Platelet Inhibition by Insulin Is Absent in Type 2 Diabetes Mellitus

18 Pages·2005·1.09 MB·English

Platelet Inhibition by Insulin Is Absent in Type 2
Diabetes Mellitus
Irlando Andrade Ferreira, Astrid I.M. Mocking, Marion A.H. Feijge, Gertie Gorter,
Timon W. van Haeften, Johan W.M. Heemskerk, Jan-Willem N. Akkerman
Objective—ADP-induced P2y signaling is crucial for formation and stabilization of an arterial thrombus. We
12
demonstratedrecentlyinplateletsfromhealthysubjectsthatinsulininterfereswithCa2(cid:1)increasesinducedbyADP-P2y
1
contact through blockade of the G-protein G, and thereby with P2y -mediated suppression of cAMP.
i 12
Methods and Results—Here we show in patients with type 2 diabetes mellitus (DM2) that platelets have lost
responsivenesstoinsulinleadingtoincreasedadhesion,aggregation,andprocoagulantactivityoncontactwithcollagen.
Using Ser473 phosphorylation of protein kinase B as output for insulin signaling, a 2-fold increase is found in
insulin-stimulatednormalplatelets,butinDMplateletsthereisnosignificantresponse.Inaddition,DM2plateletsshow
increased P2y -mediated suppression of cAMP and decreased P2y inhibition by the receptor antagonist
12 12
DDDDDDDDDDDDD AR-C69931MX.
ooooooooooooo
wwwwwwwwwwwww
nlonlonlonlonlonlonlonlonlonlonlonlonlo Conclusion—The loss of responsiveness to insulin together with increased signaling through P2y12 might explain the
aaaaaaaaaaaaa hyperactivity of platelets in patients with DM2. (Arterioscler Thromb Vasc Biol. 2006;26:000-000.)
ddddddddddddd
eeeeeeeeeeeee
ddddddddddddd
fro fro fro fro fro fro fro fro fro fro fro fro fro Key Words:
mmmmmmmmmmmmm
h h h h h h h h h h h h h
ttpttpttpttpttpttpttpttpttpttpttpttpttp
://a://a://a://a://a://a://a://a://a://a://a://a://a Platelet activation leads to release of components that agents, P2y signaling controls cAMP production through
tvtvtvtvtvtvtvtvtvtvtvtvtv 12
b.ab.ab.ab.ab.ab.ab.ab.ab.ab.ab.ab.ab.a initiate formation of a thrombus and start inflammatory adenylylcyclase.10,11cAMPinhibitsplateletsthroughcAMP-
hhhhhhhhhhhhh
aaaaaaaaaaaaa responses that contribute to atherosclerosis.1 Signaling dependent protein kinase (protein kinase A [PKA]),12 which
jojojojojojojojojojojojojo
uuuuuuuuuuuuu through the P2y receptor is crucial for formation and inhibits almost all platelet functions through blockade of
rnrnrnrnrnrnrnrnrnrnrnrnrn 12
als.orgals.orgals.orgals.orgals.orgals.orgals.orgals.orgals.orgals.orgals.orgals.orgals.org sretadbuicliezsatcioonllaogfean-tihnrdoumcbedusa.2d,3hIenshioibni,tiaogngorefgtahteioPn2ya1n2dretcheropmto-r mtourlaticptlievasttieopns,isnigpnlaatleinlegtathcrtiovuagtihonthceasmcaitdoegseinn-calcutdivinatgedrecperop--
b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ b/ bin generation.3,4 Subjects with a P2y deficiency have a tein kinases pathway, formation of thromboxane A (TxA),
yyyyyyyyyyyyy 12 2 2
gue gue gue gue gue gue gue gue gue gue gue gue gue bleedingtendency3,5andindividualswithanincreasedP2y12 andtheactivationofkeyenzymessuchasphospholipaseC(cid:2)
st onst onst onst onst onst onst onst onst onst onst onst onst on rsepcoenpstiovrenceospsytonuamgobneirsths,avaendpltahteesleetssuwbjiethctsanexipnecrrieeansceedpree-- tahnedGpr(cid:2)o(cid:3)teidnimkienrasleeaCdin(PgKtoC)th.1e3Sacetciovnatdi,ontheorfepirsottheienreklienaasseeoBf
N N N N N N N N N N N N N i
ooooooooooooo ripheral arterial thrombosis.6 The CAPRIE trial shows that (PKB/Akt), and integrin (cid:1) (cid:2) via type 1B phosphatidylino-
vvvvvvvvvvvvv IIb 3
eeeeeeeeeeeee
mmmmmmmmmmmmm long-term administration of the P2y antagonist clopidogrel sitol3-kinase(PI3-K).14Inanimalmodels,type1BPI3-Kis
bbbbbbbbbbbbb 12
er 1er 1er 1er 1er 1er 1er 1er 1er 1er 1er 1er 1er 1 ismoreeffectivethanaspirininreducingthecombinedriskof crucialforplateletactivationanditsabsenceprotectsagainst
0, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 2 ischemic stroke, myocardial infarction, or vascular death in thromboembolicvascularocclusion.15Inhumanplatelets,the
0000000000000 subjects with a prothrombotic condition such as diabetes role of type 1B PI3-K is less well understood, because
1111111111111
7777777777777
mellitus type 2 (DM2).7 These findings illustrate the crucial although being activated by ADP-P2y contact, it appears
12
role of P2y signaling in platelet activation in vitro and in under negative control by ADP-P2y binding and activation
12 1
vivo. of Src and PKC.10,16
The importance of P2y signaling is explained by its A prime example of abnormal platelet responsiveness is
12
capacity to initiate 2 pathways that directly interfere with observedinpatientswithDM2,whoarecharacterizedbyan
plateletactivatingorinhibitingmechanisms.First,thereisthe impaired responsiveness to insulin or even complete insulin
activation of the G-protein subunit G(cid:1), which inhibits ad- resistance.DM2subjectshavea2-to4-foldincreasedriskfor
i
enylylcyclaseandtherebyformationoftheplateletinhibitor cardiovascular disease and have both microvascular (ne-
cAMP.8 This property is particularly evident after treatment phropathy, retinopathy, neuropathy) and macrovascular (pe-
withprostacyclin,9andalsointheabsenceofcAMPelevating ripheral arterial disease) complications. DM2 subjects have
OriginalreceivedAugust4,2005;finalversionacceptedNovember17,2005.
