151

Brief Definitive Communications

A Functional Ryanodine-SensitiveIntracellular Ca2' Store Is Present in

Vascular Endothelial CellsRoy C. Ziegelstein, Harold A. Spurgeon, Roberto Pili, Antonino Passaniti, Linda Cheng,

Stefano Corda, Edward G. Lakatta, Maurizio C. Capogrossi

The presence of the ryanodine receptor was recently demon-strated in vascular and endocardial endothelium, but its func-tion has not been established. We investigated whether func-tional ryanodine-sensitive Ca2+ stores are present in culturedendothelial cells from rat aorta (RAECs), human aorta(HAECs), human umbilical vein (HUVECs), and bovine pul-monary artery (BPAECs) and what role these may play inintracellular Ca 2+ regulation. Under resting conditions,HAECs, BPAECs, and HUVECs demonstrated a slow in-crease in intracellular Ca2+ (indexed by indo 1 fluorescence)on exposure to 5 gmol/L ryanodine, whereas RAECs did not.However, after an initial bradykinin exposure in RAECs,ryanodine markedly blunted the rapid increase in Ca2' on a

A increase in [Ca2+] is important in mediatingmany of the functional properties of vascularendothelial cells, particularly release of va-

sodilatory prostaglandins',2 and endothelium-derivedrelaxation factor.3 Inositol 1,4,5-trisphosphate (InsP3)-mediated release of Ca21 from intracellular stores ap-pears largely responsible for the rapid transient increasein cytosolic Ca21 that occurs on agonist stimulation.4-12Although considerable information has been learnedabout endothelial Ca2' regulation, the identity andsensitivity of the intracellular compartments are notcompletely known. Further identification of these intra-cellular stores is facilitated by the use of appropriatepharmacologic probes.The plant alkaloid ryanodine has been used to study

[Ca 2ji regulation in striated muscle.13-15Ryanodine isknown to bind to the Ca2' release channels of thesarcoplasmic reticulum1617 and to thereby alter its Ca2'release properties. At low concentrations, ryanodinekeeps the Ca2' release channels in an open state,17-20causing Ca 2+ release.21 Recently, the presence of theryanodine receptor was demonstrated in vascular andendocardial endothelium.22 Whereas anti-ryanodinereceptor antibody binding sites were prominently dem-

Received February 16, 1993; accepted September 7, 1993.From the Laboratory of Cardiovascular Science (R.C.Z.,

H.A.S., L.C., S.C., E.G.L., M.C.C.) and the Laboratory of Biolog-ical Chemistry (R.P., A.P.), Gerontology Research Center, Na-tional Institute on Aging, National Institutes of Health; and JohnsHopkins University (R.C.Z.), Baltimore, Md.

Previously presented as a preliminary report in abstract form(Circulation. 1992;86[suppl 1]:I-750).Correspondence to Maurizio C. Capogrossi, MD, Laboratory of

Cardiovascular Science, Gerontology Research Center, NIA,NIH, 4940 Eastern Ave, Baltimore, MD 21224.

second exposure to bradykinin. In HUVECs, ryanodine inbuffer with 1.5 mmol/L Ca2l did not inhibit the agonist-sensitive Ca2' increase, whereas it blunted the rapid increasein Ca2' on histamine exposure in buffer with 5 mmol/L Ca2l,suggesting that increasing [Ca2+] enhances the binding ofryanodine to its receptor. Thus, functional ryanodine-sensitiveCa2+ stores are present in vascular endothelial cells. Theseappear to be involved in regulation of Ca2+ storage and releasefrom agonist-sensitive intracellular compartments. (Circ Res.1994;74:151-156.)

Key Words * endothelium * Ca2' * ryanodine .indo 1

onstrated in these cells, the function of the endothelialryanodine receptor was not established.To determine whether functional ryanodine-sensitive

intracellular Ca2' stores exist in vascular endothelialcells, we examined the effect of ryanodine on endothe-lial cells cultured from rat aorta (RAECs), human aorta(HAECs), human umbilical vein (HUVECs), and bo-vine pulmonary artery (BPAECs) and loaded with themembrane-permeant (acetoxymethyl ester) form ofindo 1. Our results demonstrate the existence of func-tional ryanodine-sensitive Ca2' stores in vascular endo-thelial cells, which appear to control, at least in part,agonist-sensitive intracellular Ca2+ pools.

