STIM1 Controls Endothelial Barrier Function Independently of Orai1 and Ca2+ Entry

Page 1

(267), ra18. [DOI: 10.1126/scisignal.2003425] 6Science SignalingPeter A. Vincent and Mohamed Trebak (19 March 2013) Alejandro P. Adam, José C. González-Cobos, Wei Zhang, Khalid Matrougui, Arti V. Shinde, Rajender K. Motiani, Xuexin Zhang, Iskandar F. Abdullaev,

Entry2+CaSTIM1 Controls Endothelial Barrier Function Independently of Orai1 and`

This information is current as of 19 March 2013. The following resources related to this article are available online at http://stke.sciencemag.org.

Article Tools http://stke.sciencemag.org/cgi/content/full/sigtrans;6/267/ra18

Visit the online version of this article to access the personalization and article tools:

MaterialsSupplemental

http://stke.sciencemag.org/cgi/content/full/sigtrans;6/267/ra18/DC1 "Supplementary Materials"

Related Content

http://stke.sciencemag.org/cgi/content/abstract/sigtrans;2005/295/re8 http://stke.sciencemag.org/cgi/content/abstract/sigtrans;2007/412/re8

http://stke.sciencemag.org/cgi/content/abstract/sigtrans;2/93/ra67 http://stke.sciencemag.org/cgi/content/abstract/sigtrans;3/148/ra82

http://stke.sciencemag.org/cgi/content/abstract/sigtrans;5/236/re4 http://stke.sciencemag.org/cgi/content/abstract/sigtrans;6/258/rs2 http://stke.sciencemag.org/cgi/content/abstract/sigtrans;6/267/pe8

's sites:ScienceThe editors suggest related resources on

References http://stke.sciencemag.org/cgi/content/full/sigtrans;6/267/ra18#BIBL

1 article(s) hosted by HighWire Press; see: cited byThis article has been

http://stke.sciencemag.org/cgi/content/full/sigtrans;6/267/ra18#otherarticlesThis article cites 70 articles, 41 of which can be accessed for free:

Glossary http://stke.sciencemag.org/glossary/

Look up definitions for abbreviations and terms found in this article:

Permissions http://www.sciencemag.org/about/permissions.dtl

Obtain information about reproducing this article:

the American Association for the Advancement of Science; all rights reserved. byAssociation for the Advancement of Science, 1200 New York Avenue, NW, Washington, DC 20005. Copyright 2008

(ISSN 1937-9145) is published weekly, except the last week in December, by the AmericanScience Signaling

on March 19, 2013

stke.sciencemag.org

Dow

nloaded from

Page 2

R E S E A R C H A R T I C L E

C E L L B I O L O G Y

STIM1 Controls Endothelial Barrier FunctionIndependently of Orai1 and Ca2+ EntryArti V. Shinde,1 Rajender K. Motiani,1 Xuexin Zhang,1,2 Iskandar F. Abdullaev,1

Alejandro P. Adam,1 José C. González-Cobos,1,2 Wei Zhang,1,2 Khalid Matrougui,3*Peter A. Vincent,1 Mohamed Trebak1,2†

Dow

nl

Endothelial barrier function is critical for tissue fluid homeostasis, and its disruption contributes tovarious pathologies, including inflammation and sepsis. Thrombin is an endogenous agonist that impairsendothelial barrier function. We showed that the thrombin-induced decrease in transendothelial electricresistance of cultured human endothelial cells required the endoplasmic reticulum–localized, calcium-sensing protein stromal interacting molecule 1 (STIM1), but was independent of Ca2+ entry across theplasma membrane and the Ca2+ release–activated Ca2+ channel protein Orai1, which is the target ofSTIM1 in the store-operated calcium entry pathway. We found that STIM1 coupled the thrombin receptorto activation of the guanosine triphosphatase RhoA, stimulation of myosin light chain phosphorylation,formation of actin stress fibers, and loss of cell-cell adhesion. Thus, STIM1 functions in pathways thatare dependent on and independent of Ca2+ entry.

oa

on M

arch 19, 2013 stke.sciencem

ag.orgded from

INTRODUCTION

The vascular endothelium forms a tightly regulated barrier between the blood-stream and the interstitial space in tissues. Thrombin is a vascular mediatorthat acts on a class of G protein (heterotrimeric guanine nucleotide–bindingprotein)–coupled receptors (GPCRs), the protease-activated receptors (PARs)(1). In endothelial cells, PAR-1 can couple to Gi, Gq, and G12/13 and disruptthe endothelial barrier (2, 3). Activation of Rho guanosine triphosphatasedownstream of PAR-1 stimulation plays a central role in controlling the actincytoskeletal rearrangements necessary for increased endothelial permeability(4, 5). The Rho guanine nucleotide exchange factors link Ga12/13 and Rhoactivation during PAR-1 signaling (6–9). However, a second pathway fromGPCR activation to endothelial barrier function has been extrapolated fromcontractility studies in muscle, in which GPCR activation stimulates Ca2+

entry across the plasma membrane through the store-operated Ca2+ entry(SOCE) pathway, and this Ca2+ entry signal (not Ca2+ release from inter-nal stores) activates myosin light chain kinase (MLCK) (10). In endothelialcells, activation of this pathway would result in MLCK-mediated phospho-rylation of MLC, formation of actin stress fibers, and endothelial “contrac-tility,” which would result in loss of intercellular barrier function (11–14).Another possible pathway throughwhichPAR-1may couple toCa2+ signalsis through activation of transient receptor potential (TRP) channels (15).

SOCE is a ubiquitous receptor-evoked Ca2+ entry pathway in non-excitable cells whereby the depletion of inositol 1,4,5-trisphosphate(IP3)–sensitive endoplasmic reticulum (ER) Ca2+ stores stimulates Ca2+

entry across the plasma membrane (16, 17). Biophysically, SOCE is me-diated through the Ca2+ release–activated Ca2+ (CRAC) current (18), a high-ly Ca2+-selective channel (19, 20). Stromal interacting molecule 1 (STIM1)and Orai1 are the molecular components of SOCE and the CRAC current(21–24). STIM1 is the Ca2+ sensor, present mainly in the ER membrane,

1The Center for Cardiovascular Sciences, Albany Medical College, Albany, NY12208, USA. 2Nanobioscience Constellation, College of Nanoscale Scienceand Engineering, University at Albany, State University of New York, Albany,NY 12203, USA. 3Department of Physiology, Hypertension and Renal Centerof Excellence, Tulane University, New Orleans, LA 70112, USA.*Present address: Department of Physiology, Eastern Virginia School of Med-icine, 700 West Olney Road, Norfolk, VA 23501, USA.†Corresponding author. E-mail: [email protected]

ww

which senses the decrease in ER Ca2+ concentrations, aggregates into punc-ta, and redistributes into discrete regions of close ER–plasma membranecontacts to activate the plasma membrane–localized Orai1 channel, whichmediates the CRAC current (25–27). STIM1 gates Orai1 channels throughdirect protein-protein interactions involving binding of a C-terminal 100–amino acid domain of STIM1 [called SOAR (STIM/Orai-activating re-gion) (28)] to the C and N termini of Orai1 (28, 29).

Earlier studies, before the discovery of the STIM1 and Orai1 proteins,proposed that SOCE in endothelial cells involved either one of the TRPCisoforms, TRPC1 or TRPC4, or a heteromultimeric channel composed ofTRPC1 and TRPC4 (13, 15, 30–35). However, TRPC proteins form non-selective cation channels, and their involvement in SOCE remains highlycontroversial (25, 36). In human umbilical vein endothelial cells (HUVECs)and human pulmonary artery endothelial cells (HPAECs), thrombin ac-tivates SOCE and the CRAC current mediated by STIM1 and Orai1, in-dependently of TRPC1 and TRPC4 (37). Here, we showed that anotherprimary endothelial cell type, human dermal microvascular endothelialcells (HDMECs), displays STIM1- and Orai1-mediated SOCE and CRACin response to either agonist stimulation with thrombin or passive storedepletion. We further showed that thrombin-mediated decrease in transen-dothelial electric resistance (TER), an indication of endothelial barrierdisruption, in HUVECs and HDMECs required STIM1 independentlyof Orai1, MLCK, and Ca2+ entry across the plasma membrane. Althoughknockdown of TRPC1, TRPC4, or TRPC6 inhibited thrombin-mediateddisruption of endothelial barrier function, the role of STIM1 in the regulationof endothelial barrier function was not due to STIM1-mediated interactionwith TRPC channels. Rather, STIM1 was required for RhoA activation,MLC phosphorylation, and formation of actin stress fibers.

RESULTS

STIM1, but not Orai1, is required for thrombin-mediateddisruption of endothelial barrier functionTo investigate the role of STIM1 and Orai1 in thrombin-mediated disrup-tion of endothelial barrier function, we knocked down STIM1 or Orai1 inHUVECs with specific small interfering RNA (siRNA) (Fig. 1A). STIM1knockdown substantially inhibited thrombin-mediated loss of HUVEC

w.SCIENCESIGNALING.org 19 March 2013 Vol 6 Issue 267 ra18 1

Page 3

R E S E A R C H A R T I C L E

on March 19, 2013

stke.sciencemag.org

Dow

nloaded from

barrier function, measured as the change in TER by electric cell-substrateimpedance sensing (ECIS) (38) (Fig. 1B). Two additional siRNAs againstSTIM1 (#3 and #6) resulted in STIM1 knockdown of varying degrees(fig. S1A), and the ability to influence the HUVEC response to thrombincorrelated with the effectiveness of knockdown (fig. S1B). However, knock-down of Orai1 failed to affect HUVEC TER in response to thrombin(Fig. 1B). Previous studies reported a lack of correlation between receptor-regulated Ca2+ signals and loss of endothelial barrier function in responseto thrombin (39, 40): Whereas 5 nM thrombin caused maximal decreasein TER, 100 nM thrombin was required to cause a maximal Ca2+ signal.Therefore, we tested whether thrombin-mediated decrease in TER mightdisplay a different requirement for STIM1 or Orai1 at high concentrationsof thrombin. Even in the presence of 100 nM thrombin, only STIM1 siRNA,but not Orai1 siRNA, impaired the thrombin-induced decrease in TER(Fig. 1C).

