Nino Muradashvili,1* Syed Jalal Khundmiri,2* Reeta Tyagi,1 Allison Gartung,4 William L. Dean,3
Menq-Jer Lee,4 and David Lominadze1
1Department of Physiology and Biophysics, School of Medicine, University of Louisville, Louisville, Kentucky; 2KidneyDisease Program, Department of Medicine, School of Medicine, University of Louisville, Louisville, Kentucky; 3Departmentof Biochemistry and Molecular Biology, School of Medicine, University of Louisville, Louisville, Kentucky; and 4Departmentof Pathology, School of Medicine, Wayne State University, Detroit, Michigan
Submitted 26 September 2013; accepted in final form 8 May 2014
Muradashvili N, Khundmiri SJ, Tyagi R, Gartung A, DeanWL, Lee MJ, Lominadze D. Sphingolipids affect fibrinogen-inducedcaveolar transcytosis and cerebrovascular permeability. Am J PhysiolCell Physiol 307: C169–C179, 2014. First published May 14, 2014;doi:10.1152/ajpcell.00305.2013.—Inflammation-induced vascular en-dothelial dysfunction can allow plasma proteins to cross the vascularwall, causing edema. Proteins may traverse the vascular wall throughtwo main pathways, the paracellular and transcellular transport path-ways. Paracellular transport involves changes in endothelial celljunction proteins, while transcellular transport involves caveolartranscytosis. Since both processes are associated with filamentousactin formation, the two pathways are interconnected. Therefore, it isdifficult to differentiate the prevailing role of one or the other pathwayduring various pathologies causing an increase in vascular permea-bility. Using a newly developed dual-tracer probing method, wedifferentiated transcellular from paracellular transport during hyper-fibrinogenemia (HFg), an increase in fibrinogen (Fg) content. Roles ofcholesterol and sphingolipids in formation of functional caveolaewere assessed using a cholesterol chelator, methyl-�-cyclodextrin,and the de novo sphingolipid synthesis inhibitor myriocin. Fg-inducedformation of functional caveolae was defined by association andcolocalization of Na�-K�-ATPase and plasmalemmal vesicle-associ-ated protein-1 with use of Förster resonance energy transfer and totalinternal reflection fluorescence microscopy, respectively. HFg in-creased permeability of the endothelial cell layer mainly through thetranscellular pathway. While M�CD blocked Fg-increased transcel-lular and paracellular transport, myriocin affected only transcellulartransport. Less pial venular leakage of albumin was observed inmyriocin-treated HFg mice. HFg induced greater formation of func-tional caveolae, as indicated by colocalization of Na�-K�-ATPasewith plasmalemmal vesicle-associated protein-1 by Förster resonanceenergy transfer and total internal reflection fluorescence microscopy.Our results suggest that elevated blood levels of Fg alter cerebrovas-cular permeability mainly by affecting caveolae-mediated transcytosisthrough modulation of de novo sphingolipid synthesis.
cholesterol; protein leakage; Förster resonance energy transfer mi-croscopy; total internal reflection fluorescence microscopy; functionalcaveolae
BLOOD PLASMA PROTEINS can cross the endothelial cell (EC) layervia transcellular and paracellular transport pathways. Whileparacellular transport of proteins occurs via EC junction pro-teins, transcellular transport involves formation of functional
caveolae. Caveolae are distinct flask-shaped, invaginated struc-tures present at the surface of many cell types, including ECs(57). Their walls are enriched with sphingolipids and choles-terol (50). Therefore, caveolae are defined as specialized,morphologically distinct sphingolipid- and cholesterol-rich mi-crodomains. It has been suggested that caveolar mobility canbe affected by sphingolipids or cholesterol (46). However, aspecific role of one or the other is not well established. Forexample, it was shown that caveolae movement can be in-creased in the absence cholesterol (48, 62), while in anotherstudy, it was found that cholesterol stimulated endocytosis ofcaveolae markers (54). Exogenous addition of glycosphingo-lipids was shown to dramatically induce caveolar endocytosis(54). However, the underlying mechanism, as well as the roleof other sphingolipids, in caveolar motility remains to beelucidated.
Caveolae have more sphingomyelin (SPM) and other sph-ingolipids than the bulk plasma membrane (44). Caveolae aredefined by the presence of caveolin-1 (Cav-1), a main buildingcomponent of the caveolar wall (58, 63). Another marker ofcaveolae is plasmalemmal vesicle-associated protein-1 (PV-1).PV-1 is an integral membrane-associated protein, and its ex-pression is associated with caveolae formation (8, 22, 57). It isalso considered a functional biomarker for altered vascularpermeability following disruption of the blood-brain barrier(55). It has been shown that Na�-K�-ATPase is associatedwith caveolae (7, 49) and is involved in regulation ofblood-brain barrier function (1). Recent evidence suggeststhat cells contain two functionally different pools of Na�-K�-ATPase, an ion-pumping pool and a signaling pool (29).Moreover, the nonpumping Na�-K�-ATPase has beenshown to colocalize in caveolae and interacts directly withmultiple proteins, such as protein kinases, ion transporters,and structural proteins, to exert its nonpumping functions,including regulation of Cav-1 membrane trafficking (7).
The combination and the functional balance of transcellularand paracellular pathways govern the net transvascular trans-port of substances in the microcirculation. Since both processesare associated with filamentous actin formation, the two path-ways are interconnected. Therefore, it is difficult to differen-tiate the prevailing role of one pathway or the other duringvarious pathologies causing an increase in vascular permeabil-ity. Consequently, it is not clear which of these pathways hasa prevailing role during certain pathologies. Caveolae-medi-ated transcytosis and paracellular transport may be intercon-nected (5), which makes differentiation of their function diffi-cult. We developed a dual-tracer probing method to differen-
* N. Muradashvili and S. J. Khundmiri contributed equally to this work.Address for reprint requests and other correspondence: D. Lominadze, Dept.
of Physiology & Biophysics, School of Medicine, Univ. of Louisville, Bldg.A, Rm. 1115, 500 South Preston St., Louisville, KY 40202 (e-mail:[email protected]).
Am J Physiol Cell Physiol 307: C169–C179, 2014.First published May 14, 2014; doi:10.1152/ajpcell.00305.2013.
tiate the relative involvement of transcellular and paracellulartransport pathways in protein crossing the EC layer and vas-cular wall under pathological conditions (43). To test forinflammation-induced changes in transcellular and paracellulartransport, we used hyperfibrinogenemia (HFg), an elevatedblood level of fibrinogen (Fg), as the experimental pathology.Elevated blood content of Fg is a biomarker of inflammationand a risk factor for many cardiovascular (14) and cerebrovas-cular disorders (16). While normal blood concentration of Fg is�2 mg/ml (32), it reaches 4 mg/ml during various inflamma-tory diseases (33). HFg accompanies many inflammatory dis-eases, such as hypertension (28, 33), diabetes (25), stroke (13,16, 17), and traumatic brain injury (9, 45, 60). It has beenshown that enhanced blood Fg content increases permeabilityof brain pial venules, involving transcellular and paracellulartransport but mainly affecting transcytosis (43).
In the present study, we hypothesized that, at elevated levelsof Fg, protein crossing of pial venules increases mainly viacaveolar transcytosis. To address this hypothesis, we used adual-tracer probing method (43) that allows differentiation ofparacellular and transcellular transport of proteins and evalu-ated the role of Fg in formation of functional caveolae. Todefine the mechanism of Fg-induced caveolar transcytosis, wetested if chelation of cholesterol by methyl-�-cyclodextrin(M�CD) or inhibition of sphingolipid de novo synthesis bymyriocin can alter Fg-induced permeability of cultured mousebrain ECs (MBECs) and pial venular macromolecular leakage.The effect of increased Fg content on levels of sphingolipidssuch as SPM, ceramide (Cer), sphingosine-1-phosphate (S1P),dihydrosphingosine, dihydro-S1P, sphingosin, and glucosylce-ramide (GlcCer) in MBECs and in mouse plasma samples wasassessed using a liquid chromatography-tandem mass spec-trometry (LC-MS/MS)-based lipidomics method. The resultsof the study confirm our hypothesis and suggest that elevatedlevels of Fg increase EC layer permeability mainly by affectinglevels of sphingolipids and, to a lesser extent, cholesterol.Moreover, our results suggest that Fg-generated sphingolipidmetabolites stimulate the formation of functional caveolae and,thus, exacerbate the Fg-mediated transcytosis.