Materials and MethodsCell Culture

Unless otherwise noted, all cell lines were grown at 37°C ina humidified atmosphere of 95% air-5% CO2 in medium with(,ug/mL) penicillin, 50; streptomycin, 50; and neomycin, 100(GIBCO, Grand Island, NY). BPAECs (American Type Cul-ture Collection, Rockville, Md) were grown to confluence inDulbecco's modified Eagle's medium (National Institutes ofHealth [NIH] Medium Center, Bethesda, Md) with 10% fetalbovine serum (Paragon, Baltimore, Md) to passages 19 and 20.HAECs (Clonetics, San Diego, Calif) were obtained as prolif-erating quaternary cultures and grown to confluence to pas-sages 5 to 7 in endothelial cell growth medium with 2% fetalbovine serum, 50 gg/mL gentamicin, 50 ng/mL amphotericinB, and approximately 12 ,ug/mL bovine brain extract. HUVECswere prepared from freshly obtained human umbilical cordveins23 and maintained to passages 4 to 6 in medium 199 (NIHMedium Center) with 10% fetal calf serum (Paragon), 100,g/mL endothelial mitogen (Biomedical Technologies, Inc,Stoughton, Mass), and 16 U/mL heparin (Fisher Scientific,Fair Lawn, NJ). RAECs were cultured by the primary explanttechnique24 and then grown in culture to passage 3 to 7 in

by guest on July 3, 2018http://circres.ahajournals.org/

Dow

nloaded from

152 Circulation Research Vol 74, No 1 January 1994

minimum essential medium with D-valine supplemented with10% fetal calf serum, 100 1£g/mL endothelial mitogen, and 16U/mL heparin. HUVECs and RAECs were identified bydemonstrating specific immunofluorescent staining for factorVIII-related antigen (DAKO Corp, Carpinteria, Calif). Aftertreatment with 0.25% trypsin (NIH Medium Center) and 0.5mmol/L EGTA (Sigma Chemical Co, St Louis, Mo), cells wereplated on glass coverslips (BPAECs and HAECs) or in 1-mm2glass capillary tubes (HAECs, HLVECs, and RAECs; VitroDynamics, Rockaway, NJ) precoated with 1% gelatin (Sigma)1 to 2 days before experimental use.

Indo 1 Fluorescence MeasurementsEndothelial monolayers were loaded with growth medium

containing 12 ,umol/L of the ester derivative (AM form) of thefluorescent Ca2+ probe indo 1 (Molecular Probes, Eugene,Ore) and kept in a 95% air-5% CO2 incubator at roomtemperature for 60 minutes. Monolayers were then gentlywashed with HEPES-buffered saline solution and maintainedan additional 60 minutes to allow for deesterification of theindicator before the beginning of the experiment. Cells werethen placed on the stage of a modified inverted fluorescencemicroscope (Zeiss IM-35) as previously described.25 Briefly,indo 1 fluorescence was excited at 350+5 nm using a xenon

strobe lamp (model 236 stroboscope and model 35S lamp,Chadwick-Helmuth Electronics, El Monte, Calif), and band-pass interference filters (Andover, Lawrence, Mass) selectedwavelength bands of emitted fluorescence at 391 to 434 nm("410 channel") and 457 to 507 nm ("490 channel") corre-sponding to the Ca2'-bound and Ca21-free forms of theindicator, respectively. The 410- to 490-nm ratio was used asan index of [Ca21Jj. Emitted indo 1 fluorescence was collectedfrom a field of approximately 15 confluent cells. Autofluores-cence of a comparable field of unloaded cells was generally<5% of indo 1-loaded cells. Data are reported as emitted indo1 fluorescence ratio and not as [Ca2+], since a reliable mul-tipoint intracellular Ca21 calibration in intact monolayers ofendothelial cells was not possible to obtain. This type ofcalibration is dependent on making repeated measurementsover a range of [Ca 2+1 values using ionophores to equilibrateCa2' between the intracellular and extracellular spaces. It alsoassumes complete hydrolysis of the indicator to a singleCa2'-sensitive fluorescent species. An intracellular Ca2' cali-bration is superior to the calibration of the free acid form ofthe probe in solution or to the calibration performed on cellsloaded with the ester form of the probe, which are then lysedsince the spectral properties of the indicator may be influencedby the intracellular environment. A variable Mn2+-insensitivefluorescent component and the apparent low efficacy of Ca 2+ionophores in the endothelial cell plasma membrane pre-

vented accurate intracellular calibration.

