 1) (CaBP1)2.8812/17(70%)41%Yes
110–123
1729977Transketolase3.188/14 (57%)17%Yes
382–395
14249134, 6840951, 37087655Thioredoxin-like 2 protein, PKC-interacting cousin of thioredoxin, PICOT3.239/31 (29%)42%Yes
1–22
a Ratio shows protein expression change compared to control estimated from silver staining intensity.
b Matches show how many of the mass spectrum peaks match the database data.
c Coverage shows how much of the amino acid sequence is covered by the peptides of the detected peaks.
d PSD shows if the post source decay analysis successfully confirms the MS data, which improves the reliability of the protein identification. The numbers show the position of the analysed peptide in the protein.

Both agents caused changes in the levels of different shock induced proteins: after nicotine treatment we observed a slight realignment of chaperonin containing TCP subunits, and an increase in the Hsp70/Hsp90-organizing protein (Hop), while phenanthrene caused a rise in the Hsp90 level of the cells.

We also noted changes in the levels of certain proteins involved in intracellular signaling. Nicotine increased the calcium binding protein 1 level, while phenanthrene treatment greatly diminished it. Nicotine also increased the abundance of the PICOT protein, which possibly plays a role in regulating stress induced signalization through the inhibition of PKC theta. Phenanthrene induced an increase in the expression of zyxin, which is a component of focal adhesion plaques and it has been shown to shuttle to the nucleus where it interacts with a variety of proteins.

Both nicotine and phenanthrene treatment increased the levels of particular proteins involved in replication, transcription and translation: Elf4A and–to a lesser extent–a protein similar to Tardbp in nicotine treatment, minichromosome maintenance protein 7 and transcription elongation factor elongin A in phenanthrene-treated cells.

Phenanthrene decreased the level of the proteasome subunit PSMC2 which is one of the six putative ATPases in the regulatory complex of the proteasome. Interestingly, the expression of GMP synthase–the key enzyme of guanine nucleotide synthesis, and a stimulator of histone deubiquitylation–was also increased by phenanthrene.

Furthermore, several enzymes have also been identified among the proteins which showed expressional changes after nicotine or phenanthrene treatment. Nicotine decreased the level of dihydrolipoamide dehydrogenase, slightly increased that of glucose-6-phosphate dehydrogenase and lactate dehydrogenase B, while markedly increasing that of transketolase. On the other hand, phenanthrene increased the expression of dihydrolipoamide acetyltransferase.

Other proteins affected by nicotine treatment were: GDP dissociation inhibitor 3, similar to dihydropyrimidinase related protein-2 and–to a greater extent–metavinculin. Phenanthrene decreased the expression of annexin I which is a member of the annexins implicated in several cellular processes, including modulation of phospholipase A2 (PLA2) activity and inflammation, immune response, proliferation, blood coagulation, differentiation, exocytosis, membrane skeletal linkage and intracellular signal transduction.

4 Discussion

Although smoking is known to be involved in the pathogenesis of several neurological and vascular diseases, few data are available about the effect of cigarette smoke components on brain endothelial cells. Since cerebral endothelium is located at the blood-CNS interface, it can be the first target of different substances affecting the brain.

Nicotine and PAHs are abundant and biologically active components of cigarette smoke. We carried out a complex investigation on the effect of nicotine and PAHs on cerebral endothelial cells in order to dissect the physiopathological consequences of smoking on the cerebral endothelium. The experiments were focused on changes affecting the tight and adherens junction proteins. Moreover, a broader range of proteins was also analyzed using 2D electrophoresis.

Treatments of nicotine, phenanthrene or methylanthracene for 15min to 5h did not affect the transmembrane proteins of the tight junctions (occludin, claudin-5). However, 24h treatment of CECs with relatively high concentration nicotine led to a decrease in occludin, cadherin and ZO-1 expression. Similar but less pronounced effects were observed after 24h treatment with phenanthrene.

Results of the immunofluorescent analysis confirmed western-blot data showing that the most sensitive tight junction protein to nicotine was ZO-1. Similar results were previously reported on primary bovine brain microvessel endothelial cells and also cerebral microvessels isolated from nicotine-treated rats (Abbruscato et al., 2002; Hawkins et al., 2004). We observed a less pronounced sensitivity of endothelial ZO-1 and cadherin to PAHs. Combination of nicotine and phenanthrene treatment did not potentiate the alterations seen with nicotine alone in the localization of junctional proteins.