METHODS
Reagents and antibodies. Human Fg (FIB-3, depleted of plasmin-ogen, von Willebrand factor, and fibronectin) was purchased fromEnzyme Research Laboratories (South Bend, IN); Alexa Fluor647-conjugated BSA (BSA-647) from Invitrogen (Carlsbad, CA);Lucifer yellow (LY), M�CD, and myriocin from Sigma Aldrich(St. Louis, MO); and Cer, GlcCer, and SPM from Cayman Chem-ical (Ann Arbor, MI).
Animals. In accordance with National Institutes of Health Guide-lines for animal research, all animal procedures were reviewed andapproved by the Institutional Animal Care and Use Committee of theUniversity of Louisville.
Black C57BL/B6 wild-type (WT) mice were obtained from Jack-son Laboratories. HFg transgenic mice were purchased from theMutant Mouse Regional Resource Center at the University of NorthCarolina at Chapel Hill. For genotyping of HFg mice, DNA wasextracted from the tail tip of mice and amplified by PCR using specificprimer sequences according to the protocol described previously (43).
EC culture. MBECs (American Type Culture Collection, Manas-sas, VA) were cultured in complete DMEM according to the recom-mendations of the American Type Culture Collection at 37°C with 5%
CO2-air in a humidified environment and used for the experiments atpassage 5 or 6.
EC layer permeability by the dual-tracer probing method. Incultured MBECs, the roles of transcellular and paracellular transportduring HFg were studied using a dual-tracer probing method (43). Themethod is based on comparison of transport of a low-molecular-weight molecule (i.e., LY) with that of a high-molecular-weightmolecule (i.e., BSA-647) through the cell layer (43). Since low-molecular-weight molecules leak mainly via paracellular transport(between EC junctions) and high-molecular-weight molecules movethrough paracellular (only when gaps between the cells are wideenough) and transcellular (caveolar transcytosis) pathways, the dif-ference in transport rates of these molecules would indicate theprevailing transport pathway involved in overall protein crossing of avascular wall (43).
MBECs were grown on Transwell permeable supports (Corning,Corning, NY) with polycarbonate membranes (Nuclepore Track-Etched membrane, 6.5 mm diameter, 0.4-�m pore size, 108/cm2 poredensity) coated with fibronectin until they formed a complete mono-layer, as described previously (43). Cell confluence and the presenceof an intact monolayer on the membranes were confirmed in eachseries of experiments (47, 64). Hirudin (0.1 U/ml) was added to eachwell in all experiments to inhibit possible effects of thrombin. Cellswere washed with PBS and treated with 4 mg/ml Fg, 4 mg/ml Fg �100 �M M�CD, 100 �M M�CD, or the same volume of PBS as Fgin medium (control) in the presence of LY (0.3 mg/ml) and BSA-647(0.2 mg/ml). In another set of experiments, besides treatment with 4mg/ml Fg and PBS in medium (control), cells were treated with 4mg/ml Fg � 500 nM myriocin and with myriocin alone. Mediasamples (50 �l) were collected from lower chambers of the Transwellsystem after 20, 40, 60, and 120 min and replaced with the samevolume of the sample added to each respective upper well. Fluores-cence intensity of each dye was measured by a microplate reader(SpectraMax M2e, Molecular Devices, Sunnyvale, CA) with excita-tion at 488 nm and emission at 520 nm for LY and excitation at 650nm and emission at 668 nm for BSA-647. Thus we tested a concen-tration gradient-induced movement of tracers through the EC layer,which can be defined by Fick’s law: J � �PS(Co � Ci), where P ismembrane and endothelium permeability to the substance, S is per-meable surface area, Ci is concentration of substance in the upperchamber, and Co is concentration of substance in the lower chamber.The variables P, S, and Ci did not change in our experiments.Therefore, changes in EC layer permeability are presented as a ratioof fluorescence intensity of each dye in the lower chamber to fluo-rescence intensity of the respective dye in the original sample at theend of the experiment.
Cerebrovascular permeability. Twelve-week-old HFg mice (26–29 gbody wt) were treated with myriocin (0.5 mg/kg body wt ip) every 12h for 3 days. The same volume of PBS was used for the control group.Mice were anesthetized with pentobarbital sodium (70 mg/kg ip). Aheating pad was used to maintain body temperature of the mice at37 � 1°C. Mean arterial blood pressure and heart rate were monitoredthrough a carotid artery cannula connected to a transducer and a bloodpressure analyzer (CyQ 103/302, Cybersense, Lexington, KY). Cra-nial windows were prepared and changes in pial venular permeabilitywere observed as described previously (41). After surgical prepara-tion, following a 1-h equilibration period, a mixture of 100 �l ofFITC-BSA (300 �g/ml) and 20 �l of myriocin or PBS in therespective control group was infused through the carotid artery can-nula and allowed to circulate for 10 min (34, 41). Brain pial circula-tion was observed with a microscope (model BXG61WI, Olympus,Tokyo, Japan) equipped with a 10/0.40 (UPlanSApo, Olympus)objective. After the baseline reading was obtained, images of theselected venular segments were recorded at 10, 20, 40, 60, and 120min.
An epi-illumination system was used to observe intravascular andextravascular FITC-BSA. The area of interest was exposed to blue
C170 SPHINGOLIPIDS IN Fg-INDUCED CEREBROVASCULAR PERMEABILITY
light (488 nm) for 10–15 s with a power density of 3.5 �W/cm2. Themicroscope images were acquired by an electron-multiplying charge-coupled device camera (Quantem 512SC, Photometrics, Tucson, AZ)and image acquisition system (SlideBook 5.0, Intelligent ImagingInnovations, Philadelphia, PA). The lamp power and camera gainsettings were held constant during the experiments. Images of the pialvenular circulation were analyzed by image analysis software (Image-Pro Plus 7.0, Media Cybernetics, Bethesda, MD). In each image, a30-�m-long line profile probe was positioned in the interstitiumadjacent to the venule (parallel to the vessel) and then in the middleof the venule, parallel to the bloodstream. Leakage of FITC-BSA wasassessed by changes in the ratio of fluorescence intensity in theinterstitium to that inside the vessel. The results were averaged and arepresented as percentage of baseline.
Fg effect on level of sphingolipids in vivo and in vitro. MBECsgrown in 100-mm-diameter petri dishes were treated in triplicateswith 2 mg/ml Fg, 4 mg/ml Fg, or PBS in phenol-red free medium(control) for 2 or 24 h. The treatment medium was collected and keptat �80°C until analysis. Also, to define plasma content of sphingo-lipids, blood samples were collected from anesthetized WT and HFganimals, and the plasma samples were kept at �80°C until analysis.
Content of sphingolipids in cell media and plasma samples wasassessed by LC-MS/MS, as described elsewhere (6). Briefly, sampleswere fortified with 5 ng of a mixture of internal standards: C17 baseD-erythrosphingosine, C17 base D-erythrosphinganine, C17 base D-ribophytosphingosine, C17-Cer, C17-S1P, and C17-SPM. Sampleswere extracted into an organic solvent system [85:15 (vol/vol) ethylacetate-isopropanol], evaporated, and reconstituted in 1 mM ammo-nium formate and 0.2% formic acid in methanol. Extracted lipids wereresolved by a reverse-phase HPLC using C8 columns (100 2.1 mm,2.4 mm; BDS Hypersil, Thermo Scientific). The HPLC eluent wasdirectly introduced to a mass spectrometer (QTRAP 5500, ABSCIEX, Framingham, MA) equipped with an electrospray ion source(ESI) that is used for ESI-MS/MS. The ESI-MS/MS test was carriedout in the positive-ion mode with ESI. Chromatographic data wereanalyzed by MultiQuant (AB SCIEX) to integrate the chromatogramsfor each multiple reaction-monitoring and/or selected reaction-moni-toring transition.
