Microvascular Research 97 (2015) 1–5
Contents lists available at ScienceDirect
Microvascular Research journal homepage: www.elsevier.com/locate/ymvre
Short Communication
Thrombomodulin regulation in human brain microvascular endothelial cells in vitro: Role of cytokines and shear stress Keith D. Rochfort, Philip M. Cummins ⁎ School of Biotechnology, Dublin City University, Glasnevin, Dublin 9, Ireland Centre for Preventive Medicine, Dublin City University, Glasnevin, Dublin 9, Ireland
a r t i c l e
i n f o
Article history: Accepted 13 September 2014 Available online 22 September 2014 Keywords: Blood–brain barrier Thrombomodulin Cytokines Shear stress Endothelium
a b s t r a c t Thrombomodulin (TM), an important determinant of blood vessel homeostasis, is expressed on the luminal vascular endothelial cell surface and is released into serum in response to circulatory signals. This includes the cerebrovascular endothelium, where the anti-coagulant and anti-inflammatory properties of TM are thought to be critical to the brain microcirculation and blood–brain barrier (BBB) integrity. Much is still unknown however about how circulatory stimuli may regulate TM activity within the brain microvasculature. To address this, the current short paper investigated the effects of opposing regulatory signals, namely cytokines (TNF-α, IL-6) and laminar shear stress, on the cellular levels and release of TM in cultured human brain microvascular endothelial cells (HBMvECs). Treatment of confluent HBMvECs with either TNF-α or IL-6 (100 ng/ml, 18 h) reduced TM protein levels by up to 70%, whilst inducing TM release into media by up to 4.4 and 5.5 fold, respectively. The effects of either cytokine (0–100 ng/ml) on TM protein levels (6 or 18 h) and release (0–18 h) were also found to be concentration- and time-dependent. Either cytokine (100 ng/ml, 24–72 h) also reduced TM mRNA levels by N 50%. When exposed to laminar shear stress for 24 h at 8 dyn/cm2 (SI unit equivalent = 0.8 Pa), TM protein levels were upregulated by 65% in parallel with a 2-fold increase in TM mRNA levels. Shear stress also proved to be a much more potent stimulus for TM release from HBMvECs, yielding media TM levels of 1000 pg/105 cells, when compared to 175 and 210 pg/105 cells for TNF-α and IL-6, respectively, after parallel 18 h treatments. Finally, shear-conditioned media was found to completely block thrombin-induced permeabilization of HBMvECs, confirming the functional efficacy of released TM. In summary, our data indicate that TM is differentially regulated within cultured HBMvECs by humoral and biomechanical signals. © 2014 Elsevier Inc. All rights reserved.
Introduction Thrombomodulin (TM), an integral membrane receptor constitutively expressed on the luminal surface of vascular endothelial cells, is an important determinant of blood vessel homeostasis (for review see Martin et al., 2013). Studies have confirmed the expression of TM within the cerebrovascular endothelium, where its anti-coagulant and antiinflammatory properties are thought to be critical to the brain microcirculation and blood–brain barrier (BBB) integrity (Tran et al., 1996; Wang et al., 1997). In this respect, soluble TM-based therapeutics have been used to treat BBB-associated neurological disorders such as stroke (Wenzel et al., 2014). Moreover, elevated TM release or “shedding” into serum invariably accompanies the cerebrovascular endothelial activation typically associated with neurodegenerative disorders (Festoff et al., 2012), stroke (Hassan et al., 2003), cerebral small vessel disease (Giwa et al., 2012) and traumatic brain injury (Yokota et al., 2002). Much is still unknown however about how common pathological and physiological stimuli may regulate TM activity within the brain ⁎ Corresponding author. Fax: +353 1 7005412. E-mail address: [email protected] (P.M. Cummins).
http://dx.doi.org/10.1016/j.mvr.2014.09.003 0026-2862/© 2014 Elsevier Inc. All rights reserved.
microvasculature. Proinflammatory cytokines for example, have been strongly linked to the pathological effects of stroke and other neurological diseases, imparting anti-barrier and pro-coagulant effects on the brain microvascular endothelium (Behling-Kelly et al., 2007; Tuttolomondo et al., 2008; Rochfort et al., 2014). The precise effects of cytokines, and particularly IL-6, on TM levels and release within the brain microvasculature however, are poorly documented. In contrast to proinflammatory cytokines, physiological laminar shear stress typically has a protective anti-inflammatory impact on the vascular endothelium (Traub and Berk, 1998). Moreover, the ability of shear stress to enhance BBB phenotype and to protect against cytokine-induced BBB injury has previously been reported (Krizanac-Bengez et al., 2006; Clark et al., 2011; Walsh et al., 2011). Whilst various studies have demonstrated the ability of shear stress to upregulate endothelial TM expression (for review see Martin et al., 2013), to our knowledge, there are no existing reports documenting the precise effects of shear stress on TM regulation within the brain microvasculature. To address these knowledge gaps, the current short paper investigated the effects of opposing regulatory influences, namely cytokines (TNF-α, IL-6) and laminar shear stress, on the cellular protein/mRNA levels and release of TM in cultured HBMvECs. Both time- and
2
K.D. Rochfort, P.M. Cummins / Microvascular Research 97 (2015) 1–5
concentration-/force-dependency studies were conducted, whilst the putative impact of TM release on thrombin-induced HBMvEC permeabilization was also investigated.
(250 bp): Forward 5′-CAGCCACCCGAGATTGAGCA-3′; Reverse 5′-TAGT AGCGACGGGCGGTGTG-3′. Enzyme-linked immunosorbent assay (ELISA)
Materials and methods Materials Unless otherwise stated, all reagents were purchased from Sigma-Aldrich (Dublin, IRL). Cytokines (TNF-α, IL-6) were purchased from Millipore (Cork, IRL). Primary antisera included mouse antithrombomodulin IgG ab6980 (Abcam, Cambridge, UK) and rabbit antiGAPDH IgG (Millipore). Secondary antisera included HRP-conjugated goat anti-mouse IgG and goat anti-rabbit IgG (Cell Signalling, MA, USA). Cell culture Culture of primary-derived HBMvECs has been described previously (Rochfort et al., 2014). HBMvECs, primary cultured from a single adult male donor post-mortem, were purchased from Cell Systems Corporation (WA, USA — Cat No. ACBRI 376). Cells were routinely grown in EndoGRO™ MV Basal Medium (Millipore) containing 5% foetal bovine serum, antibiotics (100 mg/ml Mycozap™), and relevant supplements. All cells (passages 5–12) were grown on tissue culture grade plasticware coated with Attachment Factor™ (Life Technologies, UK) and maintained in a humidified atmosphere of 5% CO2/95% air at 37 °C. For experimental purposes, cells were routinely subjected to treatment with either TNF-α or IL-6 (0–100 ng/ml, 0–72 h), or to laminar shear stress (0 or 8 dyn/cm2, 0–72 h), the latter employing an orbital rotation shear model (Walsh et al., 2011). Post-treatment, both cells and conditioned media were harvested for analysis. Total cell protein lysate preparation has been described previously (Rochfort et al., 2014). Conditioned media samples were routinely centrifuged at 700 ×g for 15 min to remove any cellular debris. Media samples were routinely assayed by ELISA prior to freezing. All samples were ultimately stored at −80 °C. Western immunoblotting Total cell protein lysates were resolved by 10% SDS-PAGE under reducing conditions and electroblotted as previously described (Rochfort et al., 2014). Membranes were blocked for 60 min in tris-buffered saline (TBS: 10 mM Tris pH 8.0, 150 mM NaCl) containing 5% w/v bovine serum albumin (BSA) before being incubated overnight in primary antisera with gentle agitation at 4 °C. Primary antisera were prepared in TBST (+1% BSA, +0.1% Tween-20): 1 μg/ml anti-thrombomodulin mouse monoclonal IgG and 0.2 μg/ml anti-GAPDH rabbit monoclonal IgG. Secondary antisera, applied to washed membranes for 3 h with gentle agitation at room temperature, were prepared in TBST: 1:2000 HRP-conjugated goat anti-mouse IgG (thrombomodulin) and 1:3000 HRP-conjugated goat anti-rabbit IgG (GAPDH). Membranes were developed using a Luminata™ Western HRP Kit (Millipore). Scanning densitometry of Western blots was routinely performed using NIH ImageJ software, with GAPDH routinely employed as a loading control to facilitate densitometric normalization of bands.