Experimental ProtocolCells were exposed to ryanodine (Penick Corp, Lyndhurst,

NJ), bradykinin (Sigma), or histamine (Sigma) in a buffer ofthe following composition (mmol/L): NaCl, 137; KCI, 4.9;MgSO4, 1.2; NaH2PO4, 1.2; D-glucose, 15; HEPES, 20; andCaCI2, 1.5; pH 7.40 (23°C). When experiments were per-formed on cells grown in capillary tubes, the tubes wereinternally perfused using a syringe pump (Harvard Apparatus,South Natick, Mass) at a flow rate (0.5 to 1.0 mL/min) thatproduced negligible shear stress calculated from flow rate andtube geometry26 (<2 dyne * cm-2) and did not affect [Ca 2+'i.Since more significant and rapid intracellular loss of indo 1occurs at 37°C than at 230C,25 all experiments were performedat 23°C.

StatisticsData are reported as mean+SEM. The ryanodine concen-

tration-response relation for RAECs was compared by a

22 A.

0.7

2.2 B.

ZmE,-wc

0.7 2nmln

FIG 1. Tracings showing the effect of ryanodine (RY) exposureon [Ca2+], in human aortic endothelial cells (HAECs). A, Arepresentative example of fluorescence measurements was ob-tained from an indo 1-loaded HAEC monolayer exposed to 10gmol/L histamine (HIST). An initial rapid increase in [Ca2+1l wasfollowed by a more sustained [Ca2+J elevation, which retumedto baseline levels after several minutes. B, A representativeexample of fluorescence measurements was obtained from anindo 1-loaded HAEC monolayer exposed to 5 &moI/L RY inbuffer with 1.5 mmol/L Ca2+. RY exposure produced a slowlydeveloping, sustained increase in [Ca2+1] that was smaller inmagnitude than that due to HIST stimulation. There was nosignificant reversibility of the [Ca2+]j increase on RY washout. Asimilar result was obtained in 14 of 16 HAEC monolayers.

one-way ANOVA, followed by a Bonferroni test. A differencewas considered significant at P<.05.

ResultsEffect of Ryanodine on [Ca2+]1 UnderBaseline ConditionsTo determine the effect of ryanodine on endothelial

[Ca`2+] under "resting" (ie, unstimulated) conditions,indo 1-loaded endothelial monolayers were exposed to5 gmol/L ryanodine for 10 to 15 minutes. These exper-imental conditions were chosen on the basis of previouswork in suspensions of intact cardiac myocytes, whichshowed that ryanodine evoked a slow release of Ca2+into the myoplasmic space in a concentration-depen-dent (up to micromolar) and time-dependent (up to 10minutes) manner.21 HAECs (Fig 1B), BPAECs, andHUVECs (not shown) showed a slow sustained increasein [Ca2+]i without significant reversibility on ryanodinewashout. Both the rate of increase and the amplitude ofthe change were reduced when compared with theresponse to histamine stimulation (Fig 1A). Fig 1 alsoshows that the ryanodine effect in HAECs was moreprolonged than that of histamine, which produced atypical rapid transient increase in [Ca2+]i followed by asustained elevation of [Ca2+]i, which eventually re-turned to control levels after several minutes. In con-trast to the sustained increase in [Ca2+]i produced by a10-minute ryanodine exposure in most HAEC, BPAEC,and HUVEC monolayers, RAECs did not demonstratethis response under these resting conditions (Fig 2).

Effect of Ryanodine After Agonist Stimulation inRAECs and HUVECs

Since ryanodine had no effect on RAEC [Ca2+]i andonly a small effect on HUVEC [Ca2+]i, the ability ofryanodine to influence refilling of the agonist-sensitiveintracellular Ca 2+ pool was examined in these two celltypes. In these experiments, RAECs were exposed to

by guest on July 3, 2018http://circres.ahajournals.org/

Dow

nloaded from

Ziegelstein et al Ryanodine-Sensitive Ca2' Store in Endothelial Cells 153

E

0

z

0.24-

0.20-

0.16-

0.12

0.08

0.04

0.00o

0..04l

T

HAEC BPAEC HUVEC RAEC

FIG 2. Bar graph showing the effect of ryanodine on [Ca2+], inunstimulated endothelial cells. HAEC indicates human aorticendothelial cells; BPAEC, bovine pulmonary artery endothelialcells; HUVEC, human umbilical vein endothelial cells; andRAEC, rat aortic endothelial cells. The change in the indo 1fluorescence ratio from baseline levels to that seen after 10minutes of exposure to 5 pmol/L ryanodine is plotted for each ofthe cell types studied. Each bar represents mean±SEM of 13 to16 experiments for each group. *P<.02 vs baseline.