Na�-K�-ATPase activity. MBECs were treated with Fg (2 or 4mg/ml) or PBS in medium (control) and then washed with PBS.Protease inhibitor cocktail in PBS was added to the cells on petridishes, which were kept on ice, and the cells were scraped andcollected. The cells were counted, suspended in PBS-protease inhib-itor cocktail at 106 cells/ml, and the sonicated five times, for 10 s each,on ice. The samples were centrifuged at 1,000 g for 10 min, thesediment was discarded, and the supernatant was centrifuged at100,000 g for 30 min at 4°C. The cell membrane fraction wassuspended in 200 �l of PBS-protease inhibitor cocktail and frozen inliquid nitrogen for further analysis.
Na�-K�-ATPase activity was measured as ouabain-sensitive ATPhydrolysis, as described previously (15). Briefly, Na�-K�-ATPaseactivity was measured by addition of membrane material isolatedfrom MBECs to a buffer containing 100 mM NaCl, 5 mM KCl, 40mM histidine, 3 mM MgCl2, 1 mM EGTA, and 11 mg/ml alamethicin.The reaction was started by addition of 5 mM MgATP and carried outin the presence or absence of 1 mM ouabain (Na�-K�-ATPaseactivity inhibitor). The amount of phosphate released was measuredby a colorimetric procedure described by Taussky and Shorr (61).Ouabain-inhibited ATPase activity was expressed as nmol Pi
released·min�1·mg protein�1.Plasmids and cell transfection. mCherry-tagged rat Na�-K�-
ATPase 1-subunit was provided by Dr. Thomas A. Pressley (TexasTech University). Yellow fluorescence protein-tagged Cav-1 wasprovided by Dr. Zijian Xie (University of Toledo). Green fluorescenceprotein (GFP)-labeled PV-1 (a marker for caveolae) was purchasedfrom OriGene Technologies (Rockville, MD). MBECs were trans-
fected with the indicated plasmids using Lipofectamine 2000 (Invit-rogen) in Opti-MEM, as described previously (24).
Sensitized Förster resonance energy transfer. Förster resonanceenergy transfer (FRET) imaging experiments were performed inliving cells. MBECs were transfected with GFP-PV-1 (donor) andmCherry-Na�-K�-ATPase 1-subunit (acceptor) for 24 h and thenviewed and analyzed by an Olympus microscope with FRET and totalinternal reflection fluorescence (TIRF) capabilities before and aftertreatment with 4 mg/ml Fg. FRET image acquisition and analysiswere performed as described previously (12) using SlideBook 4.2software (Olympus, Center Valley, PA), which is based on thethree-filter “micro-FRET” image subtraction method described byJiang and Sorkin (23). Briefly, three images (100- or 250-ms exposuresets, 2 2 binning) were obtained: a mCherry excitation/mCherryemission image, a GFP excitation/GFP emission image, and a GFPexcitation/mCherry emission image (raw, uncorrected FRET). Afterthis initial imaging, background images were obtained. BackgroundmCherry and GFP images were fractionally subtracted from rawFRET images based on measurements for GFP bleed-through (0.50–0.56) and mCherry cross-excitation (0.015–0.02). This fractionalsubtraction generates corrected FRET images. The corrected FRETimages are represented in pseudocolor (gated to mCherry acceptorlevels) showing sensitized FRET within cells. Pearson’s coefficientsfor the subtraction were rounded up from average cross-bleed valuesdetermined in cells expressing GFP- or mCherry-tagged constructsalone. Thus these coefficients result in underestimation of correctedFRET signals for true FRET partners but prevent false-positivedetection of FRET. Sensitized donor- or acceptor-normalized FRETwas calculated using SlideBook 4.2 software.
TIRF microscopy. Membrane TIRF microscopy was performed asdescribed by Blaine et al. (6a) with slight modifications. Briefly,MBECs were grown to 60% confluence in a dish with a collagen-coated coverslip bottom (no. 1.5, MatTek, Ashland, MA). Cells weretransfected with the indicated plasmids (see RESULTS). Samples wereobserved using an Olympus TIRF microscope equipped with a 60/1.45 numerical aperture objective under the control of SlideBook 4.2software. Laser excitation was derived from a multiline argon ionlaser (458/488/515 nm) run at the same current setting for all exper-iments. The power at the sample was controlled by a neutral densityfilter wheel. Excitation and emission wavelengths were selected usingfilter sets for mCherry and GFP. The laser was aligned according tothe manufacturer’s instructions to achieve TIRF illumination. Imageswere taken using a Hamamatsu camera operating with 2 2 binning.O2 was provided by the ambient air, which was supplemented with5% CO2 and warmed to 37°C in an environmental chamber surround-ing the specimen. Association between Na�-K�-ATPase and PV-1was calculated by Pearson’s coefficient using SlideBook 4.2 software.Caveolae were quantified using ImageJ software as the number ofGFP (PV-1) and mCherry (Na�-K�-ATPase) particles.
Data analysis. Values are means � SE. The experimental groupswere compared by one-way ANOVA with repeated measures. IfANOVA indicated a significant difference (P � 0.05), Tukey’smultiple comparison test was used to compare group means. Differ-ences were considered significant if P � 0.05.
RESULTS
Fg-induced transcellular vs. paracellular transport. Wetested a concentration gradient-induced movement of tracersthrough the endothelial layer. The solution levels in the upperand lower chambers were kept equal. The tracers’ concentra-tions in the upper chambers were maintained during the exper-iment. Therefore, the process of diffusion can be defined by asimplified Fick’s law of diffusion. It can easily be defined bychanges in tracers’ concentrations in the lower chambers rel-ative to those in the respective upper chambers. To present an
C171SPHINGOLIPIDS IN Fg-INDUCED CEREBROVASCULAR PERMEABILITY
example of actual permeability of the EC layer to the testtracers under control and treatment conditions, amounts oftracers in the lower chambers at 10, 20, 40, 60, and 120 minwere calculated and are presented in Table 1.
Fg-induced leakage of LY (a marker of paracellular trans-port) through the MBEC layer was greater than that throughMBECs treated with PBS in medium (control) at 20 and 40 min(Fig. 1A). However, it was no longer different from the controlat 60 and 120 min (Fig. 1A). On the other hand, EC layercrossing of BSA-647 induced by 4 mg/ml Fg was greater thanthat in the control group starting from 40 min and continued toincrease steadily over the observation period (Fig. 1B). M�CDameliorated the effect of high Fg content on leakage of bothdyes (Fig. 1, A and B). M�CD alone did not have a significanteffect on MBEC layer permeability (Fig. 1, A and B). Incontrast to the effect of M�CD, myriocin did not affect leakageof LY induced by 4 mg/ml Fg (Fig. 1C). However, it blockedFg-induced BSA leakage (Fig. 1D). Myriocin alone did notaffect permeability of MBECs (Fig. 1, C and D).
Cerebrovascular permeability. Body weight of animals usedin the study varied from 26 to 29 g. Mean arterial bloodpressure alterations were minimal (�3 � 0.5 mmHg) and pialvenular diameters (40 � 5 �m) did not change after FITC-BSAand its mixture with myriocin infusions. We found a robusttime-dependent increase in FITC-BSA leakage in HFg mice,which was ameliorated in the presence of myriocin (Fig. 2).