To confirm that the differential effects of STIM1 knockdown were notthe result of differences in the abilities of the monolayers to establish abarrier, we monitored the effect of siRNA against STIM1 and Orai1 onbarrier establishment of cells seeded on ECIS plates. Cells in which STIM1or Orai1 was knocked down established the same maximal TER in a sim-ilar period as did cells transfected with control siRNA (fig. S2A). Further-more, control nontargeting siRNA did not render endothelial monolayersmore susceptible to changes in TER because vehicle addition to monolay-ers failed to cause changes in TER, whereas thrombin produced the typicaldrop in TER (fig. S2B).

We previously showed that STIM1 and Orai1 mediate SOCE andCRAC currents in HUVECs independently of TRPC1 and TRPC4 (37).However, another study suggested that the involvement of Orai1 in SOCEmay be unique to HUVECs, and other endothelial cell types, includingmouse microvascular endothelial cells, might mediate SOCE throughSTIM1-dependent regulation of TRPC4 channels (41). Therefore, we in-vestigated the involvement of STIM1 and Orai1 in SOCE and CRAC cur-rents in another endothelial cell type, primary HDMECs. siRNA against

ww

either STIM1 or Orai1 caused significant and specific knockdown of theirrespective proteins in HDMECs (Fig. 2A) and essentially abrogated SOCEactivated by thrombin (Fig. 2B). We measured CRAC in HDMECs ac-tivated by passive store depletion with the calcium chelator BAPTA[1,2-bis-(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid] dialyzedthrough the patch pipette in both Ca2+-containing and divalent-free (DVF)bath solutions (Fig. 2C). In DVF bath solutions, Na+ ions become the chargecarrier. The first switch to a DVF solution, which was performed right af-ter whole-cell break-in, at the time when CRAC current was negligible,was used to gauge background currents before store depletion. BAPTA di-alysis produced in control cells a small, slowly developing inward Ca2+

current that was fully activated within 4 to 5 min. This current was potent-ly amplified by replacing the Ca2+-containing bath solution with a DVFsolution and was blocked by lanthanides (Gd3+). This CRAC current wasabsent in the HDMECs in which STIM1 or Orai1 was knocked down (Fig.2C). The current/voltage (I/V ) curves for siRNA control, siRNA STIM1,and siRNAOrai1 based on the traces shown in Fig. 2C showed an inward-ly rectifying current with a positive reversal potential, which is indicative ofhigh Ca2+ selectivity (Fig. 2D). Analyses of the CRAC current measured at−100 mV in DVF bath solutions, conditions in which the current is carriedby Na+ ions instead of Ca2+ ions, from HDMECs transfected with controlnontargeting siRNA, siRNA STIM1, or siRNA Orai1 confirmed the re-quirement of STIM1 and Orai1 in mediating this current (Fig. 2E). Al-though either STIM1 or Orai1 knockdown inhibited SOCE and CRACin HDMECs, only STIM1 knockdown inhibited thrombin-mediated de-crease in TER, suggesting that thrombin-induced change in endothelialbarrier function may not require Orai1 or involve SOCE (fig. S3).

Thrombin-mediated disruption of endothelial barrierfunction is independent of Ca2+ entryThe data suggested that Orai1 did not participate in the control of thrombin-induced decrease in TER, questioning the need for Ca2+ entry in this pro-cess. One could suggest that Orai1 knockdown was not 100% and that

0 1 2 3 4

0.6

0.7

0.8

0.9

1.0

1.1

1.2

TE

R(n

orm

aliz

ed)

Time (hours)

siControl

siOrai1

Thrombin (5 nM)

0 1 2 3 4

0.6

0.7

0.8

0.9

1.0

1.1

1.2

TE

R(n

orm

aliz

ed)

Time (hours)

siControl

siSTIM1

Thrombin (5 nM)

0.0

0.5

1.0

**si

Ora

i1

siS

TIM

1

Pro

tein

/b-a

ctin

siC

on

tro

l

**

0 1 2 3 4

0.50.60.70.80.91.01.11.2

Time (hours)

TE

R(n

orm

aliz

ed)

siControl

siSTIM1

siOrai1

Thrombin (100 nM)A B C

50

37

10075

siControlsiSTIM1

siOrai1

ST

Thrombin

IM1

Orai1

b-actin

– + – –+ +

Fig. 1. STIM1, but not Orai1, knockdowninhibits thrombin-induced decrease inTER of HUVECs. (A) Western blots showspecific knockdown of STIM1 and Orai1with siRNA. Graph shows the efficiencyof knockdown quantified from indepen-dent transfections (STIM1, n = 10; Orai1,n = 6). (B) Effect of STIM1 (n = 10) orOrai1 (n = 6) knockdown on the thrombin-induced decrease in TER. (C) Effects ofSTIM1 or Orai1 knockdown on the de-

w.SCIENCESIGNALING

crease in TER induced by a high concentration of thrombin (n = 3). n indicates the number of ECIS experiments from independenttransfections; cells from each transfection were assayed at least in triplicate. **P < 0.01.

.org 19 March 2013 Vol 6 Issue 267 ra18 2

Page 4

R E S E A R C H A R T I C L E

even a small amount of Ca2+ entering the endothelial cells might be suf-ficient to drive the decrease in TER. Removal or chelation of Ca2+ fromthe outside milieu as a means to assess the role of Ca2+ entry in endothe-lial barrier function is not practical because Ca2+ ions are required for celladhesion and maintenance of endothelial monolayer integrity, and results

ww

obtained from such protocol would be difficult to interpret. Therefore, weused an approach in which Ca2+ entry in response to thrombin wascompletely abrogated in HDMECs by physiological Na+-based bath solu-tions containing 2 mM Ca2+ and 10 µM Gd3+, which inhibits Ca2+ entrythrough SOCE triggered by thapsigargin [an inhibitor of the ER-localized

w.SCIENCESIGNALING.org 19 M

on March 19, 2013

stke.sciencemag.org

Dow

nloaded from

calcium ATPase (adenosine triphos-phatase)] or thrombin in endothelialcells (37). Preincubation of cells with10 µM Gd3+ inhibited Ca2+ entry in re-sponse to thrombin with no effect onCa2+ release. Ca2+ release was mea-sured in physiological Na+-based bathsolutions nominally free of Ca2+, andCa2+ entry was measured in the samecells with the same solutions subse-quently supplemented with 2 mMCa2+ (Fig. 3A and fig. S4). We mea-sured the TER of the HDMECmono-layer under basal conditions and inresponse to thrombin in the same pop-ulation of cells under the same con-ditions, and found that the cellsexhibited a reversible decrease inTER in response to thrombin. Underconditions in which Ca2+ entry wasinhibited by Gd3+ in HDMECs, the de-crease in TER produced by thrombinwas similar to the decrease in the ab-sence of Gd3+ (Fig. 3B), stronglyarguing that the thrombin-induced de-crease in TERwas independent of Ca2+

entry and thus SOCE.Because experiments with thapsi-

gargin to disrupt endothelial cellmonolayers were interpreted as evi-dence that SOCE is important fordecreasing endothelial barrier func-tion (32), we measured the TER ofHUVECs and HDMECs exposed tothapsigargin and incubated in Na+-based bath solutions containing 2 mMCa2+ in the absence or presence of10 µM Gd3+ to block SOCE. The ef-fects of thapsigargin on TER were dis-tinct from those of the physiologicalagonist thrombin. Initially, thapsigarginenhanced TER, followed by a steadydecrease in TER (Fig. 3C), which oc-curred in the presence or absence ofGd3+. To decipher the response tothapsigargin with increased time reso-lution, we measured TER at shorter in-tervals and found that thapsigargintriggered an initial, immediate, andsharp decrease in TER, which wasfollowed by a phase during whichTER increased (fig. S5A), in agreementwith Cioffi et al. (32). This early tran-sient decrease in TER that occurred inresponse to thapsigarginwas insensitive

0 5 10 15 20

1.0

1.5

2.0

2.5

siOrai1siSTIM1siControl

Thrombin (100 nM)

Ca2+

Rat

io34

0/38

0Time (min)

0.00.20.40.60.81.01.2

siO

rai1

siS

TIM

1

siC

on

tro

l

****

DR

atio

340/

380

–2.0

–1.5

–1.0

–0.5

0.0–100 0 100

**

pA

/pF

mV

* 0.0

0.2

0.4

0.6

0.8

1.0

Na+

CR

AC

(pA

/pF

)

siC

on

tro

l

siO

rai1

siS

TIM

1

******

*

DVF

5 mM Gd3+

20 mM BAPTA in pip

siControl

2 min0.2 pA/pF *

DVF DVF

5 mM Gd3+ 5 mM Gd3+20 mM BAPTA in pip

siSTIM1

*

20 mM BAPTA in pip

siOrai1

0.00.20.40.60.81.0

siO

rai1

siS

TIM

1

**

Pro

tein

/ b-a

ctin

**siC

on

tro

lsiControl

siSTIM1

siOrai1

STIM1

b-actin

Orai1

100 75

50

37

A B

C

D E

Fig. 2. STIM1 and Orai1 mediate SOCE and CRAC currents in HDMECs. (A) Western blots show knockdown of

STIM1 and Orai1 in HDMECs. Graph shows the efficiency of knockdown quantified from three independenttransfections. (B) Ca2+ imaging traces showing average from several cells assayed simultaneously from thesame coverslip of each experimental condition (nontargeting control siRNA, n = 11; siSTIM1, n = 10; siOrai1,n = 21). Quantification of Ca2+ entry in response to thrombin (plotted as the Fura2 ratio normalized to control) incells transfected with the indicated siRNA is shown (control, n = 6 coverslips, 105 cells; STIM1, n = 4 coverslips,53 cells; Orai1, n = 7 coverslips, 164 cells). (C) Whole-cell CRAC currents elicited by store depletion with20 mM BAPTA in the patch pipette (pip) from HDMECs transfected with the indicated siRNA. Currents weremeasured in bath solutions containing 20 mM Ca2+, DVF solution, or 5 µM Gd3+ as indicated. (D) I/V relation-ships of CRAC carried by Na+ in DVF solutions obtained with voltage ramps (from −140 to +150 mV) from tracesC, taken where indicated by the color-coded asterisks. (E) Graph shows the Na+ CRAC current densities(pA/pF) measured at −100 mV from HDMECs transfected with either nontargeting control siRNA (n = 4), STIM1siRNA (n = 5), or Orai1 siRNA (n = 5). **P < 0.01; ***P < 0.001.

arch 2013 Vol 6 Issue 267 ra18 3

Page 5

R E S E A R C H A R T I C L E

to 10 µM Gd3+ (fig. S5B), suggesting that it is independent of SOCE andCa2+ entry.

