of February 17, 2018.This information is current as

through Glutamate Receptor SignalingPlatelets Contribute to Allograft Rejection

William M. Baldwin III and Craig N. MorrellAnneMarie F. Swaim, David J. Field, Karen Fox-Talbot,

http://www.jimmunol.org/content/185/11/6999doi: 10.4049/jimmunol.1000929October 2010;

2010; 185:6999-7006; Prepublished online 20J Immunol 

MaterialSupplementary

9.DC1http://www.jimmunol.org/content/suppl/2010/10/20/jimmunol.100092

Referenceshttp://www.jimmunol.org/content/185/11/6999.full#ref-list-1

, 16 of which you can access for free at: cites 41 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. All rights reserved.1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on February 17, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on February 17, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

Platelets Contribute to Allograft Rejection throughGlutamate Receptor Signaling

AnneMarie F. Swaim,*,1 David J. Field,† Karen Fox-Talbot,‡ William M. Baldwin, III,x and

Craig N. Morrell†

Platelets recruit leukocytes and mediate interactions between leukocytes and endothelial cells. Platelets have been long described as

markers of transplant rejection, but the contribution of platelets to transplant rejection has not been critically examined. We dem-

onstrate in this study that following T cell initiation of allograft rejection, platelets contribute to T cell recruitment and increased

plasma inflammatory mediators and accelerate T cell-meditated skin graft rejection. Prior work from our laboratory has shown

that platelets secrete glutamate when activated, which then induces platelet thromboxane production by signaling through platelet-

expressed ionotropic glutamate receptors. Glutamate receptor antagonists therefore represent, to our knowledge, novel inhibitors of

platelet-accelerated inflammation. We have found that plasma glutamate is increased in mice that receive skin grafts and that mice

treated with glutamate receptor antagonists have improved graft survival and decreased plasma thromboxane, platelet factor 4

(CXCL4), and IFN-g. Taken together, our work now demonstrates that subsequent to T cell initiation of skin graft rejection,

platelets contribute to further T cell recruitment and that by blunting glutamate-mediated platelet activation, graft survival is

improved. The Journal of Immunology, 2010, 185: 6999–7006.

According to the National Organ Procurement and Trans-plantation Network, ∼23,000 Americans received anorgan transplant in 2008. Current immune suppression

protocols have greatly improved transplant survival, but despitemany therapeutic advances, allograft rejection continues. Plateletshave key roles in the recruitment of immune cells and can accel-erate vascular inflammatory diseases (1–5). We have implicatedplatelets in amplifying alloantibody-mediated transplant inflamma-tion (6). However, the role of platelets in T cell-directed immuneresponses in general, and transplant rejection in particular, has notbeen explored fully.Early pathological descriptions of transplants recognized the

presence of platelets in kidney transplants (7) and a number ofstudies followed in which platelets were labeled with indium totrack their accumulation in renal allografts (8, 9). These studiesled to the general observation that platelets accumulate in renalallografts and may be markers of graft rejection. Recent studies

have also identified the presence of prominent intravascular plate-let aggregates in experimental and clinical transplants that undergoAb-mediated rejection (10–13). Kirk and colleagues (14) have gonebeyond these more observational studies and made a critical findingthat CD154 (CD40 ligand) shed from platelets can serve as a co-stimulatory molecule remote from the transplant to induce rejec-tion of murine cardiac allografts. Using a skin transplant model, wedemonstrated that platelets have a key role in increasing leukocytetrafficking and graft vascular inflammation (6). These studies set thestage for a deeper investigation into platelets not only as markers oftransplant vascular injury, but also as mediators contributing to thepathogenesis of graft rejection.Platelets can recruit lymphocytes to the site of the inflammation

through contact-dependent and independent mechanisms. T cellsexpress P-selectin glycoprotein ligand 1 that interacts with P-selectin expressed by activated platelets. Platelets may also in-teract with T cells through CD154 on platelets. In experiments ofischemia-reperfusion injury to the liver, platelets augmented therecruitment of CD4+ T cells to hepatic sinusoids (15). Activatedplatelets also enhance the intrahepatic accumulation of cytotoxicT lymphocytes in murine models of viral hepatitis (16). Plateletand T cell interactions may do more than just localize T cells; theymay also augment T cell immune responses. CD154 on plateletscan augment the help delivered by low levels of CD4+ T cells forgerminal center development and isotype switching to IgG Abproduction (17). We have found that the platelet-derived chemokineplatelet factor 4 (PF4/CXCL4) can increase T cell CXCR3 expres-sion and T cell trafficking in a model of neurovascular inflammation(18). These studies provide a rationale to consider the poorly ex-plored role of platelets in T cell-mediated allograft rejection.Cyclooxygenase (COX) products, such as PGs and thromboxane,

have important roles in stimulating and shaping acquired immuneresponses. Platelets are a major source of these proinflammatorymolecules. Receptors for COX products are expressed by manyimmune cells, including T cells. Recent work has demonstratedthat some PGs, such as PGE2, acting on T cells facilitate Th1 celldifferentiation and amplify IL-23–mediated Th17 cell expansion

*Department of Molecular and Comparative Pathobiology and ‡Department ofPathology, Johns Hopkins University School of Medicine, Baltimore, MD 21205;†Aab Cardiovascular Research Institute, University of Rochester School of Medicineand Dentistry, Rochester, NY 14642; and xDepartment of Immunology, Lerner Re-search Institute, Cleveland Clinic Foundation, Cleveland, OH 44195

1Current address: Purdue School of Veterinary Medicine, Purdue University, WestLafayette, IN.