FromtheLaboratoryforThrombosisandHaemostasis(I.A.F.,A.I.M.M.,G.G.,J.-W.N.A.),DepartmentofHematology,UniversityMedicalCenter
UtrechtandtheInstituteforBiomembranes(I.A.F.,A.I.M.M.,G.G.,J.-W.N.A.),UtrechtUniversity,theNetherlands;DepartmentofBiochemistryand
Human Biology (M.A.H.F., J.W.M.H.), University of Maastricht, the Netherlands; Department of Internal Medicine (T.W.v.H.), University Medical
CenterUtrecht,theNetherlands.
Correspondence to Dr J.W.N. Akkerman, Thrombosis and Haemostasis, Department of Hematology, University Medical Center Utrecht,
Heidelberglaan100,3584CXUtrecht,[email protected]
©2005AmericanHeartAssociation,Inc.
ArteriosclerThrombVascBiol.isavailableathttp://www.atvbaha.org DOI:10.1161/01.ATV.0000199519.37089.a0
1
2 Arterioscler Thromb Vasc Biol. February 2006
platelets that show increased adhesion, aggregation, TxA
2
production, and P-selectin expression.17 In general, DM2
subjectshavemarkedinsulinresistance,mostlyexplainedby
their obesity.18 There is indirect evidence that the hyperac-
tivityoftheirplateletsmaybecausedbyinsulinresistance.In
healthy individuals, platelets are inhibited by insulin leading
to reduced Ca2(cid:1) mobilization and aggregate formation.9,17
Interestingly, a euglycemic hyperinsulinemic clamp fails to
trigger platelet inhibition in obese insulin-resistant subjects
Figure1.IncreasedresponsivenessofDM2plateletstoaggre-
even in the absence of DM.19 The hyperactivity is likely to gatingagents.Aggregationofnormal(H)andDM2(D)PRPwas
havepathologicalconsequences,becausetheincreasedadhe- initiatedwith0.1to2.5(cid:4)g/mLcollagenand1to10(cid:4)mol/L
ADP.Themaximalaggregationwasmeasuredanddatawere
sion and aggregation will accelerate the formation of a fittedbynonlinearregression(H:R2(cid:6)0.91and0.79;DM2:
thrombusandenhancetheprocoagulantactivitythathelpsto R2(cid:6)0.92and0.86forcollagenandADP,respectively).
stabilize the thrombus.20 Means(cid:2)SD;n(cid:6)7healthyand7DM2subjects;(pleasesee
http://atvb.ahajournals.orgforthedetailedMaterialsand
In the present study, we investigated whether platelet
Methods).
hyperactivityobservedinDM2isassociatedwithadefectin
P2y signaling.Wedemonstratedrecentlyinhealthysubjects
12 insulin was absent in DM2 platelets (Figure I, available
that insulin interferes with ADP- and thrombin-induced
D online at http://atvb.ahajournals.org). These results indicate
ow platelet functions through interference with the P2y12- thatDM2plateletshaveadisturbedCa2(cid:1)homeostasisthatis
nloa mediated regulation of Gi9. After receptor binding, insulin unresponsive to inhibition by insulin.
d activates the insulin receptor substrate-1 (IRS-1) through
e
d
h from tGyir(cid:1)o-ssiunbeunpiht.osTphhoeryrleastuioltn,iswihnihcihbitiinointiatoefsGais(cid:1)socaicattiivointywaintdh IAngcgrreeagsaedtinRgeAspgoennstisveness of DM2 Platelets to
ttp impaired suppression of adenylyl cyclase through P2y12, Aggregation studies were performed to assess whether the
://a introducing a phenotype that resembles platelets with a disturbed Ca2(cid:1) homeostasis in DM2 platelets affected the
tvb.a congenital P2y12 defect or platelets from normal individuals responsivenesstocollagenorADP.Aggregationwasinitiated
hajourn HwehroehwaveedbeemenontrsetaratetedwthiatthpthlaetePl2eyt1h2yapnetargacotnivisittycloinpiDdoMgr2eli.s9 w(Fiitghur0e.11)t.oC2u.r5ve(cid:4)sgw/merLefciottleladgbeynnaonndlin1etaorr1e0gr(cid:4)esmsiooln/L,wAhDicPh
als.org lwikheilcyhtionsbuelincainutseerdfebreysawditehfescigtnianlitnhgebmyetchheaPn2isym12srethcreoputogrh. 0re.4su4ltaenddi0n.3E5C(cid:4)50gv/malLuefsorfonrocrmolalalgaennd-iDndMuc2edplaatgegleretsg,arteiospneoc-f
b/ tively. For ADP-induced aggregation, these data were 1.52
y g Materials and Methods and 0.79 (cid:4)mol/L, respectively. The aggregation response to
u
est o Ththtpe:/f/uatlvlbm.aehtahjoodusrncaalsn.obreg.found in the online data supplement at collagen and ADP was higher in DM2 platelets (P(cid:4)0.05).
n N These results indicate that the responsiveness of DM2 plate-
ov Results lets is increased especially for ADP.
e
m
be Disturbed Ca2(cid:1) Homeostasis in DM2 Absent Inhibition of Collagen- and ADP-Induced
r 10, 20 Irnelehaesaelthfryomsubinjetecrtsn,atlhsetoinrecsretahsaetiancccoymtopsoanliicesCpa2l(cid:1)at,e[lCeta2a(cid:1)c]tii,vbay- PAlgagtreelgeattiAonggstruedgieastiwonerebypeIrnfosrumlienditno DinMves2tigate whether
1
7 tion is easily disturbed by factors that interfere with P2y12 theunresponsivenesstoinsulinobservedintheregulationof
signalingsuchasinsulin.9Todetectpossibleabnormalitiesin Ca2(cid:1) had an effect on the role of insulin on platelet func-
this mechanism in DM2 subjects, Ca2(cid:1) regulation was mea- tions.9,22 Platelets were treated with 1 nmol/L insulin and
sured in platelets from healthy subjects and from DM2 aggregationwasinitiatedwithcollagenandADP.Inhealthy
subjects (normal and DM2 platelets, respectively; please see subjects, 1 nmol/L insulin inhibited collagen- and ADP-
http://atvb.ahajournals.orgforphysicalandbiochemicalchar- induced aggregation to 76(cid:2)11% and 75(cid:2)8%, respectively
acteristics of study subjects in Table I) in the absence of (P(cid:4)0.05).Theinhibitionbyinsulinwascompletelyabsentin
extracellular Ca2(cid:1). The mean [Ca2(cid:1)]i in unstimulated DM2 platelets from DM2 subjects (Figure II, available online at
platelets (64.4(cid:2)21.8 nmol/L) was (cid:3)2-fold higher than in http://atvb.ahajournals.org). Thus, DM2 platelets are unre-
normal controls (27.2(cid:2)5.9 nmol/L; P(cid:4)0.01). After stimula- sponsive to insulin.