A thrombomodulin/BDCA-3 DuoSet® ELISA Kit (R&D Systems, MN, USA) was employed as per manufacturer instructions (with minor modifications) to measure absolute TM levels in HBMvEC conditioned media samples. Briefly, 96-well plates were coated with 50 μl/well of the capture antibody and incubated overnight. The plate was then blocked for 1 h by adding 150 μl of Reagent Diluent to each well. Media samples and human recombinant TM standards (31.25– 2000 pg/ml) were added in duplicate at 50 μl/well, with incubations proceeding for 2 h. 50 μl of detection antibody was then added to each well for a further 2 h, followed by 50 μl of streptavidin–HRP to each well for 20 min (in the dark). 50 μl of substrate solution was then added to each well for a further 20 min (in the dark). Finally, reactions were terminated with the addition of 25 μl of Stop Solution to each well and the plate luminescence subsequently read at both 570 nm and 450 nm (wave correction was used to subtract the readings at 570 nm from 450 nm to control for optical imperfections in the plate). Transendothelial permeability assay For analysis of HBMvEC monolayer permeability, the transwell method of Rochfort et al. (2014) was employed. Briefly, HBMvECs were plated at high density (5 × 105 cells/well) into Millicell hanging cell culture inserts placed within 6-well dishes (Millipore; 0.4 μm pore size, 24 mm filter diameter). Complete media was added to the upper (luminal, 2 ml) and lower (abluminal, 4 ml) chambers of the Millicell insert within the 6-well dish and the cells were allowed to grow to confluency. Confluent HBMvECs within the upper chamber were next incubated for 30 min with either unconditioned media (0 ng of TM) or shear-conditioned media (3 ng or 6 ng of TM — i.e. 1.5–3.0 ng/ml) in the absence and presence of 2 units of thrombin (i.e. 1 U/ml) (note: conditioned media was pre-concentrated by centrifugal filtration to contain either 3 ng or 6 ng of released thrombomodulin for co-incubations with thrombin). Post-treatment, media in the upper and lower chambers was replenished, fluorescein isothiocyanate (FITC)-labelled 40 kDa dextran was added to the upper chamber (giving a final concentration of 250 μg/ml), and transwell diffusion allowed to proceed. Media samples (28 μl) were collected from the lower chamber after 3 h, diluted to a final volume of 400 μl with complete media, and monitored in 96-well format for FITC–dextran fluorescence. A TECAN Safire 2 fluorospectrometer was used with excitation and emission wavelengths set at 490 and 520 nm, respectively. Permeability is presented as % transendothelial exchange of FITC–dextran 40 kDa (%TEE FD40). Statistical analysis Results are expressed as mean ± s.d. Experimental points were typically performed in triplicate with a minimum of three independent experiments (n = 3). Statistical comparisons between control and experimental groups were by ANOVA in conjunction with a Dunnett's post-hoc test for multiple comparisons. A Student's t-test was also routinely employed for pairwise comparisons. A value of P ≤ 0.05 was considered significant.
Quantitative real-time PCR Results Following experiments, endothelial cells were harvested for extraction of total RNA and analysis of TM mRNA expression as previously described (Guinan et al., 2013). Ribosomal subunit S18 was routinely used for normalization purposes. Primer pairs were screened for correct product size by 1% agarose gel electrophoresis and underwent meltcurve analysis for primer-dimers. TM (107 bp): Forward 5′-ACCTTCCT CAATGCCAGTCAG-3′; Reverse 5′-GCCGTCGCCGTTCAGTAG-3′; S18
Effect of proinflammatory cytokines on TM protein/mRNA levels and release in HBMvECs Treatment of confluent HBMvECs with either TNF-α or IL-6 reduced TM protein levels by up to 70% at 100 ng/ml cytokine for 18 h (Fig. 1A). Reductions in TM mRNA levels of 65% and 53% were also observed
K.D. Rochfort, P.M. Cummins / Microvascular Research 97 (2015) 1–5
following 24 h of treatment with TNF-α and IL-6, respectively, reductions which still persisted after 72 h of treatment (Fig. 1B). TNF-α and IL-6, under the same treatment conditions, also moderately induced TM release by up to 4.4 and 5.5 fold, respectively (Fig. 1C). The effects of either cytokine on TM protein levels and release were also found to be concentration- and time-dependent. Moreover, cytokine treatments had no significant effect on cell viability (data not shown). Effect of shear stress on TM protein/mRNA levels and release in HBMvECs Exposure of confluent HBMvECs to laminar shear stress at 8 dyn/cm2 for 24 h significantly increased TM protein levels by up to 65% (Fig. 2A). A 2-fold increase in TM mRNA levels was also observed after 24 h, an increase which still persisted after 72 h of shearing (Fig. 2B). Shear stress also strongly induced TM release in a timedependent manner by up to 11.8 fold after 24 h (Fig. 2C). Finally, in order to assess the functional efficacy of released TM, the effect of shear-conditioned media on thrombin-induced permeabilization of static HBMvEC reporter cultures was investigated. Conditioned media, containing released TM (volume adjusted to 3 or 6 ng TM), was found to completely block thrombin-induced permeabilization of HBMvECs to FITC–dextran 40 kDa (Fig. 2D).