Received for publication March 24, 2010. Accepted for publication September 20,2010.

This work was supported by National Institutes of Health Grants R01HL093179,R01HL093179-02S109, and R01HL094547 (to C.N.M.). W.M.B. is supported byNational Institutes of Health Grant P01AI087586-01.

Address correspondence and reprint requests to Dr. Craig N. Morrell, Aab Cardio-vascular Research Institute, University of Rochester School of Medicine and Den-tistry, 601 Elmwood Avenue, Box CVRI, Rochester, NY 14642. E-mail address:Craig_Morrell@URMC.Rochester.edu

The online version of this article contains supplemental material.

Abbreviations used in this paper: AMPA, 2-amino-5-hydroxy-5-methyl-4-isoxazole-propion acid; CNQX, 6-cyano-7-nitroquinoxaline-2,3-dione; COX, cyclooxygenase;PF4, platelet factor 4; TBX2, thromboxane B2; TP, thromboxane receptor.

Copyright� 2010 by The American Association of Immunologists, Inc. 0022-1767/10/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1000929

by guest on February 17, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

(19). The same investigators had found earlier that the bindingactivity for thromboxane receptors (TPs) is high in T cells, but notin B cells, suggesting that thromboxane–TP signaling may mod-ulate peripheral T cell immune responses (20, 21). Studies specificto transplantation have demonstrated that TP2/2 mice had reducedimmune-mediated tissue injury following cardiac transplantation,showing that thromboxane augments cellular immune responses totransplants and inflammatory tissue injury (22).We have recently demonstrated that platelets have an important

role in alloantibody-induced innate immune responses to skin grafts(6). Platelets support leukocyte and endothelial cell interactionsand contribute proinflammatory molecules that sustain leukocyterecruitment (23). We have also demonstrated that platelets con-tain and release large concentrations of glutamate upon activa-tion. Platelets express ionotropic glutamate receptors including 2-amino-5-hydroxy-5-methyl-4-isoxazolepropion acid (AMPA) andkainate type receptors (24). Glutamate promotes platelet activationby signaling via platelet glutamate receptors, leading to COX acti-vation, the production of thromboxane, and amplification of plateletactivation (25). Glutamate-induced platelet thromboxane pro-duction may do more than just support platelet activation andthrombus growth; it may also promote immune activation, includingT cell-driven transplant rejection.We demonstrate in this study that platelets have a role in

transplant rejection. Using a skin graft model, we found that fol-lowing T cell initiation of graft rejection, platelets are activated,leading to increased plasma glutamate and platelet thromboxaneproduction. Glutamate receptor antagonists demonstrate therapeu-tic potential in reducing T cell-mediated transplant rejection.

Materials and MethodsReagents

PF4, TNF-a, and IFN-g ELISA kits were purchased from R&D Systems(Minneapolis, MN). Thromboxane B2 (TBX2) ELISA kits were purchasedfrom Cayman Chemical (Ann Arbor, MI). Glutamate concentration wasdetermined using a Glutamate Oxidase kit from Invitrogen (Carlsbad, CA).The 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) was purchased fromSigma-Aldrich (St. Louis, MO).

Platelet-depletion Ab and its control IgG were purchased from EmfretAnalytics (Wurzburg, Germany). The depleting Abs are a mixture of ratmAbs directed against mouse GPIb (CD42b). This is a platelet-specificreceptor and results in the depletion of only platelets. At the concen-trations we used (1 mg/g), platelet depletion is sustained for no more than3 d (Supplemental Fig. 1). mAbs to the TCR b-chain (clone H57-597), CD4,CD8, and CD14 were purchased from BD Pharmingen (San Diego, CA).

Transplants

All mouse studies were performed under protocols approved by the Uni-versity of Rochester Animal Care and Use Committee using procedures wehave published (6). Briefly, B6 nude mice (Taconic Farms, Germantown,NY) were anesthetized with ketamine and xylazine (80 and 13 mg/kg,respectively) and transplant beds prepared by removal of the epidermisand dermis. The thin skin from ears of H-2 incompatible B10.A mice(H-2Kk, The Jackson Laboratory, Bar Harbor, ME) were grafted into fittedbeds on 10–12-wk-old C57BL/6 nude mice (H-2Kb). Mice were bandagedfor 1 wk to allow for graft healing and establishment of vascular con-nections.

Immunohistochemistry

Harvested transplants were fixed in methanol/water/acetic acid (60:30:10%).Tissue was then embedded and sectioned and immunohistochemistry per-formed using protocols and procedures described previously with Abs toCD3 (26, 27).

T cell isolations

T cells were isolated from the spleen and lymph nodes of control C57BL/6mice using a negative selection T cell enrichment kit (StemCell Tech-nologies, Vancouver, British Columbia) i.v. via the retro-orbital plexus intograft recipients.

