The Proatherogenic Role of T Cells Requires Cell Divisionand Is Dependent on the Stage of the Disease

Jamila Khallou-Laschet, Giuseppina Caligiuri, Emilie Groyer, Emanuel Tupin, Anh-Thu Gaston,Bruno Poirier, Mitchell Kronenberg, Jose L. Cohen, David Klatzmann, Srini V. Kaveri, Antonino Nicoletti

Objective—The mechanism by which T cells exert a proatherogenic potential is unclear. In order to determine whether thispotential requires their replication, we crossed atherosclerosis-prone apolipoprotein E knockout mice (ApoE°) withtransgenic mice in which exclusive and conditional ablation of dividing T cells relies on their specific expression of theherpes simplex type 1 thymidine kinase (TK) suicide gene.

Methods and Results—We first showed that conalbumin-immunized ApoE°TK mice mounted a significant immuneresponse to the antigen that was fully and specifically blocked by an in vivo ganciclovir (GCV) treatment. Next,ApoE°TK mice and ApoE° mice were treated or not with GCV either during the first 4 weeks (GCV 1 to 4w), the last4 weeks (GCV 5 to 8w), or during 8 weeks (GCV 1 to 8w). Strikingly, ApoE°TK mice displayed a dramatic decreasein lesion development in the GCV 1 to 8w and GCV 5 to 8w groups, whereas the GCV had no effect when administeredduring the first 4 weeks. In protected mice, the inflammatory parameters in lesions, the percentage of CD69�CD3�

splenocytes, and the circulating natural killer T cells were reduced.Conclusions—The present study, therefore, shows that the proatherogenic potential of T cells is crucial in the progression

of fatty streaks to mature plaques and requires cell division. (Arterioscler Thromb Vasc Biol. 2006;26:353-358.)

Key Words: immune response � NKT lymphocytes � regulatory T cells � thymidine kinase

Innate and adaptive immunities are critically involvedboth in atherogenesis and plaque complication.1,2 We3

and others4 have shown that the predominant atheromodu-lating activity of the B-cell compartment is atheroprotec-tive. Although it has recently been proposed that certainT-cell subsets might regulate negatively the inflammatoryprocess,5–7 several lines of evidence indicate that the T-cellcompartment is prevalently proatherogenic.8 –16 The mech-anism by which T cells exert a proatherogenic potential is,however, unclear, because it has not been possible to dateto target specifically and expansively the T cells involvedin the disease process. In the present study, we developeda model in which the T-cell pool dividing during athero-genesis can be conditionally ablated at any stage of thedisease process.

We took advantage of a transgenic mouse model allow-ing exclusive and conditional ablation of dividing T cellsbased on their specific expression of a suicide gene, theherpes simplex type 1 thymidine kinase (TK). Cellsexpressing TK can metabolize the nucleoside analog gan-ciclovir (GCV) into toxic triphosphated GCV that, byblocking DNA elongation, kills dividing cells. That only

dividing cells are killed by triphosphated GCV is a keyproperty of this system. We used transgenic mice specifi-cally expressing the TK gene in both CD4� and CD8� Tcells. Thus, in these TK transgenic mice, although all Tcells express TK, only the dividing ones are eliminatedduring GCV administration, whereas quiescent T cells andall other cells are spared. Furthermore, T-cell proliferationis the only biological process targeted by the GCV in thismodel. The efficiency of this system has been exemplifiedby its apt control of T cell–mediated pathologies, such asgraft-versus-host disease or cardiac and skin allograftrejections.17–19 We crossed TK transgenic mice with ath-erosclerosis-prone apolipoprotein E knockout mice(ApoE°). Because initiation time and duration of GCVtreatment can be controlled, ApoE°TK transgenic miceoffer a unique possibility to explore the role of T-celldivision at various stages of atherosclerotic lesiondevelopment.

In the present study, we confirm that T cells predominantlyplay a proatherogenic role. We show that this effect requiresT-cell division and is time restricted during disease. Thisimplies that the involvement of T cells in the atheroscleroticprocess is not constant and uniform but, rather, dynamic.

Original received April 28, 2005; final version accepted November 16, 2005.From the INSERM U681 (J.K-L., G.C., E.G., E.T., A-T.G., B.P., S.V.K., A.N.), Institut des Cordeliers/Universite Paris 6 UPMC, Paris, France;

Division of Developmental Immunology (E.T., M.K.), La Jolla Institute for Allergy and Immunology, San Diego, Calif; and Centre National de laRecherche Scientifique–Unite Mixte de Recherche 7087 (J.L.C., D.K.), Hopital Pitie-Salpetriere, Paris, France.