3
Discussion The current study employed cultured HBMvECs to address the impact of injurious and protective signals on the cellular levels and release of TM, a thrombin-binding receptor with potent anti-coagulant and anti-inflammatory properties that is known to be expressed within the brain microvasculature (Wang et al., 1997). Our initial experiments demonstrate for the first time concentration-dependent downregulation of TM protein levels (up to 70%) following either TNF-α or IL-6 treatment. Parallel experiments also demonstrated cytokine-dependent downregulation of TM mRNA levels, with maximal decrease (N50%) observed after 24 h of treatment with either cytokine. It can be noted that the ability of both of these cytokines to downregulate the expression of interendothelial junction proteins (VE-cadherin, occludin, claudin-5, and ZO-1) in a concentration- and time-dependent manner, with injurious consequences for HBMvEC barrier phenotype, has also recently been reported by our laboratory (Rochfort et al., 2014). The current observations are also consistent with an earlier study by Wang et al. (1997) who reported downregulation of TM mRNA levels in response to direct treatment of bovine brain capillaries with TNF-α. A more recent study by Behling-Kelly et al. (2007) has also demonstrated how Haemophilus somnus infection of bovine brain microvascular endothelial cells can
A TM GAPDH
TNF-α
fold change to control
TM mRNA Levels
B
**
Protein Levels
fold change to control
Time (hrs)
TNF-
0
6
IL-6
**
18
* ** ** *
24
48
72
C *
*
*
TNF-
*
*
*
*
* *
TNF-
*
Protein Levels
fold change to control
* *
* *
* *
IL-6
*
*
*
*
*
*
IL-6
*
*
* *
IL-6
* *
*
*
*
*
Fig. 1. Effect of cytokines on TM protein/mRNA levels and release in HBMvECs. (A) Confluent cells were treated with TNF-α or IL-6 (0–100 ng/ml, 6 or 18 h). Post-treatment, whole cell protein lysates were harvested for Western blotting. Histograms represent the densitometric fold change in relative protein levels for TM in response to increasing concentration of cytokine. *P ≤ 0.05 versus untreated control. All gels are representative and shown for 18 h treatments only. (B) Confluent cells were treated with TNF-α or IL-6 (100 ng/ml, 0–72 h). Post-treatment, mRNA was harvested for qRT-PCR. Histograms represent the fold change in TM mRNA levels in response to cytokine treatment. *P ≤ 0.05 versus 0 h. (C) Confluent cells were treated with TNF-α or IL-6 (0–100 ng/ml, 0–18 h). Post-treatment, HBMvEC-conditioned media samples were harvested for ELISA. Histograms represent the change in TM levels in response to increasing cytokine concentration (left hand side) and increasing time (right hand side). *P ≤ 0.05 versus untreated control. *P ≤ 0.05 versus 0 h.
4
K.D. Rochfort, P.M. Cummins / Microvascular Research 97 (2015) 1–5
C *
*
fold change to control
Protein Levels
A
* *
B
*
*
*
*
*
D
fold change to control
* TM mRNA Levels
*
*
*
*
**
*
*
**
*
*
TM (ng)
0
3
6
0
3
6
Thrombin (2U)
-
-
-
+
+
+
Fig. 2. Effect of shear stress on TM protein/mRNA levels and release in HBMvECs. (A) Confluent cells were treated with laminar shear (0 or 8 dyn/cm2, 24 h). Post-treatment, whole cell protein lysates were harvested for Western blotting. Histograms represent the densitometric fold change in relative protein levels for TM in response to shear. *P ≤ 0.05 versus static control. All gels are representative. (B) Confluent cells were treated with laminar shear (8 dyn/cm2, 0–72 h). Post-treatment, mRNA was harvested for qRT-PCR. Histograms represent the fold change in TM mRNA levels in response to cytokine treatment. *P ≤ 0.05 versus 0 h. (C) Confluent cells were treated with laminar shear (0 or 8 dyn/cm2, 0–24 h). Post-treatment, HBMvEC-conditioned media samples were harvested for ELISA. Histograms represent the change in released TM levels in response to shear. *P ≤ 0.05 versus 0 h control. (D) Confluent HBMvECs were monitored for permeability following 30 min of incubation with either unconditioned media (containing 0 ng of TM) or shear-conditioned media (containing 3 ng or 6 ng of released TM) in the absence and presence of thrombin (2 U). *P ≤ 0.05 versus TM (0 ng)/thrombin (–) control. **P ≤ 0.05.
lead to induced expression and release of proinflammatory cytokines in parallel with reduced TM mRNA levels, although a causal relationship between the cytokine induction and the TM mRNA reduction was not established in that paper. Further experiments also demonstrated the ability of both cytokines to moderately induce release of TM from HBMvECs into media in a concentration-dependent manner. This contrasts with an earlier study by Boehme et al. (1996) showing how TNF-α could only induce TM release from human umbilical vein endothelial cells (HUVECs) in the presence of neutrophils. Lack of TNF-α effect on TM release from human aortic endothelial cells (HAECs) has also recently been demonstrated in our laboratory (Martin et al., 2014), and may point to a fundamental difference in the responsiveness of micro- and macrovascular endothelial cells to cytokine stimulation in vitro. Finally, to our knowledge, there have been no studies documenting the effects of IL-6 on TM release from the cerebrovascular endothelium, although various in vivo models of endothelial injury have reported co-elevated circulating levels of soluble TM and IL-6 (Bouchama et al., 2008). Our studies went on to confirm shear-dependent induction of TM protein levels (up to 65%) and TM mRNA levels (up to 2-fold) in cultured HBMvECs. This is consistent with related studies in endothelial cells of macrovascular and retinal microvascular origin (Martin et al., 2013, 2014) and can likely be attributed to the presence of a CACCC motif within the 5′ untranslated region of the TM promoter, which is known to facilitate positive gene regulation by Krüppel-like factor 2 (KLF2), a transcription factor that is activated by shear stress (Dekker et al., 2002). Moreover, shear stress proved to be a much more potent
stimulus than cytokines for TM release from HBMvECs (over 5-fold). Whilst beyond the scope of the current short communication, this may reflect stimulus-dependent differences in endothelial activation leading to alternate mechanisms of TM release (for example, proteolytic versus microvesicular pathways) (Redl et al., 1995; Satta et al., 1997; Vion et al., 2013). Moreover, although the precise physiological relevance of this shear-induced TM release is currently unknown, we speculate that it could contribute to the shear-dependent stabilization of the cerebrovascular endothelium in vivo via reduction of thrombininduced permeabilization. In summary, we present novel data demonstrating the contrasting regulation of TM cellular protein/mRNA levels and TM release by pathological and physiological signals within the brain microvascular endothelium in vitro. This provides a useful foundation for further investigations using more advanced cellular models of closer physiological consistency with the BBB (e.g. EC/astrocyte/pericyte co-culture models). Abbreviations BBB ELISA FITC GAPDH HAEC HBMvEC HRP