tionwithcollagen,peak[Ca2(cid:1)] levelsdidnotdifferbetween
i
the 2 groups (378.2(cid:2)139.0 and 305.4(cid:2)208.2 nmol/L, Inhibition of Collagen-Induced Platelet Deposition
P(cid:5)0.05), respectively, but the time to peak was shorter in Under Flow by Insulin
DM2 platelets than normal platelets (80.7(cid:2)18.4 and Apartfromtheformationofaggregates,adhesionandgener-
103.6(cid:2)13.5 seconds, P(cid:4)0.03). In normal platelets, addition ation of a pro-coagulant surface by exposure of phosphati-
of insulin failed to change the basal [Ca2(cid:1)] but led to a dylserine(PS)areimportantstepsinplateletdepositionunder
i
dose-dependent reduction of collagen-induced Ca2(cid:1) mobili- flow. Earlier studies revealed an important role of P2y
12
zation (expressed as 100%) to 66(cid:2)11% at 100 nmol/L signaling in these processes.3 We determined platelet depo-
(P(cid:4)0.001). This agrees with earlier findings in platelets sition and binding of annexin V-fluorescein isothiocyanate
stimulatedwithADPandthrombin.9Incontrast,inhibitionby (FITC) to PS-exposing platelets after perfusion over a
Ferreira et al Impaired Insulin Signaling in DM2 Platelets 3
Figure3.Downstreamsignalingoftheinsulinreceptor/IRS-1
complex.Plateletswereincubatedwithinsulinfor15minutes
andsubjectedtoSDS-PAGE.A,BlotswereprobedwithP-PKB-
Ser473.B,Thebasalphosphorylationofindividualswas
expressedas100%.FurtherdetailsasinFigure1.
signalingtoG(cid:1),theabsentactivationofPKBsuggeststhat
i 2
Figure2.Inhibitionofcollagen-inducedplateletdeposition acommonstepintheinsulinsignalingmachineryisaffected.
underflowbyinsulin.Bloodwasperfusedatanintermediate
shearrate(1000s(cid:7)1)overacollagen-coveredsurface.A,Shown Insulin Inhibits Platelet Activation by Collagen by
are2typicalphasecontrastimagesofnormalbloodperfused
D Interfering With P2y Signaling to
o overcollagenintheabsence(upperpanel)andpresence(lower 12
w
n panel)of100nmol/Linsulin.Thearrowdepictstheflowdirec- cAMP-Dependent PKA
loa tionintheperfusionchamber.B,Surfacecoveragemeasuredin In platelets from healthy subjects, insulin interferes with the
d
e phasecontrastimages.C,Surfacecoveragewithannexin
d P2y -mediated suppression of cAMP formation, thereby
fro V-FITCstainingplatelets.FurtherdetailsasinFigure1. atten12uatingCa2(cid:1)increasesandreducingaggregationinduced
m
http collagen-coated surface at a shear rate of 1000 s(cid:7)1. Normal by ADP and thrombin.9 As shown in the present study, a
://a platelets rapidly adhered to collagen and formed aggregates similarinhibitionisobservedonstimulationbycollagen.H89
tvb (Figure 2A). The basal surface coverage by DM2 platelets isaspecificinhibitorofPKA.24PretreatmentwithH89fully
.ah (18.8(cid:2)1.3%) was higher than by normal platelets abolished the inhibition by insulin, confirming that insulin
a
jou (14.5(cid:2)1.3%, P(cid:4)0.001), which is in agreement with the reducedplateletaggregationviainterferencewiththecAMP-
rn dependent activation of PKA. To address the question
a hyperactivity of DM2 platelets observed in stirred suspen-
bls.org/ bsiyonnsor(mFiagluprela1te;lFetisguinreaId).o23seI-ndseuplienndreednutcmedansnuerrfa(c1e0.c7o(cid:2)ve1r.a4g%e wabhseetnhceeroifnasuctliivnataolrosnoercinhhainbgiteodrstohfealdeevneyllyolfcyccAlaMseP,rienstitnhge
y platelets were incubated with different concentrations of
g at100nmol/L;P(cid:4)0.001),whereasinhibitionbyinsulinwas
ue insulin followed by analysis of the phosphorylation of
st o absentinDM2platelets(Figure2Aand2B).Also,bindingof vasodilator-stimulatedphosphoprotein(VASP),amajorsub-
n N annexin V-FITC in perfusates with DM2 platelets strateofcAMP-dependentPKA.13NophosphorylatedVASP
ov (20.5(cid:2)3.4%) was higher than in controls (12.9(cid:2)2.0%,
e couldbedetectedinplateletstreatedwithinsulin.Incontrast,
m P(cid:4)0.001), probably as a result of the increased adhesion by
b additionofPGI thatthroughtheIPreceptorandG activates
er 1 DM2 platelets. In normal platelets, insulin reduced the adenylyl cyclas2e induced a rapid increase in phospshorylated
0, 2 binding of annexin V-FITC to 6.2(cid:2)2.1% at 100 nmol/L VASP (Figure III, available online at http://atvb.ahajournal-
01 (P(cid:4)0.001),butinDM2plateletsnoeffectofinsulincouldbe
7 s.org). Thus, although it interferes with the regulation of
detected (Figure 2C). Together, these results indicate that
cAMP and PKA, insulin is unable to change these signaling
DM2 platelets have an increased responsiveness to a
moleculesintheabsenceofplateletagonistsandantagonists.
collagen-coated surface under flow and that this property is
insensitive to the presence of insulin. Increased P2y Signaling in DM2 Platelets
12
Because in normal platelets insulin interferes with platelet
Downstream Signaling of the Insulin
functions through inhibition of P2y signaling,9 the unre-
12
Receptor/IRS-1 Complex
sponsivenesstoinsulininDM2plateletsmightbecausedby
In adipocytes, ineffective insulin signaling or insulin resis- disturbances in signal transduction from P2y to cAMP. To
12
tancehasbeenattributedtoabnormalitiesinIRS-1activation. address this issue, platelets were incubated with the stable
To investigate whether the loss of insulin signaling to Ca2(cid:1) PGI analog iloprost to raise cAMP and thereafter treated
2
regulatingmechanismsinDMplateletswasaccompaniedby with increasing concentrations ADP to induce different ex-
abnormal signaling initiated by the insulin receptor/IRS-1 tents of P2y signaling. Basal cAMP levels did not differ
12
complex, the phosphorylation of Ser473 on PKB was mea- betweennormalandDM2platelets.Innormalplatelets,ADP
sured. Platelets were incubated with insulin for 15 minutes dose-dependentlyreducediloprost-inducedcAMP(expressed
andsamplesweresubjectedtoSDS-PAGE.Innormalplate- as 100%) to 68.1(cid:2)8.5% (P(cid:4)0.001) at 10 (cid:4)mol/L ADP.