bradykinin, and HUVECs were exposed to histamine,since these agonists reproducibly produce a rapid in-crease in [Ca2"]i in these cell types. The response ofRAECs to histamine and of HUVECs to bradykinin issomewhat less consistent, which may reflect cell type-specific effects of tissue culture conditions on receptornumber or function. Although a different agonist was

used for each cell type, both bradykinin and histamineare thought to cause an initial rapid rise in [Ca2+]i dueto receptor-mediated activation of phospholipase C andto generation of InsP3.69,10When RAECs were exposed to bradykinin (Fig 3A),

an initial rapid increase in [Ca2+]i was followed by a

more prolonged second phase; in contrast to the initialrapid increase, this second phase is thought to be due tothe influx of Ca2' from the extracellular space.8 Aperiod in the absence of bradykinin appears to berequired for restoration of the sensitivity of the brady-kinin receptor.8 Removal of extracellular Ca2' 5 min-utes before a second bradykinin exposure in RAECs

1.7 0Ca+IEGTAui

ZE

im°Sto. 2 min

~0.-

............

E

abolishes the sustained increase in [Ca2"]i (Fig 3A).When RAECs were instead exposed for 15 minutes to 5,mol/L ryanodine after an initial bradykinin exposure(Fig 3B), the rapid increase in [Ca2+]i on a secondbradykinin exposure was markedly blunted. There wasno difference in the inhibition of the bradykinin-in-duced increase in [Ca2"], whether or not Ca2' wasremoved from the buffer 5 minutes before the secondbradykinin exposure (n=8 with and n=7 without bufferCa2'). These data suggest that under these experimen-tal conditions the effect of ryanodine on the bradykinin-sensitive intracellular Ca2' pool is incomplete. Further-more, the increase in [Ca2+] on the second bradykininexposure occurred despite the absence of an increase in[Ca2+]i during ryanodine exposure and may reflect slowCa2' release into the cytosol, which is balanced by Ca2'extrusion from the cell.

In the absence of a prior bradykinin exposure, ryan-odine had no effect on the increase in [Ca2"], due tobradykinin (n=8, not shown). Ryanodine may thusprevent refilling of the bradykinin-sensitive Ca2' store.Alternatively, a prior agonist-induced increase in [Ca2"Iimay be required to render the cells sensitive to ryano-dine. Since caffeine also has been shown to release Ca2'from a ryanodine-sensitive intracellular Ca2' store inother cell types,21 we exposed RAECs for 5 minutes to5 ,mol/L ryanodine alone and then simultaneously toryanodine and 10 mmol/L caffeine for 10 minutesbefore bradykinin stimulation. The inability of ryano-dine to inhibit the bradykinin-induced increase in[Ca 2+]i in the absence of a prior bradykinin stimulationwas not affected by simultaneous exposure to 10mmol/L caffeine and 5 ,umol/L ryanodine (n=8, notshown).To determine the concentration dependence of the

ryanodine effect in RAECs, experiments using sequen-tial bradykinin stimulation were performed in which theconcentration of ryanodine was varied without alteringthe duration of exposure. Fig 4 shows that the magni-tude of the increase in [Ca 2+]i due to the secondbradykinin exposure after a 15-minute period undercontrol conditions without ryanodine was just less thanone half (0.49+0.08, n=7) that of the first exposure.This is similar to the findings of others.8 Although 0.05gmol/L ryanodine had no effect on the bradykinin-induced rapid [Ca2+]i increase, higher concentrations

A.FIG 3. Tracings showing the effect of ryanodine (RY) onthe bradykinin (BK)-induced rapid increase in rat aorticendothelial cell (RAEC) [Ca2+1J. A, Representative fluo-rescence measurements were obtained from an indo1-loaded RAEC monolayer during sequential exposures

-v~-- to BK in the presence and absence of extracellular Ca2 .The sustained second phase of the increase in [Ca2+], isabolished in the absence of buffer Ca2+. B, Represen-tative fluorescence measurements were obtained froman indo 1-loaded RAEC monolayer exposed to BK after

B a 15-minute exposure to 5 ,mol/L RY and a 5-minuteexposure to Ca2M-free EGTA buffer (n=7). ExtracellularCa2+ was removed after only 10 minutes of RY exposureto avoid an effect of extracellular Ca2+ removal on theinitial interaction of RY with its receptor. RY markedlyblunted the increase in [Ca2+Ji due to the second BKexposure.