blood–brain barrier enzyme-linked immunosorbent assay fluorescein isothiocyanate glyceraldehyde phosphate dehydrogenase human aortic endothelial cell human brain microvascular endothelial cell horseradish peroxidase
K.D. Rochfort, P.M. Cummins / Microvascular Research 97 (2015) 1–5
HUVEC IL-6 KLF2 TBS TBST TM TNF-α ZO-1
human umbilical vein endothelial cell interleukin-6 Krüppel-like factor-2 tris-buffered saline tris-buffered saline/Tween-20 thrombomodulin tumour necrosis factor-α zonula occludens-1
Acknowledgments This work was funded by the National Development Plan/Higher Education Authority of Ireland Programme for Research in Third Level Institutes — NDP/HEA-PRTLI Cycle 4: Targeted Therapeutics & Theranostics (PMC). References Behling-Kelly, E., Kim, K.S., Czuprynski, C.J., 2007. Haemophilus somnus activation of brain endothelial cells: potential role for local cytokine production and thrombosis in central nervous system (CNS) infection. Thromb. Haemost. 98, 823–830. Boehme, M.W., Deng, Y., Raeth, U., Bierhaus, A., Ziegler, R., Stremmel, W., Nawroth, P.P., 1996. Release of thrombomodulin from endothelial cells by concerted action of TNF-alpha and neutrophils: in vivo and in vitro studies. Immunology 87, 134–140. Bouchama, A., Kunzelmann, C., Dehbi, M., Kwaasi, A., Eldali, A., Zobairi, F., Freyssinet, J.M., de Prost, D., 2008. Recombinant activated protein C attenuates endothelial injury and inhibits procoagulant microparticles release in baboon heatstroke. Arterioscler. Thromb. Vasc. Biol. 28, 1318–1325. Clark, P.R., Jensen, T.J., Kluger, M.S., Morelock, M., Hanidu, A., Qi, Z., Tatake, R.J., Pober, J.S., 2011. MEK5 is activated by shear stress, activates ERK5 and induces KLF4 to modulate TNF responses in human dermal microvascular endothelial cells. Microcirculation 18, 102–117. Dekker, R.J., van Soest, S., Fontijn, R.D., Salamanca, S., de Groot, P.G., VanBavel, E., Pannekoek, H., Horrevoets, A.J., 2002. Prolonged fluid shear stress induces a distinct set of endothelial cell genes, most specifically lung Krüppel-like factor (KLF2). Blood 100, 1689–1698. Festoff, B.W., Li, C., Woodhams, B., Lynch, S., 2012. Soluble thrombomodulin levels in plasma of multiple sclerosis patients and their implication. J. Neurol. Sci. 323, 61–65. Giwa, M.O., Williams, J., Elderfield, K., Jiwa, N.S., Bridges, L.R., Kalaria, R.N., Markus, H.S., Esiri, M.M., Hainsworth, A.H., 2012. Neuropathologic evidence of endothelial changes in cerebral small vessel disease. Neurology 78, 167–174. Guinan, A.F., Rochfort, K.D., Fitzpatrick, P.A., Walsh, T.G., Pierotti, A.R., Phelan, S., Murphy, R.P., Cummins, P.M., 2013. Shear stress is a positive regulator of thimet oligopeptidase
5
(EC3.4.24.15) in vascular endothelial cells: consequences for MHC1 levels. Cardiovasc. Res. 99, 545–554. Hassan, A., Hunt, B.J., O'Sullivan, M., Parmar, K., Bamford, J.M., Briley, D., Brown, M.M., Thomas, D.J., Markus, H.S., 2003. Markers of endothelial dysfunction in lacunar infarction and ischaemic leukoaraiosis. Brain 126, 424–432. Krizanac-Bengez, L., Mayberg, M.R., Cunningham, E., Hossain, M., Ponnampalam, S., Parkinson, F.E., Janigro, D., 2006. Loss of shear stress induces leukocyte-mediated cytokine release and blood–brain barrier failure in dynamic in vitro blood–brain barrier model. J. Cell. Physiol. 206, 68–77. Martin, F.A., Murphy, R.P., Cummins, P.M., 2013. Thrombomodulin and the vascular endothelium: insights into functional, regulatory, and therapeutic aspects. Am. J. Physiol. Heart Circ. Physiol. 304, H1585–H1597. Martin, F.A., McLoughlin, A., Rochfort, K.R., Davenport, C., Murphy, R.P., Cummins, P.M., 2014. Regulation of thrombomodulin expression and release in human aortic endothelial cells by cyclic strain. PLoS One 9, e108254. Redl, H., Schlag, G., Schiesser, A., Davies, J., 1995. Thrombomodulin release in baboon sepsis: its dependence on the dose of Escherichia coli and the presence of tumor necrosis factor. J. Infect. Dis. 171, 1522–1527. Rochfort, K.D., Collins, L.E., Murphy, R.P., Cummins, P.M., 2014. Downregulation of blood– brain barrier phenotype by proinflammatory cytokines involves NADPH oxidasedependent ROS generation: consequences for interendothelial adherens and tight junctions. PLoS One 9, e101815. Satta, N., Freyssinet, J.M., Toti, F., 1997. The significance of human monocyte thrombomodulin during membrane vesiculation and after stimulation by lipopolysaccharide. Br. J. Haematol. 96, 534–542. Tran, N.D., Wong, V.L., Schreiber, S.S., Bready, J.V., Fisher, M., 1996. Regulation of brain capillary endothelial thrombomodulin mRNA expression. Stroke 27, 2304–2310. Traub, O., Berk, B.C., 1998. Laminar shear stress: mechanisms by which endothelial cells transduce an atheroprotective force. Arterioscler. Thromb. Vasc. Biol. 18, 677–685. Tuttolomondo, A., Di Raimondo, D., di Sciacca, R., Pinto, A., Licata, G., 2008. Inflammatory cytokines in acute ischemic stroke. Curr. Pharm. Des. 14, 3574–3589. Vion, A.C., Ramkhelawon, B., Loyer, X., Chironi, G., Devue, C., Loirand, G., Tedgui, A., Lehoux, S., Boulanger, C.M., 2013. Shear stress regulates endothelial microparticle release. Circ. Res. 112, 1323–1333. Walsh, T.G., Murphy, R.P., Fitzpatrick, P., Rochfort, K.D., Guinan, A.F., Murphy, A., Phelan, S., Murphy, R.P., Cummins, P.M., 2011. Stabilization of brain microvascular endothelial barrier function by shear stress involves VE-cadherin signaling leading to modulation of pTyr-occludin levels. J. Cell. Physiol. 226, 3053–3063. Wang, L., Tran, N.D., Kittaka, M., Fisher, M.J., Schreiber, S.S., Zlokovic, B.V., 1997. Thrombomodulin expression in bovine brain capillaries. Anticoagulant function of the blood–brain barrier, regional differences, and regulatory mechanisms. Arterioscler. Thromb. Vasc. Biol. 17, 3139–3146. Wenzel, J., Assmann, J.C., Schwaninger, M., 2014. Thrombomodulin — a new target for treating stroke at the crossroad of coagulation and inflammation. Curr. Med. Chem. 21, 2025–2034. Yokota, H., Naoe, Y., Nakabayashi, M., Unemoto, K., Kushimoto, S., Kurokawa, A., Node, Y., Yamamoto, Y., 2002. Cerebral endothelial injury in severe head injury: the significance of measurements of serum thrombomodulin and the von Willebrand factor. J. Neurotrauma 19, 1007–1015.