lets, insulin increased the phosphorylation of PKB-Ser473 to Addition of insulin partially reversed the effect of ADP in
192.0(cid:2)70.3% (P(cid:4)0.03). DM2 platelets showed a complete normal platelets leading to a reduction of cAMP to
lack of PKB phosphorylation. (Figure 3). Together with the 85.1(cid:2)12.7% (P(cid:4)0.02). The decline of iloprost-induced
absent interference with Ca2(cid:1), probably reflecting impaired cAMP was significantly steeper in DM2 platelets showing
4 Arterioscler Thromb Vasc Biol. February 2006
ADP-inducedaggregationbyinsulinisabolishedbythePKA
inhibitor H89. This is in line with the concept that insulin
interfereswithP2y -inducedsuppressionofadenylylcyclase
12
and cAMP formation through tyrosine phosphorylation of
G(cid:1), resulting in its inhibition. In DM2 platelets, P2y
i 2 12
signaling is present and functional, but the pathway appears
tobeupregulatedandlesssensitivetoP2y inhibition.These
12
findings indicate that insulin interferes with platelet activa-
tionbycollagenthroughthesamemechanismasinplatelets
stimulated with ADP and thrombin.9 They also indicate that
the cause for the loss of insulin sensitivity in platelets from
DM2 subjects must be sought in a defect in a pathway that
triggers the inhibition of G(cid:1).
i 2
Ahumanplateletcontains(cid:3)570insulinreceptors.25After
receptoractivation,IRS-1isrecruitedandtyrosinephosphor-
ylated leading to activation of pathways involving PKB and
p38 mitogen-activated protein kinases in addition to inhibi-
Figure4.IncreasedP2y -signalinginDM2.Plateletswerestim-
D ulatedwith1to10(cid:4)mol1/2L(A)or10(cid:4)mol/L(B)ADPinthepres- tionofGi(cid:1)29,18,26.Possibilitiesforabnormalinsulinsignaling
o enceorabsenceofiloprost.Iloprost-inducedcAMPaccumula- inDM2plateletsaredefectsintheinsulinreceptor(cid:2)-subunit,
w
nloa t0io.1ntwoa5s00exnpmreosls/LedARas-C160909%31.MCX,,PalantdeleCtsa2w(cid:1)emreobinilcizuabtiaotnedwwasith IRS-1,andthetyrosinephosphorylationofGi(cid:1)2togetherwith
de inducedby10(cid:4)mol/LADP.D,Plateletswereincubatedwith1 the different tyrosine kinases and phosphatases that control
d
fro to50nmol/LAR-C69931MX,andstimulatedwith10(cid:4)mol/L these processes.9,18,27,28 Defects in the insulin receptor are
m ADP.FurtherdetailsasinFigure1. associated with severe abnormalities such as growth disor-
h
ttp ders, lipodystrophy, and acanthosis nigrans, which were
://a reduction to 48.8(cid:2)11.1% (P(cid:4)0.001) at the same ADP absentintheDM2studygroup.Amorelikelyexplanationfor
tvb concentration(Figure4A),whichwasunresponsivetoinsulin the loss of insulin sensitivity in DM2 platelets is a defect in
.aha (Figure 4B). These results indicate that P2y12 signaling is IRS-1. In transfected cells, the IRS-1 gene G972R variant
jou increased in DM2 platelets and not affected by the presence causes impaired activation of PI3-K and PKB.29 The same
rna of insulin. mutation is found in obese patients with DM2,30 where it is
ls.o To investigate whether this abnormality affected the sen- associatedwithanincreasedriskofcardiovasculardisease.31
brg/ sitivity to the P2y12 receptor antagonist AR-C69931MX, However,thegenevariantisalsofoundinahealthyindivid-
y g dose-inhibition studies were performed in normal platelets. uals, indicating that it is not the decisive factor that makes
uest on cToonttahcits aenndd,dCifaf2e(cid:1)remntobciolinzcaetniotrnatwioanssiAndRu-cCe6d9b9y31AMDXP-wP2eyre1 pinladtievliedtusasltsimdiualbaetetidc.w32itAhn1a0ly0snismoofl/ILRSin-1sualinndrGevi(cid:1)e2aliendthreedDuMce2d
No addedtointerferewiththeP2y12-mediatedsupportof[Ca2(cid:1)]i and often absent tyrosine phosphorylation of these interme-
v
em increases (Figure 4C). Optimal inhibition was observed at diates.Unfortunately,thenormalcontrolsalsoshowedvary-
be (cid:3)100 nmol/L AR-C69931MX, at which point the inhibition inglevelsoftyrosinephosphorylation,whichmadeitdifficult
r 10 amounted to (cid:3)70%. Using 2 suboptimal concentrations of to identify a precise block in insulin signaling. Instead,
, 20 AR-C69931MX that induced (cid:3)50% and 25% inhibition, analysis of PKB activation consistently showed strongly
1
7 studies were repeated in DM2 platelets. At 50 nmol/L impairedSer473phosphorylationinDM2plateletbutnotinthe
AR-C69931MXnormalandDM2plateletsshowedasimilar controls. Thus, loss of insulin signaling to G(cid:1) and Ca2(cid:1) is
i 2
inhibition of ADP-P2y Ca2(cid:1) mobilization, but at 10 nmol/L accompaniedwithlossofsignalingtoPKB.Alikelycauseis
1
AR-C69931MX the inhibition in DM2 platelets was signifi- a defect in IRS-1, which is an upstream regulator of both
cantly lower than in normal controls (23.6(cid:2)4.3% and pathways. The nature of the defect in IRS-1 regulation
39.3(cid:2)11.9% respectively; P(cid:4)0.05; Figure 4D). These find- remainstobeelucidatedbutapparentlyitishasagreateffect
ings suggest that in DM2 platelets the ADP-P2y -mediated onG(cid:1).Gain-andloss-of-functiondefectsofG(cid:1) havebeen
12 i i 2
supportofADP-P2y inducedCa2(cid:1)increaseshasadecreased described in bipolar affective and bleeding disorders,3,33
1
sensitivity to the P2y receptor antagonist AR-C69931MX. whereas defects in tyrosine protein phosphorylation have
12
been described in congenital thrombocytopenia, Wiscott-
Discussion Aldrich syndrome, and the Scott syndrome.34–36
ThepresentdatashowthatplateletsfromDM2subjectshave In agreement with earlier studies in DM2 subjects,23,37,38
lost their responsiveness to insulin and show increased we found an increased basal [Ca2(cid:1)] in DM2 platelets before
i
responsivenesstoADP,whichsupportsplateletactivationby stimulation with collagen. Studies with thapsigargin, an
many agonists.8 In the presence of insulin, DM2 platelets inhibitor of sarco/endoplasmatic Ca2(cid:1)-ATPase (SERCA),
show increased collagen-induced Ca2(cid:1) mobilization. made clear that the Ca2(cid:1) level is determined by a constant
Collagen- and ADP-induced aggregation are slightly higher release from and re-uptake of Ca2(cid:1) by the endoplasmatic
than controls and DM2 platelets show increased adhesion to reticulum.39 Apparently, in DM2 platelets these processes
surface-coated collagen under flow with a concomitant in- have reached a new steady-state with a 2-fold higher [Ca2(cid:1)]
i
crease in PS exposure. In normal platelets, inhibition of than in normal platelets. This did not lead to a higher Ca2(cid:1)
Ferreira et al Impaired Insulin Signaling in DM2 Platelets 5
response after collagen stimulation, but the response was activation,thrombusgrowth,andthrombusstabilityininjuredarteries.