J.K.-L. and G.C. contributed equally to this work.Correspondence to Antonino Nicoletti, INSERM U681, Institut des Cordeliers, 15, rue de l’Ecole de Medicine, Paris, France. E-mail

antonino.nicoletti@upmc.fr© 2006 American Heart Association, Inc.

Arterioscler Thromb Vasc Biol. is available at http://www.atvbaha.org DOI: 10.1161/01.ATV.0000198401.05221.13

353 by guest on April 20, 2016http://atvb.ahajournals.org/Downloaded from by guest on April 20, 2016http://atvb.ahajournals.org/Downloaded from by guest on April 20, 2016http://atvb.ahajournals.org/Downloaded from by guest on April 20, 2016http://atvb.ahajournals.org/Downloaded from by guest on April 20, 2016http://atvb.ahajournals.org/Downloaded from

Materials and Methods

Generation of ApoE°TK MiceHomozygous ApoE°TK transgenic mice were generated by crossingApoE° mice with TK transgenic B6 mice (EpCD4�TK).20 ApoE°littermates nontransgenic for the TK were used as controls. In TKtransgenic mice, the TK transgene is under control of the humanCD4 promoter coupled to the murine proximal enhancer (EpCD4). Inthese mice, the TK transgene is expressed in the vast majority of Tcells, regardless of the CD4 or CD8 coreceptor expression.21 This islikely because of the absence in the EpCD4 transgene of the intronicsequence silencing the expression of the CD4 promoter in CD8cells.22,23 Limiting dilution analyses demonstrated that, in thesemice, �98% of dividing TK� T cells are ablated by GCV.24 Micewere maintained on a regular chow diet and kept in standardconditions. All of the experiments were approved by our institutionalethical committee.

In Vivo GCV AdministrationApoE° and ApoE°TK female mice (7 weeks old) were included in aprotocol lasting 8 weeks. For each genotype, a first group of micereceived GCV during the 8 weeks (GCV 1 to 8w, n�8), a secondgroup received GCV the first 4 weeks (GCV 1 to 4w, n�7), a thirdgroup received the GCV the last 4 weeks (GCV 5 to 8w, n�7), anda last group was left untouched (Ctrl, n�7). GCV (Sigma) wasinjected IP twice a week (50 mg/mouse per injection). In a separateprotocol lasting 4 weeks, ApoE° and ApoE°TK female mice weretreated (7 ApoE° and 9 ApoE°TK mice) or not (7 ApoE° and 9ApoE°TK) with GCV. In both protocols, blood was collected bycardiac puncture at sacrifice, and the spleen and the aorta weredissected. Coagulated blood was centrifuged at 4°C for 5 minutes at12 000g, and serum was stored at �20°C until analysis. Total serumcholesterol level was measured using a Boehringer Mannheim kit(France-methode “CHOD-PAP”). A third protocol was also per-formed to analyze the effect of the GCV treatment on T-cellsubpopulations. One group (n�8) of ApoE°TK mice was treatedwith GCV twice a week (50 mg/mouse per injection) during 8 weeks,and a control group (n�8) of ApoE°TK mice was left untouched. Atsacrifice, blood, liver, spleen, and mesenteric and inguinal lymphnodes were collected, and cell suspensions were prepared from thesetissues for immunostaining and flow cytometry.

Quantitation of Atherosclerotic LesionsThe aortic root was dissected, and lesions were quantitated asdescribed previously.25 Briefly, serial cryostat sections were cut fromthe proximal 1 mm of the aortic root. The mean lesion size in eachanimal was determined after measuring four hematoxylin/oil redO–stained sections cut at 200, 400, 600, and 800 �m from the cusporigin.

Immunohistological AnalysesImmunohistochemistry and lesion morphology analysis were per-formed on consecutive aortic sections cut between 500 and 700 �mfrom the cusp origin. Slides were air dried, fixed in cold acetone, andwashed in water and PBS. Rat primary antibodies specific for themouse vascular cell adhesion molecule 1, MAC-3, and CD4 (BDBiosciences) were applied for 1 hour at room temperature. Thesecondary antibody used was the Alexa 546–conjugated goat anti-rat antibody (Molecular Probes). Slides were covermounted with theVectashield Hard set mounting medium with 4�,6-diamidino-2-phenylindole (Vector). Intensity of positively stained cells wassemiquantitatively evaluated by 2 independent investigators in ablinded fashion (If the staining was covering 0 to �25% of lesionarea, the score was set to 1, 25 to �50% to 2, 50 to �75% to 3, and�75% to 4. CD4� T cells were counted). The cellular/extracellularcomponents of plaques were analyzed on Masson’s trichrome stainedsections.

