lable at ScienceDirect

Journal of Autoimmunity 34 (2010) 87–95

Contents lists avai

Journal of Autoimmunity

journal homepage: www.elsevier .com/locate/ jaut imm

Bcl-xL is required for the development of functional regulatory CD4 cellsin lupus-afflicted mice following treatment with a tolerogenic peptide

Amir Sharabi, Smadar Lapter, Edna Mozes*

The Weizmann Institute of Science, Rehovot 76100, Israel

a r t i c l e i n f o

Article history:Received 4 May 2009Received in revised form10 June 2009Accepted 14 June 2009

Keywords:Bcl-xLImmunosuppressive moleculesRegulatory T cellsSystemic lupus erythematosusTolerogenic peptides

* Corresponding author. Department of ImmunologScience, 240 Hertzl Street, Rehovot 76100, Israel. Tel.: þ934 4173.

E-mail address: edna.mozes@weizmann.ac.il (E. M

0896-8411/$ – see front matter � 2009 Elsevier Ltd.doi:10.1016/j.jaut.2009.06.002

a b s t r a c t

Dysregulated expression of Bcl-xL and Bcl-2 may initiate the development of autoimmune diseasesincluding systemic lupus erythematosus (SLE). A tolerogenic peptide designated hCDR1 was shown toameliorate manifestations of spontaneous and induced murine SLE. Recently, we demonstrated thatBcl-xL plays a critical role in the modulating effects of hCDR1, as manifested by reducing the state ofactivation of lymphocytes and by down-regulating the secretion of the pathogenic cytokines, IFN-g andIL-10. Here we studied the role of Bcl-xL in the development and function of CD4 regulatory T-cells (Treg)from hCDR1-treated, SLE-afflicted (New-Zealand-Black � New-Zealand-White) F1 mice. We report thatBcl-xL was up-regulated in CD4 Treg of tolerized mice, where it played a role in inducing the regulatory/inhibitory molecules Foxp3, CTLA-4, and TGF-b and in repressing PD-1. Further, Bcl-xL mediated theinduction of CTLA-4 and TGF-b in effector T cells (Teff) by CD4 Treg of the tolerized mice. The induction ofBcl-xL in Teff by Treg was TGF-b dependent and CTLA-4 independent, leading to inhibition of prolifer-ation and to a decrease in activated Teff. We conclude that Bcl-xL is required for the development andfunction of CD4 Treg, which ameliorate lupus following treatment with a tolerogenic peptide.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction (CDR)-1 of an anti-DNA monoclonal antibody (mAb) that bears the

Bcl-xL and Bcl-2 belong to the Bcl-2 family of proteins [1].Dysregulated expression of either of the two molecules may resultin hematological malignancies as well as in autoimmunity [2].Bcl-xL is induced in T cells once the T-cell receptor and CD28 arestimulated, thereby affecting mainly activated cells, whereas Bcl-2is predominantly expressed in resting cells [3,4]. Bcl-xL also playsa role in the development, differentiation, and clonal selection of Bcells [5,6]. Transgenic expression of Bcl-xL or Bcl-2 in murine B cellswas shown to modify the repertoire of B cells and to result in theproduction of pathogenic antibodies and the development ofautoimmune diseases including systemic lupus erythematosus(SLE) and the exacerbation of collagen-induced arthritis [7–9].In contrast, when Bcl-2 was transgenically expressed in T cells, themice were found to be long-lived and without evidence of auto-immunity, and they were also resistant to the development ofcollagen-induced arthritis and SLE [10–12].

Our laboratory designed and prepared a peptide, namely,hCDR1, which is based on the complementarity determining region

y, The Weizmann Institute of972 8 934 3646; fax: þ972 8

ozes).

All rights reserved.

common idiotype (Id), namely, 16/6Id [13,14]. Treatment withhCDR1 ameliorated the serological and clinical (e.g. renal) mani-festations of spontaneously developed or experimentally inducedSLE [14–20]. Treatment with hCDR1 up-regulated regulatory T cells(Treg) [17–19] and reduced the rate of apoptosis [15,16,18,20].Recently, we were able to show that Bcl-xL plays a critical role inthe immunomodulating effects of hCDR1 on lupus-relatedresponses [20]. In particular, the expression of Bcl-xL, which wasenhanced in B cells and diminished in T cells of SLE-afflicted mice,was reversed following treatment with hCDR1, in association withclinical improvement [20]. The expression of Bcl-xL was up-regu-lated in both the CD4 Treg and effector T cells (Teff) of hCDR1-treated mice. However, the contribution of Bcl-xL expression to theinduction of Treg has not yet been explored. The purpose of thisreport was to study the role of Bcl-xL in the induction and functionof CD4 Treg in SLE-afflicted (New-Zealand-Black � New-Zealand-White)F1 (BWF1) mice following treatment with the tolerogenicpeptide, hCDR1. We show here that Bcl-xL is directly involved ininducing the expression of Foxp3 in the CD4 Treg of tolerized mice,in association with down-regulated expression of PD-1. In addition,Bcl-xL plays a direct role in the expression of TGF-b and CTLA-4 inboth CD4 Treg and Teff. Further, the induced expression of Bcl-xL inTeff by CD4 Treg is TGF-b dependent and CTLA-4 independent.Finally, Bcl-xL induction in Teff by CD4 Treg of tolerized mice led tothe inhibition of proliferation and a decrease in activated Teff.

A. Sharabi et al. / Journal of Autoimmunity 34 (2010) 87–9588

2. Materials and methods

2.1. Mice

Female BWF1 mice were purchased from The Jackson Labora-tory (Bar Harbor, ME, USA). This study was approved by the AnimalCare and Use Committee of the Weizmann Institute of Science.

2.2. Synthetic peptides

A peptide [14], GYYWSWIRQPPGKGEEWIG (hCDR1), based onthe CDR1 of the human anti-DNA mAb, bearing the 16/6Id [13], wassynthesized by Polypeptide Laboratories (CA, USA).