*

1'. --- --- ..

by guest on July 3, 2018http://circres.ahajournals.org/

Dow

nloaded from

154 Circulation Research Vol 74, No 1 January 1994

0.6

0.5

0.4 koJm

02

pm

I0.3

0.2

o.1_ +0

0 0.05 0.5 5.0

RYANODINE (lAM)

FIG 4. Concentration dependence of the effect of ryanodine onthe bradykinin (BK)-induced rapid [Ca2+]J increase in rat aorticendothelial cells (RAECs). The change in the indo 1 ratio frombaseline conditions to the peak of the BK response of thesecond exposure (ABK2) is compared with that of the firstexposure (ABK1) and is plotted against ryanodine concentration.In each case, the second 0.5 f.mol/L BK exposure was sepa-rated from the first by a 15-minute period either in control bufferor in buffer with ryanodine. Ryanodine inhibited the BK-inducedrapid increase in RAEC [Ca21]1 in a concentration-dependentmanner. Data represent mean+SEM of six to nine experimentsat each ryanodine concentration. *P<.05 vs control.

resulted in increasing inhibition, with significant inhibi-tion noted with 5 gmol/L ryanodine. Thus, ryanodinereceptors appear to control, at least in part, bradykinin-sensitive Ca2' pools in RAECs.

In HUVECs, histamine stimulation resulted in a typ-ical biphasic increase in [Ca2"]i that was not different inmagnitude on a second histamine exposure after a 15-minute superfusion with control buffer. In contrast to theeffect of ryanodine on agonist stimulation in RAECs, a

15-minute exposure of HUVECs to 5 ,umol/L ryanodineafter histamine stimulation did not inhibit the rapidincrease in [Ca'+]i on a second histamine exposure (Fig5). The absence of an effect of 5 ,umol/L ryanodine wasnoted whether or not ryanodine exposure was continuedduring exposure to histamine. Such an apparent lack ofeffect of ryanodine may be due to inefficient binding ofthe drug to its receptor, an interaction known to have an

absolute dependence on the presence of Ca2', which hasbeen suggested to form a complex with the ryanodinereceptor as a prerequisite to ryanodine binding.16 Thatincreasing [Ca'+]i enhances ryanodine binding and mod-ulates the effect of the drug was inferred in studiesdemonstrating a "use-dependent" effect of the alkaloidin ferret cardiac muscle27 and was more directly demon-strated in studies with rat cardiac myocytes in which theryanodine effect was attenuated when extracellular Ca21entry was inhibited.21To determine whether bath Ca21 ([Ca21]0) could

modulate the effect of ryanodine in HUVECs, theidentical protocol using sequential histamine stimula-tion was performed in 5 mmol/L [Ca2']0. When HUVECsin buffer with 1.5 mmol/L Ca2' were then exposed tobuffer with 5 mmol/L Ca2 , there was a slight increase inbasal [Ca2]i. Although the response of sequential hista-mine exposures in 5 mmol/L [Ca2l]0 in the absence ofryanodine was not affected (not shown), a 15-minute

FIG 5. Tracings showing the effect of ryanodine (RY) on thehistamine-induced rapid increase in [Ca2+1J in human umbilicalvein endothelial cells (HUVECs). A, A representative example offluorescence measurements was obtained from an indo 1-load-ed HUVEC monolayer exposed sequentially to histamine (HIST).A typical biphasic increase in [Ca2+1J is noted on both the initialand subsequent exposure to 100 ,umol/L HIST. B, A represen-tative example of fluorescence measurements was obtainedfrom an indo 1-loaded HUVEC monolayer during HIST stimula-tion before and after RY exposure. After a 15-minute period inbuffer with 1.5 mmol/L Ca2+ and 5 1,mol/L RY, cells wereexposed a second time to 100 ,umol/L HIST (n= 10). RY :had noeffect on the rapid increase in [Ca2+]j due to HIST. Continuingthe exposure to RY throughout exposure to HIST (n=5) did notalter the results (not shown).

exposure to ryanodine substantially inhibited the rapidincrease in [Ca2+]i on a second histamine exposure in 5mmol/L [Ca2`]0 in five of eight monolayers studied (Fig 6).The magnitude of the increase in indo 1 ratio of thesecond histamine exposure after 15 minutes of ryanodine(AHIS2) compared with that of the first histamine expo-sure before ryanodine (AHIS1) was significantly less in 5mmol/L [Ca21]0 than in 1.5 mmol/L [Ca2+]0: AHIS2AH IS1was 0.54+0.11 in 5 mmol/L [Ca2]0 (n=8) and 0.830.06in 1.5 mmol/L [Ca2+']0 (n=10, P<.05).

DiscussionThe present study is the first to document the exis-

tence of a ryanodine-sensitive Ca'+ store that is in-

2.4

OW

~0~ILSO-oIL

0.7.