To quantify T cell infiltrates in the skin grafts, we harvested skin from thecenter of the transplant tissue and placed it in 3 ml RPMI media with 5%FBS for 1 h. The base of the graft was then gently scraped with a razorblade, the skin then sectioned into small pieces, passed in and out of an 18-gauge needle, and vortexed vigorously to dissociate cells. The cells weretransferred into a 15-ml tube, 1 ml Percoll added, mixed, and centrifugedat 1300 3 g for 30 min to isolate a mononuclear cell layer at the interface.Mononuclear cells were then washed before incubating with a mAb andthe number of cells as percent total mononuclear cells determined by flowcytometry.

Statistical methods

Data are expressed as means 6 SD. Statistical comparisons between twogroups were performed using Student t test. Graft survival was analyzedusing a log-rank test.

ResultsTo determine the role of platelets in T cell-dependent transplantrejection, we used a skin transplant model in which the thin skinfrom an ear of a B10.A mouse (H-2Kk) is grafted onto the flank ofa C57BL/6 nude mouse (H-2Kb). After allowing 7 d for grafthealing and establishment of vascular connections, bandages wereremoved, and mice were reconstituted with 1 3 106 T cells iso-lated from naive wild-type C57Bl6 mice. Using this model in ourprior published work, we have demonstrated that the skin graftsare vascularized, and we have imaged both platelet and WBCperfusion of the grafts (23). Other have also demonstrated thatvessels of the host dermis and the skin graft connect by 7 d post-transplantation and that endothelial cells within the graft are ofdonor origin until ∼21 d (28). The time of T cell reconstitution isreferred to as study day 0.To establish that platelets were activated following T cell re-

constitution and T cell initiation of graft rejection, plasma wasisolated from control ungrafted nude mice, mice that received skingrafts but no T cells, and mice that were given skin grafts and wereT cell reconstituted. PF4 is a platelet-specific chemokine and wasused as a plasma marker of platelet activation. Mice that receivedskin grafts but no T cells did not have significantly increasedplasma PF4 compared with control nude mice (Fig. 1A, whiteversus gray bars). However, T cell-reconstituted mice had a largeincrease in plasma PF4 that was sustained over 4 d after T cellreconstitution (Fig. 1A, black versus white bars), indicating thatT cells are needed to initiate platelet activation. T cell re-constitution in the absence of a skin allograft had no effect onplatelet activation (Supplemental Fig. 2).To determine whether platelet activation plays a role in graft

rejection, mice that had received skin grafts were T cell recon-stituted, and on study day 0, mice were either treated with controlIgG or a mixture of mAbs (1 mg/g) against mouse GPIb (CD42b)to deplete platelets. This treatment greatly reduces platelet num-bers (Supplemental Fig. 1) but has no effect on other cell pop-ulations (18, 23, 29). Skin graft survival was then evaluated on day7 using visual parameters. Total rejection was manifested by scabformation and graft retraction (Fig. 1B). Approximately 50% ofcontrol IgG-treated mice had visible evidence of graft rejection,whereas ,20% of platelet-depleted mice had rejected skin grafts(Fig. 1C). To look at graft survival over a longer time, we alsofollowed the grafts for 10 d after T cell reconstitution. In contrastto Fig. 1C, to achieve sustained platelet depletion, mice weretreated with control IgG or platelet-depleting Ab on days 0 and 4.By day 10, control graft survival was only 11% (Fig. 1D, whitediamonds). In contrast, mice made platelet deficient had 50% graftsurvival on day 10 (Fig. 1D, black squares). These data demon-strate that following initiation of skin graft inflammation byT cells, platelets have a significant role in accelerating rejection.

7000 PLATELETS PROMOTE ALLOGRAFT REJECTION

by guest on February 17, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

Platelets have well-described roles in promoting monocyte traf-ficking across sites of vascular inflammation (1–3), but the role ofplatelets and platelet-derived inflammatory mediators in supportingT cell trafficking is less well understood. On study day 7, a largenumber of CD3-positive cells were noted infiltrating skin grafts(Fig. 2A, left panels). However, the number of CD3+ cells wasdecreased in platelet-depleted mice (Fig. 2A, right panels). Todetermine whether platelets increase T cell trafficking into trans-planted tissue, skin grafts from control and platelet-depleted micewere collected on study day 5, and mononuclear cell infiltrateswere isolated from the graft followed by density centrifugation.T cells were quantified by flow cytometry with Abs to TCR (Sup-plemental Fig. 3). Approximately 35% of the mononuclear cellsisolated from the control skin grafts were TCR positive comparedwith only 25% in platelet-depleted mice (Fig. 2B). These datademonstrate that platelets increase T cell infiltration into skin grafts.Not only may platelets augment T cell recruitment, they may

also influence cytokine production. To examine this, plasma fromcontrol and platelet-depleted micewas collected on study day 5 andTNF-a and IFN-g measured using an ELISA for each. Mice thatare platelet depleted have significantly less TNF-a (Fig. 3A) andIFN-g (Fig. 3B) compared with control mice.Much past and emerging research has demonstrated the impor-

tant role of COX products, including thromboxane, in augmentingvascular inflammation and directing T cell immune responses (20,22). Platelets are a major source of thromboxane. To determinewhether platelets contribute to plasma thromboxane production du-ring graft rejection skin grafted mice were treated with platelet-