2.3. mAbs

Anti-CD4-PE (clone GK1.5), anti-CD4-allophycocyanin (cloneL3T4), anti-CD25-FITC (clone 7D4), anti-CTLA-4-PE (clone 1B8),anti-CTLA-4 neutralizing mAb (clone MR1), and their matchedisotype controls were obtained from Southern BiotechnologyAssociates (Birmingham, AL). Anti-TGF-b1-PE (clone TB21) and itsmatched isotype control were purchased from IQ Products(Groningen, The Netherlands). Anti-Foxp3-FITC (clone FJK-16s Set),and anti-PD-1-FITC (clone J43) were purchased from eBioscience(San Diego, CA). Anti-Bcl-xL-PE (clone H-5) and its isotype controlwere purchased from Santa Cruz Biotechnology (Santa Cruz, CA).Anti-CD45RB-PE (clone 16A), and its matched isotype controls werepurchased from PharMingen (San Diego, CA, USA). Anti-TGF-b neutralizing mAb (clone 1D11) was obtained from R&D Systems.Bcl-2 inhibitor was obtained from Calbiochem� (San Diego, CA).

2.4. Treatment of SLE-afflicted BWF1 mice with hCDR1

Eight-mo-old BWF1 female mice with established manifesta-tions of lupus (e.g. high titers of anti-dsDNA antibodies and levels ofproteinuria �2 g/L) were given 10 weekly s.c. injections of hCDR1(50 mg/mouse) or the vehicle alone (Captisol�, sulfobutylether betacyclodextrin, CyDex, Inc., KS, USA).

2.5. Enrichment of CD4þCD25þ cells

The procedure for depletion and enrichment of CD25þ cells wasdescribed elsewhere [17].

2.6. FACS analysis

Cells were incubated with the relevant Ab and analyzed by FACS,with forward and side scatter gates adjusted to include all cells andto exclude debris. For intracellular staining, the cells were incu-bated with a fixation solution, washed, and resuspended inpermeabilization solution (Serotec; Oxford, UK).

2.7. Protein extraction and Western blot analysis

The procedure for protein extraction and Western blotting isdescribed elsewhere [20]. Anti-Bcl-xL (Santa Cruz Biotechnology,Santa Cruz, CA), anti-neuropilin (Santa Cruz Biotechnology) andanti-tubulin (Sigma–Aldrich) antibodies were used.

2.8. In vitro assays

Cells (5 � 106/well) were incubated for 24 h in the presence ofBcl-xL inhibitor (Calbiochem�, San Diego, CA), as described else-where [20]. Rates of apoptosis (detected by annexin V/PI staining)in cultures were found to be 13.9 � 0.6%.

2.9. Statistical analysis

Student’s t-test and the Mann–Whitney U test were used. Valuesof p < 0.05 were considered significant.

3. Results

3.1. Bcl-xL affects the intensity of Foxp3 and the frequencyof Foxp3-expressing CD4þCD25þ cells

We determined the levels of spleen cells that co-express CD4,CD25, Foxp3, and Bcl-xL in SLE-afflicted BWF1 mice (n ¼ 9/group)following 10 weekly s.c. injections of hCDR1 or the vehicle. Fig. 1A1shows that both the percentages and absolute numbers ofCD4þCD25þFoxp3þBcl-xLþ cells were significantly higher in thespleens of hCDR1-treated mice as compared with the vehicle-treated mice. The results of further analyses are presented inFig. 1A2, and show that most CD4þCD25þBcl-xLþ cells in thehCDR1-treated mice co-expressed Foxp3 in comparison to thevehicle-treated mice (78� 11% vs. 50� 12%, respectively; p¼ 0.02).Also, the expression of Bcl-xL in CD4þCD25þFoxp3þ cells wassignificantly elevated in the SLE-afflicted mice following treatmentwith hCDR1 (72 � 2.1% vs. 6.2 � 2.4%, respectively; p ¼ 0.007).

Next, CD4þCD25þ cells were isolated from the spleens andincubated (5�106/ml) for 24 h in the presence of two doses (25 and100 mM) of Bcl-xL inhibitor. As shown in Fig. 1(B1 and B2), a higherpercentage of CD4þCD25þ cells derived from hCDR1-treated miceexpressed Foxp3 in comparison with cells from vehicle-treatedmice. Importantly, the expression of Foxp3 in CD4þCD25þ cellsfrom both the vehicle- and hCDR1-treated mice diminishedsignificantly in a dose-dependent manner in the presence of theBcl-xL inhibitor. Apoptosis rates in the enriched culture that con-tained the higher dose of the inhibitor were less than 15%, thussupporting a genuine effect of Bcl-xL inhibition that was not relatedto apoptosis. The mean fluorescence intensity (MFI) of Foxp3 inCD4þCD25þ cells from hCDR1-treated mice was higher than that incells of vehicle-treated mice, but the addition of Bcl-xL inhibitor tothe cultures substantially down-regulated the MFI levels in bothtreatment groups (Fig. 1, B1 and B2). The effect of Bcl-xL on theinduced Treg was also confirmed by Western blot analysis of neu-ropilin, a specific marker for CD4 Treg [21]. As shown in Fig. 1C, thelevels of neuropilin in CD4þCD25þ cells of hCDR1-treated micewere significantly higher than those in cells of vehicle-treated mice,and inhibition of Bcl-xL resulted in a substantial reduction in theprotein levels in both treatment groups.