Contents lists available at ScienceDirect
Microvascular Research journal homepage: www.elsevier.com/locate/ymvre
Short Communication
Thrombomodulin regulation in human brain microvascular endothelial cells in vitro: Role of cytokines and shear stress Keith D. Rochfort, Philip M. Cummins ⁎ School of Biotechnology, Dublin City University, Glasnevin, Dublin 9, Ireland Centre for Preventive Medicine, Dublin City University, Glasnevin, Dublin 9, Ireland
a r t i c l e
i n f o
Article history: Accepted 13 September 2014 Available online 22 September 2014 Keywords: Blood–brain barrier Thrombomodulin Cytokines Shear stress Endothelium
a b s t r a c t Thrombomodulin (TM), an important determinant of blood vessel homeostasis, is expressed on the luminal vascular endothelial cell surface and is released into serum in response to circulatory signals. This includes the cerebrovascular endothelium, where the anti-coagulant and anti-inflammatory properties of TM are thought to be critical to the brain microcirculation and blood–brain barrier (BBB) integrity. Much is still unknown however about how circulatory stimuli may regulate TM activity within the brain microvasculature. To address this, the current short paper investigated the effects of opposing regulatory signals, namely cytokines (TNF-α, IL-6) and laminar shear stress, on the cellular levels and release of TM in cultured human brain microvascular endothelial cells (HBMvECs). Treatment of confluent HBMvECs with either TNF-α or IL-6 (100 ng/ml, 18 h) reduced TM protein levels by up to 70%, whilst inducing TM release into media by up to 4.4 and 5.5 fold, respectively. The effects of either cytokine (0–100 ng/ml) on TM protein levels (6 or 18 h) and release (0–18 h) were also found to be concentration- and time-dependent. Either cytokine (100 ng/ml, 24–72 h) also reduced TM mRNA levels by N 50%. When exposed to laminar shear stress for 24 h at 8 dyn/cm2 (SI unit equivalent = 0.8 Pa), TM protein levels were upregulated by 65% in parallel with a 2-fold increase in TM mRNA levels. Shear stress also proved to be a much more potent stimulus for TM release from HBMvECs, yielding media TM levels of 1000 pg/105 cells, when compared to 175 and 210 pg/105 cells for TNF-α and IL-6, respectively, after parallel 18 h treatments. Finally, shear-conditioned media was found to completely block thrombin-induced permeabilization of HBMvECs, confirming the functional efficacy of released TM. In summary, our data indicate that TM is differentially regulated within cultured HBMvECs by humoral and biomechanical signals. © 2014 Elsevier Inc. All rights reserved.
Introduction Thrombomodulin (TM), an integral membrane receptor constitutively expressed on the luminal surface of vascular endothelial cells, is an important determinant of blood vessel homeostasis (for review see Martin et al., 2013). Studies have confirmed the expression of TM within the cerebrovascular endothelium, where its anti-coagulant and antiinflammatory properties are thought to be critical to the brain microcirculation and blood–brain barrier (BBB) integrity (Tran et al., 1996; Wang et al., 1997). In this respect, soluble TM-based therapeutics have been used to treat BBB-associated neurological disorders such as stroke (Wenzel et al., 2014). Moreover, elevated TM release or “shedding” into serum invariably accompanies the cerebrovascular endothelial activation typically associated with neurodegenerative disorders (Festoff et al., 2012), stroke (Hassan et al., 2003), cerebral small vessel disease (Giwa et al., 2012) and traumatic brain injury (Yokota et al., 2002). Much is still unknown however about how common pathological and physiological stimuli may regulate TM activity within the brain ⁎ Corresponding author. Fax: +353 1 7005412. E-mail address: [email protected] (P.M. Cummins).
http://dx.doi.org/10.1016/j.mvr.2014.09.003 0026-2862/© 2014 Elsevier Inc. All rights reserved.
microvasculature. Proinflammatory cytokines for example, have been strongly linked to the pathological effects of stroke and other neurological diseases, imparting anti-barrier and pro-coagulant effects on the brain microvascular endothelium (Behling-Kelly et al., 2007; Tuttolomondo et al., 2008; Rochfort et al., 2014). The precise effects of cytokines, and particularly IL-6, on TM levels and release within the brain microvasculature however, are poorly documented. In contrast to proinflammatory cytokines, physiological laminar shear stress typically has a protective anti-inflammatory impact on the vascular endothelium (Traub and Berk, 1998). Moreover, the ability of shear stress to enhance BBB phenotype and to protect against cytokine-induced BBB injury has previously been reported (Krizanac-Bengez et al., 2006; Clark et al., 2011; Walsh et al., 2011). Whilst various studies have demonstrated the ability of shear stress to upregulate endothelial TM expression (for review see Martin et al., 2013), to our knowledge, there are no existing reports documenting the precise effects of shear stress on TM regulation within the brain microvasculature. To address these knowledge gaps, the current short paper investigated the effects of opposing regulatory influences, namely cytokines (TNF-α, IL-6) and laminar shear stress, on the cellular protein/mRNA levels and release of TM in cultured HBMvECs. Both time- and
2
K.D. Rochfort, P.M. Cummins / Microvascular Research 97 (2015) 1–5
concentration-/force-dependency studies were conducted, whilst the putative impact of TM release on thrombin-induced HBMvEC permeabilization was also investigated.
(250 bp): Forward 5′-CAGCCACCCGAGATTGAGCA-3′; Reverse 5′-TAGT AGCGACGGGCGGTGTG-3′. Enzyme-linked immunosorbent assay (ELISA)
Materials and methods Materials Unless otherwise stated, all reagents were purchased from Sigma-Aldrich (Dublin, IRL). Cytokines (TNF-α, IL-6) were purchased from Millipore (Cork, IRL). Primary antisera included mouse antithrombomodulin IgG ab6980 (Abcam, Cambridge, UK) and rabbit antiGAPDH IgG (Millipore). Secondary antisera included HRP-conjugated goat anti-mouse IgG and goat anti-rabbit IgG (Cell Signalling, MA, USA). Cell culture Culture of primary-derived HBMvECs has been described previously (Rochfort et al., 2014). HBMvECs, primary cultured from a single adult male donor post-mortem, were purchased from Cell Systems Corporation (WA, USA — Cat No. ACBRI 376). Cells were routinely grown in EndoGRO™ MV Basal Medium (Millipore) containing 5% foetal bovine serum, antibiotics (100 mg/ml Mycozap™), and relevant supplements. All cells (passages 5–12) were grown on tissue culture grade plasticware coated with Attachment Factor™ (Life Technologies, UK) and maintained in a humidified atmosphere of 5% CO2/95% air at 37 °C. For experimental purposes, cells were routinely subjected to treatment with either TNF-α or IL-6 (0–100 ng/ml, 0–72 h), or to laminar shear stress (0 or 8 dyn/cm2, 0–72 h), the latter employing an orbital rotation shear model (Walsh et al., 2011). Post-treatment, both cells and conditioned media were harvested for analysis. Total cell protein lysate preparation has been described previously (Rochfort et al., 2014). Conditioned media samples were routinely centrifuged at 700 ×g for 15 min to remove any cellular debris. Media samples were routinely assayed by ELISA prior to freezing. All samples were ultimately stored at −80 °C. Western immunoblotting Total cell protein lysates were resolved by 10% SDS-PAGE under reducing conditions and electroblotted as previously described (Rochfort et al., 2014). Membranes were blocked for 60 min in tris-buffered saline (TBS: 10 mM Tris pH 8.0, 150 mM NaCl) containing 5% w/v bovine serum albumin (BSA) before being incubated overnight in primary antisera with gentle agitation at 4 °C. Primary antisera were prepared in TBST (+1% BSA, +0.1% Tween-20): 1 μg/ml anti-thrombomodulin mouse monoclonal IgG and 0.2 μg/ml anti-GAPDH rabbit monoclonal IgG. Secondary antisera, applied to washed membranes for 3 h with gentle agitation at room temperature, were prepared in TBST: 1:2000 HRP-conjugated goat anti-mouse IgG (thrombomodulin) and 1:3000 HRP-conjugated goat anti-rabbit IgG (GAPDH). Membranes were developed using a Luminata™ Western HRP Kit (Millipore). Scanning densitometry of Western blots was routinely performed using NIH ImageJ software, with GAPDH routinely employed as a loading control to facilitate densitometric normalization of bands.