Fluorescence-Activated Cell Sorter AnalysisSpleen cell suspensions were prepared by teasing the tissue on100-�m nylon filters and by lysing erythrocytes with the ammoniumchloride kalium buffer. Splenocytes were stained for 30 minutes at4°C with various combinations of monoclonal antibodies. T and Bcells were identified with a CD45-CD3-CD19 triple staining, T-cellsubsets with a CD3-CD4-CD8 triple staining, and activated T cellswith a CD45-CD3-CD69 triple staining. In the experiment designedto analyze the effect of the GCV treatment on T-cell subpopulations,splenocytes and lymph node cells were prepared with the sameprocedure described above. The lysis of erythrocyte was omitted forlymph node cells. Peripheral blood mononuclear cell suspensionswere prepared from 500 �L of blood collected on sodium citrate andseparated by centrifugation on a Ficoll gradient. Hepatic leukocyteswere obtained by meshing the liver cut into small pieces on 70-�mnylon filters and were separated by centrifugation in an isotonicPercoll solution (P-1644, Sigma). Erythrocytes were eliminated witha red blood cell–lysing buffer (Hybri-Max, Sigma). Natural killer(NK) T cells were identified as lymphocytes (gated on the sidescatter-forward light scatter scatterplot) expressing the V� chain ofthe T-cell receptor and binding the CD1d-�GalCer tetramer.26

CD4�CD25� regulatory T cells were identified as CD4�CD25�

lymphocytes expressing high levels of CD62L and low levels ofCD45RB. Cells were analyzed with a FACSCalibur (Becton Dick-inson) flow cytometer.

Immunization ProtocolA separated set of ApoE°TK female mice (9 weeks old) wereimmunized (n�12) or not (n�12) with conalbumin (20 �g/mouseemulsified 50/50 vol/vol in complete Freund’s adjuvant injected SCin a volume of 100 �L). Half of the mice in each group received a50-mg GCV (Sigma) IP injection at day 0 and day 3 after immuni-zation. Mice were euthanized at day 7, and lymph node cells wereprepared by teasing the draining lymph nodes on 100-�m nylonfilters. We cultured 2.105 cells/well at 37°C in 5% CO2 in RPMI1640 supplemented with FCS 5% into 96-well flat bottom plates(triplicates), with or without concanavalin A (0.5, 2.5, or 5 �g/mL,Sigma) or with conalbumin (0.1, 10, or 50 �g/mL). After 48 hours,0.5 �Ci of 3[H]-thymidine was added to each well. After 1 more dayof incubation, incorporated 3[H]-thymidine was counted in a Micro-beta counter (Wallac). Results are expressed as counts per minute.

Statistical AnalysisResults are expressed as mean�SEM. Nonparametric Mann-Whitney and ANOVA tests were performed using Statview 5.0software (SAS Institute Inc). Differences between groups wereconsidered significant if P�0.05.

Results and DiscussionComplex interactions of immune cells between each otherand with nonimmune cells represent a major difficulty ininterpreting effects induced by immunointerventions. Unrav-eling the role of T cells in atherosclerosis has been faced withsuch an obstacle. For instance, total depletion of a cellpopulation, such as the CD4� T-helper cells, also has reper-cussion on the CD8� T cells and the B-cell compartment.This is because of the fact that lymphocyte ablation is notconditional and is not restricted to disease-related cells. TheApoE°TK transgenic mouse model that we have developed isunique because it allows us to block division of T cells atdefinite stages of the disease, whereas quiescent T cells andall other cells are unaffected.

The T-Cell Compartment of ApoE°TK Mice IsFunctional and Can Be Controlled by the GCVBecause apolipoprotein E might have per se immunomodu-lating capacities,27 it was important to verify that the TK

354 Arterioscler Thromb Vasc Biol. February 2006

by guest on April 20, 2016http://atvb.ahajournals.org/Downloaded from

transgene system transposed to ApoE° mice was functioningas expected. ApoE°TK mice were immunized with conalbu-min. As shown in Figure 1 (left), the capacity of T cells fromGCV-treated mice to proliferate in vitro in response to apolyclonal mitogen such as the concanavalin A was notaffected, thereby showing that the T-cell repertoire remainsfully functional after in vivo GCV treatment. At variance,conalbumin-immunized mice mounted a significant immuneresponse to the antigen that was fully and specifically blockedby an in vivo GCV treatment. Proliferation of spleen cellsfrom immunized mice treated with GCV (Figure 1, bottomright) was similar to proliferation of spleen cells from controlnonimmunized mice (Figure 1, top right). This demonstratesthat the in vivo GCV treatment effectively targets dividing Tcells and can, therefore, control an ongoing T-cell–dependentimmune response in ApoE°TK mice, as reported for singletransgenic TK mice.17–19 GCV treatment had no effect onproliferation of spleen cells from ApoE° mice, regardless ofimmunization (data not shown).