Effect of Fg on sphingolipid levels in vitro and in vivo.Myriocin is a highly selective inhibitor of serine palmitoyl-transferase (SPT) (27, 51), the first and rate-limiting enzyme inthe de novo sphingolipid biosynthesis pathway that condenses
serine and palmitoyl-CoA to produce 3-ketodihydrosphin-gosine (21, 30). Therefore, we examined the effect of Fg onlevels of sphingolipids. We have developed LC-MS/MS meth-ods that can simultaneously quantitate �40 species of sphin-golipids. MBECs were treated with or without Fg (2 or 4mg/ml) for 2 or 24 h, culture media were collected, andsphingolipid levels were measured by LC-MS/MS. While wedid not observe alterations in levels of S1P, dihydrosphin-gosine, or dihydro-S1P, we found that Fg dose dependentlyincreased levels of most sphingolipid species of Cer, SPM, andGlcCer in MBECs (Fig. 3). In addition, lipid profiling ofplasma samples from WT and HFg transgenic mice showedhigher levels of SPM, Cer, and GlcCer in HFg than WT mice(Fig. 4).
Fg-induced activation of Na�-K� ATPase. Activity of Na�-K�-ATPase in MBECs treated with 2 mg/ml Fg was notdifferent from that in cells treated with PBS in medium(control) (Fig. 5). However, 4 mg/ml Fg significantly increasedNa�-K�-ATPase activity (Fig. 5). These results indicate that ahigh Fg content can activate Na�-K�-ATPase in ECs.
FRET and TIRF. To determine if the increase in transcellulartransport described above is due to caveolae in the plasmamembrane, formation and motion of caveolae were measuredin the presence and absence of sphingolipids or Fg by FRETand TIRF microscopy. To identify caveolae, cells were trans-fected with GFP-labeled PV-1 and/or mCherry-labeled Na�-K�-ATPase 1-subunit. As shown in Fig. 6A, epifluorescenceshowed both proteins to be expressed in cells. To determine theassociation between the two proteins, FRET was performed inthe cells in the presence of PBS in medium (control), and thecells were imaged after 15 min of treatment with sphingolipidsor Fg. Figure 6B shows a basal association between PV-1 andNa�-K�-ATPase 1-subunit. The association significantly in-creased when the cells were treated with Cer and Fg, but notwith GlcCer or sphingomyelin. Figure 6C shows three-channelsensitized FRET after photobleaching. To determine if theassociation is in the plasma membranes, the same cells wereimaged by TIRF microscopy. In Fig. 6D, TIRF microscopyshows that Na�-K�-ATPase 1-subunit and PV-1 are associ-ated with each other and are localized in the plasma membrane.Treatment with Fg, Cer, and SPM increased the number ofcaveolae in the plasma membrane, as shown by the increase inthe number of mCherry and GFP particles in the plasmamembrane (Fig. 6, E and F). GlcCer had no effect on thenumber of caveolae in the plasma membrane. However, theassociation between Na�-K�-ATPase 1- az’s correlation didnot change with any of the treatments (data not shown). Similarresults were observed in cells transfected with mCherry-la-beled Na�-K�-ATPase 1-subunit and yellow fluorescenceprotein-labeled Cav-1, another marker for caveolae (data notshown). These results point to an exciting and novel singulartarget in Fg-induced caveolae formation.
DISCUSSION
We found that enhanced formation of functional caveolaeduring HFg was associated with increased content of sphingo-lipids, particularly SPM, Cer, and GlcCer. Inhibition of sphin-golipid synthesis ameliorated the Fg-induced increase in per-meability of ECs without affecting paracellular transport. Fur-ther studies indicated that the effect of Cer alone on formation
Table 1. Amounts of tracers in lower chambers of theTranswell system at 10, 20, 40, 60, and 120 min
Amount of Lucifer yellow and BSA-Alexa Fluor 647 that crossed the mousebrain endothelial cell layer treated with PBS (control), 4 mg/ml fibrinogen(Fg4), 4 mg/ml fibrinogen in the presence of methyl-��cyclodextrin (Fg4 �M�CD), or myriocin (Fg4 � myriocin), or M�CD or myriocin alone.
C172 SPHINGOLIPIDS IN Fg-INDUCED CEREBROVASCULAR PERMEABILITY
of functional caveolae in MBECs was similar to that of HFg.These data indicate that a high level of Fg increases EC layerpermeability mainly by affecting the transcellular transportpathway, which involves caveolar transcytosis. Mechanisti-cally, our results suggest that Fg activates the sphingolipidbiosynthetic pathway, and the derived sphingolipid metabolites
play a critical role in Fg-triggered caveolar transcytosis and ECpermeability.
Previously, we found that elevated levels of Fg enhanceformation of functional caveolae (40), leading to increasedpermeability of the EC layer (42, 64) and microvessels (41). Ithas been shown that Fg affects EC junction proteins, increasingthe cell layer permeability (42, 64). We also found that a highlevel of Fg increases cerebrovascular permeability mainly byaffecting the transcellular transport pathway (43). This in vivofinding was confirmed for MBECs (43). However, while themechanism of Fg-induced paracellular transport involvingchanges in junction proteins and formation of filamentous actinhas been shown (47, 64), the mechanism of the Fg effect oncaveolar transcytosis (40) was not clear.
Caveolae are flask-shaped, invaginated structures present atthe surface of various cell types, including ECs (57). As theirwalls are enriched with sphingolipids and cholesterol (50),caveolar endocytosis can be stimulated by addition of exoge-nous sphingolipids or cholesterol (46). SPM and other sphin-golipid levels are higher in the caveolae than the bulk plasmamembrane, and the density of lipids was found to be higher inthe caveolae than in the plasma membrane fraction from whichthe caveolae were isolated (44).
We used a dual-tracer probing method that allows differen-tiation between paracellular and transcellular transport path-ways (43). LY, a well-known marker of paracellular trans-port (18, 35, 36), was used as a low-molecular-weight tracer(43). The Stokes-Einstein radius of LY (0.5 nm) is signifi-cantly less than that of BSA (3.5 nm) (43). Therefore, innormal conditions, low-molecular-weight substances, such as
A B
0 20 40 60 80 100 120
0.0
0.2
0.4
0.6
0.8
††
†**
Time, minB
SA le
akag
e,re
lativ
e flu
ores
cenc
e
Control Fg4 Fg4 + MβCD MβCD
*
0 20 40 60 80 100 120
0.0
0.2
0.4
0.6
0.8
††
Control Fg4 Fg4 + MβCD MβCD
**
Time, min
Luci
fer y
ello
w le
akag
e,re
lativ
e flu
ores
cenc
e
0 20 40 60 80 100 120
0.0
0.2
0.4
0.6
0.8
Time, min
Luci
fer y
ello
w le
akag
e,re
lativ
e flu
ores
cenc
e
Control Fg4 Fg4 + Myriocin Myriocin
*
*
C D
0 20 40 60 80 100 120
0.0
0.2
0.4
0.6
0.8
††
†
Time, min
BSA
leak
age,
rela
tive
fluor
esce
nce
Control Fg4 Fg4 + Myriocin Myriocin
*
**
Fig. 1. Fibrinogen (Fg)-induced permeabilityof mouse brain endothelial cells (MBECs). Aand B: permeability of MBECs to Luciferyellow (LY) and BSA tagged with AlexaFluor 647 (BSA-647) in the presence of PBSin medium (control), 4 mg/ml fibrinogen(Fg4), 4 mg/ml Fg � 100 �M methyl-�-cyclodextrin (Fg4 � M�CD), or 100 �MM�CD. C and D: permeability of MBECs toLY and BSA-647 in the presence of PBS inmedium (control), 4 mg/ml Fg (Fg4), 4 mg/mlFg � 500 nM myriocin (Fg4 � myriocin), or500 nM myriocin. Fluorescence intensity ofeach dye in samples collected from lowerchambers of the Transwell system after 20, 40,60, and 120 min was measured by a microplatereader (488-nm excitation and 520-nm emis-sion for LY; 650-nm excitation and 668-nmemission for BSA-647). Results are expressedas ratio of fluorescence intensity of each dye inthe lower chamber to fluorescence intensity ofthe respective dye in the original sample at theend of the experiment. Values are means �SE; n � 4. *P � 0.05 vs. control. †P � 0.05vs. Fg4 � M�CD or Fg4 � myriocin.