Co-immunoprecipitation studies showed that 3×10−5M phenanthrene treatment for 24h did not induce the dissociation of the catenin-cadherin complex, supporting the notion that the adherens junction is not appreciably affected by phenanthrene.

The data suggest that only relatively high concentrations of CSCs are able to induce changes in the structure of the junctional complex. Only above peak plasma levels of nicotine influenced the expression of junctional proteins. In considering the damaging effect of nicotine, however, it should be noted that rats are significantly more resistant to nicotine toxicity than humans. Furthermore, the in vitro model system allowed only relatively short-term monitoring (max. 48h) of nicotine effects.

The observed changes did not lead to a significant decrease in the transendothelial electric resistance. This is in line with data obtained on bovine CECs (Abbruscato et al., 2002), showing that 12–24h nicotine/cotinine treatment had no effect on TEER in a normoxic environment. However, when the opening of the paracellular route of the BBB was assessed by [14C]sucrose permeability, statistically significant increase was observed after 100ng/ml nicotine treatment for 10–24h (Abbruscato et al., 2002). This is probably due to the fact that different aspects of junctional permeability are regulated by distinct mechanisms. Furthermore, an in vivo study has shown that only toxic concentrations of nicotine (10−4–10−3M) were able to increase the permeability to 70kDa FITC-dextran (Schilling et al., 1992).

Junctional proteins were much more affected when nicotine treatment was used in combination with DMNQ. Oxidative stress is known to alter the integrity of the junctions (Krizbai et al., 2005; Witt et al., 2003). We observed a pronounced decrease of TEER compared to DMNQ treatment alone, suggesting a cumulative effect of nicotine abuse and hypoxia. This was accompanied by morphological changes: an even more pronounced alteration in ZO-1 membrane staining compared to nicotine or DMNQ alone. This finding suggests that nicotine abuse may aggravate the clinical status of patients affected by hypoxia via an impaired BBB function. Our results support those of Abbruscato et al. (2002, 2004), who showed that there is a cumulative negative effect of nicotine and hypoxia/aglycemia on the BBB regarding permeability and potassium transport.

In order to explore the cellular proteins responsive to nicotine and PAHs in brain endothelial cells two-dimensional electrophoresis was performed. Proteomic analysis of changes induced by cigarette smoke components has already been performed in various cell types (Piubelli et al., 2005; Raveendran et al., 2005), indicating the up- and down-regulation of the expression of diverse proteins. Proteomic analysis of striatum from nicotine-treated rats revealed altered expression of a number of proteins involved in cellular signaling and apoptosis (Yeom et al., 2005). We observed different responses of the cells to both nicotine and phenanthrene treatment, resulting in altered expression of shock induced proteins, metabolic enzymes, signaling molecules. This confirms the cerebral endothelium as being a target of cigarette smoke components.

The effect of nicotine may be mediated by the presence of nicotinic acetylcholine receptors (subunits alpha 3, 5, 7, beta 2) on the surface of CECs (Abbruscato et al., 2002; Hawkins et al., 2005). However, non-receptor mediated mechanisms are also possible (Tonnessen et al., 2000).

Cerebral endothelial cells are also affected by polycyclic aromatic hydrocarbons. To our knowledge, the effect of PAHs has not so far been investigated on cerebral endothelial cells. However, human coronary artery endothelial cells exposed to PAHs showed important alterations: PLA2 activation, fatty acid release and apoptosis (Tithof et al., 2002). In human aortic endothelial cells stimulated by cigarette smoke condensate rich in PAHs, there was a rapid induction of the production of IL-4 and IL-8, which might reflect a damaging effect of cigarette smoke on endothelial cells (Nordskog et al., 2005). The mechanisms by which PAHs may exert their effect on endothelial cells are largely unknown, but PAH receptors may play an important role in this process (van Grevenynghe et al., 2006).

The results of our complex investigation on the effect of nicotine and polyaromatic hydrocarbons on cerebral endothelial cells suggest that cigarette smoke components probably do not cause acute alterations in the principal functional properties of the cerebral endothelial cells. However, 24h exposure of these cells to cigarette smoke components, especially in combination with other damaging effects like oxidative stress, may lead to a significantly impaired BBB function. Our results provide important information for the evaluation of smoking associated risk in different neurological disorders like cerebral ischemia.