3.2. Bcl-xL affects the expression of CTLA-4 and TGF-b

in CD4þCD25þ cells

To study the direct role of Bcl-xL in the expression of CTLA-4 andTGF-b in CD4þCD25þ cells, we incubated enriched CD4þCD25þ cellsfrom SLE-afflicted BWF1 mice that were treated with hCDR1 or thevehicle, in the presence or absence of Bcl-xL inhibitor. Whereascomparable percentages of CD4þCD25þ cells were stained forintracellular expression of CTLA-4 in both treatment groups, thepercentage of CD4þCD25þ cells that expressed CTLA-4 on theirsurface was higher after having been treated with hCDR1 (Fig. 2, A1and A2). Moreover, treatment with hCDR1 significantly up-regu-lated MFI of intracellular and surface CTLA-4 in the CD4þCD25þ

cells. Inhibition of Bcl-xL resulted in a significantly reduced rate ofintracellular, and to a lesser extent the surface levels of CTLA-4 onCD4þCD25þ cells of both hCDR1- and vehicle-treated mice ina dose-dependent manner. Nevertheless, inhibition of Bcl-xLresulted in reduced surface and intracellular MFI levels of CTLA-4 inhCDR1-derived CD4þCD25þ cells, similar to the levels observed for

Fig. 1. Bcl-xL is required for the development of Foxp3-expressing CD4þCD25þ cells. SLE-afflicted BWF1 mice (n ¼ 9/group) were treated with 10 weekly s.c. injections ofhCDR1 (50 mg/mouse) or the vehicle. (A1) Mean (�SD) percentage and absolute count in spleens of individual mice. (A2) Representative dot plots. (B1) CD4þCD25þ cellswere enriched from the spleens of the experimental mice and incubated (5 � 106/ml) for 24 h in the presence of Bcl-xL inhibitor. Representative histograms are shown.(B2) Mean (�SD) percentages and MFI for Foxp3 in CD4þCD25þ cells of three experiments. (C) Western blot analysis of neuropilin and tubulin in CD4þCD25þ cells in thepresence of Bcl-xL inhibitor (100 mM) or solvent. Results are also expressed as mean (�SD) of two experiments based on densitometry. *p < 0.05 compared with thevehicle-treated group.

A. Sharabi et al. / Journal of Autoimmunity 34 (2010) 87–95 89

cells of vehicle-treated mice. The percentage of CD4þCD25þ cellsthat express TGF-b intracellularly and even more prominent on thesurface, was significantly higher in hCDR1-treated mice incomparison with vehicle-treated mice, although the MFI of TGF-b was similar in both treatment groups (Fig. 2, B1 and B2). Inhibi-tion of Bcl-xL resulted in down-regulated percentages ofCD4þCD25þ cells that express TGF-b both intracellularly and on thecell surface. Inhibition of Bcl-xL in CD4þCD25þ cells had no effecton the intensity of expression of intracellular TGF-b in either of thetreatment groups, although the high dose of the inhibitor dimin-ished the MFI of surface TGF-b in the vehicle-derived CD4þCD25þ

cells (Fig. 2B2).

3.3. Bcl-xL is required by hCDR1-induced CD4 Treg to enable theup-regulation of CTLA-4 and TGF-b in Teff

We investigated the association between Bcl-xL expression andthe induction of the expression of CTLA-4 and TGF-b in Teff fromSLE-afflicted BWF1 mice. To this end, enriched CD4þCD25þ cells(0.5�106/ml) from hCDR1- and vehicle-treated SLE-afflicted BWF1mice were incubated for 24 h together with CD4þCD25� Teff(5 � 106/ml) from SLE-afflicted mice, in the presence or absence ofBcl-xL inhibitor (100 mM). As shown in Fig. 3, the addition ofCD4þCD25þ cells to the culture from hCDR1 but not from thevehicle-treated mice resulted in a significant up-regulation of Teff

Fig. 2. Bcl-xL affects the expression of CTLA-4 and TGF-b in CD4þCD25þ cells. Enriched CD4þCD25þ cells (5 � 106/ml) were incubated for 24 h in the presence of Bcl-xL inhibitor.Representative histograms (A1 and B1) and mean results (�SD) of three experiments (A2 and B2) for surface and intracellular expression of CTLA-4 and TGF-b, in CD4þCD25þ cells.*p < 0.05 compared with the vehicle-treated group.

A. Sharabi et al. / Journal of Autoimmunity 34 (2010) 87–9590

that expressed high levels of CTLA-4 and TGF-b. However, thiseffect was abrogated when the Bcl-xL inhibitor was added to theculture mixture.

3.4. Induced expression of Bcl-xL in Teff is TGF-b dependent andCTLA-4 independent

To characterize the mutual effects between Bcl-xL, CTLA-4 andTGF-b, splenocytes (5 � 106/ml) from SLE-afflicted BWF1 mice(n ¼ 9/group) that were treated with hCDR1 or the vehicle wereincubated in the presence of neutralizing mAb against CTLA-4(20 mg/ml) or TGF-b (10 mg/ml). The protein levels of Bcl-xL,determined by Western blotting, are shown in Fig. 4A. It can be seenthat the levels of Bcl-xL were elevated in the diseased micefollowing treatment with hCDR1, in comparison with the vehicle-treated mice. Also, the levels of induced Bcl-xL in the hCDR1-treated splenocytes were significantly down-regulated afterneutralization of TGF-b; however, neutralization of CTLA-4 had nosubstantial diminishing effect on the expression of Bcl-xL.

Next, we incubated enriched CD4þCD25þ cells (0.5 � 106/ml)from hCDR1- and vehicle-treated SLE-afflicted BWF1 mice (n ¼ 9/group) for 24 h together with CD4þCD25� cells (5 � 106/ml) from

SLE-afflicted mice, in the presence of neutralizing mAb againsteither CTLA-4 (20 mg/ml) or TGF-b (10 mg/ml). As shown in Fig. 4B,CD4þCD25þ cells derived from hCDR1-treated mice induced theexpression of Bcl-xL in Teff of the SLE-afflicted mice, unlike Tregderived from vehicle-treated mice, which had no additive effect onthe expression of Bcl-xL. The elevated expression of Bcl-xL in Tefffollowing incubation with hCDR1-induced Treg was abolished inthe presence of neutralizing mAb specific to TGF-b but not toCTLA-4. The isotype control of the neutralizing Abs had no effect onthe Bcl-xL expression (data not shown).