depleting Ab or control IgG at the time of T cell reconstitution.Plasma was then isolated on day 5, and the stable thromboxanebreakdown product TXB2 was measured by ELISA. Mice that wereskin grafted and T cell reconstituted had greatly increased plasmaTXB2 compared with control ungrafted mice (Fig. 4A, black barversus white bar). Nude mice with skin grafts but not T cell re-constituted have the same plasma TXB2 as control mice (Supple-mental Fig. 4). Platelet depletion abrogated the increased plasmaTXB2 found in T cell reconstituted and skin-grafted mice (Fig. 4A,gray bar versus white bar), indicating that platelets are an importantsource of thromboxane during skin graft rejection.Platelets release glutamate upon activation (24, 30), and we have

shown that glutamate increases platelet reactivity and thrombox-ane production through platelet ionotropic glutamate receptors(24, 25). Our recent work demonstrated that this is in part dueto glutamate-mediated stimulation of platelet thromboxane pro-duction (25). To determine whether plasma glutamate is increasedin skin graft recipients, plasma was isolated from control un-grafted nude mice, mice that were skin grafted but not T cellreconstituted, and mice that were skin grafted and T cell recon-stituted. Using an enzymatic colorimetric assay, glutamate con-centration was determined. Similar to plasma PF4, a skin graftalone did not increase plasma glutamate; however, with initiationof rejection, plasma glutamate was significantly increased 2 and5 d after T cell reconstitution (Fig. 4B).Because glutamate drives platelet thromboxane production, glu-

tamate receptor antagonists, such as CNQX, may be beneficial inimproving graft survival. Our past work has demonstrated that

FIGURE 1. Activated platelets accelerate

transplant rejection. A, Platelets are activated

early in graft rejection. Plasma PF4 was

measured in mice without skin grafts, control

skin-grafted mice (no T cells), and skin-

grafted mice that were T cell reconstituted

(n = 5 6 SD). *p , 0.03. B, Platelet-depleted

mice have improved graft survival. Repre-

sentative skin grafts on study day 7. C, Day 7

graft survival. Mice were T cell reconstituted

and treated with control IgG or T cell re-

constituted and platelet depleted by treat-

ment with mAbs to GP1b. *p, 0.01. D, Graft

survival curve. Mice were platelet depleted on

days 0 and 4 after T cell reconstitution, and

graft rejection was monitored daily. *p, 0.03.

The Journal of Immunology 7001

by guest on February 17, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

AMPA receptor antagonists decrease platelet activation, in part byblunting agonist-induced thromboxane production (31, 32). Basedon these data, glutamate-driven platelet thromboxane production

may be very important in graft rejection, and CNQX is a poten-

tially valuable compound to ameliorate these effects. To test this,

we treated skin graft recipients with either control PBS or 2 mg/kg

CNQX daily by i.p. injection, and graft survival was monitored for

7 d. Similar to our preceding experiments, control PBS-treated

mice had ∼65% graft rejection (Fig. 5A, white bar). In contrast,

mice treated with CNQX had much improved graft survival, with

∼35% of skin grafts rejecting (Fig. 5A, black bar). Graft rejection

was also observed over a longer time, and mice treated with

CNQX had significantly improved graft survival (Fig. 5B).We have reported previously that glutamate receptor signaling

helps augment platelet activation and that CNQX is an effective

platelet antagonist (24). To determine whether the protective effect

of CNQX is in part be due to platelet inhibition, plasma was

isolated from control mice, platelet-depleted mice, and CNQX-

treated mice. All of these mice were skin grafted and T cell

reconstituted. As expected, at early (day 2) and late (day 6) time

points after T cell reconstitution, control mice had elevated plasma

PF4 concentrations (Fig. 5C). Platelet-depleted mice had greatly

reduced plasma PF4 at day 2, and as platelet numbers rebound, the

PF4 concentration increased by day 6 (Fig. 5C). CNQX-treated

mice had significantly reduced plasma PF4 compared with control

mice, indicating that its protective effect may in part be due to

platelet inhibition (Fig. 5C).Similar to platelet depletion studies, the delay in transplant

rejection was also reflected in decreased T cell infiltrates. Staining

FIGURE 2. Platelets promote graft inflammation.

A, Immunohistochemistry. Skin grafts were immunos-

tained with anti-CD3 Ab. Left panels represent control

mice and right panels platelet-depleted mice. B, T cell

recruitment. On study day 5, mononuclear cells were

isolated from control and platelet-depleted skin grafts

and the number of TCR-positive cells quantified by

FACS analysis. *p , 0.03.

FIGURE 3. Platelets promote cytokine production. A, Plasma TNF-a.

Plasma was isolated from control and platelet-depleted mice on day 5, and

TNF-a was measured by ELISA (n = 5 6 SD). *p , 0.01. B, Plasma

IFN-g. Plasma was isolated from control and platelet-depleted mice on

day 5, and IFN-g was determined by ELISA (n = 5 6 SD). *p , 0.05.