2 min

FIG 6. Tracings showing the effect of ryanodine (RY) on thehistamine (HIST)-induced rapid increase in human umbiical veinendothelial cell [Ca2+1] in buffer with 5 mmol/L Ca2+. The slightincrease in indo 1 ratio that occurred when the buffer wfith 1.5mmol/L Ca21 was replaced with the buffer with 5 mmoltL Ca2+ isnot shown. Cells were stimulated with 100 gmol/L HIST before andafter a 15-minute exposure to RY (5 ,umol/L). Under conditions ofelevated [Ca2+]0, RY inhibited the HIST-induced rapid increase in[Ca2+1] in five of eight monolayers.

¢^o_2.0

zw~09-

zt.0.9_

A.

2 .,.

I

B.

by guest on July 3, 2018http://circres.ahajournals.org/

Dow

nloaded from

Ziegelstein et al Ryanodine-Sensitive Ca2' Store in Endothelial Cells 155

volved in intracellular Ca2' regulation in vascular endo-thelial cells. The study thus extends the recent work ofLesh et al,22 who demonstrated the presence of theryanodine receptor in vascular and endocardial endo-thelium. In HAECs, BPAECs, and HUVECs, ryano-dine exposure resulted in a slow sustained increase in[Ca211i. Whereas ryanodine did not increase [Ca2"]i inunstimulated RAECs, a 15-minute exposure inhibitedCa2' release from the agonist-sensitive intracellularpool in this cell type. The maximal concentration usedin the present study (5 ,umol/L) is in the range used byothers to probe the Ca 2+ uptake and release propertiesof the sarcoplasmic reticulum. Lattanzio et al19 foundthat similar concentrations (0.01 to 10 gmol/L) in-creased the Ca21 permeability of both skeletal andcardiac muscle sarcoplasmic reticulum membranes, andHansford and Lakatta2' found that 0.01 to 1 ,umol/Lryanodine caused a slow and prolonged increase in[Ca2'i in suspensions of intact cardiac myocytes.The characteristics of the ryanodine-induced Ca2'

elevation in HAECs, BPAECs, and HUVECs are dif-ferent from that due to agonist stimulation of endothe-lial cells. Exposure of indo 1-loaded endothelial cells toeither bradykinin (Fig 3) or histamine (Fig 1) results ina rapid increase in [Ca 2+]i, which then decreases to alevel above baseline for several minutes before return-ing to control values. In contrast, ryanodine exposure inHAECs (Fig 1), BPAECs, and HUVECs results in aslower, more sustained elevation in [Ca2+]i that issmaller in magnitude than that due to agonist exposure.This gradual increase in [Ca'+], may be the result ofCa2' release from the endoplasmic reticulum exceedingthe rate of extrusion from the cytosol. Alternatively, itmay reflect activation of Ca 2+ influx from the extracel-lular space by depletion of an internal Ca21 store. Theactivation of Ca21 influx by depletion of an intracellularCa21 store in cultured vascular endothelial cells hasrecently been reported,28,29 although the precise mannerin which the regulation of extracellular Ca21 influxoccurs remains unclear. Although the mechanism ofaction of ryanodine is likely the same in all endothelialtypes studied, the rate of Ca2' release and the Ca21dependence of ryanodine binding to its receptor maydiffer among them.Although ryanodine did not affect [Ca2+]i in unstim-

ulated RAECs, a 15-minute exposure to ryanodine afterinitial agonist stimulation considerably inhibited theincrease in [Ca 2+]i induced by a second agonist stimu-lation. This indicates that at least some of the ryanodineeffect in this cell type is on an agonist- and InsP3-sensitive intracellular Ca 2+ pool. In a similar manner,Giannini et a130 recently reported the existence of afunctional ryanodine receptor in mink lung epithelialcells treated with transforming growth factor-,8. Al-though exposure of these cells to ryanodine (100 to 200,mol/L for up to 20 minutes) also did not alter [Ca2+]1,10 gmol/L ryanodine prevented the increase in [Ca2'iinduced by bradykinin, as in the present study.The effects of ryanodine in vascular endothelial cells

appear to vary with species, the vascular bed of origin ofthe cell type, or both. This is similar to the variation inthe effect of ryanodine on inhibition of the twitch incardiac muscle preparations from different species.31-34Although the etiology of this species variation for bothvascular endothelium and cardiac muscle is unknown, it

may relate to a species difference in the extent ofendoplasmic or sarcoplasmic reticulum loading. Alter-natively, it may relate to differences in membranepermeability to ryanodine. Our results also indicatesome variability even within a given cell type.