A thrombomodulin/BDCA-3 DuoSet® ELISA Kit (R&D Systems, MN, USA) was employed as per manufacturer instructions (with minor modifications) to measure absolute TM levels in HBMvEC conditioned media samples. Briefly, 96-well plates were coated with 50 μl/well of the capture antibody and incubated overnight. The plate was then blocked for 1 h by adding 150 μl of Reagent Diluent to each well. Media samples and human recombinant TM standards (31.25– 2000 pg/ml) were added in duplicate at 50 μl/well, with incubations proceeding for 2 h. 50 μl of detection antibody was then added to each well for a further 2 h, followed by 50 μl of streptavidin–HRP to each well for 20 min (in the dark). 50 μl of substrate solution was then added to each well for a further 20 min (in the dark). Finally, reactions were terminated with the addition of 25 μl of Stop Solution to each well and the plate luminescence subsequently read at both 570 nm and 450 nm (wave correction was used to subtract the readings at 570 nm from 450 nm to control for optical imperfections in the plate). Transendothelial permeability assay For analysis of HBMvEC monolayer permeability, the transwell method of Rochfort et al. (2014) was employed. Briefly, HBMvECs were plated at high density (5 × 105 cells/well) into Millicell hanging cell culture inserts placed within 6-well dishes (Millipore; 0.4 μm pore size, 24 mm filter diameter). Complete media was added to the upper (luminal, 2 ml) and lower (abluminal, 4 ml) chambers of the Millicell insert within the 6-well dish and the cells were allowed to grow to confluency. Confluent HBMvECs within the upper chamber were next incubated for 30 min with either unconditioned media (0 ng of TM) or shear-conditioned media (3 ng or 6 ng of TM — i.e. 1.5–3.0 ng/ml) in the absence and presence of 2 units of thrombin (i.e. 1 U/ml) (note: conditioned media was pre-concentrated by centrifugal filtration to contain either 3 ng or 6 ng of released thrombomodulin for co-incubations with thrombin). Post-treatment, media in the upper and lower chambers was replenished, fluorescein isothiocyanate (FITC)-labelled 40 kDa dextran was added to the upper chamber (giving a final concentration of 250 μg/ml), and transwell diffusion allowed to proceed. Media samples (28 μl) were collected from the lower chamber after 3 h, diluted to a final volume of 400 μl with complete media, and monitored in 96-well format for FITC–dextran fluorescence. A TECAN Safire 2 fluorospectrometer was used with excitation and emission wavelengths set at 490 and 520 nm, respectively. Permeability is presented as % transendothelial exchange of FITC–dextran 40 kDa (%TEE FD40). Statistical analysis Results are expressed as mean ± s.d. Experimental points were typically performed in triplicate with a minimum of three independent experiments (n = 3). Statistical comparisons between control and experimental groups were by ANOVA in conjunction with a Dunnett's post-hoc test for multiple comparisons. A Student's t-test was also routinely employed for pairwise comparisons. A value of P ≤ 0.05 was considered significant.
Quantitative real-time PCR Results Following experiments, endothelial cells were harvested for extraction of total RNA and analysis of TM mRNA expression as previously described (Guinan et al., 2013). Ribosomal subunit S18 was routinely used for normalization purposes. Primer pairs were screened for correct product size by 1% agarose gel electrophoresis and underwent meltcurve analysis for primer-dimers. TM (107 bp): Forward 5′-ACCTTCCT CAATGCCAGTCAG-3′; Reverse 5′-GCCGTCGCCGTTCAGTAG-3′; S18
Effect of proinflammatory cytokines on TM protein/mRNA levels and release in HBMvECs Treatment of confluent HBMvECs with either TNF-α or IL-6 reduced TM protein levels by up to 70% at 100 ng/ml cytokine for 18 h (Fig. 1A). Reductions in TM mRNA levels of 65% and 53% were also observed
K.D. Rochfort, P.M. Cummins / Microvascular Research 97 (2015) 1–5
following 24 h of treatment with TNF-α and IL-6, respectively, reductions which still persisted after 72 h of treatment (Fig. 1B). TNF-α and IL-6, under the same treatment conditions, also moderately induced TM release by up to 4.4 and 5.5 fold, respectively (Fig. 1C). The effects of either cytokine on TM protein levels and release were also found to be concentration- and time-dependent. Moreover, cytokine treatments had no significant effect on cell viability (data not shown). Effect of shear stress on TM protein/mRNA levels and release in HBMvECs Exposure of confluent HBMvECs to laminar shear stress at 8 dyn/cm2 for 24 h significantly increased TM protein levels by up to 65% (Fig. 2A). A 2-fold increase in TM mRNA levels was also observed after 24 h, an increase which still persisted after 72 h of shearing (Fig. 2B). Shear stress also strongly induced TM release in a timedependent manner by up to 11.8 fold after 24 h (Fig. 2C). Finally, in order to assess the functional efficacy of released TM, the effect of shear-conditioned media on thrombin-induced permeabilization of static HBMvEC reporter cultures was investigated. Conditioned media, containing released TM (volume adjusted to 3 or 6 ng TM), was found to completely block thrombin-induced permeabilization of HBMvECs to FITC–dextran 40 kDa (Fig. 2D).