Ablation of Replicating T CellsAffects AtherogenesisNext, we designed an experiment to address whether ablationof replicating T cells had an effect on atherogenesis andwhether the impact of this immunointervention was timedependent. ApoE°TK mice and ApoE° mice were included inan 8-week protocol. Mice were treated with GCV eitherduring the first 4 weeks (GCV 1 to 4w), the last 4 weeks(GCV 5 to 8w), or during 8 weeks (GCV 1 to 8w). Controlmice of both genotypes were left untouched (Ctrl).

The GCV treatment had no effect on mouse weight andserum total cholesterol measured at sacrifice, regardless ofthe GCV administration schedule (Table I, available online athttp://atvb.ahajournals.org). The GCV also had no effect on

lesion development in ApoE° mice (Figure 2, left), providingevidence that T-cell proliferation is the only biologicalprocess targeted by the GCV in this experimental model.Control ApoE°TK mice developed similar lesions as com-pared with ApoE° mice. Most importantly, ApoE°TK micedisplayed a dramatic decrease in lesion development whenGCV was administered either continuously during the 8weeks (GCV 1 to 8w, 55% reduction versus Ctrl) or during

Figure 1. In vitro proliferation of lymphnode cells from ApoE°TK mice, immu-nized (bottom) or not (top) with conalbu-min. In each group, half of the micereceived an IP injection of GCV at day 0and day 3 after immunization (closedsymbols), and all mice were euthanizedat day 7. Lymph node cells were cul-tured with increasing doses of con-canavalin A (ConA, left) or conalbumin(ConALB, right) as indicated, and3[H]-thymidine incorporation was evalu-ated after 3 days of culture. Results areexpressed as mean counts perminute�SEM.

Figure 2. Extent of lesion development was analyzed bycomputer-assisted morphometry on oil red-O–stained andhematoxylin counterstained sections of the aortic root fromApoE° (left) and ApoE°TK (right) mice treated or not with GCVaccording to different schedules. All mice were euthanized 8weeks after the initiation of the treatments. Serial cryostat sec-tions were cut from the proximal 1 mm of the aortic root. Thelesion density (surface area of lesions/surface area of vessel, %)was determined in each aortic root on 4 hematoxylin/oil redO–stained sections cut at 200, 400, 600, and 800 �m from theappearance of the first cusp. Results are expressed asmean�SEM. §, P�0.001 GCV 1 to 8w vs Ctrl; †, P�0.001 GCV5 to 8w vs Ctrl.

Laschet-Khallou et al T-Cell Dynamic in Atherogenesis 355

by guest on April 20, 2016http://atvb.ahajournals.org/Downloaded from

the last 4 weeks (GCV 5 to 8w, 49% reduction versus Ctrl),whereas it had no effect when administered during the first 4weeks (GCV 1 to 4w versus Ctrl, ns). Furthermore, inApoE°TK mice protected by the GCV treatment, the inflam-matory parameters in the lesions were reduced. Fewer mac-rophages and CD4� T cells were detected, and vascular celladhesion molecule 1 expression level was also reduced in thelesions of GCV 1 to 8w and GCV 5 to 8w mice (Figure 3).

In the GCV 1 to 4w group of mice, it was conceivable thatthe GCV treatment cessation during the last 4 weeks allevi-ated the control on remaining autoreactive T cells, whichmight have then exerted their proatherogenic potential again.We, therefore, designed a shorter separate experiment lasting4 weeks. We could not detect any effect of the GCV treatmenton lesion development in this experiment (ApoE° Ctrl:2.7�1.4%; ApoE° GCV 1 to 4w: 3.4�1.7%; ApoE°TK Ctrl:3.2�1.5%; ApoE°TK GCV 1 to 4w: 2.8�1.1%; mean lesiondensity�SEM).

These 2 sets of experiments clearly demonstrate that theeffect of the T-cell division blockade on atherogenesis iscritically dependent on the stage of the disease. They indicatethat T-cell division is not proatherogenic at the early stage offatty streak but rather at a later stage. This is in agreementwith our previous observation that a blockade of theproatherogenic T-helper 1 pathway had no repercussion onearly lesion development.12 Histological analyses showedthat the fibrous cap is likely formed between 12 and 16 weeksof age (data not shown). Given that mice of the GCV 1 to 8wand GCV 5 to 8w had lesions of the fatty streak type (Figure3), it is tempting to speculate that T cells play a keyregulatory role in the progression from fatty streak to maturelesions. Mice were included in the present study at the age of7 weeks, whereas the atherosclerotic process likely initiatesearlier. Although we cannot formally rule out a proathero-genic role for T-cell replication during the very early steps ofthe disease (before 7 weeks of age), the fact that the lesiondevelopment in the 1 to 4 w group (7 to 11 weeks of age) wasnot affected by the GCV treatment argues against thispossibility.