In summary, we have documented the existence offunctional ryanodine-sensitive Ca2+ stores in culturedvascular endothelial cells, which largely deplete theInsP3-sensitive intracellular Ca2' pool. The intracellularCa2' pool in endothelial cells plays an important role inthe rapid [Ca2+]i rise seen on exposure to severalphysiological agonists such as adenine nucleotides,4bradykinin,5-8 histamine,9-12 and thrombin210 throughgeneration of InsP3. Like InsP3 receptors, endothelialryanodine receptors are involved in Ca2+ regulation ofagonist-sensitive intracellular compartments. The devel-opment of pharmacologic probes of the intracellularCa21 pool in vascular endothelial cells will enhance ourknowledge of endothelial Ca2' homeostasis.

AcknowledgmentsThe statistical expertise of Frances O'Connor and the secre-

tarial assistance of Miriam Glaser are greatly appreciated.

References1. Hallam TJ, Pearson JD, Needham LA. Thrombin-stimulated ele-

vation of human endothelial-cell cytoplasmic free calcium concen-tration causes prostacyclin production. Biochem J. 1988;251:243-249.

2. Jaffe EA, Grulich J, Weksler BB, Hampel G, Watanabe K. Cor-relation between thrombin-induced prostacyclin production andinositol trisphosphate and cytosolic free calcium levels in culturedhuman endothelial cells. J Biol Chem. 1987;262:8557-8565.

3. Long CJ, Stone TW. The release of endothelium-derived relaxantfactor is calcium dependent. Blood Vessels. 1985;22:205-208.

4. Pirotton S, Raspe E, Demolle D, Erneux C, Boeynaems J-M.Involvement of inositol 1,4,5-trisphosphate and calcium in theaction of adenine nucleotides on aortic endothelial cells. J BiolChem. 1987;262:17461-17466.

5. Colden-Stanfield M, Schilling WP, Ritchie AK, Eskin SG, NavarroLT, Kunze DL. Bradykinin-induced increases in cytosolic calciumand ionic currents in cultured bovine aortic endothelial cells. CircRes. 1987;61:632-640.

6. Lambert TL, Kent RS, Whorton AR. Bradykinin stimulation ofinositol polyphosphate production in porcine aortic endothelialcells. J Biol Chem. 1986;261:15288-15293.

7. Freay A, Johns A, Adams DJ, Ryan US, van Breemen C.Bradykinin and inositol 1,4,5-trisphosphate-stimulated calciumrelease from intracellular stores in cultured bovine endothelialcells. Pflugers Arch. 1989;414:377-384.

8. Morgan-Boyd R, Stewary JM, Vavrek RJ, Hassid A. Effects ofbradykinin and angiotensin II on intracellular Ca2` dynamics inendothelial cells. Am J Physiol. 1987;253:C588-C598.

9. Lo WWY, Fan T-PD. Histamine stimulates inositol phosphateaccumulation via the H1-receptor in cultured human endothelialcells. Biochem Biophys Res Commun. 1987;148:47-53.

10. Pollock WK, Wreggett KA, Irvine RF. Inositol phosphate pro-duction and Cal' mobilization in human umbilical-vein endo-thelial cells stimulated by thrombin and histamine. Biochem J.1988;256:371-376.

11. Resink TJ, Grigorian GY, Moldabaeva AK, Danilov SM, BuihlerFR. Histamine-induced phosphoinositide metabolism in culturedhuman umbilical vein endothelial cells: association with throm-boxane and prostacyclin release. Biochem Biophys Res Commun.1987;144:438-446.

12. Jacob R, Merritt JE, Hallam TJ, Rink TJ. Repetitive spikes incytoplasmic calcium evoked by histamine in human endothelialcells. Nature. 1988;335:40-45.

13. Fairhurst AS, Jenden DJ. Effect of ryanodine on the calciumuptake system of skeletal muscle. Proc NatlAcad Sci USA. 1962;48:807-813.

14. Jenden DJ, Fairhurst AS. The pharmacology of ryanodine.Pharmacol Rev. 1969;21:1-25.

by guest on July 3, 2018http://circres.ahajournals.org/

Dow

nloaded from

156 Circulation Research Vol 74, No 1 January 1994

15. Fairhurst AS, Hasselbach W. Calcium efflux from a heavy sar-cotubular fraction: effect of ryanodine, caffeine and magnesium.EurJBiochem. 1970;13:504-509.

16. Pessah IN, Waterhouse AL, Casida JE. The calcium-ryanodinereceptor complex of skeletal and cardiac muscle. Biochem BiophysRes Commun. 1985;128:449-456.