faster and accompanied with increased aggregation. In addi- JClinInvest.2003;112:398–406.
3. RemijnJA,WuYP,JeningaEH,IJsseldijkMJ,vanWilligenG,deGroot
tion, adhesion to collagen under flow was higher with DM2
PG,SixmaJJ,NurdenAT,NurdenP.RoleofADPreceptorP2Y(12)in
plateletsthanwithnormalplateletssuggestingthatabnormal- plateletadhesionandthrombusformationinflowingblood.Arterioscler
itiesinCa2(cid:1)regulationmadeplateletsmorereactivetowarda ThrombVascBiol.2002;22:686–691.
collagen-coated surface. At high shear platelet adhesion to 4. Leon C, Alex M, Klocke A, Morgenstern E, Moosbauer C, Eckly A,
SpannaglM,GachetC,EngelmannB.PlateletADPreceptorscontribute
collagen through interaction of glycoprotein (platelet glyco-
totheinitiationofintravascularcoagulation.Blood.2004;103:594–600.
protein [GP]) Ib with von Willebrand factor bound to colla- 5. Gachet C. ADP receptors of platelets and their inhibition. Thromb
gen is followed by integrin (cid:1)(cid:2) mediated firm adhesion, Haemost.2001;86:222–232.
2 1
6. FontanaP,GaussemP,AiachM,FiessingerJN,EmmerichJ,RenyJL.
which halts platelet rolling and allows collagen to interact
P2Y H2 haplotype is associated with peripheral arterial disease: a
with GPVI.40,41 Signaling mediated via GPVI increases the case-1c2ontrolstudy.Circulation.2003;108:2971–2973.
binding affinity of (cid:1)(cid:2) to collagen and induces release of 7. CAPRIESteeringCommittee.Arandomised,blinded,trialofclopidogrel
2 1
ADP.42,43 Adhesion to collagen of platelets deficient of the versusaspirininpatientsatriskofischaemicevents(CAPRIE).Lancet.
P2y receptor is (cid:5)20% lower than with normal platelets, 1996;348:1329–1339.
12 8. YangJ,WuJ,JiangH,MortensenR,AustinS,ManningDR,WoulfeD,
illustrating the potent enhancement by P2y12 signaling BrassLF.SignalingthroughGifamilymembersinplatelets.Redundancy
through released ADP.3 The binding of GPVI to collagen is andspecificityintheregulationofadenylylcyclaseandothereffectors.
independentofintracellularcontrolindicatingthatchangesin JBiolChem.2002;277:46035–46042.
9. Ferreira IA, Eybrechts KL, Mocking AIM, Kroner C, Akkerman JW.
D adhesion to collagen reflect modulation of (cid:1)2(cid:2)142. IRS-1mediatesinhibitionofCa2(cid:1)mobilizationbyinsulinviatheinhib-
ow Another abnormality found in DM2 platelets is the in- itoryG-proteinGi.JBiolChem.2004;297:3254–3264.
nlo creasedsignalingcapacityoftheP2y12pathway.DM2plate- 10. Hardy AR, Jones ML, Mundell SJ, Poole AW. Reciprocal cross-talk
a betweenP2Y andP2Y receptorsatthelevelofcalciumsignalingin
d lets induced a steeper fall in cAMP in iloprost-treated 1 12
ed humanplatelets.Blood.2004;104:1745–1752.
fro platelets than their normal counterparts. A similar enhance- 11. Feijge MA, Ansink K, Vanschoonbeek K, Heemskerk JW. Control of
m ment has been found in carriers of the P2y -H2 haplotype, plateletactivationbycyclicAMPturnoverandcyclicnucleotidephos-
http://atvb waatrhetihccehhpaislraacschmtaearraimzceteedmribzbeyradnpbeey.r6iapCnhaeirnrracielrresaarsoteefdrtihaPel2yPd12i2sy1er21ea2c-sHeep2tdohuraenpultoomtybthpeeer 12. pdp2hel0ano0tde2Dli;ee2ets7kti7ken:reh2ari9sbEe3i,2ttiGy1op–noe2rb-t9e3y3r.2cGB9Ai.,oMHchePeemmdusrPkinehrgakrmmJWeagc,aoAkl.akr2ky0eo0rcm4y;at6on7p:Jo1Wi5e5.s9iDs–.e1Jv5eB6lo7iop.lmCehnetmof.