Our data show that replication of T cells does not influenceearly atherogenesis. Whether the presence of T cells is neededat all in this early stage of inflammation is a matter of debate.For instance, 1 group has shown that CD4� T-cell deficiencydoes not change lesion development at the aortic sinus inApoE° mice. Indeed, it led to an increase at the level of thedescending thoracic and abdominal aorta.16 However, in thesame study, mice deficient for T-cell receptor �� T cellsshowed significant smaller plaques throughout the aorta,additionally documenting the proatherogenic role of T cells.Another group has found that the absence of CD4� T cells inApoE° mice leads to reduced atherosclerosis in the aorticsinus, indicating that CD4� T cells constitute a majorproatherogenic cell population.15 These 2 studies illustratethat the very atherosclerosis model and the site specificity oflesion development are confounding factors in immunointer-vention approaches in experimental atherosclerosis. This isadditionally compounded by another caveat, the gendereffect.28

T-Cell Division Blockade Impacts on the Pool ofActivated T CellsIn order to determine whether the blockade of the T-celldivision had an impact on the other immune compartments,spleen cell populations were analyzed by flow cytometry. Asshown in Figure I (available online at http://atvb.ahajournal-s.org), the percentage of B cells, total T cells, helper T cells,and cytotoxic T cells were not affected by the GCV treatment.

We also sought to determine whether blockade of T-cellreplication reduced overall disease-related T-cell activation.Because activation does not always require proliferation, anacute immunointervention targeting cell replication has fewchances to modify detectably the whole pool of activated Tcells. However, T-cell proliferation always witnesses T-cellactivation. If the stimulation persists, replicated cells will berecruited to the next activation round. Consequently, achronic immunointervention targeting T-cell division is ex-pected to affect the pool of activated disease–related T cells.In fact, we found that the expression of the CD69 activation

Figure 3. Immunohistochemistry andlesion morphology analysis were per-formed on cryosections of the aortic root(cut between 500 and 700 �m from thecusp origin) from ApoE° and ApoE°TKmice. Vascular cell adhesion molecule(VCAM) 1, MAC-3, and CD4 immunola-belings (red fluorescence) were observedunder fluorescence microscopy. The bluefluorescence corresponds to the nucleistained with the 4�,6-diamidino-2-phenylindole dye. Semiquantitative esti-mations (mean�SEM) of staining areindicated in each frame and wereobtained as described in the Methodssection. The cellular/extracellular compo-nents of plaques were analyzed on Mas-son’s trichrome stained sectionsobserved with light microscopy. Arrow-heads point at CD4� cells. L indicateslumen. Original magnification: 200.

356 Arterioscler Thromb Vasc Biol. February 2006

by guest on April 20, 2016http://atvb.ahajournals.org/Downloaded from

marker was specifically and significantly decreased on CD3�

T lymphocytes in the GCV 1 to 8w and GCV 5 to 8w groupsof ApoE°TK mice (Figure I). The percentage of CD69�CD3�

lymphocytes, which is 2 to 3 times higher in ApoE° ascompared with wild-type C57Bl/6 mice (data not shown),was also decreased in these groups. This actually shows thatthe blockade of the T-cell division by the GCV treatment (1to 8w or 5 to 8w) resulted in a significantly reduced pool ofactivated T cells. The corollary to this observation is thatatherosclerosis induces the expression of CD69 on a substan-tial fraction of T cells, that is, it induces the activation of alarge T-cell pool. Of note, this is not incompatible with afocused oligoclonal T-cell response, which is a feature of thisdisease.29,30

GCV treatment reduced the CD69� activated population ofT cells in the spleen by 20% to 30%. The ablation of thesecells has not been more complete, because spleen cells maybe partially protected from the GCV as compared withcirculating cells, the administration protocol may be notoptimal to fully prevent T-cell activation, or not all of theCD69� T cells may be dividing. An indication that the latter

possibility was likely came from the experiment in whichApoE°TK mice treated with GCV for 4 weeks and euthanizedat 4 weeks had only a modestly reduced number of CD69� Tcells as compared with untouched controls (GCV:18.8�0.5%; Ctrl: 21.5�0.9%; mean percentage of spleen Tcells�SEM; P�0.0236).