0
100
200
300
****
5 10 20 40 60
Fluo
resc
ence
inte
nsity
,%
of b
asel
ine
Time, min
- HFg + PBS - HFg + Myriocin
*
Fig. 2. Cerebrovascular permeability to macromolecules in hyperfibrinogenic(HFg) mice. Pial venular permeability to FITC-BSA was assessed in HFg micetreated with myriocin (0.5 mg·kg�1·day�1) or PBS for 3 days. Fluorescenceintensity changes in an area of interest adjacent to the venular segment weremeasured as described in METHODS. Venular permeability was assessed bychanges in the ratio of fluorescence intensity measured in the interstitiumadjacent to the venule to that inside the vessel. Values (means � SE) are shownas percent change in fluorescence compared with PBS alone (control); n � 4.*P � 0.05 vs. HFg � PBS.
C173SPHINGOLIPIDS IN Fg-INDUCED CEREBROVASCULAR PERMEABILITY
LY, traverse the EC layer via mainly cell junctions, whilehigh-molecular-weight substances, such as albumin, hardlypass cell junction proteins and leak between the cells. In thecase of an increased level of Fg, which affects EC junctionsand caveolae formation (40, 47), LY should leak mainlythrough the cell junctions (31), while BSA should leak throughthe cell junctions when they open wide enough and via thetranscellular pathway when caveolae become functional. It has
long been known that albumin crosses the endothelial barriervia vesicular transcytosis (37, 52). However, when EC junc-tional gaps open wider in response to various inflammatorystimuli (i.e., Fg), albumin can easily move between the cells. Ithas been shown that activation of ICAM-1 on the EC surfaceleads to an increase in albumin transport via the paracellulartransport pathway (59). We have shown that an increased levelof Fg activates ICAM-1 on the EC surface (53), suggesting
Ceramides Sphingomyelins
- Control- 2 mg/ml Fg- 4 mg/ml Fg
*
**
** *
*
*
**
*
*
*
* *
**
**
*
***†
*†
*† *
†
*†
*†
*†
Sphi
ngol
ipid
s(n
g/m
l)
A
Ceramides Sphingomyelins
- Control- 2 mg/ml Fg- 4 mg/ml Fg
**
*
*
**
**
*
*
**
*
*
*†
*†
Sphi
ngol
ipid
s(n
g/m
l)
B
Fig. 3. Fg-induced increase in sphingolipid syn-thesis in MBECs. MBECs were treated withPBS in medium (control) or 2 or 4 mg/ml Fg for2 h (A) or 24 h (B). Levels of sphingolipids wereassessed in cell culture medium by LC-MS/MS.Fg markedly increased levels of ceramide andsphingomyelin species in MBECs. Sphingolip-ids with different length and degree of saturationof fatty acids were measured. C18:2, fatty acidwith 18 carbons and 2 double bonds; DH, di-hydro; Sph, sphingosine; S1P, sphingosine-1-phosphate. Values are means � SE; n � 3. *P �0.05 vs. control. †P � 0.05 vs. 2 mg/ml Fg.
C174 SPHINGOLIPIDS IN Fg-INDUCED CEREBROVASCULAR PERMEABILITY
that, during HFg, albumin can traverse the EC layer viatranscellular and paracellular transport pathways.
At elevated levels, Fg induced robust leakage of LY at thebeginning of the observation period compared with cellstreated with PBS. However, this leakage was no longer differ-ent between the cell groups during the last hour of observation,suggesting that HFg-mediated gap opening in MBECs is notdifferent from that in control cells after �1 h. In contrast,albumin traversing the MBEC layer was greater throughout theperiod of observation in the presence of a high level of Fg thanin the control group. These results suggest that while junctionalcanals between the cells were closing after their initial robustopening, protein (albumin) crossing of the cell layer was stillgreater in the presence of a high Fg content. Therefore, albu-min was traversing the cell layer through paracellular andtranscellular pathways, but mainly via the transcellular trans-port pathway.
Na�-K�-ATPase utilizes �30% of the total cellular energyto maintain intracellular ionic concentration, which allowstransport of various ions, glucose, and amino acids against theirconcentration gradient. An increase in Na�-K�-ATPase activ-ity would strongly suggest an increase in overall transportactivity in cells with a high Fg content. Thus our functionalstudy, in agreement with previous work, showed that increasedFg concentration mainly induces transcellular transport whiletransiently activating paracellular transport (43).
Transient opening of EC junctional gaps has been observedduring many neuropathological disorders (e.g., central nervoussystem inflammation, Alzheimer’s disease, Parkinson’s dis-ease, and epilepsy) (56). Transcellular transport during HFgmay depend on how long the blood level of Fg is elevated. Ithas been shown that, after a moderate inflammatory stimulus,the blood level of Fg remains elevated for �15 days (19). Afterinflammation subsides, this prolonged elevation of Fg contentcan cause a lasting effect on EC activation and possibleformation of functional caveolae (40). The present data con-firm this notion. Moreover, formation of functional caveolaecan result in caveolar transcytosis of proteins, as found in ourprevious studies (41–43).
To differentiate the roles of cholesterol and sphingolipids inFg-induced caveolar transcytosis, we tested the effects of the
cholesterol chelator MbCD and the sphingolipid de novo syn-thesis inhibitor myriocin on permeability of MBECs to twotracers. Fg-induced BSA crossing of the MBEC layer wasinhibited by cholesterol chelation. While M�CD did not havean effect on LY leakage at later times, it blocked Fg-inducedenhanced leakage of LY at 20 and 40 min, when cell junctionalgaps were more open. This suggested its predominant effect onparacellular transport. On the other hand, myriocin did not alterFg-induced leakage of LY, but it blocked the effect of Fg onBSA crossing of the cell layer. These results suggest thatinhibition of sphingolipid synthesis has a lesser effect on celljunction openings but a greater effect on Fg-induced functionalcaveolae formation and the resultant caveolar transcytosis.This effect of sphingolipid synthesis inhibition was confirmedin vivo, when cerebrovascular permeability to BSA in HFgmice was ameliorated in the presence of myriocin. Thus ourdata indicate that while cholesterol can affect caveolar trans-cytosis and EC junctional gap openings, sphingolipids canaffect formation of functional caveolae and their motility.Recent evidence suggests that sphingolipids and their metab-olites have important roles in signal transduction (3, 4, 26, 65).These reports, together with our present data, suggest thatsphingolipids can have a signaling role in Fg-induced micro-vascular permeability.
The de novo sphingolipid biosynthesis pathway is initiatedby SPT, which synthesizes 3-ketosphinganine from L-serineand palmitoyl CoA (21, 30). Subsequently, 3-ketosphiganine isconverted to Cer, SPM, GlcCer, and other sphingolipids, viareactions catalyzed by a series of enzymes, including ceramidesynthase. Since it was suggested that local production of Cercan increase vascular permeability leading to tissue edema (20,38), we tested if inhibition of de novo sphingolipid synthesisby the SPT inhibitor myriocin (39) can affect MBEC layerpermeability. Myriocin inhibited Fg-induced BSA leakage butdid not affect leakage of LY. This exciting observation sug-gests that Fg activates the de novo sphingolipid biosyntheticpathway, and metabolites of the sphingolipid biosyntheticpathway play a critical role in caveolae-involved transcytosisand affect paracellular transport to a lesser degree. Theseresults were confirmed in vivo: treatment with myriocin de-creased cerebrovascular leakage in HFg mice. In combination,these data suggest that Fg activates the de novo sphingolipid
1.0
1.1
1.2
1.3
1.4Li
pid
conc
entr
atio
n,fo
ld in
crea
se o
ver t
he W
T
*
*
*
- SPM - Cer - GlcCer
Fig. 4. Comparison of sphingomyelin (SPM), ceramide (Cer), and glucosyl-ceramide (GlcCer) content in plasma of hyperfibrinogenic (HFg) and wild-type(WT) mice. Plasma concentration of SPM, Cer, and GlcCer sphingolipids wasmeasured by LC-MS/MS. Content of these lipids was higher in HFg than WTmice. Values are means � SE; n � 4. *P � 0.05 vs. WT.