3.5. Up-regulation of Bcl-xL in hCDR1-induced Treg is associatedwith down-regulation of PD-1

Since PD-1 was reported to affect Bcl-xL expression [22,23], westudied the possible interaction between PD-1 and Bcl-xL inhCDR1-induced Treg. To this end, splenocytes (5 � 106/ml) fromSLE-afflicted mice were incubated following in vitro addition ofeither hCDR1 (25 mg/ml; n ¼ 3) or the vehicle (n ¼ 3) for 24 h. Theexpressions of surface and intracellular PD-1 as well as Bcl-xL weredetermined in CD4þCD25þ gated cells using flow cytometry. Fig. 5Ashows representative histograms and Fig. 5B presents the mean

Fig. 3. Bcl-xL plays a role in the up-regulation of CTLA-4 and TGF-b in CD4 Teff by hCDR1-induced CD4 Treg. CD25� Teff (5 � 106/ml) from vehicle-treated, SLE-afflicted mice wereincubated alone or together with CD4þCD25� cells (0.5 � 106/ml) from vehicle- or hCDR1-treated, SLE-afflicted BWF1 mice for 24 h with Bcl-xL inhibitor (100 mM), or with theinhibitor’s solvent. Shown are the mean (�SD) percentages of CD4þCD25þ cells expressing CTLA4 (A1) or TGF-b (B1) in three experiments, and representative histograms (A2 andB2). *p < 0.05 compared with incubation of vehicle-derived CD25� Teff alone.

A. Sharabi et al. / Journal of Autoimmunity 34 (2010) 87–95 91

percentages of CD4þCD25þ cells that express the various mole-cules. Accordingly, Treg in the cultures that contained hCDR1 up-regulated the expression of Bcl-xL whereas it down-regulatedmainly the surface expression of PD-1 in comparison to the cultureswith the vehicle. These effects were more prominent when hCDR1(as compared with the vehicle) was administered in vivo to SLE-afflicted mice (n ¼ 5/group), as shown in Fig. 5C. Accordingly,hCDR1-treated mice had significantly higher percentages ofCD4þCD25þ cells that expressed Bcl-xL, whereas the intracellular,

Fig. 4. The induced expression of Bcl-xL in Teff is TGF-b dependent and CTLA-4 independeincubated for 24 h in the presence of neutralizing mAb against CTLA-4 (20 mg/ml) or TGF-b (1is shown based on densitometry. (B) CD25� Teff (5 � 106/ml) from vehicle-treated, SLE-afflivehicle- or hCDR1-treated, SLE-afflicted BWF1 mice for 24 h with or without anti-CTLA-4 (CD4þCD25� Teff was determined by FACS. Mean (�SD) percentages of cells of three experimthe effect of hCDR1-induced CD25þ cells.

and to a lesser extent, the surface expression of PD-1 was down-regulated. The results of the in vitro and in vivo experiments werereproduced in two independent experiments.

We further studied the effects of Bcl-xL on the simultaneousexpression of PD-1 (at the surface and intracellular) and Foxp3 inthe hCDR1-induced Treg. Fig. 5D and E shows the results of in vitroexperiments, in which splenocytes of SLE-afflicted mice werecultured (5 � 106/ml) for 24 h in the presence of hCDR1 (25 mg/ml)and Bcl-xL inhibitor at different concentrations, or in the presence

nt. Splenocytes (5 � 106/ml) from hCDR1- or vehicle-treated SLE-afflicted mice were0 mg/ml). (A) Western blot analysis. Mean (�SD) percent expression of two experimentscted mice were incubated alone or together with CD4þCD25þ cells (0.5 � 106/ml) from20 mg/ml) or anti-TGF-b (10 mg/ml) neutralizing mAb. The expression of Bcl-xL in theents are shown. *p < 0.05 compared with the effect of hCDR1. yp < 0.05 compared with

Fig. 5. Up-regulated expression of Bcl-xL is associated with Foxp3 induction and PD-1 down-regulation in CD4þCD25þ cells. (A) Splenocytes (5 � 106/ml) from SLE-afflicted micewere incubated with hCDR1 (25 mg/ml; n ¼ 3), or the vehicle (n ¼ 3) for 24 h. (A) Representative histograms and (B) Mean (�SD) percentages of CD4þCD25þ gated cells for theexpression of Bcl-xL and PD-1. Gray contours indicate staining with the isotype control. (C) In vivo effects of treatment with hCDR1 (50 mg/mouse; 10 weekly s.c. injections; n ¼ 5), orwith the vehicle (n ¼ 5). The results of in vitro and in vivo experiments were reproduced in two different experiments. (D) Representative dot plots and (E) mean (�SD) absolutenumbers for the expression of Foxp3 and surface PD-1 in CD4þCD25þ gated cells of splenocytes (5 � 106/ml) from SLE-afflicted mice following incubation with different doses ofBcl-xL inhibitor or its solvent together with hCDR1 (25 mg/ml; n ¼ 3) for 24 h. Results are expressed as means (�SD) of absolute numbers of cells per culture. Results werereproduced in two independent experiments. (F) In vivo effects of treatment with hCDR1 (50 mg/mouse; 10 weekly s.c. injections; n ¼ 5), or with the vehicle (n ¼ 5). Results werereproduced in two independent experiments.