3
Discussion The current study employed cultured HBMvECs to address the impact of injurious and protective signals on the cellular levels and release of TM, a thrombin-binding receptor with potent anti-coagulant and anti-inflammatory properties that is known to be expressed within the brain microvasculature (Wang et al., 1997). Our initial experiments demonstrate for the first time concentration-dependent downregulation of TM protein levels (up to 70%) following either TNF-α or IL-6 treatment. Parallel experiments also demonstrated cytokine-dependent downregulation of TM mRNA levels, with maximal decrease (N50%) observed after 24 h of treatment with either cytokine. It can be noted that the ability of both of these cytokines to downregulate the expression of interendothelial junction proteins (VE-cadherin, occludin, claudin-5, and ZO-1) in a concentration- and time-dependent manner, with injurious consequences for HBMvEC barrier phenotype, has also recently been reported by our laboratory (Rochfort et al., 2014). The current observations are also consistent with an earlier study by Wang et al. (1997) who reported downregulation of TM mRNA levels in response to direct treatment of bovine brain capillaries with TNF-α. A more recent study by Behling-Kelly et al. (2007) has also demonstrated how Haemophilus somnus infection of bovine brain microvascular endothelial cells can
A TM GAPDH
TNF-α
fold change to control
TM mRNA Levels
B
**
Protein Levels
fold change to control
Time (hrs)
TNF-
0
6
IL-6
**
18
* ** ** *
24
48
72
C *
*
*
TNF-
*
*
*
*
* *
TNF-
*
Protein Levels
fold change to control
* *
* *
* *
IL-6
*
*
*
*
*
*
IL-6
*
*
* *
IL-6
* *
*
*
*
*
Fig. 1. Effect of cytokines on TM protein/mRNA levels and release in HBMvECs. (A) Confluent cells were treated with TNF-α or IL-6 (0–100 ng/ml, 6 or 18 h). Post-treatment, whole cell protein lysates were harvested for Western blotting. Histograms represent the densitometric fold change in relative protein levels for TM in response to increasing concentration of cytokine. *P ≤ 0.05 versus untreated control. All gels are representative and shown for 18 h treatments only. (B) Confluent cells were treated with TNF-α or IL-6 (100 ng/ml, 0–72 h). Post-treatment, mRNA was harvested for qRT-PCR. Histograms represent the fold change in TM mRNA levels in response to cytokine treatment. *P ≤ 0.05 versus 0 h. (C) Confluent cells were treated with TNF-α or IL-6 (0–100 ng/ml, 0–18 h). Post-treatment, HBMvEC-conditioned media samples were harvested for ELISA. Histograms represent the change in TM levels in response to increasing cytokine concentration (left hand side) and increasing time (right hand side). *P ≤ 0.05 versus untreated control. *P ≤ 0.05 versus 0 h.
4
K.D. Rochfort, P.M. Cummins / Microvascular Research 97 (2015) 1–5
C *
*
fold change to control
Protein Levels
A
* *
B
*
*
*
*
*
D
fold change to control
* TM mRNA Levels
*
*
*
*
**
*
*
**
*
*
TM (ng)
0
3
6
0
3
6
Thrombin (2U)
-
-
-
+
+
+
Fig. 2. Effect of shear stress on TM protein/mRNA levels and release in HBMvECs. (A) Confluent cells were treated with laminar shear (0 or 8 dyn/cm2, 24 h). Post-treatment, whole cell protein lysates were harvested for Western blotting. Histograms represent the densitometric fold change in relative protein levels for TM in response to shear. *P ≤ 0.05 versus static control. All gels are representative. (B) Confluent cells were treated with laminar shear (8 dyn/cm2, 0–72 h). Post-treatment, mRNA was harvested for qRT-PCR. Histograms represent the fold change in TM mRNA levels in response to cytokine treatment. *P ≤ 0.05 versus 0 h. (C) Confluent cells were treated with laminar shear (0 or 8 dyn/cm2, 0–24 h). Post-treatment, HBMvEC-conditioned media samples were harvested for ELISA. Histograms represent the change in released TM levels in response to shear. *P ≤ 0.05 versus 0 h control. (D) Confluent HBMvECs were monitored for permeability following 30 min of incubation with either unconditioned media (containing 0 ng of TM) or shear-conditioned media (containing 3 ng or 6 ng of released TM) in the absence and presence of thrombin (2 U). *P ≤ 0.05 versus TM (0 ng)/thrombin (–) control. **P ≤ 0.05.
lead to induced expression and release of proinflammatory cytokines in parallel with reduced TM mRNA levels, although a causal relationship between the cytokine induction and the TM mRNA reduction was not established in that paper. Further experiments also demonstrated the ability of both cytokines to moderately induce release of TM from HBMvECs into media in a concentration-dependent manner. This contrasts with an earlier study by Boehme et al. (1996) showing how TNF-α could only induce TM release from human umbilical vein endothelial cells (HUVECs) in the presence of neutrophils. Lack of TNF-α effect on TM release from human aortic endothelial cells (HAECs) has also recently been demonstrated in our laboratory (Martin et al., 2014), and may point to a fundamental difference in the responsiveness of micro- and macrovascular endothelial cells to cytokine stimulation in vitro. Finally, to our knowledge, there have been no studies documenting the effects of IL-6 on TM release from the cerebrovascular endothelium, although various in vivo models of endothelial injury have reported co-elevated circulating levels of soluble TM and IL-6 (Bouchama et al., 2008). Our studies went on to confirm shear-dependent induction of TM protein levels (up to 65%) and TM mRNA levels (up to 2-fold) in cultured HBMvECs. This is consistent with related studies in endothelial cells of macrovascular and retinal microvascular origin (Martin et al., 2013, 2014) and can likely be attributed to the presence of a CACCC motif within the 5′ untranslated region of the TM promoter, which is known to facilitate positive gene regulation by Krüppel-like factor 2 (KLF2), a transcription factor that is activated by shear stress (Dekker et al., 2002). Moreover, shear stress proved to be a much more potent
stimulus than cytokines for TM release from HBMvECs (over 5-fold). Whilst beyond the scope of the current short communication, this may reflect stimulus-dependent differences in endothelial activation leading to alternate mechanisms of TM release (for example, proteolytic versus microvesicular pathways) (Redl et al., 1995; Satta et al., 1997; Vion et al., 2013). Moreover, although the precise physiological relevance of this shear-induced TM release is currently unknown, we speculate that it could contribute to the shear-dependent stabilization of the cerebrovascular endothelium in vivo via reduction of thrombininduced permeabilization. In summary, we present novel data demonstrating the contrasting regulation of TM cellular protein/mRNA levels and TM release by pathological and physiological signals within the brain microvascular endothelium in vitro. This provides a useful foundation for further investigations using more advanced cellular models of closer physiological consistency with the BBB (e.g. EC/astrocyte/pericyte co-culture models). Abbreviations BBB ELISA FITC GAPDH HAEC HBMvEC HRP