STIM1 functions independently of EB1 and focal adhesionkinase in the thrombin-induced decrease in TERBecause STIM1 and the plus-endmicrotubule-binding protein EB1 interact(42), we investigated whether the STIM1-EB1 interaction contributed toSTIM1-mediated loss of TER in response to thrombin. We performed“erase and replace” experiments in which endogenous STIM1was depletedfrom the HUVECs by siRNA followed by rescue with either wild-typeSTIM1 or STIM1 with a mutation in the EB1 binding domain (EB1 mut)

ww

on March 19, 2013

stke.sciencemag.org

Dow

nloaded from

that cannot interact with EB1 and thus cannot bind microtubules (42).Both versions of STIM1 rescued the thrombin-induced decrease in TER(Fig. 3D), which indicates that the interaction of STIM1 with EB1 wasnot essential for disruption of endothelial barrier function. STIM1knockdown significantly increased basal phosphorylation of focal adhesionkinase (FAK) and paxillin in HUVECs (fig. S6). Thrombin stimulation ofHUVECs in which STIM1 was knocked down caused a further increasein FAK and paxillin phosphorylation (fig. S6A). However, quantification ofdata from four independent experiments revealed that the thrombin-mediatedincrease in paxillin phosphorylation in STIM1knockdown cellswasmarginalcompared to control (fig. S6B).

STIM1 functions independently of TRPC channels in thethrombin-induced decrease in TERTRPC1 (32, 35), TRPC4 (30, 31, 33, 34), and TRPC6 (43–46) channelshave been proposed to play a role in endothelial barrier function, andSTIM1 stimulates TRPC channels through a mechanism involving electro-static interactions between two positively charged residues in the STIM1 Cterminus (Lys684, Lys685) and two negatively charged residues in TRPC1(Asp639, Asp640) (47). Individual knockdown of TRPC1, TRPC4, orTRPC6 significantly inhibited the thrombin-mediated decrease in TERin HUVECs, with TRPC4 and TRPC6 knockdown each showing a greatereffect than TRPC1 knockdown (Fig. 4A), which correlated with the dif-ferent efficiencies of TRPC isoform knockdown (fig. S7).

To determine whether the role of STIM1 in thrombin-mediated disrup-tion of endothelial barrier function was mediated through activation ofTRPC channels, we used an enhanced yellow fluorescent protein (eYFP)–tagged K684E, K685E mutant STIM1 (KK/EE) that does not supportelectrostatic interactions with TRPC channels but does activate Orai1channels (47). Overexpression of this STIM1 KK/EE in either HUVECs(Fig. 4, B and C) or HDMECs (Fig. 4, D and E) enhanced Ca2+ entryupon stimulation with either thrombin or thapsigargin, confirming thatSOCE in endothelial cells is mediated through STIM1 activation of Orai1channels. Erase and replace experiments in HUVECs showed that STIM1KK/EE rescued the thrombin-mediated decrease in TER in a manner sim-ilar to that of wild-type eYFP-STIM1 (Fig. 5, A and B). Thus, both TRPCchannels and STIM1 contributed to the control of endothelial barrierfunction, but the effects of STIM1 on TER involve other downstream tar-gets in addition to TRPC channels.

STIM1 contributes to the thrombin-induced decrease inTER through RhoA but not MLCKThe thrombin-induced decrease in TER in endothelial cell monolayers re-quires cytoskeletal rearrangements, including the rearrangement of the ac-tin cytoskeleton from peripheral actin to stress fibers and phosphorylationof MLC to mediate contraction of the actin cytoskeleton. Depletion ofSTIM1 prevented the thrombin-mediated increase in stress fibers contain-ing phosphorylated MLC in HUVECs (Fig. 6A), suggesting that STIM1is required for thrombin-induced stress fiber formation. Control cells hadmore stress fibers and substantial disruption of vascular endothelial cadherin(VE-cadherin) cell-cell junction in response to thrombin (Fig. 6A, white arrowsin panels 2 and 4) compared to the STIM1 knockdown cells. Furthermore,control cells had a more substantial staining and colocalization of actinstress fibers and phosphorylated MLC compared to STIM1 knockdowncells (Fig. 6A, panels 6 and 8).

Phosphorylation of MLC can be mediated by ROCK, a kinase activatedby RhoA (48), or by MLCK, a kinase activated by calcium-calmodulin(49). Because SOCE was not required for thrombin-induced reductionin TER in endothelial cell monolayers, we hypothesized that MLCK wouldnot be involved in this process because MLCK is activated by Ca2+ sig-

0 5 10 15 20

1.0

1.5

2.0

2.5

Gd3+

Ca2+

Thrombin (100 nM)

Rat

io34

0/38

0

Time (min)

Control

A

0.00

0.25

0.50

0.75

1.00

1.25

Gd3+

Co

ntr

ol

***

DR

atio

340/

380

B

0 1 2 3 4 5

0.40.50.60.70.80.91.01.11.2

H2O

Gd3+

+thrombin (5 nM)

Gd3+

+H2OT

ER

(no

rmal

ized

)

Time (hours)

Thrombin (5 nM)

C

0 1 2 3

0.7

0.8

0.9

1.0

1.1

1.2

TE

R(n

orm

aliz

ed)

Time (hours)

ControlGd3+

Thapsigargin (2 mM)

siControl siSTIM1 siSTIM1+EB1Mut. siSTIM1+STIM1

0 1 2 3 4 50.50.60.70.80.91.01.11.2

Time (hours)

TE

R(n

orm

aliz

ed)

Thrombin (5 nM)

lort

no

Cis siS

TIM

1

siS

TIM

1+.t

um 1

BE si

ST

IM1+

ST

IM1

STIM1

b-actin

10075

50

D

Fig. 3. Ca2+ entry is not required for thrombin-induced decrease in TER. (A)2+

Thrombin-induced SOCE in the presence or absence of the Ca entry in-

hibitor Gd3+ (10 µM). HDMECs were preincubated with Gd3+ (or control),and Ca2+ release was assayed in the absence of extracellular Ca2+, thenCa2+ entry was assayed in Na+-containing salt solutions supplementedwith 2 mM Ca2+. Quantification of calcium entry (plotted as the increaseof Fura2 ratio normalized to control) in the control or presence of Gd3+

(n = 3 experiments and 63 cells for control and n = 4 experiments and 98cells for Gd3+) is shown. (B) The same HDMECs depicted in (A) were as-sayed using ECIS to determine TER response to thrombin on the same daywith the same Na+-containing solutions supplemented with 2 mM Ca2+ andthe same Gd3+ stock that blocked SOCE in the Ca2+ imaging protocols. (C)The TER of HDMECs in response to thapsigargin was assayed in the ab-sence and presence of Gd3+ under the same buffer conditions as used forthe data shown in (B). (D) Western blots from erase and replace exper-iments show control, knockdown of STIM1, and expression of either wild-type STIM1 or EB1 mut. STIM1 in HUVECs. The TER of HUVECs in responseto thrombin was assayed in all four conditions shown. Data shown in (B)and (C) are representative of six independent experiments, and data in (D)are representative of four independent experiments. ***P < 0.001.

w.SCIENCESIGNALING.org 19 March 2013 Vol 6 Issue 267 ra18 4

Page 6

R E S E A R C H A R T I C L E

naling through Ca2+/calmodulin. Knockdown of MLCK did not preventthe thrombin-mediated decrease in HUVEC TER (Fig. 6B). We used twoindependent siRNAs targeting MLCK (siMLCK#2 and siMLCK#3; fig.S8A). A third siRNA targeting MLCK (siMLCK#1; fig. S8A) was less ef-ficient at reducing MLCK abundance and affected cell viability and, there-fore, was not used for TER experiments. Similar to HUVECs, we knockeddown MLCK in HDMECs with two independent siRNAs (fig. S8B), and thisfailed to affect the thrombin-induced decrease in TER in these endothelialcell monolayers (fig. S8C).

ww

To determine whether the thrombin-mediated reduction in TER thatinvolved STIM1 was mediated by RhoA, which could influence MLCphosphorylation through ROCK, we measured RhoA activity in thrombin-treated HUVECs in which STIM1, Orai1, or MLCK was knockeddown. Only STIM1 knockdown significantly reduced RhoA activityin response to thrombin (Fig. 6C), indicating that STIM1 functions asan upstream effector coupling the PAR-1 receptor to RhoA duringthrombin-mediated decrease in TER and disruption of endothelial barrierfunction.

w.SCIENCESIGNALING.org 19 M

on March 19, 2013

stke.sciencemag.org

Dow

nloaded from

DISCUSSION

In a previous study (37), we reportedthat STIM1 and Orai1 mediate SOCEand CRAC currents in HUVECs. Inthis study, we showed that SOCEand CRAC currents in HDMECswere also mediated by STIM1 andOrai1, indicating that the involve-ment of STIM1 and Orai1 in SOCEis not unique to HUVECs (41). UsingHUVEC and HDMEC monolayersand ECIS, we found that STIM1and Orai1 did not function togetherin thrombin-mediated disruption ofendothelial barrier function. STIM1was required for thrombin-mediateddecrease in TER of HUVECs andHDMECs, whereas Orai1 was not.The thrombin-induced reduction inTER independent of Orai1, along withthe thrombin TER response inde-pendent of Ca2+ entry, indicated thatSOCE was not involved in mediatingdisruption of endothelial barrier func-tion in response to the physiologicalagonist thrombin. These results agreewith previous studies indicating thatthe thrombin-induced disruption inendothelial barrier function is inde-pendent of Gq-mediated Ca2+ signals(39). However, the SOCE-independentnature of the thrombin response con-trasts with the conclusions of studiesthat used thapsigargin, which activatesSOCE, to disrupt endothelial barrierfunction (32, 50).