Acknowledgements

This work was supported by OTKA T037956 and Philip Morris Inc, USA.

References

Abbruscato TJ, Lopez, SP, Mark, KS, Hawkins, BT, Davis, TP. Nicotine and cotinine modulate cerebral microvascular permeability and protein expression of ZO-1 through nicotinic acetylcholine receptors expressed on brain endothelial cells. J Pharm Sci 2002:91:2525-38
Crossref   Medline   1st Citation   2nd   3rd   4th   5th   6th   7th  

Abbruscato TJ, Lopez, SP, Roder, K, Paulson, JR. Regulation of blood–brain barrier Na,K,2Cl-cotransporter through phosphorylation during in vitro stroke conditions and nicotine exposure. J Pharmacol Exp Ther 2004:310:459-68
Crossref   Medline   1st Citation  

Conklin BS, Zhao, W, Zhong, DS, Chen, C. Nicotine and cotinine up-regulate vascular endothelial growth factor expression in endothelial cells. Am J Pathol 2002:160:413-8
Medline   1st Citation   2nd  

Cucina A, Sapienza, P, Borrelli, V, Corvino, V, Foresi, G, Randone, B. Nicotine reorganizes cytoskeleton of vascular endothelial cell through platelet-derived growth factor BB. J Surg Res 2000:92:233-8
Crossref   Medline   1st Citation   2nd  

Dehouck MP, Meresse, S, Delorme, P, Fruchart, JC, Cecchelli, R. An easier, reproducible, and mass-production method to study the blood–brain barrier in vitro. J Neurochem 1990:54:1798-801
Crossref   Medline   1st Citation  

Furuse M, Hirase, T, Itoh, M, Nagafuchi, A, Yonemura, S, Tsukita, S. Occludin: a novel integral membrane protein localizing at tight junctions. J Cell Biol 1993:123:1777-88
Crossref   Medline   1st Citation  

Furuse M, Fujita, K, Hiiragi, T, Fujimoto, K, Tsukita, S. Claudin-1 and -2: novel integral membrane proteins localizing at tight junctions with no sequence similarity to occludin. J Cell Biol 1998:141:1539-50
Crossref   Medline   1st Citation  

Hawkins BT, Abbruscato, TJ, Egleton, RD, Brown, RC, Huber, JD, Campos, CR. Nicotine increases in vivo blood–brain barrier permeability and alters cerebral microvascular tight junction protein distribution. Brain Res 2004:1027:48-58
Crossref   Medline   1st Citation   2nd  

Hawkins BT, Egleton, RD, Davis, TP. Modulation of cerebral microvascular permeability by endothelial nicotinic acetylcholine receptors. Am J Physiol Heart Circ Physiol 2005:289:H212-9
Crossref   Medline   1st Citation  

Hoheisel D, Nitz, T, Franke, H, Wegener, J, Hakvoort, A, Tilling, T. Hydrocortisone reinforces the blood–brain barrier properties in a serum free cell culture system. Biochem Biophys Res Commun 1998:244:312-6
Crossref   Medline   1st Citation  

Krizbai IA, Bauer, H, Bresgen, N, Eckl, PM, Farkas, A, Szatmari, E. Effect of oxidative stress on the junctional proteins of cultured cerebral endothelial cells. Cell Mol Neurobiol 2005:25:129-39
Crossref   Medline   1st Citation   2nd  

Martin-Padura I, Lostaglio, S, Schneemann, M, Williams, L, Romano, M, Fruscella, P. Junctional adhesion molecule, a novel member of the immunoglobulin superfamily that distributes at intercellular junctions and modulates monocyte transmigration. J Cell Biol 1998:142:117-27
Crossref   Medline   1st Citation  

Nordskog BK, Fields, WR, Hellmann, GM. Kinetic analysis of cytokine response to cigarette smoke condensate by human endothelial and monocytic cells. Toxicology 2005:212:87-97
Crossref   Medline   1st Citation  