A. Sharabi et al. / Journal of Autoimmunity 34 (2010) 87–9592

of hCDR1 and the solvent of the inhibitor. It can be seen that theinhibition of Bcl-xL affected the CD4þCD25þ cells in the cultures byprogressively enhancing the expression of PD-1, along witha simultaneous decrease in the expression of Foxp3 (Fig. 5D). It canbe seen that the number of CD4þCD25þ cells that express Foxp3was down-regulated in association with the inhibition of Bcl-xL;

however, the number of CD4þCD25þ cells that express PD-1 wasup-regulated (Fig. 5E). The effects of Bcl-xL inhibition on cells fromhCDR1-treated (in vivo) SLE-afflicted mice are shown in Fig. 5F, andare similar to those determined for the in vitro cultures except thatthe numbers of non-Foxp3-expressing CD4þCD25þPD-1þ cellscontinued to increase even at a high-dose inhibition of Bcl-xL,

A. Sharabi et al. / Journal of Autoimmunity 34 (2010) 87–95 93

suggesting that in vivo treatment with hCDR1, as compared with invitro incubation, conferred a stronger resistance to apoptosis.

3.6. Bcl-xL mediates the suppressive activity of hCDR1-induced CD4Treg on proliferation and the state of activation of Teff derived fromSLE-afflicted mice

To determine whether Bcl-xL plays a role in the suppressivefunction of hCDR1-induced Treg, vehicle-derived CD25� Teff werepooled from SLE-afflicted mice (n ¼ 5–7/group), and then labeledwith CFSE and put into a 36 h culture with enriched CD4þCD25þ cells(at 2 ratios) from mice that received 10 weekly s.c. injections ofeither hCDR1- or the vehicle, with or without the addition of Bcl-xLinhibitor. Fig. 6A shows two distinct populations in CD4þ cells ofwhich CD4high cells were the main subset to proliferate. The latterwere previously reported to represent the helper-inducer subset oflymphocytes [24–26]. It is shown in the figure that the baselineproliferative levels of CFSE-labeled, vehicle-derived CD25� Teff fromthe diseased mice were inhibited following incubation withCD4þCD25þ cells from hCDR1-treated mice (35 � 4% in CD25� Teffalone vs. 28� 3% and 15�4% in the presence of hCDR1-induced Tregat ratios of 1/100 and 1/10, respectively; p ¼ 0.02). The addition ofBcl-xL inhibitor to the cultures abrogated the suppressive effects ofhCDR1-induced Treg on proliferation, and led to proliferation in cellcultures that contained either hCDR1- or vehicle-derivedCD4þCD25þ cells. It is noteworthy that incubation of Teff in thepresence of Bcl-xL inhibitor resulted in the up-regulation of

Fig. 6. Bcl-xL mediates the suppressive functions of hCDR1-induced Treg on proliferation anweekly s.c. injections of hCDR1 (50 mg/mouse) or with the vehicle. Pooled splenocytes from thcells were then labeled with CFSE, and thereafter incubated for 36 h with enriched CD4þCD2addition of Bcl-xL inhibitor (100 mM). (A) Representative dot plots showing the effects on CFand (C) CD4þCD45RBhigh cells among the CFSE-labeled, vehicle-derived CD25� Teff. Resultsbation of vehicle-derived CD25� Teff alone.

proliferating cells as compared with incubation excluding theinhibitor (77 � 8% versus 41 � 4%, respectively), in accordance withour previous report [20]. Likewise, the addition of Treg from hCDR1-and not vehicle-treated mice to cultures of CFSE-labeled CD25� Tefffrom vehicle-treated, SLE-afflicted mice resulted in a significantdown-regulation in CD4þCD69þ cells (Fig. 6B) and CD4þCD45RBhigh

cells (Fig. 6C), which represent activated T cells [27,28]. The lattereffect was abrogated in the presence of Bcl-xL inhibitor. Theseresults were reproduced in three independent experiments.

4. Discussion

The main findings of the present study are that the elevatedexpression of Bcl-xL in T cells of SLE-afflicted BWF1 mice, treatedwith the tolerogenic peptide hCDR1, plays a key role in the inductionof CD4 Treg. The hCDR1-induced expression of Bcl-xL in the cellsresulted in down-regulation of PD-1 and in up-regulation of Foxp3and CTLA-4. Further, Bcl-xL mediated the influence of Treg on Teff ofthe diseased mice by inducing two inhibitory molecules, CTLA-4 andTGF-b, in the latter cells. It is thus suggested that Bcl-xL is pivotal forthe induction of T-cell-associated immunosuppressive moleculesthat are important for tolerance. To the best of our knowledge, thisunrecognized role of Bcl-xL in the T-cell compartment of mice withan autoimmune disease is reported here for the first time.

Several studies reported increased expression of Bcl-xL andBcl-2 in CD4þCD25þ cells but only to explain the superior survivalrates or a numerical increase in the latter cells [12,29–31]. In fact,

d the state of activation of Teff. SLE-afflicted mice (n ¼ 5–7/group) were treated with 10e vehicle-treated group were depleted of CD4þCD25þ cells. The vehicle-derived CD25�

5þ cells (at 2 ratios) from either hCDR1- or the vehicle-treated mice with or without theSE-labeled, vehicle-derived CD25� Teff. (B) The effects on the rates of CD4þCD69þ cellswere reproduced in three independent experiments. *p < 0.05 compared with incu-

A. Sharabi et al. / Journal of Autoimmunity 34 (2010) 87–9594

the association between the expression of Bcl-xL and Foxp3 couldbe indirectly suspected, based on retrospective data. Hence, theexpression of CD28, which was reported to be essential forthe development of Treg [32,33], was also reported to initiate theexpression of Bcl-xL [34]. Likewise, STAT-5, which was shown to beimportant for the development of Treg [35], was also capable ofinducing the expression of Bcl-xL [36]. Herein, we demonstratedthat the hCDR1-induced expression of Foxp3 and neuropilin inCD4þCD25þ cells was directly linked to the expression of Bcl-xL(Fig. 1). It is noteworthy that T cells from mice expressing OVAsystematically and therefore are chronically exposed to the self-Ag(i.e. OVA protein) remained anergic even when apoptosis wasprevented, in association with the induction, though transiently, ofFoxp3 and Bcl-xL [37].