7002 PLATELETS PROMOTE ALLOGRAFT REJECTION

by guest on February 17, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

for CD3-positive T cells demonstrated fewer T cells in CNQX-treated skin grafts compared with grafts on untreated controlmouse grafts (Fig. 6A). Mononuclear cells were also isolatedfrom control and CNQX-treated mouse skin grafts and TCR-positive cells as the percent of total mononuclear cells wasdetermined by flow cytometry. Approximately 25% of isolatedmononuclear cells from control mice were TCR positive (Fig. 6B).The number of infiltrating T cells was greatly reduced in micetreated with CNQX (Fig. 6B). We also further defined the typesof cellular infiltrates by quantifying CD4, CD8 T cell subsets andmacrophages (CD14 positive) by flow cytometry. Platelet-depletedmice had reduced numbers of CD4-, CD8-, and CD14-positivecells, but only CD8+ cells were significantly reduced comparedwith control skin graft mice (Fig. 6C, gray bar versus white bar).All types of cells were significantly reduced in mice treated withCNQX (Fig. 6C, black bar versus white bar).If AMPA receptor antagonists increase graft survival in part by

blunting glutamate-mediated platelet thromboxane production andimmune stimulation, we predicted that CNQX-treated mice wouldhave reduced plasma thromboxane and T cell-derived cytokines. Todetermine this, we isolated plasma from control and CNQX-treatedmice and measured TXB2 and IFN-g. CNQX significantly reducedplasma TXB2 (Fig. 6D) and IFN-g (Fig. 6E) compared with con-trol mice. These data indicate that inhibition of platelet glutamatereceptors increased graft survival in part by directly or indirectlyreducing plasma proinflammatory mediators.

DiscussionPlatelets are dynamic cells with important hemostasis and immuneregulatory functions. Our past work used a model in which allo-antibody initiated inflammation in skin grafts to demonstratean important role for platelets in leukocyte recruitment (6). Thepresent experiments demonstrate that platelets themselves do not

initiate skin graft inflammation and rejection, but with T cell re-constitution and T cell-initiated graft rejection, platelets are acti-vated, leading to an increase in plasma thromboxane, IFN-g,augmented T cell infiltrates, and graft rejection. Whether plateletsare activated by T cell-driven graft vascular inflammation or bydirect T cell interactions is not clear in this study, but our datasupport a role for platelets in sustaining T cell recruitment andgraft rejection. We also demonstrated that plasma glutamate isincreased in mice that received skin grafts and that glutamatereceptor antagonists improve graft survival and blunt T cell re-sponses. These data point to platelet-mediated mechanisms oftransplant rejection and glutamate-mediated platelet COX activa-tion and a potential clinical utility for glutamate receptor antag-onists in prolonging graft survival.Platelet interactions with innate immune cells are much better

defined than platelet and T cell interactions. However, plateletinteractions with T cells may be just as important. Platelets expressnumerous surface receptors for contact-dependent interactions withT cells, and platelets also secrete a large number of chemokines,cytokines, and other inflammatory molecules with the potentialto increase T cell responses in a contact-independent manner.Our data indicate that platelets, perhaps via glutamate-mediated

FIGURE 4. Glutamate drives thromboxane production. A, Platelets are

a major source of plasma thromboxane. On study day 5, plasma samples

were collected from control mice (no graft) and skin-grafted mice that

were T cell reconstituted with or without platelet depletion. TXB2 was

quantified by ELISA (n = 5–6 6 SD). *p , 0.01 versus control. B, Plasma

glutamate is increased in transplanted mice. Plasma samples were col-

lected from control mice (no skin graft), mice with skin grafts only (no

T cells), and mice with skin grafts plus T cells. Glutamate was quantified

by an enzymatic assay. *p , 0.05.

FIGURE 5. Glutamate receptor antagonist improves graft survival. A,

T cell-reconstituted mice treated with glutamate receptor antagonist have

improved skin graft survival on day 7. *p , 0.01. B, Graft survival curve.

*p , 0.05. C, CNQX reduced platelet activation as measured by PF4 in

T cell-reconstituted mice with skin grafts. Plasma was isolated from

control mice, mice that were platelet depleted, and mice treated with

CNQX (n = 5 6 SD). *p , 0.05; **p , 0.01 versus control.

The Journal of Immunology 7003

by guest on February 17, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

thromboxane release, may contribute to allograft rejection. It isclear that platelet-depleted mice and mice treated with glutamatereceptor antagonists have reduced plasma IFN-g. This may in-dicate an important role for platelets in sustaining T cell re-cruitment and activation and a role for platelets in enhancing IFN-g production in the cycle of allograft inflammation (33, 34). IFN-gresults in endothelial cell activation and, with it, more plateletactivation, T cell recruitment, and continued IFN-g production.Therefore, by blunting platelet activation, this cycle with IFN-g asa central player may be broken.We have found that as early as 2 d post-reconstitution with

unsensitized T cells, there is vascular inflammation and plateletactivation. Studies have indicated that within 48 h posttransfer,T cells may undergo as many as three rounds of division (35), andas many as 20% of T cells may recognize alloantigen (36). Fair-child and coworkers (37) have demonstrated that endogenousCD8 memory T cells in nonsensitized recipients infiltrate cardiacallografts within 24 h of reperfusion. In addition, the nude mousecan make IgM alloantibodies that may cross-link MHC and acti-vate complement prior to T cell reconstitution, helping to rapidlylocalize T cells to the graft.In this work, we present data to demonstrate a potential role for

platelet glutamate receptor signaling stimulation of thromboxaneproduction in transplant-recipient mice. This in turn may assist inT cell recruitment, activation, and transplant rejection. Glutamatehas long been known to be an important signaling molecule in the