To explore this discrepancy, weused ECIS to monitor the natureand kinetics of the change in TERin endothelial monolayers exposedto thapsigargin stimulation and foundthat the effect of thapsigargin on TERwas different from that of the physio-logical agonist thrombin. Whereasthrombin caused a substantial decreasein TER that was reversible within 2 to3 hours, thapsigargin triggered morecomplex effects: Thapsigargin causeda sharp, low-amplitude, and brief

0 1 2 30.6

0.7

0.8

0.9

1.0

1.1

TE

R (

no

rmal

ized

)

TE

R (

no

rmal

ized

)

TE

R (

no

rmal

ized

)Time (hours)

siControl siTRPC1

Thrombin (5 nM)

0 1 2 30.6

0.7

0.8

0.9

1.0

1.1

Time (hours)

siControl siTRPC4

Thrombin (5 nM)

0 1 2 30.6

0.7

0.8

0.9

1.0

1.1

siControl siTRPC6

Time (hours)

Thrombin (5 nM)

0 5 10 15 20 25

1.0

1.5

2.0

2.5

3.0

Time (min)

Rat

io34

0/38

0

Ca2+ (2 mM)Thapsigargin (2 mM)

ControlKK/EE

0.00.51.01.52.02.53.03.5 **

TGD3

40/ 3

80

Co

ntr

ol

KK

/EE

0 5 10 15 20 25

1.0

1.5

2.0

2.5

KK/EE

Time (min)

Rat

io34

0/38

0

Control

Thrombin (100 nM)

Ca2+ (2 mM)

0.00.51.01.52.02.53.03.5 ***

Th

r.D 3

40/3

80

KK

/ EE

Co

ntr

ol

0 5 10 15 20 25

1.0

1.5

2.0

2.5

3.0

3.5

4.0

ControlKK/EE

Rat

io34

0/38

0

Time (min)

Thapsigargin (2 mM)

Ca2+ (2 mM)

0.0

0.5

1.0

1.5

2.0

2.5 **

KK

/EEC

on

tro

l

TGD3

40/3

80

0 5 10 15 20 25

1.0

1.5

2.0

2.5

3.0Thrombin (100 nM)

Time (min)

Rat

io34

0/3 8

0

KK/EEControl

Ca2+ (2 mM)

0.00.51.01.52.02.5 ***

Th

r.D3

40/3

80

Co

ntr

ol

KK

/EE

A

B C

D E

Fig. 4. TRPC channels and TER in response to thrombin. (A) TER in response to thrombin in HUVECs in which

the indicated siRNAs were expressed. Results are representative of three independent transfections (n = 3).(B) Effect of the STIM1 KK/EE mutant on SOCE when expressed in HUVECs exposed to thapsigargin. (C)Effect of the STIM1 KK/EE mutant on SOCE when expressed in HUVECs exposed to thrombin. (D) Effectof the STIM1 KK/EE mutant on SOCE when expressed in HDMECs exposed to thapsigargin. (E) Effect ofthe STIM1 KK/EE mutant on SOCE when expressed in HDMECs exposed to thrombin. Quantification of theincrease in calcium entry in (B) to (E) are from three to five independent experiments with a total of 17 to 99 cells.Data are expressed as the change in the fluorescence ratio in response to thapsigargin (TG) or thrombin (Thr.),normalized to control (untransfected cells). **P < 0.01; ***P < 0.001.

arch 2013 Vol 6 Issue 267 ra18 5

Page 7

Thr

ombi

n

2 4 6 8

R E S E A R C H A R T I C L E

(lasting less than 15 min) decrease in TER, followed by an increase inTER within 1 hour, and then a slow decrease in TER and loss of mono-layer integrity. Thapsigargin is an inhibitor of the ER calcium ATPase,which disrupts not only calcium signaling but also ER function, and theloss of monolayer integrity likely reflects the toxic nature of long-term ex-posure to thapsigargin. We found that the response of endothelial mono-layers to thapsigargin was the same when Ca2+ entry was blocked withlow concentrations of lanthanides, indicating that the thapsigargin-induced changes in TER were independent of Ca2+ entry and SOCE.

TRPC1, TRPC4, and TRPC6 channels have all been implicated in thecontrol of endothelial barrier function (30–35, 43–46). Our data showingthat TRPC1, TRPC4, or TRPC6 knockdown inhibited thrombin-induceddecrease in TER agree with these studies and highlight the importance ofTRPC channels in endothelial barrier function. However, we found thatTRPC channel–mediated disruption of endothelial barrier function wasindependent of Ca2+ entry and that the STIM1-mediated reductionin HUVEC or HDMEC TER that occurred in response to thrombinwas independent of the STIM1-TRPC interaction. Indeed, introduction into

ww

either HUVECs or HDMECs of a mutant version of STIM1 that cannotinteract with TRPC channels but can interact with Orai1 (47) failed to in-hibit but instead potentiated SOCE in response to thapsigargin and throm-bin. These data indicated that SOCE in endothelial cells is mediated byCRAC channels composed of STIM1 and Orai1, independently of TRPC

0.0

0.2

0.4

0.6

0.8

1.0

*Thrombin

Rh

oA

acti

vity

s iC

on

t ro

l

siS

TI M

1

siO

r ai 1

siM

LC

K

Co

nst

it. R

ho

A

No stimulus

***

siControl siSTIM1siControl siSTIM1

No

stim

ulus

1 3 5 7

0 1 2 3

0.6

0.7

0.8

0.9

1.0

1.1

1.2

TE

R(n

orm

aliz

ed)

Time (hours)

siControl siMLCK#2 siMLCK#3

Thrombin (5 nM)

CB

on March 19, 2013

stke.sciencemag.org

Dow

nloaded from

0 1 2 3 4

0.6

0.7

0.8

0.9

1.0

1.1

1.2

Time (hours)

TE

R (

no

rmal

ized

)

Thrombin (5 nM) siControl

siSTIM1

siSTIM1+KK/EE

siSTIM1+STIM1

A

B

eYFP-STIM1

50

75

100

150

siCon

trol

siSTI

M1

siSTI

M1+

KK/EE M

ut.

siSTI

M1+

YFP-S

TIM

1

STIM1

b-actin

Fig. 5. STIM1 controls TER in response to thrombin independently of TRPC

channels. (A) Western blots from erase and replace experiments in HUVECsshowing knockdown of endogenous STIM1 and expression of eYFP-taggedversions of STIM1: wild-type STIM1 (YFP-STIM1) or STIM1 KK/EE (KK/EE Mut.).(B) ECIS experiments from the same cells in (A) show the TER responses tothrombin in the indicated cells. Data are representative of three inde-pendent experiments from independent transfections performed on sep-arate days.

Phalloidin/ /VE-cadherin DAPI ppMLC/ /phalloidin DAPIA

Fig. 6. STIM1 controls thrombin-mediated decreasein TER through RhoA but not MLCK. (A) Immuno-fluorescence of HUVEC monolayers transfected

with either control nontargeting siRNA (panels 1, 2, 5, and 6) orSTIM1 siRNA (panels 3, 4, 7, and 8). Nonstimulated HUVECs and cells stimulated with thrombin (10 nM for 5 min) were stained with eitherphalloidin (green) and an antibody recognizing VE-cadherin (red; panels1 to 4) or phalloidin (red) and diphosphorylated (Thr18/Ser19) MLC (ppMLC;green). White arrows point to areas of disrupted VE-cadherin junctions.Data are representative of three independent transfections performed onseparate days. The number of disrupted VE-cadherin junctions wasdetermined as average ± SE of disrupted sites per field and are 9 ± 3.7 insiControl versus 1 ± 0.4 in siSTIM1; P < 0.05. (B) Effect of siRNA targetingMLCK on thrombin-induced reduction in TER (data are representative ofthree independent transfections performed on separate days). (C) RhoAactivity was measured in HUVECs transfected with the indicated siRNAin the presence or absence of thrombin (10 nM for 5 min). Cells expressingconstitutively active RhoA (Constit. RhoA) serve as a positive control. Re-sults are from three independent experiments representing three inde-pendent transfections performed on separate days. *P < 0.05; **P < 0.01.

w.SCIENCESIGNALING.org 19 March 2013 Vol 6 Issue 267 ra18 6

Page 8

R E S E A R C H A R T I C L E

on March 19, 2013

stke.sciencemag.org

Dow

nloaded from

channels, and that TRPC channels were not responsible for the STIM1-mediated response.

In vitro Förster resonance energy transfer between coimmunoprecipi-tated proteins, coimmunoprecipitation assays, and electrophysiologicalstudies in pulmonary artery endothelial cells in which Orai1 was knockeddown suggested that Orai1 interacts with heteromeric TRPC1 and TRPC4channels in endothelial cells to confer their Ca2+ selectivity (32). However,the electrophysiological data (32) are difficult to interpret because the con-ditions used would have caused store depletion, which would activateSOCE through Orai1 and currents through TRPC channels in response toCa2+ released from the ER or Ca2+ entry through Orai1 (36). The effect ofOrai1 knockdown onwhole-cell currents is consistent with the activation ofseveral conductances that would be detected under the conditions used. Thefinding that at high Ca2+ concentrations the current is increased only whenOrai1 is present can also be explained by the ability of Ca2+ to block Orai1 atlowexternal Ca2+ concentrations (51, 52). Therefore, the data of Cioffi et al.(32) are fully compatible with the presence of two independent, but closelyassociated, channels in endothelial cells: a store depletion–activated Ca2+-selective channel composed of Orai1 and a Ca2+-activated nonselective chan-nel composed of TRPC. Indeed, both Orai1 and TRPC are activated byphospholipase C–coupled agonists (53, 54), and local Ca2+ entry throughOrai1 stimulates TRPC1 and TRPC5 channel activity (55, 56).

How TRPC channels couple to endothelial barrier function remains apuzzle. TRPC channels may contribute to a Ca2+ signal, may function asadaptor proteins, or may trigger membrane depolarization by mediatingNa+ entry. Membrane depolarization activates RhoA and disrupts barrierfunction (57). Indeed, we identified activation of RhoA as a potentialmechanism for the STIM1-mediated contribution to thrombin-mediateddisruption of endothelial barrier function. Consistent with ROCK as a ki-nase that phosphorylates MLC (48), we found that STIM1 knockdowncaused increased basal phosphorylation of FAK and paxillin and inhibitedRhoA activity, MLC phosphorylation, and actin stress fiber formation inresponse to thrombin, providing a mechanism for STIM1 action.