Several signaling pathways can mediate the ability of Bcl-xL tolead to the development of hCDR1-induced Treg. JNK was reportedpreviously to affect the induction of Bcl-xL [38]. Therefore, theability of hCDR1 to reduce the levels of pJNK [15] may impair theformation of AP-1 [39], and as a result, NFAT will become moreavailable for association with Foxp3 instead of with AP-1, therebypromoting the differentiation of anergized/regulatory cells [40,41].

The presence of Bcl-xL in Treg of tolerized SLE-afflicted miceaffected two additional immunosuppressive molecules, CTLA-4 andTGF-b, however, in different manners (Fig. 2). Whereas treatmentwith hCDR1 did not affect the frequency of CTLA-4-expressingCD4þCD25þ cells, it did up-regulate both the surface and intracel-lular expression of CTLA-4, and this effect was Bcl-xL dependent.CTLA-4 is a molecule that has been shown to enable both the TGF-b-mediated expansion [42] and the in vivo suppressive function[43] of CD4 Treg. As for TGF-b, the intensity of its expression wasnot affected by hCDR1 (Fig. 2B). In contrast, treatment with hCDR1resulted in a higher rate of TGF-b-expressing CD4þCD25þ cells, aneffect that was dependent on Bcl-xL expression. Therefore, thesedata indicate that Bcl-xL is involved in the ability of TGF-b toexpand the population of hCDR1-induced CD4þCD25þ cells.

Bcl-xL was shown here to also play a role in up-regulating theexpression of CTLA-4 and TGF-b, in Teff of the SLE-afflicted BWF1mice by hCDR1-induced Treg (Fig. 3). Our findings, supported byprevious reports showing that anergized T cells express Bcl-xL [44],may suggest that Bcl-xL could also suppress activated T cells byinducing a state of anergy through the induction of CTLA-4 and TGF-b (Fig. 3). The hCDR1-induced expression of Bcl-xL in Teff was TGF-b dependent. In agreement, it was previously demonstrated that theinduction of Bcl-xL in activated Tcells was dependent on TGF-b [45].

The inhibition of Bcl-xL prevented the up-regulation of CTLA-4in Teff by hCDR1-induced Treg (Fig. 3A). However, neutralization ofCTLA-4 did not interfere with the ability of the Treg to up-regulateBcl-xL in Teff (Fig. 4B). Therefore, it is likely that the inducedexpression of Bcl-xL occurred prior to the up-regulation of CTLA-4.In accordance with a previous report [46], the fact that Bcl-xL wasstill expressed when CTLA-4 was already induced may suggest thatBcl-xL by itself contributes to the reduction of T-cell activation.

The up-regulated levels of Bcl-xL in the hCDR1-induced CD4 Tregcontributed to the inhibitory effect on PD-1 expression togetherwith the induction of Foxp3 expression (Fig. 5). In line with ourfindings, Hahn’s group demonstrated that treatment of murinelupus with an artificial tolerogenic peptide resulted in a decrease inPD-1 and an increase in Foxp3, in CD8 suppressor cells [47]. It thusappears that the expression of PD-1 is down-regulated in hCDR1-induced Treg in order to potentiallyacquire a suppressive phenotypein the cells. Because PD-1 was shown to inhibit the expression ofBcl-xL [23,48], it is likely that the down-regulation of PD-1 could alsotrigger the induction of Bcl-xL in Treg of the tolerized mice.

Two distinct populations in CD4þ cells were shown in SLE-afflicted BWF1 mice (Fig. 6). However, activated proliferating CD4þ

cells were determined to be within the CD4high subset of cells.These observations are in agreement with previous reportsdemonstrating that CD4 cell activation and proliferation, which arehallmarks of T-helper cells occur exclusively in the CD4high subset ofcells [24–26]. Our previous report demonstrated that the reductionin the state of activation of the Teff and the down-regulatedsecretion of pathogenic cytokines (IFN-g and IL-10), followingtreatment of SLE-afflicted mice with hCDR1, were both mediated byBcl-xL [20]. Here we showed that Bcl-xL played an important role inthe ability of hCDR1-induced Treg to suppress proliferation and toreduce the state of activation of CD4þCD25� Teff in SLE-afflictedmice (Fig. 6). Apparently, the Bcl-xL suppressive effects wereaccomplished because Bcl-xL mediated the up-regulation of TGF-b and CTLA-4, which are essential for hCDR1 activity [17,18], on Teffof the tolerized mice (Fig. 3).

Altogether, the mechanisms underlying the therapeutic effectsof hCDR1 on SLE-afflicted mice involve different functions of thesurvival molecule, Bcl-xL, in tolerance induction. The consequencesof up-regulation of Bcl-xL in the tolerized mice are as follows:1) CD4þCD25þ cells acquire regulatory/suppressor activity; 2)CD4þCD25� Teff up-regulate inhibitory molecules; 3) CD4þCD25�

Teff are suppressed by CD4 Treg. This unrecognized role of Bcl-xL inthe T-cell compartment of an autoimmune background providesa novel mechanism of induction of Treg. Therefore, manipulationsaimed at immunomodulating the expression of Bcl-xL in the T-cellcompartment might be of importance for controlling and treatinga wide range of diseases that are affected by Treg.

References

[1] Cory S. Regulation of lymphocyte survival by the bcl-2 gene family. Annu RevImmunol 1995;13:513–43.

[2] Hughes P, Bouillet P, Strasser A. Role of Bim and other Bcl-2 family members inautoimmune and degenerative diseases. Curr Dir Autoimmun 2006;9:74–94.

[3] Grillot DA, Merino R, Nunez G. Bcl-XL displays restricted distribution during Tcell development and inhibits multiple forms of apoptosis but not clonaldeletion in transgenic mice. J Exp Med 1995;182:1973–83.