CNS, but its role in the periphery is becoming better appreciated.Glutamate receptors are expressed on non-CNS tissue includingplatelets, and we have shown that platelet glutamate receptors havea role in platelet activation and thromboxane production (24, 25,30, 38). We have also demonstrated that in a large cohort of plateletdonors, a block of polymorphisms in the kainate-type glutamatereceptor is associated with a decrease in patient plasma throm-boxane and platelet aggregation (25). These data demonstrate apotentially important clinical relevance for this class of receptors.COX enzymes are expressed by many cells, but platelets are amajor source of COX-derived inflammatory molecules that con-tribute to vascular inflammatory diseases such as atherosclerosis. TPsignaling has also been shown to be important in graft rejection, asTP receptor knockout mice have improved heart allograft survival(22, 39).Thromboxane represents a single platelet-derived inflammatory

molecule that may have a role in further T cell recruitment. Otherplatelet-derived inflammatory molecules may be just as important.For example, PF4 alone or in concert with other chemokines suchas CCL5 (RANTES) might also increase T cell infiltrates into theskin grafts. Platelet alpha-granules also contain numerous cytokinessuch as IL-1b, TGF-b, and TNF-a. These cytokines have multipleeffects in transplants, and of particular relevance to platelets arethe effects of these cytokines on vascular endothelial cells. Forexample, TNF-a stimulates the exocytosis of vonWillebrand factorand P-selectin from Weible-Palade storage granules in endothelial

FIGURE 6. Glutamate receptor an-

tagonist reduced T cell recruitment

and inflammation. A, CNQX-treated

mice have fewer T cell infiltrates.

Immunohistochemistry with anti-CD3

Ab. B, Quantification of T cell infil-

trates. *p , 0.05. C, CD4-, CD8-,

and CD14-positive cell infiltrates by

flow cytometry (n = 5 6 SD). *p ,0.05 versus control. D, Plasma TXB2

(n = 5 6 SD). *p , 0.03. E, Plasma

IFN-g (n = 5 6 SD). *p , 0.03.

7004 PLATELETS PROMOTE ALLOGRAFT REJECTION

by guest on February 17, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

cells (40, 41). The increased expression of these adhesion mole-cules promotes platelet activation and attachment that may thenincrease T cell recruitment and trafficking.Based on these data, we propose that in our model, T cells initiate

vascular inflammation resulting in platelet recruitment, plateletactivation, elaboration of glutamate, platelet thromboxane pro-duction, and sustained platelet activation and T cell recruitment.Despite some attention in the 1970s and early 1980s, the role ofplatelets in transplant rejection has not been closely examined.Platelets are much more numerous than other immune cells and,when stimulated, release or produce large quantities of proin-flammatory molecules. Because of their vital role in hemostasis,this important function has been greatly overlooked. There are fewplatelet antagonists currently in clinical use. These are used mainlyfor people at risk for heart attack or stroke, and platelet inhibi-tors are now a major part of reducing the risk of cardiovascularevents. Our work demonstrates a potential role for platelet-derivedthromboxane in T cell recruitment and graft rejection, furtherunderscoring the possible utility of common drugs like aspirin andibuprofen and perhaps glutamate receptor antagonists to blockCOX upstream signaling. Thromboxane and other COX productssuch as PGs are known to direct T cell responses (19, 20). One ofthe largest and most rapid sources for induction of thromboxaneand PGs is platelets. Our work implies that this may be an im-portant area for future clinical study, and platelet inhibitors suchas glutamate receptor antagonists may be of potential clinical uti-lity in targeting diverse vascular inflammatory diseases, includingtransplantation.Although small and anucleate, platelets are dynamic cells ca-

pable of diverse functions in vascular biology. Our studies indicateplatelets may have a role in enhancing T cell allograft immuneresponses. Further basic science and clinical studies are needed tofully elucidate the most relevant pathways, but this work indicatesthat inhibiting platelet glutamate receptor signaling may be onepathway of note.

DisclosuresThe authors have no financial conflicts of interest.

References1. Huo, Y., A. Schober, S. B. Forlow, D. F. Smith, M. C. Hyman, S. Jung,

D. R. Littman, C. Weber, and K. Ley. 2003. Circulating activated platelets ex-

acerbate atherosclerosis in mice deficient in apolipoprotein E. Nat. Med. 9: 61–

67.2. Huo, Y., and K. F. Ley. 2004. Role of platelets in the development of athero-

sclerosis. Trends Cardiovasc. Med. 14: 18–22.3. Burger, P. C., and D. D. Wagner. 2003. Platelet P-selectin facilitates athero-

sclerotic lesion development. Blood 101: 2661–2666.4. Smyth, S. S., R. P. McEver, A. S. Weyrich, C. N. Morrell, M. R. Hoffman,

G. M. Arepally, P. A. French, H. L. Dauerman, and R. C. Becker. 2009. Platelet

functions beyond haemostasis. J. Thromb. Haemost. 7: 1759–1766.5. Kirk, A. D., C. N. Morrell, and W. M. Baldwin, III. 2009. Platelets influence

vascularized organ transplants from start to finish. Am. J. Transplant. 9: 14–22.6. Morrell, C. N., K. Murata, A. M. Swaim, E. Mason, T. V. Martin,

L. E. Thompson, M. Ballard, K. Fox-Talbot, B. Wasowska, and W. M. Baldwin,

III. 2008. In vivo platelet-endothelial cell interactions in response to major

histocompatibility complex alloantibody. Circ. Res. 102: 777–785.7. Porter, K. A. 1965. Morphological aspects of renal homograft rejection. Br. Med.