.ah presence of hyper-responsive platelets that are less sensitive 13. Schwarz UR, Walter U, Eigenthaler M. Taming platelets with cyclic
a
journals tsoucphhaarsmcalcoopliodgoigcarell.s6trDatMeg2iespltahtaetletisnhsibhiotwPe2dy1a2 dsiegcnraelaisnegd, 14. npLuaotcvhlaewoPatyi,dPiesas.greaBqniuionicriheSedm,fSoPrinhaiagcrtaimgvlaaitaciooFln.,o2Bf0a0tlh1de;u6is2nm:i1aC1ll5,3GT–oT1rP1tia6s1Me..RAapG1Bi-dienphenudmeannt
.o responsiveness to a suboptimal concentration of the P2y platelets.JBiolChem.2002;277:12009–12015.
brg/ antagonist AR-C69931MX but at an optimal concentratio1n2 15. HirschE,BoscoO,TropelP,LaffargueM,CalvezR,AltrudaF,Wymann
y M, Montrucchio G. Resistance to thromboembolism in PI3Kgamma-
gu the difference disappeared. Clopidogrel resistance has been deficientmice.FASEBJ.2001;15:2019–2021.
es describedinapatientpopulationwithageneralhighriskfor 16. AbramsCS,ZhangJ,DownesCP,TangX,ZhaoW,RittenhouseSE.
t on recurrent vascular events, although it was not specific for Phosphopleckstrin inhibits Gbetagamma-activable platelet
No DM.44 Our present findings might indicate that clopidogrel phosphatidylinositol-4,5-bisphosphate3-kinase.JBiolChem.1996;271:
v 25192–25197.
em resistance can be overcome by applying higher doses of the 17. WinocourPD.Plateletabnormalitiesindiabetesmellitus.Diabetes.1992;
b
e drug. 41:26–31.
r 10 PlatelethyperactivityinDM2correlateswithanincreased 18. SaltielAR,KahnCR.Insulinsignallingandtheregulationofglucoseand
, 20 risk of atherothrombotic complications.45–48 In addition to lipidmetabolism.Nature.2001;414:799–806.
1 19. WesterbackaJ,Yki-JarvinenH,TurpeinenA,RissanenA,VehkavaaraS,
7 hyperactive platelets, the coagulation mechanism shows ab-
SyrjalaM,LassilaR.Inhibitionofplatelet-collageninteraction:aninvivo
normalities in DM2 subjects with elevated levels of coagu- actionofinsulinabolishedbyinsulinresistanceinobesity.Arterioscler
lation factors factor VII, VIII, XI, and XII.49 The enhanced ThrombVascBiol.2002;22:167–172.
20. HeemskerkJW,BeversEM,LindhoutT.Plateletactivationandblood
adhesionofplateletsinDM2withaconcomitantincreasein
coagulation.ThrombHaemost.2002;88:186–193.
PS exposure that facilitates the coagulation cascade might 21. Report of the expert committee on the diagnosis and classification of
contribute to the hypercoagulable state observed in this diabetesmellitus.DiabetesCare.2003;26:S5–S20.
disease. 22. FerreiraIA,MockingAI,UrbanusRT,VarlackS,WnukM,Akkerman
JW. Glucose uptake via glucose transporter 3 by human platelets is
regulatedbyproteinkinaseB.JBiolChem.2005;280:32625–32633.
Acknowledgments
23. Mazzanti L, Rabini RA, Faloia E, Fumelli P, Bertoli E, De Pirro R.
ThesupportoftheDutchDiabetesResearchFoundation(1999.046; AlteredcellularCa2(cid:1)andNa(cid:1)transportindiabetesmellitus.Diabetes.
Amersfoort, the Netherlands) and the “Fonds voor het Hart” (The 1990;39:850–854.
Hague, the Netherlands) is gratefully acknowledged. J.W.N.A. is 24. denDekkerE,HeemskerkJW,GorterG,vandV,DonathJ,KronerC,
supportedbytheNetherlandsThrombosisFoundation.WethankDr MikoshibaK,AkkermanJW.CyclicAMPraisesintracellularCa(2(cid:1))in
S.J.Korporaal,M.vanLier,DrE.J.P.deKoning,andJ.P.vanWijk human megakaryocytes independent of protein kinase A. Arterioscler
ThrombVascBiol.2002;22:179–186.
(UMCUtrecht,theNetherlands)fortheirassistance,andthevolun-
25. HajekAS,JoistJH,BakerRK,JarettL,DaughadayWH.Demonstration
teersfortheirhelp.
andpartialcharacterizationofinsulinreceptorsinhumanplatelets.JClin
Invest.1978;63:1060–1065.
References
26. Saltiel AR. Putting the brakes on insulin signaling. N Engl J Med.
1. RossR.Atherosclerosis-aninflammatorydisease.NEnglJMed.1999; 2003;349:2560–2562.
340:115–125. 27. Obermaier-Kusser B, White MF, Pongratz DE, Su Z, Ermel B, Muhl-
2. AndreP,DelaneySM,LaRoccaT,VincentD,DeGuzmanF,JurekM, bacherC,HaringHU.Adefectiveintramolecularautoactivationcascade
Koller B, Phillips DR, Conley PB. P2Y regulates platelet adhesion/ may cause the reduced kinase activity of the skeletal muscle insulin
12
6 Arterioscler Thromb Vasc Biol. February 2006
receptor from patients with non-insulin-dependent diabetes mellitus. 38. PellegattaF,FolliF,RonchiP,CaspaniL,GalliL,VicariAM.Deranged
JBiolChem.1989;264:9497–9504. platelet calcium homeostasis in poorly controlled IDDM patients.
28. Ferreira IA, Akkerman JWN. IRS-1 and vascular complications in DiabetesCare.1993;16:178–183.
diabetesmellitus.VitaminsHorm.2005;70:25–66. 39. RosadoJA,PorrasT,CondeM,SageSO.Cyclicnucleotidesmodulate
29. FedericiM,PandolfiA,DeFilippisEA,PellegriniG,MenghiniR,Lauro store-mediatedcalciumentrythroughtheactivationofprotein-tyrosine
D,CardelliniM,RomanoM,SestiG,LauroR,ConsoliA.G972RIRS-1 phosphatasesandalteredactinpolymerizationinhumanplatelets.JBiol
variantimpairsinsulinregulationofendothelialnitricoxidesynthasein Chem.2001;276:15666–15675.
culturedhumanendothelialcells.Circulation.2004;109:399–405.
40. MiuraY,TakahashiT,JungSM,MoroiM.Analysisoftheinteractionof
30. SestiG,FedericiM,HribalML,LauroD,SbracciaP,LauroR.Defects
platelet collagen receptor glycoprotein VI (GPVI) with collagen. A
oftheinsulinreceptorsubstrate(IRS)systeminhumanmetabolicdis-
dimericformofGPVI,butnotthemonomericform,showsaffinityto
orders.FASEBJ.2001;15:2099–2111.
fibrouscollagen.JBiolChem.2002;277:46197–46204.