Replicating Blood NK T Cells Are Targeted by theGCV TreatmentIt has been proposed recently that different T-cell subsets,namely CD4�CD25�, Tr1, T-helper 3, and NK T cells, mightexhibit differential atheromodulating activities.5–7,14 Becausethese cells likely display distinct proliferative potential, theirsusceptibility to GCV might be different. We, therefore,monitored the fluctuation induced by the GCV treatment onthe NK and CD4�CD25� naturally occurring regulatoryT-cell populations in the blood, liver, lymph nodes, andspleen.

As shown in the table of Figure 4 and in agreement withprevious reports,14,26 NK T cells were mainly found in theliver. CD4�CD25� regulatory T cells were found to be more

Figure 4. Cell suspensions were obtained from blood, lymph nodes, spleen,and liver of ApoE°TK mice treated (n�8) or not (n�8) with GCV for 8weeks. The scatterplots are from representative blood samples. NK T cellswere identified as lymphocytes (forward light scatter/side scatter lympho-cyte gate) expressing the V� chain of the T-cell receptor and binding theCD1d-�GalCer tetramer. Regulatory T cells were identified as CD4�CD25�

cells among CD62LhighCD45RBlow lymphocytes. The table below shows thepercentages (mean� SEM) of each cell population within the various tis-sues analyzed in the 2 groups of mice.

Laschet-Khallou et al T-Cell Dynamic in Atherogenesis 357

by guest on April 20, 2016http://atvb.ahajournals.org/Downloaded from

abundant in the lymph nodes and the spleen. The GCVtreatment had an effect only on circulating NK T cells. Ofnote, the pool of circulating NK T cells, although present ata low percentage among blood leukocytes, represents apopulation of 80106 cells. The number of NK T cells inthe blood decreased significantly by 25% in GCV-treatedmice (60106 cells). It is, therefore, likely that NK T cellsin the blood are dividing NK T cells, whereas those in thelymph nodes, spleen, and liver are resting cells not targetedby the GCV treatment. These findings are also in agreementwith the highly compartmentalized response of NK T cells toantigenic stimulation.26 Given the proatherogenic potential ofactivated NK T cells,14 this experiment indicates that theantiatherogenic effect of the GCV treatment may partiallyrely on its capacity to target dividing NK T cells in the blood.The observation we have made in the present study does notallow us to establish the relative importance of the proathero-genic effect of conventional CD4� T cells over that of NK Tcells. Future studies with ApoE°TK mice, deficient or not inNK T cells, will help determine the contribution of each cellpopulation in the atherosclerotic process.

The present study shows that the blockade of T-celldivision after the establishment of fatty streaks is associatedwith a reduced proportion of activated T cells and smalleratherosclerotic lesions, which also display a less severeinflammatory phenotype. The proatherogenic potential of Tcells is crucial in the progression of fatty streaks to matureplaques and requires cell division.

AcknowledgmentsThis work was supported by INSERM, CNRS, and by a grant fromthe Fondation de France. B Poirier was the recipient of a fellowshipfrom GRRC and E Tupin was the recipient of a fellowship from theARCOL and received a research stipend from the Swedish Heart &Lung Foundation.

References1. Binder CJ, Chang MK, Shaw PX, Miller YI, Hartvigsen K, Dewan A,

Witztum JL. Innate and acquired immunity in atherogenesis. NatMed. 2002;8:1218–1226.

2. Hansson GK. Inflammation, atherosclerosis, and coronary artery disease.N Engl J Med. 2005;352:1685–1695.

3. Caligiuri G, Nicoletti A, Poirier B, Hansson GK. Protective immunityagainst atherosclerosis carried by B cells of hypercholesterolemic mice.J Clin Invest. 2002;109:745–753.

4. Major AS, Fazio S, Linton MF. B-lymphocyte deficiency increases ath-erosclerosis in LDL receptor-null mice. Arterioscler Thromb Vasc Biol.2002;22:1892–1898.

5. Mallat Z, Gojova A, Brun V, Esposito B, Fournier N, Cottrez F, TedguiA, Groux H. Induction of a regulatory T cell type 1 response reduces thedevelopment of atherosclerosis in apolipoprotein E-knockout mice. Cir-culation. 2003;108:1232–1237.

6. Gojova A, Brun V, Esposito B, Cottrez F, Gourdy P, Ardouin P, TedguiA, Mallat Z, Groux H. Specific abrogation of transforming growthfactor-beta signaling in T cells alters atherosclerotic lesion size andcomposition in mice. Blood. 2003;102:4052–4058.

7. Robertson AK, Rudling M, Zhou X, Gorelik L, Flavell RA, Hansson GK.Disruption of TGF-beta signaling in T cells accelerates atherosclerosis.J Clin Invest. 2003;112:1342–1350.