Pal D, Audus, KL, Siahaan, TJ. Modulation of cellular adhesion in bovine brain microvessel endothelial cells by a decapeptide. Brain Res 1997:747:103-13
Crossref   Medline   1st Citation  

Perriere N, Demeuse, P, Garcia, E, Regina, A, Debray, M, Andreux, JP. Puromycin-based purification of rat brain capillary endothelial cell cultures. Effect on the expression of blood–brain barrier-specific properties. J Neurochem 2005:93:279-89
Crossref   Medline   1st Citation  

Piubelli C, Cecconi, D, Astner, H, Caldara, F, Tessari, M, Carboni, L. Proteomic changes in rat serum, polymorphonuclear and mononuclear leukocytes after chronic nicotine administration. Proteomics 2005:5:1382-94
Crossref   Medline   1st Citation  

Raveendran M, Senthil, D, Utama, B, Shen, Y, Wang, J, Zhang, Y. Effect of water-soluble fraction of cigarette smoke on human aortic endothelial cells—a proteomic approach. Cell Biol Toxicol 2005:21:27-40
Crossref   Medline   1st Citation  

Schilling L, Bultmann, A, Wahl, M. Lack of effect of topically applied nicotine on pial arteriole diameter and blood–brain barrier integrity in the cat. Clin Investig 1992:70:210-7
Medline   1st Citation  

Shen Y, Rattan, V, Sultana, C, Kalra, VK. Cigarette smoke condensate-induced adhesion molecule expression and transendothelial migration of monocytes. Am J Physiol 1996:270:H1624-33
Medline   1st Citation  

Snajdar RM, Busuttil, SJ, Averbook, A, Graham, DJ. Inhibition of endothelial cell migration by cigarette smoke condensate. J Surg Res 2001:96:10-6
Crossref   Medline   1st Citation  

Tithof PK, Elgayyar, M, Cho, Y, Guan, W, Fisher, AB, Peters-Golden, M. Polycyclic aromatic hydrocarbons present in cigarette smoke cause endothelial cell apoptosis by a phospholipase A2-dependent mechanism. FASEB J 2002:16:1463-4
Medline   1st Citation   2nd  

Tonnessen BH, Severson, SR, Hurt, RD, Miller, VM. Modulation of nitric-oxide synthase by nicotine. J Pharmacol Exp Ther 2000:295:601-6
Medline   1st Citation  

van Grevenynghe J, Monteiro, P, Gilot, D, Fest, T, Fardel, O. Human endothelial progenitors constitute targets for environmental atherogenic polycyclic aromatic hydrocarbons. Biochem Biophys Res Commun 2006:341:763-9
Crossref   Medline   1st Citation  

Wang L, McComb, JG, Weiss, MH, McDonough, AA, Zlokovic, BV. Nicotine downregulates alpha 2 isoform of Na,K-ATPase at the blood–brain barrier and brain in rats. Biochem Biophys Res Commun 1994:199:1422-7
Crossref   Medline   1st Citation  

Witt KA, Mark, KS, Hom, S, Davis, TP. Effects of hypoxia-reoxygenation on rat blood–brain barrier permeability and tight junctional protein expression. Am J Physiol Heart Circ Physiol 2003:285:H2820-31
Medline   1st Citation  

Yamaguchi Y, Nasu, F, Harada, A, Kunitomo, M. Oxidants in the gas phase of cigarette smoke pass through the lung alveolar wall and raise systemic oxidative stress. J Pharmacol Sci 2007:103:275-82
Crossref   Medline   1st Citation  

Yeom M, Shim, I, Lee, HJ, Hahm, DH. Proteomic analysis of nicotine-associated protein expression in the striatum of repeated nicotine-treated rats. Biochem Biophys Res Commun 2005:326:321-8
Crossref   Medline   1st Citation  

Zidovetzki R, Chen, P, Fisher, M, Hofman, FM, Faraci, FM. Nicotine increases plasminogen activator inhibitor-1 production by human brain endothelial cells via protein kinase C-associated pathway. Stroke 1999:30:651-5
Medline   1st Citation  


Received 12 April 2007/18 July 2007; accepted 29 August 2007

doi:10.1016/j.cellbi.2007.08.026


ISSN Print: 1065-6995
ISSN Electronic: 1095-8355
Published by Portland Press Limited on behalf of the International Federation for Cell Biology (IFCB)