31. MariniMA,FrontoniS,MineoD,BracagliaD,CardelliniM,DeNicolais
41. SixmaJJ,vanZantenGH,HuizingaEG,vanderPlasRM,VerkleyM,
P,BaroniA,D’AlfonsoR,PernaM,LauroD,FedericiM,Gambardella
S,LauroR,SestiG.TheArg972variantininsulinreceptorsubstrate-1is WuYP,GrosP,deGrootPG.Plateletadhesiontocollagen:anupdate.
associated with an atherogenic profile in offspring of type 2 diabetic ThrombHaemost.1997;78:434–438.
patients.JClinEndocrinolMetab.2003;88:3368–3371. 42. Nieswandt B, Watson SP. Platelet-collagen interaction: is GPVI the
32. BaroniMG,D’AndreaMP,MontaliA,PannitteriG,BarillaF,Campagna centralreceptor?Blood.2003;102:449–461.
F,MazzeiE,LovariS,SeccarecciaF,CampaPP,RicciG,PozzilliP, 43. GibbinsJM.Plateletadhesionsignallingandtheregulationofthrombus
Urbinati G, Arca M. A common mutation of the insulin receptor formation.JCellSci.2004;117:3415–3425.
substrate-1geneisariskfactorforcoronaryarterydisease.Arterioscler 44. MatetzkyS,ShenkmanB,GuettaV,ShechterM,BienartR,Goldenberg
ThrombVascBiol.1999;19:2975–2980. I,NovikovI,PresH,SavionN,VaronD,HodH.Clopidogrelresistance
33. SteinMB,ChenG,PotterWZ,ManjiHK.G-proteinlevelquantification isassociatedwithincreasedriskofrecurrentatherothromboticeventsin
D inplateletsandleukocytesfrompatientswithpanicdisorder.Neuropsy- patients with acute myocardial infarction. Circulation. 2004;109:
ow chopharmacology.1996;15:180–186. 3171–3175.
nloa 34. BWaolllmterasierS,MB,oSgcehnublezregeHr,JS,trWaueslsteGK,C. hTehrrkoamoubiopKo,ieWtinittninerpNat,ieLnytsnewniSth, 45. PGaoyondealDlAAH,J,oNnaeysloCrI,AHRa.yBeesnPefDic,iaTlheofmfepcstsonofMclMop,idLoognrdeolncoNmJb,iBneedllwPiRth,
d
ed congenital thrombocytopenia and absent radii: elevated serum levels, aspirininreducingcerebralemboliinpatientsundergoingcarotidendar-
from nBolormoda.l1r9e9ce7p;9to0r:6e1x2p–r6e1ss9i.on, but defective reactivity to thrombopoietin. 46. tCeroelcwtoemllyJ.AC.irAcuslpaitriionn.th20er0a4p;1y09in:14d7ia6b–e1t4e8s.1.Diabetes Care. 2004;27:
h 35. GrossBS,WildeJI,QuekL,ChapelH,NelsonDL,WatsonSP.Regu-
ttp lationandfunctionofWASpinplateletsbythecollagenreceptor,gly- S72–S73.
://a coproteinVI.Blood.1999;94:4166–4176. 47. Collaborative overview of randomised trials of antiplatelet therapy–I:
tv 36. DekkersDW,ComfuriusP,VuistWM,BillheimerJT,DickerI,Weiss Prevention of death, myocardial infarction, and stroke by prolonged
b
.a HJ,ZwaalRF,BeversEM.ImpairedCa2(cid:1)-inducedtyrosinephosphor- antiplatelet therapy in various categories of patients. Antiplatelet Tri-
hajou ySlcaotitotnsyannddrodmefee:catisvteudliypiudsisncgraamnbilninhgibiintoerrfyotrhsrocrcaymtebslfarsoemthaatpmatiimenitcswtihthe 48. aRliinstgsl’ebCoPllAab,oBrahtaiottn.DBLM,JH.i1r9sc9h4;3A0T8,:8T1–o1p0o6l.EJ, Hacke W. Benefit of
rna defectinScottsyndrome.Blood.1998;91:2133–2138. clopidogreloveraspirinisamplifiedinpatientswithahistoryofischemic
ls.o 37. Li Y, Woo V, Bose R. Platelet hyperactivity and abnormal Ca(2(cid:1)) events.Stroke.2004;35:528–532.
rg homeostasis in diabetes mellitus. Am J Physiol Heart Circ Physiol. 49. CarrME.Diabetesmellitus:ahypercoagulablestate.JDiabetesCompli-
b/ 2001;280:H1480–H1489. cations.2001;15:44–54.
y
g
u
e
s
t o
n
N
o
v
e
m
b
e
r 1
0
, 2
0
1
7
D
o Platelet Inhibition by Insulin Is Absent in Type 2 Diabetes Mellitus
w
nlo Irlando Andrade Ferreira, Astrid I.M. Mocking, Marion A.H. Feijge, Gertie Gorter, Timon W.
a
d van Haeften, Johan W.M. Heemskerk and Jan-Willem N. Akkerman
e
d
fro
m
h
ttp Arterioscler Thromb Vasc Biol. published online December 8, 2005;
://a Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272
tv
b Greenville Avenue, Dallas, TX 75231
.a
h Copyright © 2005 American Heart Association, Inc. All rights reserved.
a
jo Print ISSN: 1079-5642. Online ISSN: 1524-4636
u
rn
a
ls
.o
rg
b/
y
g
u
e
s
t o The online version of this article, along with updated information and services, is located on the
n
N World Wide Web at:
o
ve http://atvb.ahajournals.org/content/early/2005/12/08/01.ATV.0000199519.37089.a0.citation
m
b
e
r 1
0, 2 Data Supplement (unedited) at:
01 http://atvb.ahajournals.org/content/suppl/2005/12/12/01.ATV.0000199519.37089.a0.DC1
7
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
inArteriosclerosis, Thrombosis, and Vascular Biology can be obtained via RightsLink, a service of the
Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for
which permission is being requested is located, click Request Permissions in the middle column of the Web
page under Services. Further information about this process is available in thePermissions and Rights
Question and Answerdocument.