0.0
0.5
1.0
1.5
2.0
2.5
Control Fg 2mg/ml Fg 4mg/ml
*
Oua
bain
-Sen
sitiv
e AT
P hy
drol
yzed
(nM
/min
/mg)
Fig. 5. Fg-induced Na�-K�-ATPase activation in MBECs. Ouabain-sensitiveNa�-K�-ATPase activity was measured as an indicator of active transportacross the membrane. Cells treated with PBS in medium were used as a controlgroup. Values are means � SE; n � 3. *P � 0.05 vs. control.
C175SPHINGOLIPIDS IN Fg-INDUCED CEREBROVASCULAR PERMEABILITY
Fig. 6. Effect of Cer, GlcCer, SPM, and Fg on formation of functional caveolae in MBECs. Formation of caveolae was determined by Förster resonance energytransfer (FRET) and total internal reflection fluorescence (TIRF) microscopy. Cells were transfected with green fluorescence protein (GFP)-labeled plasmalemmalvesicle-associated protein-1 (PV-1) and/or mCherry-labeled Na�-K�-ATPase, and live cells were imaged as described in METHODS. A, B, and D: representativeimages from 3 individual experiments for epifluorescence (A), FRET (B), and TIRF (D). Merged images in B are in pseudocolor (gated to mCherry acceptorlevels). Color scale shows reference spectrum: blue indicates no association (FRET and TIRF), and red indicates association (FRET and TIRF). C: data for3-channel FRET efficiency after photobleaching from 3 individual experiments. In each experiment, FRET efficiency from 30–50 cells was calculated, averaged,and considered as 1 experimental value. E and F: expression of Na�-K�-ATPase and caveolin-1 in the plasma membrane was determined by TIRF, and thenumber of caveolae was counted as individual GFP (E) and mCherry (F) particles using ImageJ software. In each experiment, values from 30–50 cells wereaveraged and considered as 1 experimental value. Values are means � SE; n � 3. *P � 0.05 vs. control.
C176 SPHINGOLIPIDS IN Fg-INDUCED CEREBROVASCULAR PERMEABILITY
biosynthetic pathway and that metabolites of the de novosphingolipid biosynthetic pathway play critical roles in medi-ating Fg-triggered cerebrovascular leakage.
To define the role of a specific metabolite of the de novosphingolipid biosynthetic pathway, we treated cells with Fgand measured levels of sphinoglipids by LC-MS/MS. Fg dosedependently increased the levels of SPM, Cer, and GlcCer inMBECs, while some other sphingolipids were unaffected.Moreover, the content of the same three sphingolipid specieswas elevated in blood plasma of mice treated with a high doseof Fg. Collectively, our results suggest a novel autocrinemechanism attributed to Fg-induced endothelial leakage: ele-vated Fg content activates the de novo sphingolipid biosyn-thetic pathway, leading to the local increased generation ofsphingolipids (particularly Cer and SPM), which results inenhanced caveolae formation and, thus, increased cerebrovas-cular permeability.
The role of SPM, Cer, and GlcCer in formation of functionalcaveolae, the caveolae that can move inside the cell and can beinvolved in caveolar trafficking, was further tested in MBECs.FRET and TIRF microscopy demonstrated that Fg and Cerincreased the number of caveolae in the plasma membrane, asshown by the increase in the number of GFP and mCherryparticles in the plasma membrane and the increase in three-channel FRET efficiency after photobleaching. Unlike Fg andCer, SPM and GlcCer had no effect on three-channel FRETefficiency. However, the number of caveolae in the plasmamembrane was increased to a lesser extent by SPM and GlcCerthan by Fg and Cer. These data would indicate that, similar toFg, Cer has a greater effect on formation of caveolae than doesSPM or GlcCer and suggest that HFg activates synthesis ofSPM, Cer, and GlcCer. Although Cer has a prevailing role incaveolar trafficking, the other two sphingolipids are also in-volved in Fg-induced formation of functional caveolae. Thiscan be concluded on the basis of the similarity of the effect ofFg and the sum of effects of the other three sphingolipidsrelative to the control group. Thus our data indicate thatFg-induced formation of functional caveolae, which results inincreased caveolar transcytosis, involves enhanced synthesis ofSPM, Cer, and GlcCer, which subsequently results in forma-tion of functional caveolae contributing to transcellular trans-port of blood proteins. Enhanced accumulation of blood pro-teins in the interstitium can lead to edema formation, which hasdevastating effects in the brain, which is enclosed in a re-stricted space. In addition, enhanced deposition of Fg in theinterstitium can contribute to formation of amyloid plaques andresult in irreversible loss of short-term memory, as in Alzhei-mer’s disease (2, 10, 11).
Thus our study shows that, at an elevated blood level, Fgenhances EC layer permeability to proteins mainly by enhanc-ing caveolar transcytosis. Fg-induced formation of functionalcaveolae can be modulated by sphingolipids and, particularly,by Cer. It is most likely that an increased Fg level enhancessynthesis of Cer, which increases formation and functionalityof caveolae. These effects inevitably lead to an increasedpermeability of the EC layer and the resultant enhanced mi-crovascular permeability. Increased permeability of cerebralmicrovessels to proteins would cause significant destructiveeffects leading to various vasculoneuronal disorders. There-fore, our data suggest that many cerebrovascular problemsleading to inflammation and the resultant increase in blood
level of Fg may be counterbalanced by inhibition of sphingo-lipid synthesis.
GRANTS
This work was supported in part by National Institutes of Health GrantsAG-047474 (to S. J. Khundmiri), HL-071071 (to M.-J. Lee), and NS-084823(to D. Lominadze).
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the authors.
AUTHOR CONTRIBUTIONS
N.M., S.J.K., R.T., and A.G. performed the experiments; N.M., S.J.K., R.T.,A.G., and D.L. analyzed the data; N.M., S.J.K., A.G., W.L.D., M.-J.L., andD.L. interpreted the results of the experiments; N.M. and S.J.K. prepared thefigures; N.M. and S.J.K. drafted the manuscript; W.L.D., M.-J.L., and D.L. areresponsible for conception and design of the research; W.L.D., M.-J.L., andD.L. edited and revised the manuscript; D.L. approved the final version of themanuscript.
REFERENCES
1. Abbruscato TJ, Lopez SP, Roder K, Paulson JR. Regulation of blood-brain barrier Na,K,2Cl-cotransporter through phosphorylation during invitro stroke conditions and nicotine exposure. J Pharmacol Exp Ther 310:459–468, 2004.
2. Ahn HJ, Zamolodchikov D, Cortes-Canteli M, Norris EH, GlickmanJF, Strickland S. Alzheimer’s disease peptide �-amyloid interacts withfibrinogen and induces its oligomerization. Proc Natl Acad Sci USA 107:21812–21817, 2010.
3. Alvarez SE, Harikumar KB, Hait NC, Allegood J, Strub GM, KimEY, Maceyka M, Jiang H, Luo C, Kordula T, Milstien S, Spiegel S.Sphingosine-1-phosphate is a missing cofactor for the E3 ubiquitin ligaseTRAF2. Nature 465: 1084–1088, 2010.
4. Ancellin N, Hla T. Differential pharmacological properties and signaltransduction of the sphingosine 1-phosphate receptors EDG-1, EDG-3,and EDG-5. J Biol Chem 274: 18997–19002, 1999.
5. Bauer PM, Yu J, Chen Y, Hickey R, Bernatchez PN, Looft-Wilson R,Huang Y, Giordano F, Stan RV, Sessa WC. Endothelial-specific ex-pression of caveolin-1 impairs microvascular permeability and angiogen-esis. Proc Natl Acad Sci USA 102: 204–209, 2005.
6. Bielawski J, Szulc ZM, Hannun YA, Bielawska A. Simultaneous quan-titative analysis of bioactive sphingolipids by high-performance liquidchromatography-tandem mass spectrometry. Methods 39: 82–91, 2006.
6a.Blaine J, Okamura K, Giral H, Breusegem S, Caldas Y, Millard A,Barry N, Levi M. PTH-induced internalization of apical membraneNaPi2a: role of actin and myosin VI. Am J Physiol Cell Physiol 297:C1339–C1346, 2009.