blood–brain barrier enzyme-linked immunosorbent assay fluorescein isothiocyanate glyceraldehyde phosphate dehydrogenase human aortic endothelial cell human brain microvascular endothelial cell horseradish peroxidase
K.D. Rochfort, P.M. Cummins / Microvascular Research 97 (2015) 1–5
HUVEC IL-6 KLF2 TBS TBST TM TNF-α ZO-1
human umbilical vein endothelial cell interleukin-6 Krüppel-like factor-2 tris-buffered saline tris-buffered saline/Tween-20 thrombomodulin tumour necrosis factor-α zonula occludens-1
Acknowledgments This work was funded by the National Development Plan/Higher Education Authority of Ireland Programme for Research in Third Level Institutes — NDP/HEA-PRTLI Cycle 4: Targeted Therapeutics & Theranostics (PMC). References Behling-Kelly, E., Kim, K.S., Czuprynski, C.J., 2007. Haemophilus somnus activation of brain endothelial cells: potential role for local cytokine production and thrombosis in central nervous system (CNS) infection. Thromb. Haemost. 98, 823–830. Boehme, M.W., Deng, Y., Raeth, U., Bierhaus, A., Ziegler, R., Stremmel, W., Nawroth, P.P., 1996. Release of thrombomodulin from endothelial cells by concerted action of TNF-alpha and neutrophils: in vivo and in vitro studies. Immunology 87, 134–140. Bouchama, A., Kunzelmann, C., Dehbi, M., Kwaasi, A., Eldali, A., Zobairi, F., Freyssinet, J.M., de Prost, D., 2008. Recombinant activated protein C attenuates endothelial injury and inhibits procoagulant microparticles release in baboon heatstroke. Arterioscler. Thromb. Vasc. Biol. 28, 1318–1325. Clark, P.R., Jensen, T.J., Kluger, M.S., Morelock, M., Hanidu, A., Qi, Z., Tatake, R.J., Pober, J.S., 2011. MEK5 is activated by shear stress, activates ERK5 and induces KLF4 to modulate TNF responses in human dermal microvascular endothelial cells. Microcirculation 18, 102–117. Dekker, R.J., van Soest, S., Fontijn, R.D., Salamanca, S., de Groot, P.G., VanBavel, E., Pannekoek, H., Horrevoets, A.J., 2002. Prolonged fluid shear stress induces a distinct set of endothelial cell genes, most specifically lung Krüppel-like factor (KLF2). Blood 100, 1689–1698. Festoff, B.W., Li, C., Woodhams, B., Lynch, S., 2012. Soluble thrombomodulin levels in plasma of multiple sclerosis patients and their implication. J. Neurol. Sci. 323, 61–65. Giwa, M.O., Williams, J., Elderfield, K., Jiwa, N.S., Bridges, L.R., Kalaria, R.N., Markus, H.S., Esiri, M.M., Hainsworth, A.H., 2012. Neuropathologic evidence of endothelial changes in cerebral small vessel disease. Neurology 78, 167–174. Guinan, A.F., Rochfort, K.D., Fitzpatrick, P.A., Walsh, T.G., Pierotti, A.R., Phelan, S., Murphy, R.P., Cummins, P.M., 2013. Shear stress is a positive regulator of thimet oligopeptidase
5
(EC3.4.24.15) in vascular endothelial cells: consequences for MHC1 levels. Cardiovasc. Res. 99, 545–554. Hassan, A., Hunt, B.J., O'Sullivan, M., Parmar, K., Bamford, J.M., Briley, D., Brown, M.M., Thomas, D.J., Markus, H.S., 2003. Markers of endothelial dysfunction in lacunar infarction and ischaemic leukoaraiosis. Brain 126, 424–432. Krizanac-Bengez, L., Mayberg, M.R., Cunningham, E., Hossain, M., Ponnampalam, S., Parkinson, F.E., Janigro, D., 2006. Loss of shear stress induces leukocyte-mediated cytokine release and blood–brain barrier failure in dynamic in vitro blood–brain barrier model. J. Cell. Physiol. 206, 68–77. Martin, F.A., Murphy, R.P., Cummins, P.M., 2013. Thrombomodulin and the vascular endothelium: insights into functional, regulatory, and therapeutic aspects. Am. J. Physiol. Heart Circ. Physiol. 304, H1585–H1597. Martin, F.A., McLoughlin, A., Rochfort, K.R., Davenport, C., Murphy, R.P., Cummins, P.M., 2014. Regulation of thrombomodulin expression and release in human aortic endothelial cells by cyclic strain. PLoS One 9, e108254. Redl, H., Schlag, G., Schiesser, A., Davies, J., 1995. Thrombomodulin release in baboon sepsis: its dependence on the dose of Escherichia coli and the presence of tumor necrosis factor. J. Infect. Dis. 171, 1522–1527. Rochfort, K.D., Collins, L.E., Murphy, R.P., Cummins, P.M., 2014. Downregulation of blood– brain barrier phenotype by proinflammatory cytokines involves NADPH oxidasedependent ROS generation: consequences for interendothelial adherens and tight junctions. PLoS One 9, e101815. Satta, N., Freyssinet, J.M., Toti, F., 1997. The significance of human monocyte thrombomodulin during membrane vesiculation and after stimulation by lipopolysaccharide. Br. J. Haematol. 96, 534–542. Tran, N.D., Wong, V.L., Schreiber, S.S., Bready, J.V., Fisher, M., 1996. Regulation of brain capillary endothelial thrombomodulin mRNA expression. Stroke 27, 2304–2310. Traub, O., Berk, B.C., 1998. Laminar shear stress: mechanisms by which endothelial cells transduce an atheroprotective force. Arterioscler. Thromb. Vasc. Biol. 18, 677–685. Tuttolomondo, A., Di Raimondo, D., di Sciacca, R., Pinto, A., Licata, G., 2008. Inflammatory cytokines in acute ischemic stroke. Curr. Pharm. Des. 14, 3574–3589. Vion, A.C., Ramkhelawon, B., Loyer, X., Chironi, G., Devue, C., Loirand, G., Tedgui, A., Lehoux, S., Boulanger, C.M., 2013. Shear stress regulates endothelial microparticle release. Circ. Res. 112, 1323–1333. Walsh, T.G., Murphy, R.P., Fitzpatrick, P., Rochfort, K.D., Guinan, A.F., Murphy, A., Phelan, S., Murphy, R.P., Cummins, P.M., 2011. Stabilization of brain microvascular endothelial barrier function by shear stress involves VE-cadherin signaling leading to modulation of pTyr-occludin levels. J. Cell. Physiol. 226, 3053–3063. Wang, L., Tran, N.D., Kittaka, M., Fisher, M.J., Schreiber, S.S., Zlokovic, B.V., 1997. Thrombomodulin expression in bovine brain capillaries. Anticoagulant function of the blood–brain barrier, regional differences, and regulatory mechanisms. Arterioscler. Thromb. Vasc. Biol. 17, 3139–3146. Wenzel, J., Assmann, J.C., Schwaninger, M., 2014. Thrombomodulin — a new target for treating stroke at the crossroad of coagulation and inflammation. Curr. Med. Chem. 21, 2025–2034. Yokota, H., Naoe, Y., Nakabayashi, M., Unemoto, K., Kushimoto, S., Kurokawa, A., Node, Y., Yamamoto, Y., 2002. Cerebral endothelial injury in severe head injury: the significance of measurements of serum thrombomodulin and the von Willebrand factor. J. Neurotrauma 19, 1007–1015.