17. Fleischer S, Ogunbunmi EM, Dixon MC, Fleer EAM. Localizationof Ca21 release channels with ryanodine in junctional terminalcisternae of sarcoplasmic reticulum of fast skeletal muscle. ProcNatlAcad Sci U SA. 1985;82:7256-7259.

18. Meissner G. Ryanodine activation and inhibition of the Ca2+release channel of sarcoplasmic reticulum. J Biol Chem. 1986;261:6300-6306.

19. Lattanzio FA, Schlatterer RG, Nicar M, Campbell KP, Sutko JL.The effects of ryanodine on passive calcium fluxes across sarco-plasmic reticulum membranes. J Biol Chem. 1987;262:2711-2718.

20. Rardon DP, Cefali DC, Mitchell RD, Seiler SM, Jones LR. Highmolecular weight proteins purified from cardiac junctional sarco-plasmic reticulum vesicles are ryanodine-sensitive calciumchannels. Circ Res. 1989;64:779-789.

21. Hansford RG, Lakatta EG. Ryanodine releases calcium from sar-coplasmic reticulum in calcium-tolerant rat cardiac myocytes.JPhysiol (Lond). 1987;390:453-467.

22. Lesh RE, Marks AR, Somlyo AV, Fleischer S, Somlyo AP. Anti-ryanodine receptor antibody binding sites in vascular andendocardial endothelium. Circ Res. 1993;72:481-488.

23. Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of humanendothelial cells derived from umbilical veins: identification bymorphologic and immunologic criteria. J Clin Invest. 1973;52:2745-2756.

24. Cole OF, Fan T-PD, Lewis GP. Isolation, characterization, growthand culture of endothelial cells from the rat aorta. Cell Biol Int Rep.1986;10:399-405.

25. Spurgeon HA, Stern MD, Baartz G, Raffaeli S, Hansfoyd RG, TaloA, Lakatta EG, Capogrossi MC. Simultaneous measurement ofCa2 , contraction, and potential in cardiac myocytes.Am JPhysiol.1990;258:H574-H586.

26. Ziegelstein RC, Cheng L, Capogrossi MC. Flowzdependentcytosolic acidification of vascular endothelial cells. Science. 1992;258:656-659.

27. Malecot CO, Katzung BG. Use-dependence of ryanodine effectson postrest contraction in ferret cardiac muscle. Circ Res. 1987;60:560-567.

28. Dolor RJ, Hurwitz LM, Mirza Z, Strauss HC, Whorton AR. Reg-ulation of extracellular calcium entry in endothelial gells: role ofintracellular calcium pool. Am J Physiol. 1992;262:C171-C181.

29. Schilling WP, Cabello OA, Rajan L. Depletion of the inositol1,4,5-trisphosphate-sensitive intracellular Ca21 store in vascularendothelial cells activates the agonist-sensitive Ca 2+-influxpathway. Biochem J. 1992;284:521-530.

30. Giannini G, Clementi E, Ceci R, Marziali G, Sorrentino V.Expression of a ryanodine receptor-Ca21 channel that is regulatedby TGF-3. Science. 1992;257:91-94.

31. Sutko JL, Willerson JT. Ryanodine alteration of the contractilestate of rat ventricular myocardium: comparison with log, cat, andrabbit ventricular tissues. Circ Res. 1980;46:332-343.

32. Kort AA, Lakatta EG. Spontaneous sarcoplasm4z reticulumcalcium release in rat and rabbit cardiac muscle: relation totransient and rested-state twitch tension. Circ Res. 1988;63:969-979.

33. Wier WG, Yue DT, Marban E. Effects of ryanodiie on intra-cellular Ca21 transients in mammalian cardiac muscle. Fed Proc.1985;44:2989-2993.

34. Smith GL, Valdeomillos M, Eisner DA, Allen DG. Effects of rapidapplication of caffeine on intracellular calcium congentration inferret papillary muscles. J Gen Physiol. 1988;92:351-368.

by guest on July 3, 2018http://circres.ahajournals.org/

Dow

nloaded from

CapogrossiR C Ziegelstein, H A Spurgeon, R Pili, A Passaniti, L Cheng, S Corda, E G Lakatta and M C

cells.A functional ryanodine-sensitive intracellular Ca2+ store is present in vascular endothelial

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1994 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.74.1.1511994;74:151-156Circ Res. 

http://circres.ahajournals.org/content/74/1/151World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Researchin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on July 3, 2018http://circres.ahajournals.org/

Dow

nloaded from