7. Cai T, Wang H, Chen Y, Liu L, Gunning WT, Quintas LE, Xie ZJ.Regulation of caveolin-1 membrane trafficking by the Na/K-ATPase. JCell Biol 182: 1153–1169, 2008.
8. Carson-Walter E, Hampton J, Shue E, Geynisman D, Pillai P, Satha-noori R, Madden S, Hamilton R, Walter K. Plasmalemmal vesicleassociated protein-1 is a novel marker implicated in brain tumor angio-genesis. Clin Cancer Res 11: 7643–7650, 2005.
9. Chhabra G, Rangarajan K, Subramanian A, Agrawal D, Sharma S,Mukhopadhayay A. Hypofibrinogenemia in isolated traumatic braininjury in Indian patients. Neurology India 58: 756–757, 2010.
10. Cortes-Canteli M, Paul J, Norris EH, Bronstein R, Ahn HJ, Zamo-lodchikov D, Bhuvanendran S, Fenz KM, Strickland S. Fibrinogen and�-amyloid association alters thrombosis and fibrinolysis: a possible con-tributing factor to Alzheimer’s disease. Neuron 66: 695–709, 2010.
11. Cortes-Canteli M, Strickland S. Fibrinogen, a possible key player inAlzheimer’s disease. J Thromb Haemost 7: 146–150, 2009.
12. Cummins TD, Mendenhall MD, Lowry MN, Korte E, Barati MT,Khundmiri SJ, Salyer SA, Klein JB, Powell DW. Elongin C is amediator of Notch4 activity in human renal tubule cells. Biochim BiophysActa 1814: 1748–1757, 2011.
13. D’Erasmo E, Acca M, Celi FS, Medici F, Palmerini T, Pisani D.Plasma fibrinogen and platelet count in stroke. J Med 24: 185–191, 1993.
14. Danesh J, Lewington S, Thompson SG, Lowe GD, Collins R, KostisJB, Wilson AC, Folsom AR, Wu K, Benderly M, Goldbourt U, Willeit
C177SPHINGOLIPIDS IN Fg-INDUCED CEREBROVASCULAR PERMEABILITY
J, Kiechl S, Yarnell JW, Sweetnam PM, Elwood PC, Cushman M,Psaty BM, Tracy RP, Tybjaerg-Hansen A, Haverkate F, de Maat MP,Fowkes FG, Lee AJ, Smith FB, Salomaa V, Harald K, Rasi R, VahteraE, Jousilahti P, Pekkanen J, D’Agostino R, Kannel WB, Wilson PW,Tofler G, Arocha-Pinango CL, Rodriguez-Larralde A, Nagy E, Mi-jares M, Espinosa R, Rodriquez-Roa E, Ryder E, Diez-Ewald MP,Campos G, Fernandez V, Torres E, Coll E, Marchioli R, Valagussa F,Rosengren A, Wilhelmsen L, Lappas G, Eriksson H, Cremer P, NagelD, Curb JD, Rodriguez B, Yano K, Salonen JT, Nyyssonen K,Tuomainen TP, Hedblad B, Lind P, Loewel H, Koenig W, Meade TW,Cooper JA, De Stavola B, Knottenbelt C, Miller GJ, Bauer KA,Rosenberg RD, Sato S, Kitamura A, Naito Y, Iso H, Rasi V, PalosuoT, Ducimetiere P, Amouyel P, Arveiler D, Evans AE, Ferrieres J,Juhan-Vague I, Bingham A, Schulte H, Assmann G, Cantin B, La-marche B, Despres JP, Dagenais GR, Tunstall-Pedoe H, WoodwardM, Ben Shlomo Y, Davey SG, Palmieri V, Yeh JL, Rudnicka A,Ridker P, Rodeghiero F, Tosetto A, Shepherd J, Ford I, Robertson M,Brunner E, Shipley M, Feskens EJ, Kromhout D, Dickinson A,Ireland B, Juzwinsin K, Kaptoge S, Lewington S, Memon A, SarwarN, Walker M, Wheeler J, White I, Wood, A. Plasma fibrinogen leveland the risk of major cardiovascular diseases and nonvascular mortality:an individual participant meta-analysis. JAMA 294: 1799–1809, 2005.
15. Dean W, Delamere N, Borchman D, Moseley A, Ahuja R. Studies onlipids and the activity of Na,K-ATPase in lens fibre cells. Biochem J 314:961–967, 1996.
16. del Zoppo GJ, Levy DE, Wasiewski WW, Pancioli AM, Demchuk AM,Trammel J, Demaerschalk BM, Kaste M, Albers GW, Ringelstein EB.Hyperfibrinogenemia and functional outcome from acute ischemic stroke.Stroke 40: 1687–1691, 2009.
17. Eidelman RS, Hennekens CH. Fibrinogen: a predictor of stroke andmarker of atherosclerosis. Eur Heart J 24: 499–500, 2003.
18. Gil ES, Li J, Xiao H, Lowe TL. Quaternary ammonium �-cyclodextrinnanoparticles for enhancing doxorubicin permeability across the in vitroblood-brain barrier. Biomacromolecules 10: 505–516, 2009.
19. Gitlin J, Colten H. Molecular biology of the acute phase plasma proteins.In: Lymphokines, edited by Pick E, Landy M. San Diego, CA: Academic,1987, p. 123–153.
20. Goggel R, Winoto-Morbach S, Vielhaber G, Imai Y, Lindner K,Brade L, Brade H, Ehlers S, Slutsky AS, Schutze S, Gulbins E, UhligS. PAF-mediated pulmonary edema: a new role for acid sphingomyelinaseand ceramide. Nat Med 10: 155–160, 2004.
21. Hanada K. Serine palmitoyltransferase, a key enzyme of sphingolipidmetabolism. Biochim Biophys Acta 1632: 16–30, 2003.
22. Hnasko R, McFarland M, Ben-Jonathan N. Distribution and character-ization of plasmalemma vesicle protein-1 in rat endocrine glands. JEndocrinol 175: 649–661, 2002.
23. Jiang X, Sorkin A. Coordinated traffic of Grb2 and Ras during epidermalgrowth factor receptor endocytosis visualized in living cells. Mol Biol Cell13: 1522–1535, 2002.
24. Khundmiri S, Amin V, Henson J, Lewis J, Ameen M, Rane M,Delamere N. Ouabain stimulates protein kinase B (Akt) phosphorylationin opossum kidney proximal tubule cells through an ERK-dependentpathway. Am J Physiol Cell Physiol 293: C1171–C1180, 2007.
25. Lee AJ, Lowe GD, Woodward M, Tunstall-Pedoe H. Fibrinogen inrelation to personal history of prevalent hypertension, diabetes, stroke,intermittent claudication, coronary heart disease, and family history: theScottish Heart Health Study. Br Heart J 69: 338–342, 2007.
26. Lee MJ, Van Brocklyn JR, Thangada S, Liu CH, Hand AR, Men-zeleev R, Spiegel S, Hla T. Sphingosine-1-phosphate as a ligand for theG protein-coupled receptor EDG-1. Science 279: 1552–1555, 1998.
27. Lee YS, Choi KM, Choi MH, Ji SY, Lee S, Sin DM, Oh KW, Lee YM,Hong JT, Yun YP, Yoo HS. Serine palmitoyltransferase inhibitor myri-ocin induces growth inhibition of B16F10 melanoma cells through G2/Mphase arrest. Cell Prolif 44: 320–329, 2011.
28. Letcher RL, Chien S, Pickering TG, Sealey JE, Laragh JH. Directrelationship between blood pressure and blood viscosity in normal andhypertensive subjects. Role of fibrinogen and concentration. Am J Med 70:1195–1202, 1981.
29. Liang M, Tian J, Liu L, Pierre S, Liu J, Shapiro J, Xie ZJ. Identifi-cation of a pool of non-pumping Na/K-ATPase. J Biol Chem 282:10585–10593, 2007.