8. Gupta S, Pablo AM, Jiang XC, Wang N, Schindler C. IFN� potentiatesatherosclerosis in apoE knock-out mice. J Clin Invest. 1997;99:2752–2561.

9. Lee TS, Yen HC, Pan CC, Chau LY. The role of interleukin 12 in thedevelopment of atherosclerosis in ApoE-deficient mice. ArteriosclerThromb Vasc Biol. 1999;19:734–742.

10. Nicoletti A, Paulsson G, Caligiuri G, Zhou X, Hansson GK. Induction ofneonatal tolerance to oxidized lipoprotein reduces atherosclerosis inApoE knockout mice. Mol Med. 2000;6:283–290.

11. Zhou X, Nicoletti A, Elhage R, Hansson GK. Transfer of CD4(�) T cellsaggravates atherosclerosis in immunodeficient apolipoprotein E knockoutmice. Circulation. 2000;102:2919–2922.

12. Laurat E, Poirier B, Tupin E, Caligiuri G, Hansson GK, Bariety J,Nicoletti A. In vivo downregulation of T helper cell 1 immune responsesreduces atherogenesis in apolipoprotein E-knockout mice. Circulation.2001;104:197–202.

13. Caligiuri G, Rudling M, Ollivier V, Jacob MP, Michel JB, Hansson GK,Nicoletti A. Interleukin-10 deficiency increases atherosclerosis,thrombosis, and low-density lipoproteins in apolipoprotein E knockoutmice. Mol Med. 2003;9:10–17.

14. Tupin E, Nicoletti A, Elhage R, Rudling M, Ljunggren HG, Hansson GK,Berne GP. CD1d-dependent activation of NKT cells aggravates athero-sclerosis. J Exp Med. 2004;199:417–422.

15. Zhou X, Robertson AK, Rudling M, Parini P, Hansson GK Lesiondevelopment and response to immunization reveal a complex role forCD4 in atherosclerosis. Circ Res. 2005;96:427–434.

16. Elhage R, Gourdy P, Brouchet L, Jawien J, Fouque MJ, Fievet C, Huc X,Barreira Y, Couloumiers JC, Arnal JF, Bayard F. Deleting TCR alphabeta� or CD4� T lymphocytes leads to opposite effects on site-specificatherosclerosis in female apolipoprotein E-deficient mice. Am J Pathol.2004;165:2013–2018.

17. Braunberger E, Cohen JL, Boyer O, Pegaz-Fiornet B, Raynal-RaschilasN, Bruneval P, Thomas-Vaslin V, Bellier B, Carpentier A, Glotz D,Klatzmann D. T-Cell suicide gene therapy for organ transplantation:induction of long-lasting tolerance to allogeneic heart without generalizedimmunosuppression. Mol Ther. 2000;2:596–601.

18. Cohen JL, Saron MF, Boyer O, Thomas-Vaslin V, Bellier B, Lejeune L,Charlotte F, Klatzmann D. Preservation of graft-versus-infection effectsafter suicide gene therapy for prevention of graft-versus-host disease.Hum Gene Ther. 2000;11:2473–2481.

19. Thomas-Vaslin V, Bellier B, Cohen JL, Boyer O, Raynal-Raschilas N,Glotz D, Klatzmann D. Prolonged allograft survival through conditionaland specific ablation of alloreactive T cells expressing a suicide gene.Transplantation. 2000;69:2154–2161.

20. Salomon B, Maury S, Loubiere L, Caruso M, Onclercq R, Klatzmann D.A truncated herpes simplex virus thymidine kinase phosphorylates thy-midine and nucleoside analogs and does not cause sterility in transgenicmice. Mol Cell Biol. 1995;15:5322–5328.

21. Salmon P, Boyer O, Lores P, Jami J, Klatzmann D. Characterization of anintronless CD4 minigene expressed in mature CD4 and CD8 T cells, butnot expressed in immature thymocytes. J Immunol. 1996;156:1873–1879.

22. Sawada S, Scarborough JD, Killeen N, Littman DR. A lineage-specifictranscriptional silencer regulates CD4 gene expression during T lym-phocyte development. Cell. 1994;77:917–929.

23. Siu G, Wurster AL, Duncan DD, Soliman TM, Hedrick SM. A transcrip-tional silencer controls the developmental expression of the CD4 gene.Embo J. 1994;13:3570–3579.

24. Bellier B, Thomas-Vaslin V, Saron MF, Klatzmann D. Turning immu-nological memory into amnesia by depletion of dividing T cells. ProcNatl Acad Sci U S A. 2003;100:15017–15022.