Unexpectedly, we found that knockdown of MLCK failed to affect thereduction in TER that occurred in response to thrombin. Despite effectiveMLCK knockdown, the residual amounts of MLCK in our cells might besufficient to decrease TER in response to thrombin, or the studies previ-ously implicating MLCK may have alternative interpretations. Earlier workthat implicated MLCK in endothelial barrier function was based on theuse of pharmacological inhibitors, such as ML7 and ML9, or peptide in-hibitors (49, 58, 59), and the results might reflect the nonspecific nature ofthe MLCK inhibitors and peptides used in these studies. ML9 efficientlyblocks STIM1 movement and reorganization in response to store deple-tion, and this effect is independent of MLCK inhibition (60). Severalstudies have implicated MLCK as contributing to disruption of endothelialbarrier function. Introduction of constitutively active MLCK into coronaryvenular endothelial cells increased phosphorylation of MLC and increasedthe transendothelial flux of albumin across endothelial monolayers (58).Mice with an endothelial-specific knockout of MLCK (MLCK210) havereduced microvascular hyperpermeability in models of atherosclerosis or inresponse to severe burns (61, 62). However, another study with endothelial-specific MLCK knockout mice found that the increase of microvascularendothelial permeability in response to lipopolysaccharides was not depen-dent on MLCK (63), in agreement with our present study. The reasons forthese discrepancies are unclear, but it seems likely that the involvement ofMLCK in regulating endothelial permeability depends on the stimulationor endothelial bed considered.

There is a minor pool of STIM1 at the plasma membrane (~10 to 15%of total STIM1) that was proposed to play a role in the activation of thestore-independent arachidonate-regulated Ca2+ (ARC) channels (64). Our

ww

data show that the effects of STIM1 on TER responses to thrombin couldbe rescued by eYFP-tagged STIM1, a fusion protein that does not reachthe plasma membrane (65), suggesting that the STIM1 in the ER is likelymediating the effects on RhoA and TER in response to thrombin. An im-portant question that future studies should address is, how does STIM1couple PAR-1 receptor to RhoA activation? Erase and replace ECIS ex-periments with STIM truncation mutants may determine the region ofSTIM1 necessary for the reduction in TER in response to thrombin. Thisregion of STIM1 could be subsequently used to identify additional bind-ing partners and signaling molecules using affinity purification or yeasttwo-hybrid studies. This knowledge will likely reveal STIM1 partnersand shed light on the molecular mechanisms controlling endothelial bar-rier function.

MATERIALS AND METHODS

ReagentsThe BCA Protein Assay Kit, SuperSignal West Pico, West Femto, and Re-store Western Blot Stripping Buffer were from Pierce Biotechnology Inc.The QuickChange II Site-Directed Mutagenesis Kit was purchased fromStratagene. iQ SYBR Green Supermix was from Bio-Rad, and RNeasyMini Kit and QIAshredder were from Qiagen. Gd3+ was purchased fromAcros Organics; thapsigargin was from Calbiochem. Na-methanesulfonate,Cs-methanesulfonate, and thrombin were purchased from Sigma. Cs-BAPTAwas from Invitrogen. siRNA sequences for MLCK were obtainedfrom Invitrogen, and all other siRNA sequences were from Dharmacon.Specific primers for STIM1, the STIM1 EB1 mutant, with point mutationsIle644 to Asn and Pro645 to Asn, and MLCK were synthesized by IntegratedDNATechnologies. All other chemical products were obtained from FisherScientific unless specified otherwise. The antibodies used for Western blotsand immunofluorescence, as well as the source, species, and dilutions used,are listed in table S1.

Cell cultureHUVECs were purchased from Cascade Biologics (Invitrogen). HUVECswere grown in endothelial cell basal medium 2 (EBM-2) BulletKit (CC-3156and CC-4176) (Lonza Walkersville Inc.) containing 2% fetal bovine serum(FBS). HDMECs were isolated from neonatal foreskins (obtained fromthe maternity ward of the Albany Medical Center Hospital with InstitutionalReview Board approval) by incubating the obtained cells with magneticbeads coated with an antibody to CD31 (Dynal) as described previously(66). Cells were assessed for VE-cadherin and lectin binding to confirmtheir endothelial origin and then frozen in liquid nitrogen. HDMECs weregrown inEGM-2MVBulletKit (CC-3156 andCC-4147) (LonzaWalkersvilleInc.) containing 5%FBS. For all experiments, HUVECs andHDMECswerebetween passages 2 and 6.

TER experimentsChanges in TER, which is a measure of endothelial barrier integrity ofHUVEC and HDMEC monolayers, were determined with an electric cell-substrate impedance sensing apparatus (ECIS, Applied BioPhysics Inc.),as described in detail previously (38). Briefly, HUVECs or HDMECs(1.2 × 105 cells) transfected with either nontargeting control siRNA, siRNAtargeting specific proteins, or STIM1 plasmids or lentiviruses were seededonto ECIS 10E culture ware (0.8 cm2 per well) precoated with 0.2% gelatin,and the cells were incubated for 3 days. The cells were serum-starved for4 hours. The electrical impedance across the monolayer was measured at1V, 4000Hzwith current flowing through 10 small gold electrodes per well

w.SCIENCESIGNALING.org 19 March 2013 Vol 6 Issue 267 ra18 7

Page 9

R E S E A R C H A R T I C L E

on March 19, 2013

stke.sciencemag.org

Dow

nloaded from

plus one large counter electrode, with the culture medium as the source ofelectrolytes. Once resistances were relatively constant, stimuli (thrombin,thapsigargin) were added directly to thewells. Impedancewasmonitoredby the lock-in amplifier, stored, and resistance and capacitance were cal-culated with the manufacturer’s software. For comparing experimentalconditions, data were normalized to the mean resistance of each conditiononce the monolayers had a constant resistance immediately before stimulusaddition.

For ECIS experiments using preincubation with 10 µM Gd3+ and thecorresponding controls, the Na+-based medium was adapted fromWaheedet al. (57) with a slight modification. The composition was as follows: 130 mMNaCl, 3 mM KCl, 1 mM MgCl2, 2 mM CaCl2, 5 mM glucose, and20 mM Na-Hepes (pH 7.4).

Immunofluorescence assaysHUVECswere transfectedwith control and STIM1 siRNAs. Cells (1.2 × 105

cells per well) were seeded on BioCoat eight-well glass culture slides(BD Falcon) precoated with 0.2% gelatin (Acros Organics). Two dayslater, cells were serum-starved in EBM-2 supplemented only with 0.3% FBSfor 16 hours before adding 10 nM thrombin for 5 min. The cells were fixedfor 30 min with 4% paraformaldehyde and processed for immuno-fluorescence with antibodies recognizing VE-cadherin or phosphorylatedMLC at Thr18 and Ser19 (67) (see table S1 for details) and with Alexa Fluor594– or Alexa Fluor 647–conjugated secondary antibodies. F-actin wasstainedwith eitherAlexa Fluor 488– orAlexa Fluor 594–conjugated phalloidin,and nuclei were stained with DAPI (4',6-diamidino-2-phenylindole).Imageswere obtainedwith a Zeiss Axio Observer Z1 inverted microscopewith the ApoTome module and Zeiss AxioVision software.

siRNA transfectionsSeveral different siRNAs per target gene were initially assessed for theirability to reduce mRNA abundance by quantitative reverse transcriptionpolymerase chain reaction (the primers used for each target mRNA andthe different siRNA sequences used in the study are listed in table S2).siRNA sequences that induced significant decreases in their target mRNA(more than 70%) without affecting other mRNAs were used, and knock-down efficiency was confirmed by Western blotting. All transfections inHUVECs and HDMECs were performed with the Nucleofector II Device(Amaxa Biosystems) with program A-034 or M-003, respectively, accord-ing to the manufacturer’s instructions. Green fluorescent protein (GFP;0.5 mg) was cotransfected with siRNA for identification of successfullytransfected cells and to serve as a control for the YFP- or GFP-tagged plas-mid constructs used in transfections. The efficiency of siRNA and plasmidtransfections typically exceeded 95 and 75%, respectively.

Production of lentivirusesPrecision LentiORF individual clone for STIM1 (STIM1 pLOC, acces-sion no. DQ895509) was from Open Biosystems. This lentiviral vectorhas nuclear-localized TurboGFP. In this system, because of a 2a self-cleaving peptide, both STIM1 and TurboGFP are translated as two distinctproteins. This lentiviral STIM1 ORF and STIM1 EB1 mut were used toproduce lentiviruses. Viral particles were generated following standardprotocols. PolyJet was used as a transfection reagent (SignaGen) to trans-fect human embryonic kidney (HEK) 293FT cells (Invitrogen). Briefly,the lentiviral constructs pCMV-VSVG, pCMV-dR8.2, and either STIM1pLOC or STIM1 pLOC STIM1 EB1mut were cotransfected into a flask of95% confluent HEK293FT cells. Cell culture mediawith viral particles werecollected at 48 and 72 hours after transfection andwere concentratedwith anAmicon Ultra-15 filter (Millipore) by centrifugation and stored at −80°C.For in vitro infections, cells were typically assayed 7 days after infection.

ww

HUVECs were infected at 50% confluence in EBM-2 supplementedwith 5% heat-inactivated FBS. The medium was changed to growth me-dium, and the infected cells were allowed to grow to confluence beforetrypsinization and reseeding for experiments. Infection efficiency wasevaluated by visualization of nuclear GFP in live cells.