[4] Kerstan A, Hunig T. Cutting edge: distinct TCR- and CD28-derived signalsregulate CD95L, Bcl-xL, and the survival of primary T cells. J Immunol2004;172:1341–5.

[5] Takahashi Y, Cerasoli DM, Dal Porto JM, Shimoda M, Freund R, Fang W, et al.Relaxed negative selection in germinal centers and impaired affinity matu-ration in bcl-xL transgenic mice. J Exp Med 1999;190:399–410.

[6] Amanna IJ, Dingwall JP, Hayes CE. Enforced bcl-xL gene expression restoredsplenic B lymphocyte development in BAFF-R mutant mice. J Immunol2003;170:4593–600.

[7] Strasser A, Whittingham S, Vaux DL, Bath ML, Adams JM, Cory S, et al. EnforcedBCL2 expression in B-lymphoid cells prolongs antibody responses and elicitsautoimmune disease. Proc Natl Acad Sci U S A 1991;88:8661–5.

[8] Lopez-Hoyos M, Carrio R, Merino J, Merino R. Regulation of B cell apoptosis byBcl-2 and Bcl-XL and its role in the development of autoimmune diseases. IntJ Mol Med 1998;1:475–83.

[9] Zheng B, Marinova E, Switzer K, Wansley D, He H, Bheekha-Escura R, et al.Overexpression of Bcl(XL) in B cells promotes Th1 response and exacerbatescollagen-induced arthritis. J Immunol 2007;179:7087–92.

[10] Strasser A, Harris AW, Cory S. Bcl-2 transgene inhibits T cell death and per-turbs thymic self-censorship. Cell 1991;67:889–99.

[11] Van Parijs L, Biuckians A, Abbas AK. Functional roles of Fas and Bcl-2-regulatedapoptosis of T lymphocytes. J Immunol 1998;160:2065–71.

[12] Gonzalez J, Tamayo E, Santiuste I, Marquina R, Buelta L, Gonzalez-Gay MA,et al. CD4þCD25þ T cell-dependent inhibition of autoimmunity in transgenicmice overexpressing human Bcl-2 in T lymphocytes. J Immunol2007;178:2778–86.

[13] Waisman A, Shoenfeld Y, Blank M, Ruiz PJ, Mozes E. The pathogenic humanmonoclonal anti-DNA that induces experimental systemic lupus eryth-ematosus in mice is encoded by a VH4 gene segment. Int Immunol 1995;7:689–96.

[14] Luger D, Dayan M, Zinger H, Liu JP, Mozes E. A peptide based on thecomplementarity determining region 1 of a human monoclonal autoantibodyameliorates spontaneous and induced lupus manifestations in correlationwith cytokine immunomodulation. J Clin Immunol 2004;24:579–90.

[15] Rapoport MJ, Sharabi A, Aharoni D, Bloch O, Zinger H, Dayan M, et al.Amelioration of SLE-like manifestations in (NZBxNZW)F1 mice followingtreatment with a peptide based on the complementarity determining region 1of an autoantibody is associated with a down-regulation of apoptosis and ofthe pro-apoptotic factor JNK kinase. Clin Immunol 2005;117:262–70.

A. Sharabi et al. / Journal of Autoimmunity 34 (2010) 87–95 95

[16] Sharabi A, Haviv A, Zinger H, Dayan M, Mozes E. Amelioration of murine lupusby a peptide, based on the complementarity determining region-1 of anautoantibody as compared to dexamethasone: different effects on cytokinesand apoptosis. Clin Immunol 2006;119:146–55.

[17] Sharabi A, Zinger H, Zborowsky M, Sthoeger ZM, Mozes E. A peptide based onthe complementarity determining region 1 of an autoantibody ameliorateslupus by up-regulating CD4þCD25þ cells and TGF-b. Proc Natl Acad Sci U S A2006;103:8810–5.

[18] Sharabi A, Azulai H, Shtoeger ZM, Mozes E. Clinical amelioration of murinelupus by a peptide based on the complementarity determining region-1 of anautoantibody and by cyclophosphamide: similarities and differences in themechanisms of action. Immunology 2007;121:248–57.

[19] Sharabi A, Mozes E. The suppression of murine lupus by a tolerogenic peptideinvolves Foxp3-expressing CD8 Cells that are required for the optimalinduction and function of Foxp3-expressing CD4 Cells. J Immunol2008;181:3243–51.

[20] Sharabi A, Luger D, Ben-David H, Dayan M, Zinger H, ad Mozes E. The role ofapoptosis in the ameliorating effects of a CDR1-based peptide on lupusmanifestations in a mouse model. J Immunol 2007;179:4979–87.

[21] Bruder D, Probst-Kepper M, Westendorf AM, Geffers R, Beissert S, Loser K,et al. Neuropilin-1: a surface marker of regulatory T cells. Eur J Immunol2004;34:623–30.

[22] Raimondi G, Shufesky WJ, Tokita D, Morelli AE, Thomson AW. Regulatedcompartmentalization of programmed cell death-1 discriminatesCD4þCD25þ resting regulatory T cells from activated T cells. J Immunol2006;176:2808–16.

[23] Chemnitz JM, Parry RV, Nichols KE, June CH, Riley JL. SHP-1 and SHP-2 asso-ciate with immunoreceptor tyrosine-based switch motif of programmeddeath 1 upon primary human T cell stimulation, but only receptor ligationprevents T cell activation. J Immunol 2004;173:945–54.

[24] Novak EJ, Masewicz SA, Liu AW, Lernmark A, Kwok WW, Nepom GT. Activatedhuman epitope-specific T cells identified by class II tetramers reside withina CD4high, proliferating subset. Int Immunol 2001;13:799–806.