Bull. 21: 171–175.8. Fenech, A., A. Nicholls, and F. W. Smith. 1981. Indium (111In)-labelled pla-

telets in the diagnosis of renal transplant rejection: preliminary findings. Br. J.

Radiol. 54: 325–327.9. Oluwole, S., T. Wang, R. Fawwaz, K. Satake, R. Nowygrod, K. Reemtsma, and

M. A. Hardy. 1981. Use of indium-111-labeled cells in measurement of cellular

dynamics of experimental cardiac allograft rejection. Transplantation 31: 51–55.10. Wasowska, B. A., Z. Qian, D. L. Cangello, E. Behrens, K. Van Tran, J. Layton,

F. Sanfilippo, and W. M. Baldwin, III. 2001. Passive transfer of alloantibodies

restores acute cardiac rejection in IgKO mice. Transplantation 71: 727–736.11. Ota, H., K. Fox-Talbot, W. Hu, Z. Qian, F. Sanfilippo, R. H. Hruban, and

W. M. Baldwin, III. 2005. Terminal complement components mediate release of

von Willebrand factor and adhesion of platelets in arteries of allografts. Trans-plantation 79: 276–281.

12. Nakashima, S., Z. Qian, S. Rahimi, B. A. Wasowska, and W. M. Baldwin, III.2002. Membrane attack complex contributes to destruction of vascular integrityin acute lung allograft rejection. J. Immunol. 169: 4620–4627.

13. Wehner, J., C. N. Morrell, T. Reynolds, E. R. Rodriguez, and W. M. Baldwin, III.2007. Antibody and complement in transplant vasculopathy. Circ. Res. 100:191–203.

14. Xu, H., X. Zhang, R. B. Mannon, and A. D. Kirk. 2006. Platelet-derived orsoluble CD154 induces vascularized allograft rejection independent of cell-bound CD154. J. Clin. Invest. 116: 769–774.

15. Khandoga, A., M. Hanschen, J. S. Kessler, and F. Krombach. 2006. CD4+ T cellscontribute to postischemic liver injury in mice by interacting with sinusoidalendothelium and platelets. Hepatology 43: 306–315.

16. Iannacone, M., G. Sitia, M. Isogawa, P. Marchese, M. G. Castro, P. R. Lowenstein,F. V. Chisari, Z. M. Ruggeri, and L. G. Guidotti. 2005. Platelets mediate cytotoxicT lymphocyte-induced liver damage. Nat. Med. 11: 1167–1169.

17. Elzey, B. D., D. L. Sprague, and T. L. Ratliff. 2005. The emerging role ofplatelets in adaptive immunity. Cell. Immunol. 238: 1–9.

18. Srivastava, K., I. A. Cockburn, A. Swaim, L. E. Thompson, A. Tripathi,C. A. Fletcher, E. M. Shirk, H. Sun, M. A. Kowalska, K. Fox-Talbot, et al. 2008.Platelet factor 4 mediates inflammation in experimental cerebral malaria. CellHost Microbe 4: 179–187.

19. Yao, C., D. Sakata, Y. Esaki, Y. Li, T. Matsuoka, K. Kuroiwa, Y. Sugimoto, andS. Narumiya. 2009. Prostaglandin E2-EP4 signaling promotes immune in-flammation through Th1 cell differentiation and Th17 cell expansion. Nat. Med.15: 633–640.

20. Narumiya, S. 2003. Prostanoids in immunity: roles revealed by mice deficient intheir receptors. Life Sci. 74: 391–395.

21. Ushikubi, F., Y. Aiba, K. Nakamura, T. Namba, M. Hirata, O. Mazda, Y. Katsura,and S. Narumiya. 1993. Thromboxane A2 receptor is highly expressed in mouseimmature thymocytes and mediates DNA fragmentation and apoptosis. J. Exp.Med. 178: 1825–1830.

22. Thomas, D. W., P. N. Rocha, C. Nataraj, L. A. Robinson, R. F. Spurney,B. H. Koller, and T. M. Coffman. 2003. Proinflammatory actions of thromboxanereceptors to enhance cellular immune responses. J. Immunol. 171: 6389–6395.

23. Morrell, C. N., K. Murata, A. M. Swaim, E. Mason, T. V. Martin,L. E. Thompson, M. Ballard, K. Fox-Talbot, B. Wasowska, and W. M. Baldwin,III. 2008. In vivo platelet-endothelial cell interactions in response to majorhistocompatibility complex alloantibody. Circ. Res. 102: 777–785.

24. Morrell, C. N., H. Sun, M. Ikeda, J. C. Beique, A. M. Swaim, E. Mason,T. V. Martin, L. E. Thompson, O. Gozen, D. Ampagoomian, et al. 2008. Glu-tamate mediates platelet activation through the AMPA receptor. J. Exp. Med.205: 575–584.

25. Sun, H., A. Swaim, J. E. Herrera, D. Becker, L. Becker, K. Srivastava,L. E. Thompson, M. R. Shero, A. Perez-Tamayo, B. Suktitpat, et al. 2009.Platelet kainate receptor signaling promotes thrombosis by stimulating cyclo-oxygenase activation. Circ. Res. 105: 595–603.