Erase and replace assaysHUVECs (1 × 106) were electroporated with either control siRNA orSTIM1 siRNA (20 µg) and incubated for 72 hours. Cells were then de-tached and electroporated again with siRNA along with plasmids encodinghuman STIM1 or STIM1KK/EE mutant. Cells were allowed to recover foran additional 48 hours, after which the cells were serum-starved in MCDB131 medium (12 hours) (Gibco, Invitrogen). Cells were stimulated withthrombin (5 nM), and TER was measured by ECIS. Cells were platedin parallel dishes to verify protein production from the STIM1 plasmidsbyWestern blots or by fluorescence microscopy (for eYFP-STIM1 plasmids).For lentiviral transfections, HUVECs (1 × 106) were electroporated witheither control siRNA or STIM1 siRNA (20 µg) and seeded to <50% con-fluence in six-well dishes. On the following day, these cells were infectedwith lentiviral particles encoding either STIM1 or STIM1 EB1 mut andallowed to recover for 3 days. The presence of nuclear TurboGFP servedas a gauge of the efficiency of infection. On the fourth day, the cells wereelectroporated again with either control or STIM1 siRNAs and seededonto the ECIS plates. Cells were allowed to recover for an additional48 hours, after which the cells were serum-starved in MCDB 131 medium(12 hours) (Gibco, Invitrogen). After serum starvation, cells were treatedwith thrombin (5 nM) and TER was measured. Cells were plated in par-allel dishes to verify protein expression of STIM1 constructs by Westernblots and fluorescence microscopy.

Calcium imagingHUVECs and HDMECs were cultured on 30-mm glass coverslips for Ca2+

imaging as previously described (37). Coverslips with cells attached weremounted in a Teflon chamber. Cells were incubated (37°C, 25 to 30 min)in culture medium containing 4 mM Fura2/AM. Upon completion of in-cubation, cells were washed three times and bathed in Hepes-buffered sa-line solution [140 mM NaCl, 1.13 mM MgCl2, 4.7 mM KCl, 2 mM CaCl2,10 mM D-glucose, and 10 mM Hepes (pH 7.4)] for at least 5 min before Ca2+

measurements were made. For Ca2+ measurements, cells were stimulatedby either thrombin or thapsigargin at the concentrations indicated in anominally Ca2+-free solution, and Ca2+ was restored to the extracellularmilieu after the Ca2+ release phase was complete. This protocol separatesdetection of intracellular Ca2+ release from Ca2+ entry. A digital fluores-cence imaging system (InCyt Im2, Intracellular Imaging Inc.) was used formeasurement of Fura2 fluorescence signal, and fluorescence images ofseveral cells were recorded and analyzed simultaneously. The 340/380 nmratio images were obtained on a pixel-by-pixel basis. Figures showingCa2+ traces are an average from several cells attached on one coverslipand are representative of several independent recordings as mentionedin the figure legends. For the Ca2+ measurements depicted in Fig. 3, thebath solutions used were the Na+-based solutions (with and without 2 mMCa2+; see composition under the “TER experiments” section) instead ofregular Hanks’ balanced salt solution (HBSS).

Western blottingCells were transfected with either control nontargeting siRNA or STIM1,Orai1, or MLCK-specific siRNAs. Three days after transfection, cells wererinsed and lysed on ice in standard radioimmunoprecipitation assay (RIPA)lysis buffer [50 mM tris-HCl (pH 8), 150 mMNaCl, 1% Triton X-100, 0.2 mMEDTA, 0.1% SDS, 0.5% sodium deoxycholate, 2 mM phenylmethylsulfonyl

w.SCIENCESIGNALING.org 19 March 2013 Vol 6 Issue 267 ra18 8

Page 10

R E S E A R C H A R T I C L E

on March 19, 2013

stke.sciencemag.org

Dow

nloaded from

fluoride, 10% protease inhibitor cocktail (Roche), 10% phosphatase in-hibitor cocktail (Roche)], and protein concentrations were determinedwith the BCA Assay Reagent (Pierce). Lysates (50 to 200 µg) were sub-jected to SDS–polyacrylamide gel electrophoresis and transferred ontoImmuno-Blot polyvinylidene difluoride membranes (Bio-Rad). The mem-branes were blocked [5% nonfat milk in tris-buffered saline (TBS)–Tween,4°C overnight], incubated with antibodies (5% nonfat milk in TBS-Tween,4°C overnight), washed, and then incubated with horseradish peroxidase(HRP)–linked secondary antibodies (2% nonfat milk in TBS-Tween for1 hour). Bound antibodies were detected by enhanced chemiluminescencewith SuperSignalWest Pico or Femto reagents (Pierce). Signal intensitywasmeasured with a Fuji LAS4000 Imaging Station. Membranes were thenstripped and reprobedwith an antibody againstb-actin toverify equal loading,and densitometric analysis was performed with ImageJ software. For phos-phorylated FAK and phosphorylated paxillin experiments, cells were serum-starved (6 hours), then placed on a platewarmer (37°C) and bathed inHBSS(with 2mMCa2+ and 10mMHepes) followedby thrombin addition (10 nM).The disheswere then transferred to ice and lysed inRIPA lysis buffer (0.2mlof lysis buffer per 35-mm dish) after various time points. Lysates werecollected into ice-cold 1.5-ml tubes, cleared by centrifugation at 14,000 rpmin a microfuge at 4°C for 10 min, snap-frozen in liquid nitrogen, and storedat −80°C until use.

RhoA activity assaysRhoA activity was measured with a Rho G-LISA Activation Assays Bio-chem Kit (Cytoskeleton Inc.) according to the manufacturer’s recommen-dations. HUVECs were electroporated with siRNAs, seeded at confluence,and grown for three more days to form mature confluent monolayers in35-mm dishes. Monolayers were washed twice at room temperature withEBM-2 medium and incubated (4 hours) before stimulation in HBSS(containing 2 mM Ca2+ and 10 mM Hepes) with thrombin (5 nM). After5 min, solutions were aspirated, and cell lysis buffer (4°C) was added toculture dishes placed on ice. Cell lysates were centrifuged (14,000 rpm4°C, 2 min in a microfuge), and an aliquot was removed for determinationof protein concentration by BCA Protein Assay Kit (Pierce). Followingadjustment for protein concentration, cell lysates were snap-frozen andstored at −80°C. After thawing, 50 ml of lysate was added to the wellsof plates coated with RhoA binding domain peptides. Additional wellswere filled with lysis buffer or constitutively active RhoA as negativeand positive controls, respectively. Plates were placed immediately onan orbital shaker (400 rpm, 4°C, 30 min) and washed twice with HBSS;antigen-presenting buffer (200 µl) was added to each well (2 min at roomtemperature). After three washes, diluted RhoA primary antibody (50 µl)was added (45 min with shaking). After three washes, secondary HRP-labeledantibody (50 µl, diluted 1:100) was added to each well (45 min), followedby three washes and the addition of HRP detection reagent (37°C, 15 min).Finally, HRP stop buffer (50 ml) was added, and the signal was measuredimmediately at 490 nm with a microplate spectrophotometer.

Whole-cell patch clamp experimentsWhole-cell patch clamp recordings were carried out with Axopatch 200Band Digidata 1440A (Axon Instruments) as previously published (68–70).Clampfit 10.1 software was used for data analysis. Pipettes were pulledfrom borosilicate glass capillaries (World Precision Instruments Inc.) witha P-97 Flaming/Brown micropipette puller (Sutter Instrument Company)and polished with DMF1000 (World Precision Instruments Inc.) to aresistance of 2 to 4 MW when filled with pipette solutions. HDMECs wereelectroporated with STIM1, Orai1, or control siRNAs 3 days before re-cordings. Cells were seeded on round coverslips 36 hours before experi-ments. Immediately before the experiments, cells were washed with bath

ww

solution. Only cells with tight seals (>16 GW) were selected to break in.Cells were maintained at a 0-mV holding potential during experimentsand subjected to voltage ramps from +150 to −140 mV lasting 250 ms,every 2 s. “Reverse” ramps were designed to inhibit Na+ channels presentin these cells. Nimodipine (3 µM) was added to the bath solution to gen-erally stabilize membrane patches and reach better seals. HighMgCl2 (8 mM)was included in the patch pipette to inhibit TRPM7 currents.

Bath solution: 135 mM Na-methanesulfonate, 10 mM CsCl, 1.2 mMMgSO4, 10 mM Hepes, 20 mM CaCl2, and 10 mM glucose (pH wasadjusted to 7.4 with NaOH).

Pipette solution: 145 mM Cs-methanesulfonate, 20 mM Cs-BAPTA,8 mM MgCl2, and 10 mM Hepes (pH adjusted to 7.2 with CsOH).

DVF bath solution: 155 mM Na-methanesulfonate, 10 mM HEDTA,1 mM EDTA, and 10 mM Hepes (pH 7.4, adjusted with NaOH).

Statistical analysisData are expressed as means ± SE, and statistical analysis using one-wayanalysis of variance was done with Origin 8.5 software (OriginLab).Throughout the text, figures, and figure legends *, **, and *** indicateP values of <0.05, <0.01, and <0.001, respectively. Differences were con-sidered significant when P < 0.05.

SUPPLEMENTARY MATERIALSwww.sciencesignaling.org/cgi/content/full/6/267/ra18/DC1Fig. S1. Effect of different STIM1 siRNAs on the thrombin-mediated decrease in TER inHUVECs.Fig. S2. STIM1 knockdown does not affect basal endothelial monolayer resistance.Fig. S3. STIM1 knockdown, but not Orai1 knockdown, inhibited the decrease in TER ofHDMECs in response to thrombin.Fig. S4. Low concentrations of Gd3+ completely abrogate Ca2+ entry in response to thrombinin HDMECs.Fig. S5. The effects of thapsigargin on TER are distinct from those of thrombin and areSOCE-independent.Fig. S6. STIM1 knockdown in HUVECs increased basal phosphorylation of FAK and paxillin.Fig. S7. Effectiveness of TRPC1, TRPC4, or TRPC6 knockdown in HUVECs.Fig. S8. MLCK knockdown in HUVECs and HDMECs has no effect on the thrombin-induceddecrease in TER.Table S1. Antibodies used in the study with corresponding dilutions.Table S2. Sequences of the primers and siRNA sequences used throughout the study.

REFERENCES AND NOTES1. S. R. Coughlin, Protease-activated receptors in hemostasis, thrombosis and vascular

biology. J. Thromb. Haemost. 3, 1800–1814 (2005).2. J. F. Vanhauwe, T. O. Thomas, R. D. Minshall, C. Tiruppathi, A. Li, A. Gilchrist, E. J. Yoon,

A. B. Malik, H. E. Hamm, Thrombin receptors activate Go proteins in endothelial cells to reg-ulate intracellular calcium and cell shape changes. J. Biol. Chem. 277, 34143–34149 (2002).