30. Linn S, Kim H, Keane E, Andras L, Wang E, Merrill AJ. Regulationof de novo sphingolipid biosynthesis and the toxic consequences of itsdisruption. Biochem Soc Trans 29: 831–835, 2001.
31. Little TL, Xia J, Duling BR. Dye tracers define differential endothelialand smooth muscle coupling patterns within the arteriolar wall. Circ Res76: 498–504, 1995.
32. Lominadze D, Dean WL, Tyagi SC, Roberts AM. Mechanisms offibrinogen-induced microvascular dysfunction during cardiovascular dis-ease. Acta Physiol (Oxf) 198: 1–13, 2010.
33. Lominadze D, Joshua IG, Schuschke DA. Increased erythrocyte aggre-gation in spontaneously hypertensive rats. Am J Hypertens 11: 784–789,1998.
34. Lominadze D, Roberts AM, Tyagi N, Tyagi SC. Homocysteine causescerebrovascular leakage in mice. Am J Physiol Heart Circ Physiol 290:H1206–H1213, 2006.
35. Luissint AC, Federici C, Guillonneau F, Chretien F, Camoin L,Glacial F, Ganeshamoorthy K, Couraud PO. Guanine nucleotide-binding protein Gi2: a new partner of claudin-5 that regulates tightjunction integrity in human brain endothelial cells. J Cereb Blood FlowMetab 32: 860–873, 2012.
36. Markoutsa E, Papadia K, Clemente C, Flores O, Antimisiaris SG.Anti-A�-MAb and dually decorated nanoliposomes: effect of A�1–42peptides on interaction with hCMEC/D3 cells. Eur J Pharm Biopharm 81:49–56, 2012.
40. Muradashvili N, Benton R, Tyagi S, Lominadze D. Elevated level offibrinogen increases caveolae formation: role of matrix metalloprotei-nase-9. Cell Biochem Biophys 69: 283–294, 2014.
41. Muradashvili N, Qipshidze N, Munjal C, Givvimani S, Benton RL,Roberts AM, Tyagi SC, Lominadze D. Fibrinogen-induced increasedpial venular permeability in mice. J Cereb Blood Flow Metab 32: 150–163, 2012.
42. Muradashvili N, Tyagi N, Tyagi R, Munjal C, Lominadze D. Fibrin-ogen alters mouse brain endothelial cell layer integrity affecting vascularendothelial cadherin. Biochem Biophys Res Commun 413: 509–514, 2011.
43. Muradashvili N, Tyagi R, Lominadze D. A dual-tracer method fordifferentiating transendothelial transport from paracellular leakage in vivoand in vitro. Front Physiol 3: 166–172, 2012.
44. Örtegren U, Karlsson M, Blazic N, Blomqvist M, Nystrom FH,Gustavsson J, Fredman P, Strålfors P. Lipids and glycosphingolipids incaveolae and surrounding plasma membrane of primary rat adipocytes.Eur J Biochem 271: 2028–2036, 2004.
45. Pahatouridis D, Alexiou G, Zigouris A, Mihos E, Drosos D, VoulgarisS. Coagulopathy in moderate head injury. The role of early administrationof low molecular weight heparin. Brain Injury 24: 1189–1192, 2010.
46. Parton RG, Simons K. The multiple faces of caveolae. Nat Rev Mol CellBiol 8: 185–194, 2007.
47. Patibandla PK, Tyagi N, Dean WL, Tyagi SC, Roberts AM, Lomi-nadze D. Fibrinogen induces alterations of endothelial cell tight junctionproteins. J Cell Physiol 221: 195–203, 2009.
48. Pelkmans L, Bürli T, Zerial M, Helenius A. Caveolin-stabilized mem-brane domains as multifunctional transport and sorting devices in endo-cytic membrane traffic. Cell 118: 767–780, 2004.
49. Pierre S, Xie Z. The Na,K-ATPase receptor complex. Cell BiochemBiophys 46: 303–315, 2006.
50. Razani B, Woodman SE, Lisanti MP. Caveolae: from cell biology toanimal physiology. Pharmacol Rev 54: 431–467, 2002.
51. Roelants F, Breslow D, Muir A, Weissman J, Thorner J. Protein kinaseYpk1 phosphorylates regulatory proteins Orm1 and Orm2 to controlsphingolipid homeostasis in Saccharomyces cerevisiae. Proc Natl AcadSci USA 108: 19222–19227, 2011.
52. Schnitzer JE. gp60 is an albumin-binding glycoprotein expressed bycontinuous endothelium involved in albumin transcytosis. Am J PhysiolHeart Circ Physiol 262: H246–H254, 1992.
53. Sen U, Tyagi N, Patibandla PK, Dean WL, Tyagi SC, Roberts AM,Lominadze D. Fibrinogen-induced endothelin-1 production from endo-thelial cells. Am J Physiol Cell Physiol 296: C840–C847, 2009.
54. Sharma DK, Brown JC, Choudhury A, Peterson TE, Holicky E,Marks DL, Simari R, Parton RG, Pagano RE. Selective stimulation ofcaveolar endocytosis by glycosphingolipids and cholesterol. Mol Biol Cell15: 3114–3122, 2004.
C178 SPHINGOLIPIDS IN Fg-INDUCED CEREBROVASCULAR PERMEABILITY
55. Shue E, Carson-Walter E, Liu Y, Winans B, Ali Z, Chen J, Walter K.Plasmalemmal vesicle associated protein-1 (PV-1) is a marker of blood-brain barrier disruption in rodent models. BMC Neurosci 9: 29, 2008.
56. Stamatovic S, Keep R, Andjelkovic A. Brain endothelial cell-cell junc-tions: how to “open” the blood brain barrier. Curr Neuropharmacol 6:179–192, 2008.
57. Stan RV, Marion K, Palade GE. PV-1 is a component of the fenestraland stomatal diaphragms in fenestrated endothelia. Proc Natl Acad SciUSA 96: 13203–13207, 1999.
58. Stan RV. Structure of caveolae. Biochim Biophys Acta 1746: 334–348,2005.
59. Sumagin R, Lomakina E, Sarelius IH. Leukocyte-endothelial cell inter-actions are linked to vascular permeability via ICAM-1-mediated signal-ing. Am J Physiol Heart Circ Physiol 295: H969–H977, 2008.
60. Sun Y, Wang J, Wu X, Xi C, Gai Y, Liu H, Yuan Q, Wang E, Gao L,Hu J, Zhou L. Validating the incidence of coagulopathy and disseminated
intravascular coagulation in patients with traumatic brain injury—analysisof 242 cases. Br J Neurosurg 25: 363–368, 2011.
61. Taussky HH, Shorr E. A microcalorimetric method for the determinationof inorganic phosphorus. J Biol Chem 202: 675–685, 1953.
62. Thomsen P, Roepstorff K, Stahlhut M, van Deurs B. Caveolae arehighly immobile plasma membrane microdomains, which are not involvedin constitutive endocytic trafficking. Mol Biol Cell 13: 238–250, 2002.
63. Tse D, Stan RV. Morphological heterogeneity of endothelium. SeminThromb Hemost 36: 236–245, 2010.
64. Tyagi N, Roberts AM, Dean WL, Tyagi SC, Lominadze D. Fibrinogeninduces endothelial cell permeability. Mol Cell Biochem 307: 13–22,2008.
65. Windh RT, Lee MJ, Hla T, An S, Barr AJ, Manning DR. Differentialcoupling of the sphingosine 1-phosphate receptors Edg-1, Edg-3, andH218/Edg-5 to the Gi, Gq, and G12 families of heterotrimeric G proteins.J Biol Chem 274: 27351–27358, 1999.
C179SPHINGOLIPIDS IN Fg-INDUCED CEREBROVASCULAR PERMEABILITY
![[Ca2+ and sphingolipids as modulators for apoptosis and cancer]](https://static.documents.pub/doc/80x56/6357fc1fa08050b05c07a540/ca2-and-sphingolipids-as-modulators-for-apoptosis-and-cancer.jpg)