26. Yamakuchi, M., N. C. Kirkiles-Smith, M. Ferlito, S. J. Cameron, C. Bao, K. Fox-Talbot, B. A. Wasowska, W. M. Baldwin, III, J. S. Pober, and C. J. Lowenstein.2007. Antibody to human leukocyte antigen triggers endothelial exocytosis.Proc. Natl. Acad. Sci. USA 104: 1301–1306.

27. Murata, K., K. Fox-Talbot, Z. Qian, K. Takahashi, G. L. Stahl, W. M. Baldwin,III, and B. A. Wasowska. 2007. Synergistic deposition of C4d by complement-activating and non-activating antibodies in cardiac transplants. Am. J. Trans-plant. 7: 2605–2614.

28. de Waal, R. M., M. J. Bogman, I. M. Cornelissen, A. N. Vermeulen, and R. A. Koene.1986. Expression of donor class I major histocompatibility antigens on the vascularendothelium of mouse skin allografts. Transplantation 42: 178–183.

29. Kisucka, J., C. E. Butterfield, D. G. Duda, S. C. Eichenberger, S. Saffaripour,J. Ware, Z. M. Ruggeri, R. K. Jain, J. Folkman, and D. D. Wagner. 2006. Pla-telets and platelet adhesion support angiogenesis while preventing excessivehemorrhage. Proc. Natl. Acad. Sci. USA 103: 855–860.

30. Aliprandi, A., M. Longoni, L. Stanzani, L. Tremolizzo, M. Vaccaro, B. Begni,G. Galimberti, R. Garofolo, and C. Ferrarese. 2005. Increased plasma glutamatein stroke patients might be linked to altered platelet release and uptake. J. Cereb.Blood Flow Metab. 25: 513–519.

31. Sun, H., A. Swaim, J. E. Herrera, D. Becker, L. Becker, K. Srivastava,L. E. Thompson, M. R. Shero, A. Perez-Tamayo, B. Suktitipat, et al. 2009.Platelet kainate receptor signaling promotes thrombosis by stimulating cyclo-oxygenase activation. Circ. Res. 105: 595–603.

32. Morrell, C. N., H. Sun, M. Ikeda, J. C. Beique, A. M. Swaim, E. Mason,T. V. Martin, L. E. Thompson, O. Gozen, D. Ampagoomian, et al. 2008. Glu-tamate mediates platelet activation through the AMPA receptor. J. Exp. Med.205: 575–584.

33. Koh, K. P., Y. Wang, T. Yi, S. L. Shiao, M. I. Lorber, W. C. Sessa, G. Tellides, andJ. S. Pober. 2004. T cell-mediated vascular dysfunction of human allografts resultsfrom IFN-gamma dysregulation of NO synthase. J. Clin. Invest. 114: 846–856.

34. Wang, Y., W. R. Burns, P. C. Tang, T. Yi, J. S. Schechner, H. G. Zerwes,W. C. Sessa, M. I. Lorber, J. S. Pober, and G. Tellides. 2004. Interferon-gammaplays a nonredundant role in mediating T cell-dependent outward vascular re-modeling of allogeneic human coronary arteries. FASEB J. 18: 606–608.

35. Min, B., and W. E. Paul. 2005. Endogenous proliferation: burst-like CD4 T cellproliferation in lymphopenic settings. Semin. Immunol. 17: 201–207.

36. Suchin, E. J., P. B. Langmuir, E. Palmer, M. H. Sayegh, A. D. Wells, andL. A. Turka. 2001. Quantifying the frequency of alloreactive T cells in vivo: newanswers to an old question. J. Immunol. 166: 973–981.

The Journal of Immunology 7005

by guest on February 17, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

37. Schenk, A. D., T. Nozaki, M. Rabant, A. Valujskikh, and R. L. Fairchild. 2008.Donor-reactive CD8 memory T cells infiltrate cardiac allografts within 24-hposttransplant in naive recipients. Am. J. Transplant. 8: 1652–1661.

38. Ferrarese, C., L. Tremolizzo, M. Rigoldi, G. Sala, B. Begni, L. Brighina,G. Ricci, M. G. Albizzati, R. Piolti, F. Crosti, et al. 2001. Decreased plateletglutamate uptake and genetic risk factors in patients with Parkinson’s disease.Neurol. Sci. 22: 65–66.

39. Rocha, P. N., T. J. Plumb, S. D. Crowley, and T. M. Coffman. 2003. Effectormechanisms in transplant rejection. Immunol. Rev. 196: 51–64.

40. Jurd, K. M., C. J. Stephens, M. M. Black, and B. J. Hunt. 1996. Endothelial cellactivation in cutaneous vasculitis. Clin. Exp. Dermatol. 21: 28–32.

41. Sun, R. J., S. Muller, X. Wang, F. Y. Zhuang, and J. F. Stoltz. 2000. Regulation ofvon willebrand factor of human endothelial cells exposed to laminar flows: anin vitro study. Clin. Hemorheol. Microcirc. 23: 1–11.

7006 PLATELETS PROMOTE ALLOGRAFT REJECTION

by guest on February 17, 2018http://w

ww

.jimm

unol.org/D

ownloaded from