3. A. B. Malik, J. W. Fenton II, Thrombin-mediated increase in vascular endothelial per-meability. Semin. Thromb. Hemost. 18, 193–199 (1992).

4. V. Spindler, N. Schlegel, J. Waschke, Role of GTPases in control of microvascularpermeability. Cardiovasc. Res. 87, 243–253 (2010).

5. S. Terry, M. Nie, K. Matter, M. S. Balda, Rho signaling and tight junction functions.Physiology 25, 16–26 (2010).

6. T. Kozasa, X. Jiang, M. J. Hart, P. M. Sternweis, W. D. Singer, A. G. Gilman, G. Bollag,P. C. Sternweis, p115 RhoGEF, a GTPase activating protein for Ga12 and Ga13. Science280, 2109–2111 (1998).

7. C. Murga, S. Fukuhara, J. S. Gutkind, Novel molecular mediators in the pathway con-necting G-protein-coupled receptors to MAP kinase cascades. Trends Endocrinol.Metab. 10, 122–127 (1999).

8. S. Fukuhara, C. Murga, M. Zohar, T. Igishi, J. S. Gutkind, A novel PDZ domaincontaining guanine nucleotide exchange factor links heterotrimeric G proteins toRho. J. Biol. Chem. 274, 5868–5879 (1999).

9. M. Aittaleb, C. A. Boguth, J. J. Tesmer, Structure and function of heterotrimeric Gprotein-regulated Rho guanine nucleotide exchange factors. Mol. Pharmacol. 77,111–125 (2010).

10. A. P. Somlyo, A. V. Somlyo, Ca2+ sensitivity of smooth muscle and nonmuscle myosinII: Modulated by G proteins, kinases, and myosin phosphatase. Physiol. Rev. 83,1325–1358 (2003).

w.SCIENCESIGNALING.org 19 March 2013 Vol 6 Issue 267 ra18 9

Page 11

R E S E A R C H A R T I C L E

on March 19, 2013

stke.sciencemag.org

Dow

nloaded from

11. E. Vandenbroucke, D. Mehta, R. Minshall, A. B. Malik, Regulation of endothelial junctionalpermeability. Ann. N. Y. Acad. Sci. 1123, 134–145 (2008).

12. Q. Shen, R. R. Rigor, C. D. Pivetti, M. H. Wu, S. Y. Yuan, Myosin light chain kinase inmicrovascular endothelial barrier function. Cardiovasc. Res. 87, 272–280 (2010).

13. D. L. Cioffi, T. Stevens, Regulation of endothelial cell barrier function by store-operatedcalcium entry. Microcirculation 13, 709–723 (2006).

14. S. M. Dudek, J. G. Garcia, Cytoskeletal regulation of pulmonary vascular permeability.J. Appl. Physiol. 91, 1487–1500 (2001).

15. C. Tiruppathi, G. U. Ahmmed, S. M. Vogel, A. B. Malik, Ca2+ signaling, TRP channels,and endothelial permeability. Microcirculation 13, 693–708 (2006).

16. J. W. Putney Jr., A model for receptor-regulated calcium entry. Cell Calcium 7, 1–12(1986).

17. J. W. Putney Jr., Capacitative calcium entry revisited. Cell Calcium 11, 611–624 (1990).18. M. Hoth, R. Penner, Depletion of intracellular calcium stores activates a calcium current in

mast cells. Nature 355, 353–356 (1992).19. I. Derler, R. Schindl, R. Fritsch, C. Romanin, Gating and permeation of Orai channels.

Front. Biosci. 17, 1304–1322 (2012).20. A. Engh, A. Somasundaram, M. Prakriya, Permeation and gating mechanisms in

store-operated CRAC channels. Front. Biosci. 17, 1613–1626 (2012).21. S. Feske, Y. Gwack, M. Prakriya, S. Srikanth, S. H. Puppel, B. Tanasa, P. G. Hogan,

R. S. Lewis, M. Daly, A. Rao, A mutation in Orai1 causes immune deficiency by abrogatingCRAC channel function. Nature 441, 179–185 (2006).

22. M. Vig, C. Peinelt, A. Beck, D. L. Koomoa, D. Rabah, M. Koblan-Huberson, S. Kraft,H. Turner, A. Fleig, R. Penner, J. P. Kinet, CRACM1 is a plasma membrane proteinessential for store-operated Ca2+ entry. Science 312, 1220–1223 (2006).

23. J. Roos, P. J. DiGregorio, A. V. Yeromin, K. Ohlsen, M. Lioudyno, S. Zhang, O. Safrina,J. A. Kozak, S. L. Wagner, M. D. Cahalan, G. Veliçelebi, K. A. Stauderman, STIM1, anessential and conserved component of store-operated Ca2+ channel function. J. Cell Biol.169, 435–445 (2005).

24. J. Liou, M. L. Kim, W. D. Heo, J. T. Jones, J. W. Myers, J. E. Ferrell Jr., T. Meyer,STIM is a Ca2+ sensor essential for Ca2+-store-depletion-triggered Ca2+ influx. Curr. Biol.15, 1235–1241 (2005).

25. M. Potier, M. Trebak, New developments in the signaling mechanisms of the store-operatedcalcium entry pathway. Pflugers Arch. 457, 405–415 (2008).

26. J. W. Putney, Capacitative calcium entry: From concept to molecules. Immunol. Rev.231, 10–22 (2009).

27. P. G. Hogan, R. S. Lewis, A. Rao, Molecular basis of calcium signaling in lymphocytes:STIM and ORAI. Annu. Rev. Immunol. 28, 491–533 (2010).

28. J. P. Yuan, W. Zeng, M. R. Dorwart, Y. J. Choi, P. F. Worley, S. Muallem, SOAR andthe polybasic STIM1 domains gate and regulate Orai channels. Nat. Cell Biol. 11,337–343 (2009).

29. C. Y. Park, P. J. Hoover, F. M. Mullins, P. Bachhawat, E. D. Covington, S. Raunser,T. Walz, K. C. Garcia, R. E. Dolmetsch, R. S. Lewis, STIM1 clusters and activatesCRAC channels via direct binding of a cytosolic domain to Orai1. Cell 136, 876–890(2009).

30. C. Tiruppathi, M. Freichel, S. M. Vogel, B. C. Paria, D. Mehta, V. Flockerzi, A. B. Malik,Impairment of store-operated Ca2+ entry in TRPC4−/− mice interferes with increase inlung microvascular permeability. Circ. Res. 91, 70–76 (2002).

31. D. L. Cioffi, S. Wu, M. Alexeyev, S. R. Goodman, M. X. Zhu, T. Stevens, Activation ofthe endothelial store-operated ISOC Ca2+ channel requires interaction of protein 4.1with TRPC4. Circ. Res. 97, 1164–1172 (2005).

32. D. L. Cioffi, S. Wu, H. Chen, M. Alexeyev, C. M. St Croix, B. R. Pitt, S. Uhlig, T. Stevens,Orai1 determines calcium selectivity of an endogenous TRPC heterotetramer channel.Circ. Res. 110, 1435–1444 (2012).

33. S. Wu, E. A. Cioffi, D. Alvarez, S. L. Sayner, H. Chen, D. L. Cioffi, J. King, J. R. Creighton,M. Townsley, S. R. Goodman, T. Stevens, Essential role of a Ca2+-selective, store-operated current (ISOC) in endothelial cell permeability: Determinants of the vascularleak site. Circ. Res. 96, 856–863 (2005).

34. M. Freichel, S. H. Suh, A. Pfeifer, U. Schweig, C. Trost, P. Weissgerber, M. Biel, S. Philipp,D. Freise, G. Droogmans, F. Hofmann, V. Flockerzi, B. Nilius, Lack of an endothelial store-operated Ca2+ current impairs agonist-dependent vasorelaxation in TRP4−/− mice.Nat. Cell Biol. 3, 121–127 (2001).

35. G. H. Brough, S. Wu, D. Cioffi, T. M. Moore, M. Li, N. Dean, T. Stevens, Contributionof endogenously expressed Trp1 to a Ca2+-selective, store-operated Ca2+ entrypathway. FASEB J. 15, 1727–1738 (2001).

36. M. Trebak, STIM1/Orai1, ICRAC, and endothelial SOC. Circ. Res. 104, e56–e57 (2009).37. I. F. Abdullaev, J. M. Bisaillon, M. Potier, J. C. Gonzalez, R. K. Motiani, M. Trebak,

Stim1 and Orai1 mediate CRAC currents and store-operated calcium entry importantfor endothelial cell proliferation. Circ. Res. 103, 1289–1299 (2008).

38. I. Giaever, C. R. Keese, Micromotion of mammalian cells measured electrically. Proc. Natl.Acad. Sci. U.S.A. 88, 7896–7900 (1991).

39. J. N. McLaughlin, L. Shen, M. Holinstat, J. D. Brooks, E. Dibenedetto, H. E. Hamm,Functional selectivity of G protein signaling by agonist peptides and thrombin for theprotease-activated receptor-1. J. Biol. Chem. 280, 25048–25059 (2005).

www

40. Z. Wang, R. Ginnan, I. F. Abdullaev, M. Trebak, P. A. Vincent, H. A. Singer, Calcium/calmodulin-dependent protein kinase II delta 6 (CaMKIId6) and RhoA involvement in thrombin-induced endothelial barrier dysfunction. J. Biol. Chem. 285, 21303–21312 (2010).

41. P. C. Sundivakkam, M. Freichel, V. Singh, J. P. Yuan, S. M. Vogel, V. Flockerzi, A. B. Malik,C. Tiruppathi, The Ca2+ sensor stromal interaction molecule 1 (STIM1) is necessary andsufficient for the store-operated Ca2+ entry function of transient receptor potential canon-ical (TRPC) 1 and 4 channels in endothelial cells. Mol. Pharmacol. 81, 510–526 (2012).

42. I. Grigoriev, S. M. Gouveia, B. van der Vaart, J. Demmers, J. T. Smyth, S. Honnappa,D. Splinter, M. O. Steinmetz, J. W. Putney Jr., C. C. Hoogenraad, A. Akhmanova,STIM1 is a MT-plus-end-tracking protein involved in remodeling of the ER. Curr. Biol.18, 177–182 (2008).