25. Nicoletti A, Kaveri S, Caligiuri G, Bariety J, Hansson GK. Immuno-globulin treatment reduces atherosclerosis in apo E knockout mice. J ClinInvest. 1998;102:910–918.

26. Matsuda JL, Naidenko OV, Gapin L, Nakayama T, Taniguchi M, WangCR, Koezuka Y, Kronenberg M. Tracking the response of natural killer Tcells to a glycolipid antigen using CD1d tetramers. J Exp Med. 2000;192:741–754.

27. Tenger C, Zhou X. Apolipoprotein E modulates immune activation byacting on the antigen-presenting cell. Immunology. 2003;109:392–397.

28. Caligiuri G, Nicoletti A, Zhou X, Tornberg I, Hansson GK. Effects of sexand age on atherosclerosis and autoimmunity in apoE- deficient mice.Atherosclerosis. 1999;145:301–308.

29. Paulsson G, Zhou X, Tornquist E, Hansson GK. Oligoclonal T cellexpansions in atherosclerotic lesions of apolipoprotein E-deficient mice.Arterioscler Thromb Vasc Biol. 2000;20:10–17.

30. Caligiuri G, Paulsson G, Nicoletti A, Maseri A, Hansson GK. Evidencefor antigen-driven T-cell response in unstable angina. Circulation. 2000;102:1114–1119.

358 Arterioscler Thromb Vasc Biol. February 2006

by guest on April 20, 2016http://atvb.ahajournals.org/Downloaded from

and Antonino NicolettiGaston, Bruno Poirier, Mitchell Kronenberg, José L. Cohen, David Klatzmann, Srini V. Kaveri

Jamila Khallou-Laschet, Giuseppina Caligiuri, Emilie Groyer, Emanuel Tupin, Anh-Thuof the Disease

The Proatherogenic Role of T Cells Requires Cell Division and Is Dependent on the Stage

Print ISSN: 1079-5642. Online ISSN: 1524-4636 Copyright © 2005 American Heart Association, Inc. All rights reserved.

Greenville Avenue, Dallas, TX 75231is published by the American Heart Association, 7272Arteriosclerosis, Thrombosis, and Vascular Biology

doi: 10.1161/01.ATV.0000198401.05221.132005;

2006;26:353-358; originally published online December 1,Arterioscler Thromb Vasc Biol. 

http://atvb.ahajournals.org/content/26/2/353World Wide Web at:

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

http://atvb.ahajournals.org/content/suppl/2005/12/02/01.ATV.0000198401.05221.13.DC1.htmlData Supplement (unedited) at:

  http://atvb.ahajournals.org//subscriptions/

at: is onlineArteriosclerosis, Thrombosis, and Vascular Biology Information about subscribing to Subscriptions:

  http://www.lww.com/reprints

Information about reprints can be found online at: Reprints: 

document. Question and AnswerPermissions and Rightspage under Services. Further information about this process is available in the

which permission is being requested is located, click Request Permissions in the middle column of the WebCopyright Clearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of theArteriosclerosis, Thrombosis, and Vascular Biologyin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on April 20, 2016http://atvb.ahajournals.org/Downloaded from

Legend to Figure I

Spleen cell suspensions obtained 8 weeks after the beginning of the treatments from mice of the

two genotypes, treated or not with GCV as indicated, were stained with various combinations of

monoclonal antibodies and were analyzed by flow cytometry. T and B cells were identified with

a CD45-CD3-CD19 triple staining and the percentages of CD19+ and CD3+ cells were counted

among CD45+ cells gated in the lymphocyte FSC/SSC region. The percentages of CD4+ and

CD8+ T-cell subsets identified with a CD3-CD4-CD8 triple staining and the percentage and the

mean fluorescence intensity (MFI) of CD69+ activated T cells identified with a CD45-CD3-CD69

triple staining was calculated among CD3+ cells gated in the lymphocyte FSC/SSC region.

Table I: Total cholesterol (TC) and weight of ApoE° and ApoE°TK mice treated or not with the GCV.

ApoE° ApoE°TK

Treatment n TC (mg/mL) Weight (g) n TC (mg/mL) Weight (g)

Ctrl 7 3.1 ± 0.1 22.7 ± 0.4 10 3.0 ± 0.1 22.7 ± 0.4

GCV 1-4w 7 3.3 ± 0.2 23.8 ± 0.6 7 3.1 ± 0.1 23.8 ± 0.6

GCV 5-8w 7 2.9 ± 0.1 22.4 ± 0.8 10 3.2 ± 0.1 22.4 ± 0.8

GCV 1-8w 8 3.0 ± 0.1 22.7 ± 0.4 12 3.2 ± 0.1 22.7 ± 0.4