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

Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online
at:
http://atvb.ahajournals.org//subscriptions/
Online Supplement of:

Platelet Inhibition by Insulin is Absent in Type 2 Diabetes Mellitus

Irlando Andrade Ferreira1*, Astrid I. M. Mocking1, Marion A. H. Feijge2, Gertie Gorter1, Timon W. van
Haeften3, Johan W. M. Heemskerk2, and Jan-Willem N. Akkerman1

From the 1Laboratory for Thrombosis and Haemostasis, Department of Hematology, University Medical Center
Utrecht, and the 1Institute for Biomembranes, Utrecht University, the Netherlands; 2Department of Biochemistry
and Human Biology, University of Maastricht, the Netherlands; 3Department of Internal Medicine, University
Medical Center Utrecht, the Netherlands

Running title: Impaired insulin signaling in DM2 platelets

Corresponding author:
Prof. Dr. J. W. N. Akkerman Thrombosis and Haemostasis, Department of Hematology, University Medical
Center Utrecht, Heidelberglaan 100, 3584 CX Utrecht, the Netherlands, phone: [phone], fax: +31 30
251 1893, e-mail: [email protected]

Word Count:
Abstract 150
Condensed abstract 48
Total word count MS 4949
Online text word count 2050
Online Supplement
Impaired insulin signaling in DM2 platelets
Materials and Methods
Subjects
The healthy (n=14) and DM2 subjects, diagnosed according to the guidelines of the Expert Committee
on the diagnosis and classification of DM 21, without clinical signs of peripheral arterial disease (n=14) recruited
for the present study did not take any platelet-inhibiting medication 10 days prior to blood-collection. All DM2
subjects were treated with oral glucose lowering agents (biguanides or sulfonylurea derivatives). The physical
and biochemical characteristics are available in the online supplement Table 1. The DM2 subjects had a higher
BMI and systolic blood pressure than healthy subjects. The biochemical characteristics did not differ between
the two study groups with the exception of fasting serum insulin, glucose and serum creatinine. The degree of
insulin resistance was estimated by homeostasis model assessment (HOMA), which is an independent marker of
CVD in DM2, by the following formula: fasting plasma glucose (mmol/L) times fasting plasma insulin (mU/L)
divided by 22.51. Low HOMA values represent a high insulin efficiency, whereas high values represent insulin
inefficiency or insulin resistance. HOMA was increased 3-fold in DM2 subjects, thus reflecting the insulin
resistant state in these subjects.
Exclusion criteria included a previous cardiovascular event, pregnancy, and the use of insulin, anti-
epileptic drugs, acetyl salicylic acid or other non-steroidal anti-inflammatory drugs. Written informed consent
was obtained after the purpose, characteristics, and potential risks of the experiments had been explained to the
subjects. No sex-based or racial/ethnic-based differences were present. The study protocol was approved by the
Medical Ethics Committee of the University Medical Center Utrecht (UMCU, the Netherlands) and the
Academic Hospital Maastricht (the Netherlands). For logistic reasons the patients group was divided in a group
for adhesion analysis under flow (n=7, University of Maastricht) and a group for the other measurements (n=7,
University Medical Center Utrecht). The characteristics of the patient groups did not differ significantly.

Materials
We obtained collagen reagent Horm (collagen in short), as native collagen type I/III fibrils from equine tendons,
from Nycomed Pharma (Munich, Germany); D-Phe-Pro-Arg-chloromethylketone (PPACK), heparin and the
monoclonal phosphospecific vasodilator phosphoprotein (VASP P-Ser157) antibody from Calbiochem (San
Diego, CA, USA); prostacyclin (PGI ) and iloprost (PGE ) from Cayman Chemical (Ann Arbor, MI, USA);
2 2
human recombinant insulin (solubilized according to the recommendations of the manufacturer in 10 mmol/L
acetic acid, 100 mmol/L NaCl, and 0.01% BSA to reach the stock concentration of 100 µmol/L), adenosin-5’-
II
Online Supplement
Impaired insulin signaling in DM2 platelets
diphosphate (ADP) and Fura 2-AM from Sigma (St. Louis, MO, USA); FITC labeled annexin V from Nexus
Research (Hoeven, the Netherlands); the isoquinolinesulfonamide compound H89 (N-[2-(p-
bromocinnamylamino)ethyl]-5-isoquinolinesulfonamide) from Alexis Biochemicals (Lausen, Switzerland);
phosphospecific protein kinase B (P-PKB) Ser473 and P-selectin (C-20) antibodies from Santa Cruz
Biotechnology (Santa Cruz, CA, USA); goat anti-mouse horseradish peroxidase (HRP) (GAMPO), rabbit anti-
goat HRP (RAGPO) from DAKO A/S (Glostrup, Denmark) and goat anti-rabbit HRP from Cell Signaling
Technology Inc. (Beverly, MA, USA). The ADP receptor P2Y antagonist, the ATP analogue N6-(2-
12
methylthioethyl)-2-(3,3,3-trifluoropropylthio)-β,γ-dichloromethylene ATP (AR-C69931MX) was a kind gift
from Astra Zeneca (Loughborough, UK) and the PGI analog iloprost was kindly provided by Schering AG
2
(Berlin, Germany). All other chemical reagents were of analytical grade.

Preparation of Washed Platelets
Freshly drawn venous blood was collected into 0.1 volume of 130 mmol/L trisodium citrate. Citrated
blood was centrifuged (150xg, 15 minutes, 20oC), and the platelet-rich plasma (PRP) was collected. The
remaining blood was centrifuged (800xg, 10 minutes, 20oC) to obtain platelet-poor plasma (PPP; see below).
PRP was supplemented with 0.1 vol of ACD (2.5% trisodium citrate, 1.5% citric acid, 2% D-glucose) and
centrifuged (330xg, 15 minutes, 20oC). The supernatant was removed and the platelet pellet was resuspended in
HEPES/Tyrode buffer (145 mmol/L NaCl, 5 mmol/L KCl, 0.5 mmol/L Na HPO , 1 mmol/L MgSO , 10 mmol/L
2 4 4
HEPES, pH6.5) containing 5 mmol/L D-glucose. Prior to centrifugation (330xg, 15 minutes, 20oC), 0.001 vol of
PGI (10 ng/mL final concentration) was added. The platelet pellet was resuspended in HEPES/Tyrode buffer
2
(pH7.25) containing 5 mmol/L D-glucose. The final platelet concentration was adjusted to 2.0x1011 cells/L. Prior
to the experiments, platelets were kept at 20oC for 45 minutes to ensure a resting state. Where indicated, cells
were incubated with 0.1-1000 nmol/L AR-C69931MX for 30 seconds, 10 µmol/L H89 for 10 minutes, 1-100
nmol/L insulin for 5 minutes, or 10 µg/L iloprost for 1 minute. These treatments were followed by stimulation
with 2.5 µg/mL collagen or 10 µmol/L ADP unless stated otherwise.

Measurement of Ca2+ Mobilization
PRP was prepared as described above and incubated with 3 µmol/L Fura 2-AM (45 minutes, 37oC,
light-protected). Then PRP was supplemented with ACD, centrifuged again (330xg, 15 minutes, 20oC) and
resuspended in HEPES/Tyrode buffer (pH7.25) containing 5 mmol/L D-glucose. The final platelet concentration
III