43. I. Singh, N. Knezevic, G. U. Ahmmed, V. Kini, A. B. Malik, D. Mehta, Gaq-TRPC6-mediated Ca2+ entry induces RhoA activation and resultant endothelial cell shapechange in response to thrombin. J. Biol. Chem. 282, 7833–7843 (2007).

44. V. Kini, A. Chavez, D. Mehta, A new role for PTEN in regulating transient receptorpotential canonical channel 6-mediated Ca2+ entry, endothelial permeability, and an-giogenesis. J. Biol. Chem. 285, 33082–33091 (2010).

45. R. Samapati, Y. Yang, J. Yin, C. Stoerger, C. Arenz, A. Dietrich, T. Gudermann, D. Adam,S. Wu, M. Freichel, V. Flockerzi, S. Uhlig, W. M. Kuebler, Lung endothelial Ca2+ andpermeability response to platelet-activating factor is mediated by acid sphingomyelinaseand transient receptor potential classical 6. Am. J. Respir. Crit. Care Med. 185, 160–170(2012).

46. N. Weissmann, A. Sydykov, H. Kalwa, U. Storch, B. Fuchs, M. Mederos y Schnitzler,R. P. Brandes, F. Grimminger, M. Meissner, M. Freichel, S. Offermanns, F. Veit, O. Pak,K. H. Krause, R. T. Schermuly, A. C. Brewer, H. H. Schmidt, W. Seeger, A. M. Shah,T. Gudermann, H. A. Ghofrani, A. Dietrich, Activation of TRPC6 channels is essentialfor lung ischaemia-reperfusion induced oedema in mice. Nat. Commun. 3, 649 (2012).

47. W. Zeng, J. P. Yuan, M. S. Kim, Y. J. Choi, G. N. Huang, P. F. Worley, S. Muallem,STIM1 gates TRPC channels, but not Orai1, by electrostatic interaction. Mol. Cell 32,439–448 (2008).

48. M. Amano, M. Ito, K. Kimura, Y. Fukata, K. Chihara, T. Nakano, Y. Matsuura, K. Kaibuchi,Phosphorylation and activation of myosin by Rho-associated kinase (Rho-kinase). J. Biol.Chem. 271, 20246–20249 (1996).

49. S. Y. Yuan, M. H. Wu, E. E. Ustinova, M. Guo, J. H. Tinsley, P. De Lanerolle, W. Xu,Myosin light chain phosphorylation in neutrophil-stimulated coronary microvascularleakage. Circ. Res. 90, 1214–1221 (2002).

50. R. Sandoval, A. B. Malik, T. Naqvi, D. Mehta, C. Tiruppathi, Requirement for Ca2+

signaling in the mechanism of thrombin-induced increase in endothelial permeability.Am. J. Physiol. Lung Cell. Mol. Physiol. 280, L239–L247 (2001).

51. M. Hoth, R. Penner, Calcium release-activated calcium current in rat mast cells. J. Physiol.465, 359–386 (1993).

52. M. Prakriya, R. S. Lewis, Regulation of CRAC channel activity by recruitment of silentchannels to a high open-probability gating mode. J. Gen. Physiol. 128, 373–386(2006).

53. M. Trebak, L. Lemonnier, J. T. Smyth, G. Vazquez, J. W. Putney Jr., PhospholipaseC-coupled receptors and activation of TRPC channels. Handb. Exp. Pharmacol. 179,593–614 (2007).

54. J. T. Smyth,W. I. Dehaven, B. F. Jones, J. C. Mercer, M. Trebak, G. Vazquez, J.W. Putney Jr.,Emerging perspectives in store-operated Ca2+ entry: Roles of Orai, Stim and TRP.Biochim. Biophys. Acta 1763, 1147–1160 (2006).

55. K. T. Cheng, X. Liu, H. L. Ong, W. Swaim, I. S. Ambudkar, Local Ca2+ entry via Orai1regulates plasma membrane recruitment of TRPC1 and controls cytosolic Ca2+

signals required for specific cell functions. PLoS Biol. 9, e1001025 (2011).56. S. A. Gross, G. A. Guzman, U. Wissenbach, S. E. Philipp, M. X. Zhu, D. Bruns,

A. Cavalié, TRPC5 is a Ca2+-activated channel functionally coupled to Ca2+-selective ionchannels. J. Biol. Chem. 284, 34423–34432 (2009).

57. F. Waheed, P. Speight, G. Kawai, Q. Dan, A. Kapus, K. Szászi, Extracellular signal-regulated kinase and GEF-H1 mediate depolarization-induced Rho activation and para-cellular permeability increase. Am. J. Physiol. Cell Physiol. 298, C1376–C1387 (2010).

58. J. H. Tinsley, P. De Lanerolle, E. Wilson, W. Ma, S. Y. Yuan, Myosin light chain kinasetransference induces myosin light chain activation and endothelial hyperpermeability.Am. J. Physiol. Cell Physiol. 279, C1285–C1289 (2000).

59. N. Norwood, T. M. Moore, D. A. Dean, R. Bhattacharjee, M. Li, T. Stevens, Store-operated calcium entry and increased endothelial cell permeability. Am. J. Physiol.Lung Cell. Mol. Physiol. 279, L815–L824 (2000).

60. J. T. Smyth, W. I. Dehaven, G. S. Bird, J. W. Putney Jr., Ca2+-store-dependent and-independent reversal of Stim1 localization and function. J. Cell Sci. 121, 762–772(2008).

61. R. Reynoso, R. M. Perrin, J. W. Breslin, D. A. Daines, K. D. Watson, D. M. Watterson,M. H. Wu, S. Yuan, A role for long chain myosin light chain kinase (MLCK-210) inmicrovascular hyperpermeability during severe burns. Shock 28, 589–595 (2007).

62. C. Sun, M. H. Wu, S. Y. Yuan, Nonmuscle myosin light-chain kinase deficiency attenu-ates atherosclerosis in apolipoprotein E-deficient mice via reduced endothelial barrierdysfunction and monocyte migration. Circulation 124, 48–57 (2011).

.SCIENCESIGNALING.org 19 March 2013 Vol 6 Issue 267 ra18 10

Page 12

R E S E A R C H A R T I C L E

Dow

n

63. Y. Yu, N. Lv, Z. Lu, Y. Y. Zheng, W. C. Zhang, C. Chen, Y. J. Peng, W. Q. He, F. Q. Meng,M. S. Zhu, H. Q. Chen, Deletion of myosin light chain kinase in endothelial cells has aminor effect on the lipopolysaccharide-induced increase in microvascular endotheliumpermeability in mice. FEBS J. 279, 1485–1494 (2012).

64. T. J. Shuttleworth, J. L. Thompson, O. Mignen, STIM1 and the noncapacitative ARCchannels. Cell Calcium 42, 183–191 (2007).

65. J. C. Mercer, W. I. Dehaven, J. T. Smyth, B. Wedel, R. R. Boyles, G. S. Bird, J. W. Putney Jr.,Large store-operated calcium selective currents due to co-expression of Orai1 orOrai2 with the intracellular calcium sensor, Stim1. J. Biol. Chem. 281, 24979–24990 (2006).

66. D. M. Ferreri, F. L. Minnear, T. Yin, A. P. Kowalczyk, P. A. Vincent, N-cadherin levelsin endothelial cells are regulated by monolayer maturity and p120 availability. CellCommun. Adhes. 15, 333–349 (2008).

67. R. T. Clements, F. L. Minnear, H. A. Singer, R. S. Keller, P. A. Vincent, RhoA andRho-kinase dependent and independent signals mediate TGF-b-induced pulmonaryendothelial cytoskeletal reorganization and permeability. Am. J. Physiol. Lung Cell.Mol. Physiol. 288, L294–L306 (2005).

68. R. K. Motiani, I. F. Abdullaev, M. Trebak, A novel native store-operated calciumchannel encoded by Orai3: Selective requirement of Orai3 versus Orai1 in estrogenreceptor-positive versus estrogen receptor-negative breast cancer cells. J. Biol. Chem.285, 19173–19183 (2010).

69. M. Potier, J. C. Gonzalez, R. K. Motiani, I. F. Abdullaev, J. M. Bisaillon, H. A. Singer,M. Trebak, Evidence for STIM1- and Orai1-dependent store-operated calcium influxthrough ICRAC in vascular smooth muscle cells: Role in proliferation and migration.FASEB J. 23, 2425–2437 (2009).

www

70. W. Zhang, K. E. Halligan, X. Zhang, J. M. Bisaillon, J. C. Gonzalez-Cobos, R. K. Motiani,G. Hu, P. A. Vincent, J. Zhou, M. Barroso, H. A. Singer, K. Matrougui, M. Trebak, Orai1-mediated I (CRAC) is essential for neointima formation after vascular injury. Circ. Res.109, 534–542 (2011).

Acknowledgments: We thank S. Muallem (National Institute of Dental and CraniofacialResearch/NIH) and J. Yuan (University of North Texas) for the STIM1 KK/EE mutant. Wethank Z. Wang and X. Wang for sharing their expertise on RhoA assays and MLCK West-ern blots. We are grateful to F. Jourd’heuil for advice and reagents during the course ofthese studies. Funding: This study was supported by NIH grant R01HL097111 to M.T.and in part by NIH grant R01HL095566 to K.M. Author contributions: A.V.S., R.K.M., X.Z.,I.F.A., A.P.A., J.C.G.-C., and W.Z. performed the experiments. K.M., P.A.V., and M.T. de-signed the experiments and supervised the research. M.T. wrote the paper with input fromco-authors. Competing interests: The authors declare that they have no competinginterests.

Submitted 23 July 2012Accepted 19 February 2013Final Publication 19 March 201310.1126/scisignal.2003425Citation: A. V. Shinde, R. K. Motiani, X. Zhang, I. F. Abdullaev, A. P. Adam,J. C. González-Cobos, W. Zhang, K. Matrougui, P. A. Vincent, M. Trebak, STIM1controls endothelial barrier function independently of Orai1 and Ca2+ entry. Sci. Signal.6, ra18 (2013).

lo

.SCIENCESIGNALING.org 19 March 2013 Vol 6 Issue 267 ra18 11

on March 19, 2013

stke.sciencemag.org

aded from