[25] Ridgway W, Fasso M, Fathman CG. Following antigen challenge, T cells up-regulate cell surface expression of CD4 in vitro and in vivo. J Immunol1998;161:714–20.

[26] Martins SL, St John LS, Champlin RE, Wieder ED, McMannis J, Molldrem JJ,et al. Functional assessment and specific depletion of alloreactive human Tcells using flow cytometry. Blood 2004;104:3429–36.

[27] Morrissey PJ, Charrier K, Braddy S, Liggitt D, Watson JD. CD4þ T cells thatexpress high levels of CD45RB induce wasting disease when transferred intocongenic severe combined immunodeficient mice. Disease development isprevented by cotransfer of purified CD4þ T cells. J Exp Med 1993;178:237–44.

[28] Powrie F, Correa-Oliveira R, Mauze S, Coffman RL. Regulatory interactionsbetween CD45RBhigh and CD45RBlow CD4þ T cells are important for thebalance between protective and pathogenic cell-mediated immunity. J ExpMed 1994;179:589–600.

[29] Chen X, Murakami T, Oppenheim JJ, Howard OM. Differential response ofmurine CD4þCD25þ and CD4þCD25- T cells to dexamethasone-induced celldeath. Eur J Immunol 2004;34:859–69.

[30] Minamimura K, Gao W, Maki T. CD4þ regulatory T cells are spared fromdeletion by antilymphocyte serum, a polyclonal anti-T cell antibody. J Immu-nol 2006;176:4125–32.

[31] Hahn BH, Singh RP, La Cava A, Ebling FM. Tolerogenic treatment of lupus micewith consensus peptide induces Foxp3-expressing, apoptosis-resistant,TGFbeta-secreting CD8þ T cell suppressors. J Immunol 2005;175:7728–37.

[32] Tang Q, Henriksen KJ, Boden EK, Tooley AJ, Ye J, Subudhi SK, et al. Cuttingedge: cD28 controls peripheral homeostasis of CD4þCD25þ regulatory T cells.J Immunol 2003;171:3348–52.

[33] Tai X, Cowan M, Feigenbaum L, Singer A. CD28 costimulation of developingthymocytes induces Foxp3 expression and regulatory T cell differentiationindependently of interleukin 2. Nat Immunol 2005;6:152–62.

[34] Boise LH, Gonzalez-Garcıa M, Postema CE, Ding L, Lindsten T, Turka LA, et al.Bcl-x, a bcl-2-related gene that functions as a dominant regulator of apoptoticcell death. Cell 1993;74:597–608.

[35] Antov A, Yang L, Vig M, Baltimore D, Van Parijs L. Essential role for STAT5signaling in CD25þCD4þ regulatory T cell homeostasis and the maintenanceof self-tolerance. J Immunol 2003;171:3435–41.

[36] Socolovsky M, Fallon AE, Wang S, Brugnara C, Lodish HF. Fetal anemia andapoptosis of red cell progenitors in Stat5a-/-5b-/- mice: a direct role for Stat5in Bcl-X(L) induction. Cell 1999;98:181–91.

[37] Barron L, Knoechel B, Lohr J, Abbas AK. Cutting edge: contributions of apoptosisand anergy to systemic T cell tolerance. J Immunol 2008;180:2762–6.

[38] Kharbanda S, Saxena S, Yoshida K, Pandey P, Kaneki M, Wang Q, et al.Translocation of SAPK/JNK to mitochondria and interaction with Bcl-x(L) inresponse to DNA damage. J Biol Chem 2000;275:322–7.

[39] Whitmarsh AJ, Davis RJ. Transcription factor AP-1 regulation by mitogen-activated protein kinase signal transduction pathways. J Mol Med1996;74:589–607.

[40] Wu Y, Borde M, Heissmeyer V, Feuerer M, Lapan AD, Stroud JC, et al. FOXP3controls regulatory T cell function through cooperation with NFAT. Cell2006;126:375–87.

[41] Ziegler SF. FOXP3: of mice and men. Annu Rev Immunol 2006;24:209–26.[42] Zheng SG, Wang JH, Stohl W, Kim KS, Gray JD, Horwitz DA. TGF-beta requires

CTLA-4 early after T cell activation to induce FoxP3 and generate adaptiveCD4þCD25þ regulatory cells. J Immunol 2006;176:3321–9.

[43] Read S, Greenwald R, Izcue A, Robinson N, Mandelbrot D, Francisco L, et al.Blockade of CTLA-4 on CD4þCD25þ regulatory T cells abrogates their functionin vivo. J Immunol 2006;177:4376–83.

[44] Boussiotis VA, Lee BJ, Freeman GJ, Gribben JG, Nadler LM. Induction of T cellclonal anergy results in resistance, whereas CD28-mediated costimulationprimes for susceptibility to Fas- and Bax-mediated programmed cell death.J Immunol 1997;159:3156–67.

[45] Chen W, Jin W, Tian H, Sicurello P, Frank M, Orenstein JM, et al. Requirementfor transforming growth factor beta1 in controlling T cell apoptosis. J Exp Med2001;194:439–53.

[46] Blair PJ, Riley JL, Levine BL, Lee KP, Craighead N, Francomano T, et al. Cuttingedge: cTLA-4 ligation delivers a unique signal to resting human CD4 T cellsthat inhibits interleukin-2 secretion but allows Bcl-X(L) induction. J Immunol1998;160:12–5.

[47] Singh RP, La Cava A, Hahn BH. pConsensus peptide induces tolerogenic CD8þ Tcells in lupus-prone (NZB x NZW)F1 mice by differentially regulating Foxp3and PD1 molecules. J Immunol 2008;180:2069–80.

[48] Parry RV, Chemnitz JM, Frauwirth KA, Lanfranco AR, Braunstein I,Kobayashi SV, et al. CTLA-4 and PD-1 receptors inhibit T-cell activation bydistinct mechanisms. Mol Cell Biol 2005;25:9543–53.